>>27021 Sauce on 2nd pic?

There was a question in the old thread. If a new one is made, it would be good to link that, since the thread doesn't come up: >>26973

>>26973 That's more of a software issue. Check official documentation (https://www.arduino.cc/reference/en/libraries/can/) then go to StackOverflow. I default to them if an issue is outside of documentation.

>>27031 It was from a post on Twitter that is lost. Nothing of value to it aside from serving as a reminder that actuators can be hidden in a base with outputs being translated via string in tubes. Picrel is a Sukabi illustration of what a potential robot could look is using such a system. Everything in a box with cables going into her to communicate with whatever sensors are in the main body and transmit motion. It's good to remember that actuators can be used in many ways.

>>27021 Just found an interesting transmission system for servo motors that has what appears to be built-in compliance: >https://www.gobilda.com/3405-series-round-belt-5mm-cord-diameter-214mm-circumference/ >https://www.gobilda.com/3400-series-hub-mount-round-belt-pulley-14mm-bore-64mm-pd/ There's stretchy 5mm round belts available on the same site I got my GT2 hub pulleys. I had been thinking of a way to put in a compliant (stretchy) mechanism as a shock absorber cable for the pony leg/foot to match the biology, and this might be just the right thing. Black loctite works for rubber belts and o-rings, but not platinum cure silicone material. Some have reported Smooth On Sil-Poxy works for all of the above but I've not tried it yet. I might end up making a finger-like round belt mechanism that is fixed at the end of the hoof. This way, the belt drive would have built-in spring and damping for when the leg hits the ground vs regular inextensible toothed timing belts.

I have some vacuous ideas, not well formed, about ferrofluids. I found a good paper on these with good references also. It's worth saving if you are interested in that sort of thing. https://pubs.rsc.org/en/content/articlehtml/2022/nr/d1nr05841j

>>27081 Your description reminds me of this: https://youtu.be/rY7IGdIvOyY

I've been trying to figure out how to convert rotational motion into linear motion for animating my papercraft waifu and learned about the Scotch yoke. It's probably one of the most basic mechanisms but I didn't see it mentioned in any of the threads so I thought I'd share it. For the joints I'm going to try using elastic bands and strings attached to little pins glued onto bamboo sticks on the back, but maybe there's a better way? I know jack shit about mechanical engineering.

>>28104 Neat! What a smooth, organic motion envelope Anon. That animation that shows that variable-position central pin, demonstrates that you can adjust such a mechanism on the fly, tuning it's leverage-advantage/power tradeoff, kinda like a transmission might. >that scoldy, Class President-chan tho Lo. A cute! :D Thanks Anon, cheers.

>>28104 I saw this before but forgot about it. It's certainly a good idea to keep in mind.

>>28104 This "Cyloidal Heringbone Rack and Pinion" could also be used for linear motion, but the motor would be in a different orientation. https://youtu.be/Cu5LvIYXbU0 While this Rigid Chain Mechanism here also does linear motion, but it's also similar to something I wanted to do. So it might come in extra handy. I was thinking of a loaded spring mechanism, where the spring would push the mechanism at the start so it would have some extra energy. This would have to go around the elbow, so it would need to be a chain. https://youtu.be/loCBrD4gRi4

>>28104 just have the output shaft of the motor reel in fishing line and you have your rotation converted to linear. can wind 2 fishing lines to same output shaft winding in different directions so that one reels in while the other extends to get opposing muscle groups assigned to each motor.

What motors could be used for the arms to achieve an A pose. Assume the arms are say 1kg and 2 feet long.

>>28143 See https://www.engineeringtoolbox.com/levers-d_1304.html Assuming you’re talking about the shoulder rotating from arms-by-side “up” to an A-pose (or even a T-pose) - You’re looking at either a class 1 or 3 lever depending on how you transfer the force. 1. Make a spreadsheet with those equations and your length/mass parameters. 2. Make another sheet that calculates torque (force required) at all theta between 0 -> 90 (or 180, or just 40 for an A-pose, as you please) 3. Decide how long it can take to go from arms-by-side to your pose’s angle. You now have X degrees over T seconds 4. Select the spreadsheet row with the highest required torque to move (worst case) in your range of angles. This is how much torque your motor (or mechanism, in general) needs to be capable of. 5. Convert degrees-over-seconds to rotations-per-minute. You now have RPM and torque requirements. Use that to calculate power (mechanical watts or HP) 6. Select a motor that can output that much mechanical power PROVIDED you have a way to reduce it to your required RPM (most motors are above 2k, you’ll probably want under 30 RPM) 7. Can’t find anything? Start thinking over alternative mechanisms— you already know your force requirements, after all. Extra: integrate force over angle (or time) to calculate total energy required. Just be sure to normalize for time with a dT(theta) or theta(T) function (depending on what you integrated over)

>>28141 Ah, you mean a Twisted String Actuator? I had to translate your text with ChatGPT. The statement is describing a mechanical arrangement using a motor and fishing lines to convert rotational motion into linear motion. Here's a breakdown of the concept: 1. Output Shaft of the Motor: - The motor has a rotating output shaft. 2. Reeling in Fishing Line: - The fishing line is wound around the output shaft of the motor. - As the motor turns the shaft, the fishing line is reeled in. 3. Conversion of Rotation to Linear Motion: - The winding of the fishing line around the output shaft converts the rotational motion of the motor into linear motion. 4. Dual Fishing Lines with Different Winding Directions: - Two fishing lines are used, and they are wound around the same output shaft but in different directions. - This means that when the motor turns in one direction, one fishing line will be reeled in (shortened), and the other will extend (lengthen). 5. Opposing Muscle Groups: - The idea is to use this mechanism to engage opposing muscle groups. - For example, if one fishing line is reeled in, it could simulate a movement that engages one set of muscles. Conversely, if the other fishing line extends, it could engage a different set of muscles. 6. Overall Function: - By having two fishing lines wound in opposite directions on the same output shaft, the system can provide both contraction and extension movements, potentially targeting different muscle groups. This concept could be utilized in various applications, such as in rehabilitation devices or exercise equipment. The use of fishing lines allows for a flexible and lightweight means of transmitting motion, and by controlling the motor's direction, you can control the type of muscle engagement.

>>28165 no, not twisted string actuator. the string reels and never twists. This video explains the concept https://youtu.be/0emPU0bIwwc?t=1135 Also, will cogley uses the concept I'm describing here: https://youtu.be/MxbX9iKGd6w

>>28130 >This "Cyloidal Heringbone Rack and Pinion" could also be used for linear motion What a cool idea Anon, thanks! Cheers. :^)

>>28181 Nice content, thanks Anon. Yes, I've planned to use silk ribbons and 2 normal 'bobbin-like' spools affixed together to provide that sort of remote, bi-directional linear actuation system before. And I'm pretty sure ol' Leo DaVinci had a similar idea too! :D

>>28196 here's a drawing by will cogly showing his version of the rotational to linear mechanism I'm talking about. In my case, I would use a brushless dc motor instead and a complex pulley system instead of a geared brushed motor servo like he's using. My method would be much more powerful but his is much simpler to implement if you have the room for it.

>>28202 Thanks! Yes Bowden-cable -like arrangements are a good way to transmit force at some distance. That's a pretty good sketch IMO. >My method would be much more powerful That sounds very interesting. I'd love to be able to actuate the robowaifu's knee & elbow joints (not to mention ankles/wrists) in this fashion, to keep the actuation motors located back inside the torso near to the pelvic volume Center of Mass. This will ofc require much more power than, say, a single finger's force to be effective. Any chance you can let us in on your designs related to this?

>>28202 heres a few images to demonstrate my design on this matter

>>28229 the pulley system here was described in this video and demonstrated: https://youtu.be/M2w3NZzPwOM?t=576

>>28229 Thanks for clarification, that's one more good idea to keep in mind. I'm highly interested in doing things with strings, so this might be useful.

>Rigid robots can be precise but struggle in environments where compliance, robustness to disturbances, or energy efficiency is crucial. This has led researchers to develop biomimetic robots incorporating soft artificial muscles. Electrohydraulic actuators are promising artificial muscles that perform comparably to mammalian muscles in speed and power density. However, their operation requires several thousand volts. The high voltage leads to bulky and inefficient driving electronics. Here, we present hydraulically amplified low-voltage electrostatic (HALVE) actuators that match mammalian skeletal muscles in average power density (50.5 watts per kilogram) and peak strain rate (971% per second) at a 4.9 times lower driving voltage (1100 volts) compared to the state of the art. HALVE actuators are safe to touch, are waterproof, and exhibit self-clearing properties. We characterize, model, and validate key performance metrics of our actuator. Last, we demonstrate the utility of HALVE actuators on a robotic gripper and a soft robotic swimmer. https://www.science.org/doi/10.1126/sciadv.adi9319 via Kiwi on Discord

>>28229 Great post. That Archimedes pulley system is something I really, really need. The video is good too. I have like four or five of these snatch blocks like he showed in the video. Harbor freight. I need to lift some really heavy stuff and have been trying to find a good way. I have some cable That Archimedes pulley is perfect.

>>28274 Wow that sounds really exciting NoidoDev, thanks. I'd love to see this tech incorporated into opensauce robowaifus! Cheers.

>>28274 hydraulically amplified low-voltage electrostatic (HALVE) actuator used to control a small tube which controls a shut off valve to a large flow tube to muscle. I've had this idea for a while but haven't talked about it because it's not completely fleshed out and people complain I write too much so... but I'll talk about it anyways. I show high pressure tube but in fact it could be the same pressure. The idea is a spring operated rigid cut off valve which pushes a small area of the mass flow tube closed. The control tube uses a large surface area flexible bladder to push the valve open. The pressure pressing open the flow tube would be large because of the large surface area of the control tube bladder. What's the point of this? If you had to control this large flow tube for each muscle it would be too big to go through the bones or thread through the limbs. But if you have one big flow tube with individual T's off of it to each muscle and then used the smaller control tube you could thread a lot of tiny control tubes down and around the body. Like nerves. The big flow tube would be like the blood stream. Since the control tube has little flow it could still act quickly and since the large flow tube has a large flow it could fill the muscles quickly. I didn't draw it but the tiny control tube could be on a pivot. If the control tube activated it would let in fluid to the muscle(flow). If it turned off a spring could pivot it to a bleed line that would bleed the muscle when it turned off.

>>28447 I forgot to add, I personally much prefer electrically operated motors but I have had a hell of a time thinking of how I could get either a micro-controller close to what it actuated or all the mass of wires needed if it were placed in the chest. It would take a lot of manual labor to place all this stuff and that's not good. It would drive up cost extensively. Though I HATE hydraulics it may very well be that if a cheap enough valve can be engineered, and I think it can, it would be better to have a central pump in the chest and a bunch of small control tubes that control the flow to the muscles as I show. The control tubes could be really small as the volume of fluid needed to open the valve would be small making everything compact and relatively fast, if much slower than electrically operated motors. If you tried to add an individual tube for each muscle that would flow enough for each muscle the whole waifu would be one big ass mass of huge tubes. So yes this is...hydraulics, the New Jersey approach. Sigh...

>>28447 >>28449 This will be really interesting if you solve this Anon!

Good overview over a lot of mechanical principles. I thought about putting that into >>24152 but it's closely related to the servos and gears.

>>28522 Super cool! Wow I wish I'd seen this as a kid, Noido Dev. :D BTW, any idea what that system he's using for all this is? Looks kinda like Legos, but probably isn't. Thanks!

>>28546 >Looks kinda like Legos, but probably isn't No, those are all definitely Lego. The motors too, they come in a small number of sets. I have the exact same motor as the first one shown in an AT/AT.

>>28522 That's very cool.

>>28546 Pretty sure this is Lego Technic. I had some of it as as child. I wasn't sure at first if it could be Fischer Technik, but no.

>>28567 Thanks!

Update on cheap motors from china: I now have 20 wife servomotors. I'll need 16 for 4 legs and a few to move the back and head. They're kind of vibraty, kind of grindy, not perfectly smooth, buuuut they only cost 100 each and I needed a lot. Also, they're weak, but I don't care for a proof of concept since it just has to move. I've got an idea for a relatively cheap diy actuator bldc+planetary that would be much more powerful if I have to upgrade and no other stock motor can do it. The motors will use CAN bus so I can get the skills for a robotics job, and I'll learn ROS, gazebo, mujoco, and isaac sim too. mujoco and isaac sim will be most useful for rapid simulation since it has elastic tendons and ROS/gazebo are useful for a robot job skill.

>>28738 Which project is this? Please post ongoing projects and updates in the prototype thread: >>21647 - This thread here should be used for general discussions on specific actuators.

for my tendon based actuator, i have PE fishing line and it is .63mm OD and I have a teflon tube that is 1mm ID that I want to thread with this fishing line. When I cut the line, the end is a bulge. How do I make the end pointed so I can thread the tube eye with it?

>>28859 Heat up the fishing line over a candle flame and stretch out the melted plastic until it breaks. With practice you can get the total stretch down to a few inches instead of a couple of feet, but either will work. This way the plastic will only taper without a bulge. For source idea, that's how we modelers make whip antennas for model military vehicles.

>>28738 what is mujoco and isaac sim?

>>28862 We have a whole thread on simulators >>155, at least mujoco is mentioned there.

>>27021 What do you guys think of these? Especially the second one where they made a model that mimics human walking efficiency? Both are from this week/last week. https://www.news-medical.net/news/20240127/Biohybrid-robot-makes-sharp-rotations-with-lab-grown-muscles.aspx https://journals.plos.org/ploscompbiol/article?id=10.1371/journal.pcbi.1011771

>>28866 https://neurosciencenews.com/robotic-human-walking-25509/

>>28860 thanks I'll try it should work I think

>>28867 Thanks, just some addition: >Summary: Researchers achieved a significant advancement in robotics by replicating human-like variable speed walking using a musculoskeletal model. This model, steered by a reflex control method akin to the human nervous system, enhances our understanding of human locomotion and sets new standards for robotic technology. >The study utilized an innovative algorithm to optimize energy efficiency across various walking speeds. This breakthrough paves the way for future innovations in bipedal robots, prosthetics, and powered exoskeletons. https://github.com/Shunsuke-KK/reflex_plos_revision

Turbines driven by air might be interesting in some edge cases or for long term for optimization. I had the idea watching a video on Tesla Turbines. These can work with air pressure, but have up and downsides: https://www.youtube.com/shorts/9XTEGrze920 I mainly think of turbines an additional booster for an electric motor, e.g. to make it start faster and reach some high torque faster under some load. Or the turbine (Tesla or other) kicks in when the motor is already at a high speed but needs some extra kick for a moment. This is of course more relevant for builds which would use some air muscles or stabilizers already. I don't want to start this whole general discussion again, I think pressurized air actuators can be useful if combined with other actuators. Problems with Tesla Turbines in particular, based on some comment on that video: - centrifugal forces warp the discs that results in touching from the discs under extreme high speed, this can cause failure but also abrasion over a longer period of time. - seems to need high speed and only works with high speed, would need a gearbox making it less efficient. - efficiency goes down to ~65% under load, similar to normal turbines - sensitive stall speed - higher rotational masses compared to other turbines

>>28910 hate turbines, this is 12th century technology i always wondered why you cant just send a balloon up into the ionosphere with a magnet, like the earth is already just one big solar magnet how hard can it be to put it to good use ffs

>>28911 Even if this was the case, how would we use this in a Robowaifu?

>>28910 We might be able to use a micro-turbine for the micro-rotary-engine, randoseru power pack? (>>28697, ...) After all Mazda made an awesome twin-turbo RX7 that was a rather-lightweight, relatively-inexpensive beast of a machine.

>>28921 With a balloon attached going up into the ionosphere?!

I feel like 20k is to much for this robot https://hello-robot.com/ do you guys have any suggestion to duplicate this telescopic approach much cheaper?

>>28943 Well, look for telescope actuators: https://youtu.be/X28CxOnxZh0[Embed] https://youtu.be/WavBXfh1giE[Embed] https://www.youtube.com/shorts/jWxARpHUVrg I recently had another one in my recommendations, but don't find it.

>>28943 >6x speed My fuggin sides are in orbit.

>>28957 ikr this is why I am thinking if there is a way to replicate this (make it faster) using pulley systems.

>>28958 why re-invent the wheel? just use rack & pinion.

>>28959 the problem is , if it lost energy in the height axis will fall, so I think a pulley system can keep it

>>28976 Screw-drive linear actuators for the main stalk, then. I love these things and use them in my mech projects. They only consume power when they move: to get them to maintain position you stop giving them power.

>>29007 that takes up a massive form factor and space is a great concern ruling such an actuator out for a humanoid with human form factor if you are going for humanlike full body articulation which requires so many more actuators that all are competing for space

>>29014 Oh, so you just want the approach, not building the same robot. I misunderstood this before. Do you mean the arm or the mechanism for up- and down? And how to put this into a human-like robot anyways?

>>29014 Not really...

you have your motor which acts as a muscle and gives the strength. look at the volume that takes up. then look at the volume taken by the gearing. if the gearing is taking up even close to as much space as the motor itself, then you are being very wasteful on space imo. that's why small pulleys to downgear like I posted earlier in this thread are WAY more efficient on space and get you your desired linear motion making them 100x superior to the type of linear actuators you all are posting now

also bonus points if the gearing can be relocated to another area than the motor and is not physically bound to a certain orientation with respect to the motor and required to be located at the same general location as the motor. this lack of flexibility in arranging motor and gearing gives you less options for negotiating fitting everything into your desired build volume of the robot. whereas pulley based downgearing gives you all the flexibility in the world and WAY smaller downgearing total volume taken up comparatively

>>29017 yes I'm building a different robot than any designs I've seen here but I am here to learn and share techniques. you say "Do you mean the arm or the mechanism for up- and down?" --- I am saying the bulky linear screw based actuator has too much hardware volume taken up by things that are not the motor itself and this volume is not something that should be given away so loosely as space is at a premium in value for a robot and must be guarded visciously if you want to fit enough stuff into the robot to do the things you want it to do (depending on how capable you want it to be of course). You say "And how to put this into a human-like robot anyways?" --- my concern is that fitting all the components one would like to fit to get the functionality one wants in the end is going to be a huge challenge we all face (unless we are creating like dumbed down super simple completely incapable robot toys that are laughably simple and incompetent in their capabilities.

>>28866 >>28867 >>28873 Good find! Adjusting power to actuators based on what is actually needed is essential for waifu efficiency! >>28910 Excellent out of the box thinking! Could be helpful for budget systems that need sudden bursts of power. Though, a proper coupling system to deliver power when needed, without imposing significant friction. >>28911 Explain how a magnet on a balloon could affect the actuation of a waifu. If you fail to do so, your post will be deleted. >>28921 >Power Randoseru Just like in Prima Doll! Though, routing kinetic energy from turbines into your waifu to animate her, is beyond my imagination at this point. >>28943 See >>29007 and >>29020 For the objectively correct answer on how to actuate the telescoping arm. >>28976 >Pulley maintaining static position sans power Are you referencing differential pulley systems (also known as differential/Chinese windlass)? https://en.wikipedia.org/wiki/Differential_pulley https://makezine.com/projects/the-chinese-windlass// Otherwise, there is too much friction in your pulley system. Without input power, the system should allow for the load to fall until it is stopped. >>29014 >>29030 >>29032 >>29034 I understand you are learning and likely have familiarity with pulleys. This being said, every method of reduction and translation of energy has its place and use. Pulleys are not universally superior in all applications. For a linear actuator based telescoping robotic arm, the precision/load capacity/simplicity of implementation of a stepper driven lead screw relative to its cost is essentially unbeatable. A pulley system can be used to achieve the same results, it will be more expensive, more complex, and require the same volume per work area. If you wish to argue your point, do so with engineering. Prove me wrong by designing and implementing a pulley based robot arm. Furthermore, try to think and breathe while your post. You are speaking with others who going through similar challenges and want to see you succeed. Be kind, always.

>>29061 >Just like in Prima Doll! Though, routing kinetic energy from turbines into your waifu to animate her, is beyond my imagination at this point. Haha, me too! :D Actually, I simply meant using the unified rotary/turbo engine coupled together with a smol electrical power generator as a means to keep her electrical systems functioning outdoors, beyond just the limits of her own batteries. > pic-related

>>28923 Kek. AS LONG AS YOU GIVE ME FREE HYDROGEN GENERATORS, ANON!111 :^) > (>>28681, >>28692)

>>29061 you say " A pulley system can be used to achieve the same results, it will be more expensive, more complex, and require the same volume per work area" --- a pulley system is not necessarily more expensive and absolutely would not require the same volume by a long shot. It would be SIGNIFICANTLY less volume taken and more importantly, the volume taken would be like a pen so long and skinny which is ideally suited for fitting into the limbs and torso of a humanoid form factor body. the screw is thick and the pulleys can be very flat so stackable. the full pulley system for a 64:1 downgear can be 4mm thick and 11cm long and less than 1cm wide and it is very easy to fit many of these in a stackable manner into any humanoid limb of one's design with the volume constraints of a human body. Attached is design imagery of a 3d model of the archimedes pulley system, then teal green boxes showing the total build volume of such a pulley system with measurements demonstrated with a metric ruler to scale all placed within a humanoid robot forearm to scale over the top of the bldc motors between the motors and where the skin will be. you can see how many you can stack easily. The pulleys themselves can be made using 1x3x1mm bearings or 2x5x2.5mm bearings for larger force areas if needed in some cases. each bearing costs $25/200 bearings on aliexpress for the smaller size I just listed. so around $0.13 so $0.91 for a full archimedes pulley 64:1 downgear system. The weight will be a MASSIVE decrease compared to a ball screw style. The outer containment flanges can be sewn on through the center of the bearing by way of needle and thread and can be made of very thin plastic. effectively free. the string/cable PE braided fishing line 70lb test or 130lb test. You say "Pulleys are not universally superior in all applications" --- I argue they are for all applications relating to humanoid robotics of substantial complexity and humanlike movement and capabilities and humanlike form factor due to their light weight, extremely low cost, and their not being needed to be physically located in the same vicinity of the motor which creates tremendous freedom for squeezing things into the human form factor with less limitations on space concerns. You say "For a linear actuator based telescoping robotic arm, the precision/load capacity/simplicity of implementation of a stepper driven lead screw relative to its cost is essentially unbeatable.' --- I disagree completely. Maybe for a 3d printer but not for a humanoid. You say "try to think and breathe while you post. You are speaking with others who going through similar challenges and want to see you succeed. Be kind, always." --- this infers I'm not being kind which is false. Prove I'm not. This also infers I neither think nor breath while I post. Both of these inferences are false and unkind and insulting making you a hypocrite.

also stepper motors have inferior power density as compared to bldc motors. a so you add weight and lose power by going that route instead of bldc. also steppers can tend to be loud-ish although I hear this can be solved in software. Also, the pulleys are essentially silent as are the bldc motors. a 2430 bldc motor pulls 200w and is $11 and the archimedes pulley system is $0.91 in parts for the bearings. it can downgear it 64:1 ratio. gears also are loud whereas this is silent way to downgear.

here's a drawing of a even more simplified version of the basic gearing pulley for such a system as I am talking about based on a small bearing.

for clarity here's a little zoomed out image showing where all these archimedes pulley systems are in the arm design

i also developed an alternative pulley design based on a M1.7x4 PCB rivet as the center shaft and a copper butt connector (basically a hollow metal cylinder) of 1.5mm ID. they are a perfect fit. just have to cut the butt connector to length of around 3mm and then insert the hollow M1.7mmx4mm PCB rivet into it then flare the rivet and you now have your pulley. very easy to make. then run your string through. will need to be greased with white lithium grease. acts similar in principle to a door hinge. This setup should cost closer to $0.05 per pulley in the materials needed to make it. So it is cheaper than the bearing method above but might wear faster not sure. Chat gpt said it would wear faster.

I don't know why everyone acts like the physical robot is a solved problem. Inmoovs arm movement is limited, its face is expressionless, it has no legs, I haven't looked into it much but it probably doesn't have a mechanism for keeping its upper half balanced either. Poppy is better however those actuators are prohibitively expensive and even then it needs to be handholded into walking. Its also a small robot. I don't know if pulleys are a good idea but the way they usually work is by leverage rope length, a robot doesn't have space to work with, quiet the opposite.

>>29065 do you really need all those wires though

>>29071 I'll see if I can record a vid of SPUD's torso copying the angle of her GY-521 gyroscope. Should be a relatively safe demonstration since she is currently sitting on a support stand rather than her own legs right now.

downgearing by way of a worm screw: the finer the thread pitch the more it is downgearing right? but that is capped I'd assume. what is the higher end to downgearing ratio that can be achieved with this method of downgearing like 6:1? we'd want more like 64:1

>>29079 I'm assuming you are called the black lines in the main model wires when in fact they are just indicators connecting parts to labels and other parts like when you draw a picture and draw little lines pointing to objects in the picture and part of the line comes near the object it is pointing to and the other end of the line goes up to the text that labels what the part is. I noticed you did a picture of gears in your post. well I explained why those preferrably are to be avoided: noisy, physically located at top of motor so less flexibility in design as far as moving them around to other areas where you have more room, etc.

>>29065 Tad odd, I'm usually defending pulleys. In this case, I would be happy to make a linear robot arm with a lead screw if you'll make an arm with pulleys. Howabout a friendly competition to see which is best in practice? >>29088 There are many factors which affect functionality of lead screws. These resources provide good information for anyone learning about screw based actuators. https://en.wikipedia.org/wiki/Leadscrew https://www.youtube.com/watch?v=0jrcs8lS7-M https://www.youtube.com/watch?v=BstZUC4tcOA

based on those videos, it looks like we might be looking at lead screws giving us a ballpark of 5:1 or 6:1 downgearing on the high end. What we want for a typical 2430 bldc motor is 64:1 at least if not 128:1 which can be achieved easily in a very small volume by way of pulleys. It does not appear lead screws offer significant down gearing at all like this

chat gpt and I discussed this just now and I'm gettign this: 64:1 is the very least downgearing we'd need. can a single lead screw give us that extreme downgearing or not? ChatGPT Achieving a 64:1 downgear ratio with a single lead screw might be challenging. Lead screws are generally better suited for lower gear ratios, typically up to 6:1 or 8:1. As you increase the gear ratio, the mechanical efficiency and precision can be compromised. For very high downgear ratios like 64:1, a pulley system, as you initially mentioned, may be more suitable. Combining multiple pulleys in a system, as per Archimedes' principles, allows for efficient torque multiplication and can achieve the higher ratios you're aiming for. In summary, while a single lead screw might struggle to provide a 64:1 downgear ratio efficiently, a pulley system could be a more viable solution for your specific needs.

interestingly, the lead screw idea involves clearly metal rubbing on metal and no bearing surfaces it seams to me. just like a nut and bolt it looks like. perhaps some are more sophisticated but I digress. I was thinking we could make pulleys also by having a shaft inside another shaft and the outer shaft rotates over the inner shaft - like the rivet drawing I made earlier, but shaped more like a bearing pulley in overal shape like the drawing I showed of a bearing based pully. This would remove the balls and so have more of a metal on metal approach but could be more robust for handling higher forces put onto it without crushing the bearing balls or warping the bearings. I noticed the load capacity of tiny bearings is startlingly and alarmingly low. like the 1x3x1mm ones are like 10lb only and the 2x5x2.5mm ones are like 35lb. the complexity of their design and larger number of moving parts they have make them vulnerable at these tiny sizes. So going with a metal on metal 2 part non-ball door hinge-like design (basically like fusing all the balls together with the core as one mass of metal) can make it able to handle way more forces evne if it loses some of the extreme frictionlessness it has. thoughts?

also for reference as a side note: your typical hobby servo motor which employs a dc brushed motor has gears integrated into the box it comes in that downgear that motor 180:1. so when I say we need at least 64:1, now you can see why. Higher speed, small, lower torque motors like the 2430 bldc motor need a ton of downgearing just like these small brushless dc motors in order for them to convert all of their inherent speed down to big torque and more moderate speeds. that is the principle at hand. lead screws primarily convert rotation to linear motion but are not really intended to be big downgearing approaches as they only offer very minor downgearing

Can't nitinol be used?

>>29071 >I don't know why everyone acts like the physical robot is a solved problem. Literally nobody does so. >pulleys are a good idea but the way they usually work is by leverage rope length I'm coming more and more to the conclusion that some soft parts should internally be made out of rope. >>29102 >Can't nitinol be used? >>1747 and >>13723

>>29092 What is this whole discussion even about?! The initial posting >>28943 showed a non-human robot. That's how I answered it, though it's kind of OT, until we get a thread for helper bots for chores. I focused on the telescope arm part. Obviously a inline screw drive or any screw-drive linear actuator might make sense for the vertical movement, though it would be worth contemplating why the hello-robot team didn't use them. More importantly, why on earth would anyone use that for the arm?! Also, how and why would anyone use a telescope element in a humanoid waifu(!) robot at all?! Could anyone picture that and maybe draw it for those who can't, so that we can all see how weird this would be looking? I assume >>28911 and >>28943 are troll attempts, and it would be worth contemplating to delete or move these together with the responses.

>>28202 That looks very useful. I had something like this in mind for some use cases, but it's good to have a few references. Especially combining it with a pulley: >>28229 >>28447 >>28449 >HALVE Thanks, very useful explanation. I guess for a lot of small valves piezo electric ones might be the right ones. I was thinking a while ago about using water bags as some kind of holding torque and a valve in between. So for example the chest could tilt sideways, but if water is in the bag inside and the valve closed it won't. If the valve is open it might tilt, in case a muscle is pulling it then it would certainly.

>Cycloidal Hub Gearbox Might require a resin printer and work better with a laser cutter https://youtu.be/IKkw4d7jyu0 https://www.thingiverse.com/thing:5437152 https://www.printables.com/model/333438-3d-printed-planetary- https://drive.google.com/drive/folders/1apuReZyJan5Dxpbf_ou8xmFl2_huDlke >What makes cycloidal gearboxes so amazing? https://youtu.be/8GDMShY8vAU >I discovered something really interesting about cycloidal gear drives that I did not expect! In this video, I use a 3D printed gearbox to demonstrate just a few of the reasons why cycloidal gearboxes are so useful. Then a video on "CNC Machined vs 3D Printed Cycloidal Drive": https://youtu.be/mPwbDrXq50Q - with the result that the CNC one is much heavier but the 3D printed one broke under heavy load and didn't have as much torque. Down to 50% under strong load. This might mean a smaller CNC machined one would be better or that we need to use galvanization for the 3D printed parts. It might also be so that the motor can't be too strong, so with a relatively weak motor it doesn't matter.

>In this electronics basics episode we will have a closer look at motor encoders. Those can not only turn any motor into an awesome and satisfying input device for your projects; but can also turn any kind of motor into a kind of stepper motor. There are many different types of such encoders. So let me show you how they work, what they do and how you can use them. https://youtu.be/d3IH8zwUVYU https://www.solomotorcontrollers.com https://www.youtube.com/@SOLOMotorControllers

Thanks for your nice inputs here, Noido Dev. Cheers.

>>25792 Hey Kiwi, > I have a new hoverboard which uses a cheap Chinese chip that no one has figured out how to hack yet. It looks like that chip is here, under the "Three Gate Driver Chips" picrel. Click on the MM32SPIN05/06 = 2 boards link. https://github.com/RoboDurden/Hoverboard-Firmware-Hack-Gen2.x/wiki I got there from here: https://github.com/RoboDurden/Hoverboard-Firmware-Hack-Gen2.x RoboDurden seems to be one of the masters at reflashing hoverboards. Hope this helps.

I found this guys video channel on YouTube where he covers an extraordinary amount of actuators and motors and damn near everything. It covers not only all types of electric motors, also, pneumatic, hydraulic and even covers all the math from the very basics to do detailed analysis of all of these. And all the parts needed plus programmable logic control and circuits to control all these things. A PILE! As I was scrolling through his set of instruction videos it blew me away the depth of material this guy covers. It would be well worth bookmarking this and when you have a question scroll through his videos as an answer, while maybe not easy, is likely there. https://www.youtube.com/@bigbadtech/videos?view=0&sort=dd&shelf_id=0

>>30048 Yeah, that looks like a goldmine, Grommet. Thanks very much! :^) yt-dlp -i https://www.youtube.com/@bigbadtech/videos Do it today, /robowaifu/ anons! :^)

I seem to have found a kind of actuator that hasn't been mentioned before but im not sure? Apparently there is something I think is called an electromagnetic feedback linear actuator. So far this is only one I could find. It seems similar to a solenoid but it apparently isn't. I assume the price when hidden like this is they price you base on how many you want to buy. I am not looking to buy anything so not looking for a quote myself so I can't be bothered to ask them. https://irisdynamics.com/products/orca-series https://www.youtube.com/watch?v=xZ7T1H1uPSc These seem ideal in sense of being fairly quiet since I find nothing sexy about a robowaifu that sounds like a power drill from screw based linear actuators.

After attempting to work with servomotors to animate a simple leg, I gave up on this idea due to the weakness of the system. With the help of ChatGPT, I realized that I needed to switch to a pneumatic system instead." Servomotors: Frequently employed in hobbyist applications such as small-scale robots and precise motion control systems like those found in 3D printers. Pneumatic Motors: for average applications like automotive assembly lines. Hydraulic Motors: powered by hydraulic fluid, utilized for heavy-duty machinery like bulldozers, providing high torque. The next challenge is figuring out how to set up a pneumatic solution, as there isn't a plug-and-play option available on marketplace. Here's a list of products that can do the job: https://www.aliexpress.com/item/1005005870528273.html - air cylinder https://www.aliexpress.com/item/4000329351824.html - valve to connect between the air cylinder and the air pump https://www.aliexpress.com/item/1005006519202716.html - basic electrical air pump for easier modification ability. https://aliexpress.com/item/1005005209160156.html - For the question of controlling airflow, solenoid valve can do the job to on/off the air flow through a dev board https://aliexpress.com/item/1005005209160156.html https://bc-robotics.com/tutorials/controlling-a-solenoid-valve-with-arduino/ Now it's time to put this brainstorm into action!

>>30222 >Irisdynamic's Orca Actuator I'm pretty sure this was mentioned before, since I knew about it. But thanks anyways! The issue was that even the smallest ones they had a few years ago were quite big and heavy. Close maybe, but still rather a bit big. Though, these might be very useful to build VR rigs moving a vehicle simulator smoothly around. So I guess and hope this company will stick around. I would't rule them out completely, it's good to keep an eye on it. >>30245 >pneumatic system Thanks, keep us updated on it please, whatever comes out of it.

>>30245 What servo were you using? What kind of leg did you set up? Length and mass of the upper and lower leg. Post pictures.

>>30251 chat logs are there: https://alogs.space/robowaifu/res/418.html#13541 I spent 100 euros on this, but upon reflection, I believe I could achieve a more cost-effective solution by using a pneumatic equipment.

>>30256 So, the leg is equivalent to a 100cm doll leg, safe to assume a steel skeleton? Moving with thick silicone providing resistance. The link to your servos doesn't work. I'd use 40kg.cm servos as a minimum if you replaced the steel skeleton with a low density plastic. 80 to 100+ if you kept the steel skeleton. Servos usually are represented with their stall torque parameters. Do the math for the leverage force needed and multiply it by 4 to find your stall torque. This is because servos are designed to reach their stall torque for brief moments. As an example; to set up a simple robot maid arm, with upper arm length of 23cm lower arm 22cm and her joints and end effector add up to 5cm. To lift a kg load, assuming the arm has a total mass of a kg, you'd multiply distance 50cm by half mass of the arm .5kg plus load 1kg to get 75kg.cm. Based on that, you'd want to use a 300kg.cm servo in her shoulder. This is just a simplification to demonstrate how high torques can be at human scales.

>>30256 Ah, you are that guy. Nice to see you back. >>30258 >80 to 100+ if you kept the steel skeleton Steel doesn't need the same diameter, a combination of plastic an steal might be a bit heavier than just plastics but much stronger. >calculation Interesting, thanks.

>>30250 >I'm pretty sure this was mentioned before, since I knew about it. Possibly. I mainly was searching for the word "magnet" and nothing came up. Didn't think of searching by manufacturer branding. >The issue was that even the smallest ones they had a few years ago were quite big and heavy. Hm looking at the Orca 6 LITE the mechanical body is 2.4 kg (5.29 lbs) and the shaft is 1.38kg (3.04lbs). Im not sure why the cheapest one is no lighter than most other models.

>>30245 >>30250 Check out Clippard Minimatic. They have pretty much been the leaders in small-scale pneumatic systems for labs and process control for production lines. Lots of their stuff is on ebay. https://www.clippard.com/

>>30258 Indeed, the skeleton was made up with steel tubes, but going forward, I'll switch to 2020 aluminum profiles for improved design customization. I was used 35kg.cm or 50kg.cm Indeed, the challenge with servos lies in the scale of torques required for human-scale applications, where the servo motor torque needs to match the demands of the human skeleton. However, the downside of this is that the motor torque capability of the servomotor is determined by its size. I am hopeful that I will not encounter this issue when dealing with a pneumatic system instead of a servo system. >>30260 Thank you, It's great to be back in the world of robotics, I thought Elon Musk's Optimus robot would dominate the field, but still waiting :| >>30266 Nice found, Bing AI shared me several cool website: https://makerindustry.com/pneumatic-robot-arms/ https://www.rowse-pneumatics.co.uk/blog/post/pros-and-cons-of-pneumatic-robots https://www.quincycompressor.com/how-a-pneumatic-robot-arm-works/

Interesting high ration gear reduction. From the most excellent Robert Murray-Smith. He is another one of those with videos on a huge amount of stuff. Mostly DIY. In this he shows the gear and where it came from plus has links to 3D print files to make your own, https://www.youtube.com/watch?v=HsFHEf-oRk4 He's another you might be interested in bookmarking. Look at his hundreds of videos. He started with a lot of battery, super capacitor, inks and then decided to spread out into all sorts of DIY gears, wind mills, materials, lots and lots on all sorts of electric motors and a ton of stuff. Of particular interest to me is his graphene production. He shows many ways to make it. One of the most recent you use a blender, graphite and whey, yes, whey, to make a liquid solution of graphene. He made some of this and put it in an old fashioned Casein glue/plastic(I believe). Casein is good stuff on it's own. They used to glue airplanes together with it before epoxies. He added 1/2% of graphene and made a bulletproof puck a centimeter or so thick(I think the plastic is casein based but not sure). They shot it with a 9mm and no change. They made a hook with it and suspended some huge guys off of it and the steel rings broke before the hook. This is one way to make extraordinarily strong plastics. It would make a fine skeleton. I wonder if it would do the same in a polyester matrix resin like they use for fiberglass but substitute graphene or glass/graphene combination? https://www.youtube.com/watch?v=YXZ1gfJMtW4 https://www.youtube.com/watch?v=PcLd4h095DQ 1943 Unbreakable Plastic. https://www.youtube.com/watch?v=qoGPRXjCwwA

BTW this new research on making graphene cheaply means in not too long you will begin to see this stuff in all sorts of things. Consumer products like sporting goods first and eventually in airplanes and stuff like that. It will bring super high efficiency because it will make things so lightweight. This has all sorts of second order effects. If you make it lighter, you need smaller motors to move stuff and smaller motors also makes it lighter so now they are even smaller, this stuff compounds making ever higher performance with ever less materials.

>>30293 >>30294 Thanks for the channel recommendation. This guy really covers a lot of different things and his video output rate is really high. I've been excited about graphene for a long time. Glad to see that more R&D effort is being spent on it.

>>30309 I wish to bring to your attention two of his videos that seem to me to be the best. He has thousands and I've seen most. Here's simple ways to make graphene. One is a method to make graphene foam. title to search for "Making Graphene Foam From Table Sugar" The next is how to add whey. This keeps the graphene from recombining. I'm not sure if you need this. Maybe you could grind up from the first method and that would be good enough. There does seem to be a problem with dispersion into powders. With the liquid method you have a liquid. He was using it to reinforce concrete and getting really strong concrete. "If" you wanted to mix with a powder the whey treated graphene you would need to dry it. He says centrifuge. None of this is beyond kitchen chemistry. The strength of this stuff is off the charts. it makes steel look weak and likely anything you can get to mix it in will become super strong. "1933 Blood Graphene - A Way To Make Graphene From Blood, Eggs Or Milk"

Has there been any robots that used a shape memory polymer as an actuator? I know nitinol has been used for tiny bots. SMPs seems to rebound a fair amount faster than nitinol from what i can see in videos. You can actually even buy SMP filaments for 3D printing. Though problem I see is even if you use a mechanism to heat it locally and controlled the state such filament regains it's original shape is increased heat which also is what is required to reshape it so it's difficult to tell how much push or pull force it can provide.

Switched reluctance motors are great for gaining a better understanding of the physics which governs motor movement. https://www.youtube.com/watch?v=1IUVjp03On8 https://www.youtube.com/watch?v=TAhF45AtsgA

>>30353 Yes, they're about 1~5% efficient, are slow, need high precision heat control, do not work in cold or hot environments without good engineering, etc... Though cheap and easy as crisps, they just have too many problems. I like them and wish they can be cheap and electroactive while requiring sub kV range voltages If you're interested, read papers published on them and try to make your own. There's plenty of room for improvements and I'd be grateful to have a viable alternative to heavy gear motors. PLA and plastic polymers based on it could be a great and cheap start to your research. https://www.youtube.com/watch?v=SiOn-LqW4KI

>>30353 I don't know why I overlooked that, but some of us want to use water cooling inside, then store it and it might be used for muscles (e.g. nylon) which are just complementary. I also think small muscles might make sense in some places, e.g. the face. > SMP filaments Thanks, I'll look into it. >>30549 Interesting, especially with that video, thanks.

>>30549 Isn't that 1-5% just for one specific kind the kind of coiled nylon muscles or I may be getting mixed up. Though I came across info on why heat is a poor activator, it is faster to heat an object than cool it so it isnt exactly even how it operates. Doing more digging I just now came across solid state actuators which includes things such as ionic polymer metal composites which seemingly can operate at low voltage. Ive yet to figure out performance and possible use of them. Im not sure why they think only make a flat object instead of a design that can have pulling force like a cross pattern or something XXXXXXXXX but maybe it would be too weak to be of use? https://www.sciencedirect.com/science/article/abs/pii/S1385894723027079 https://www.youtube.com/watch?v=E1jkAQ-pBpU >>30557 Id question using water. My thought was using electric heating wires to control the SMPs but Im second guessing using SMPs that require temperature changes.

>>30578 >Id question using water. My thought was using electric heating wires to control the SMPs We had this several times during the years, not sure if something changed. If you have wires bending then they might break. I recall some video testing these actuators and the results simply weren't good. Good look anyways, but always keep in mind it has to work 10k*X times or 100k*X times.

>>30353 You are probably the same who posted it, I'm just linking it here: Silicone shape memory polymer >>30584 - Idk, maybe this will be interesting for smaller muscles, like in the face.

> Homebuilt Electric Motors Just leaving this here for now. https://www.bavaria-direct.co.za/

>Servo Horn: Pick the right one. via >>30927

>>27021 >Boston Dynamics Atlas runs on hydraulics but it's a power guzzler and heavy Their hydraulic Atlas was retired and they have an electric one now. https://www.youtube.com/watch?v=29ECwExc-_M

Well, the idea of using a rotary cylinder fueled by air or oil seems outdated now. Thank you Boston Dynamics, for saving me time and money, I guess :p Looking at how their servos look, the idea would be to DIY our own servomotor with a large AC or DC motor :/

It's the servo motors that make up the structure of this robot, the mistake was to set a structural basis and attach servo motors to it.

>>30969 Heh, they are going to run into the whole power/mass ratio issue now, same as all the other humanoid robotics companies. Burning hydrocarbons is a very energy-dense way to motivate cars (or robots). Now they are taking the 'EV' pill, they will be on level ground with say, Optimus. >tl;dr BD has a long row to hoe now. But of course the perceived """superiority""" of their competition will probably light a bigger fire under the other vendors. This, ofc, should accelerate the eventual introduction of home-based service robots -> companion robots. All of them GH Surveillance-State Specials, sadly. >>30970 Great insight, Anon. I've been trying to think several design approaches for a while that take advantage of this very concept. Volume is certainly one of the several severe constraints we're all up against here.

>>30969 >Well, the idea of using a rotary cylinder fueled by air or oil seems outdated now. I wouldn't be so quick to rule out hydraulics. I HATE hydraulics. I've complained about them a lot but...with the right valves they are super cheap and they solve a lot of problems. I much prefer straight electric but to have lifelike waifus hydraulics would be far easier to do and MUCH easier to mass produce. In the video here, >>30956 He talks about spending 35 hours straight soldering alone. One of the main problems with hydraulics in terms of efficiency , in machinery anyway, is that the pumps run all the time and bypass flow the extra pressure not needed. The problem is if a pump is set for max pressure and flow then when not using max, it waste an a enormous amount of power. But, if we use electricity to control a pump and have some accumulator to store pressure then the efficiency becomes much better.(So that no one whines, I note, yes "some" hydraulic systems have some control over the pump but most still waste some amount of energy flowing back pressure.)

Here's an idea. First visualize one of those party toys that is a rolled up flat sheet-tube. You blow on it and it unrolls and makes a cylinder. So instead of blowing the inside and unrolling it, we have it full of fluid and compress it to push out the fluid. Roll it up. Long skinny bladder with an iron covering that can still roll up. We doing the opposite of the party blower thing. To operate it put an electric field on the end and the bladder will try to pull in the iron just like a solenoid. But since it's a long tube the only way to do so is to roll up, forcing out the fluid. This is srt of whacked "but" since all of your muscles would be made of badders and bladders could be cast "theoretically", for the whole body at one time with liquid plastics/rubber. Once you had the pattern you could churn them out.

Since actuators are all tied up with the sensors to position them, a reference to a comment I made in the skin section that may be relevant here also. >>30980

>>30974 >iron covering I wish to make this more clear because people may get the idea it's some big iron plate. No, it could be iron oxide particles embedded in silicon or rubber. If you notice while this is a different form I have not given up on the idea of inductive reactance motors. Mostly because they are cheap, fast, lightweight and powerful for their size. Yes you can do a little better with super expensive rare earth magnets but...they're super expensive and do you want a magnetic waifu that has metal sticking to it, I think not.

>>30982 force is proportional to mass, some dust isnt going to create a lot of force

https://www.zygotebody.com/ There are so many muscles/actuators to set up for just one arm, and the deltoids are composed of three smaller muscles :o Indeed, it would be much easier to use only three large servomotors that control Yaw, Pitch and Roll rotations. https://www.aliexpress.com/item/1005003988911701.html However, the issue is that the rpm is really slow for acceptable torque: 24V, 11rpm, 85kg.cm Perhaps a 1000W Brushless DC Motor coupled with a gear set would do the job. But since I cannot find the gear set, I might have to build it myself, possibly with the help of the following website for delving into gear configurations and building custom solutions: https://drivetrainhub.com/gears/

>>30988 Nice observations, Anon. Please continue! Cheers. :^)

>>30988 or you can downgear by way of pulleys instead of gears which will be a more quiet outcome and imo easier to make than gears. Just need small bearings and some scrap plastic and some string and a sewing needle and scissors as only tools.

>>30984 >force is proportional to mass, some dust isnt going to create a lot of force No F=MA You left out the A part and the A in this case is due to the electromotive force on the, as you call it, dust. I assure you if you get between the "dust" and a 20 Tesla magnet you will feel some serious force.

>>30988 >three large servomotors that control Yaw, Pitch and Roll Yes much easier but it would look too much like a robot to me. Obviously this is an issue with me, others might not mind this. Of course trying to replicate the muscles in a human body complicates things a good deal but...that's what I would find more attractive.

>>30990 Thank you :] I've tallied up four large servomotors for one leg, so it might cost around 100 euros of servos, assuming each of them has a torque of 1 Nm: www.aliexpress.com/item/1005005428153048.html Cheers >>30995 Oh like a bicycle's gearing system, sounds like an interesting challenge to take on >>30997 Interesting thought, the tactile sensation of touching filament, indeed differs from the hardened material of servomotors

>>30999 I think it would be more like a stringbike rather than a normal chain and sprocket system wouldn't it? >>30995

>>31002 well not like a bike at all imo. A bike is downgearing by way of sprocket and chain. Even string and pulley in continuous rotation is not ideal imo. You want pulley blocks like these photos but way way smaller custom made with bearings and plastic discs and fishing line and super glue

>>30996 This. The rare-earths are used to greatly 'amp up' the electromagnetic properties of the basic electro magnets themselves. Kind of like how you need trace (tiny amounts of) otherwise-often-harmful-micronutrients (like cyanide) for your cells to properly manufacture the incredibly-yuge swath of proteins vital for your metabolic processes, these trace amounts of rare earths significantly enhance magnets -- especially when they are at cold temps! (Say, around liquid nitrogen realms.) >>30999 Good luck aydoll! If I haven't yet given you the 'official' greeting yet, welcome! :^) >>31004 As a country boy, I'm quite familiar with block & tackle. It's absolutely amazing the force multiplier that can be achieved with this simple machine.

>>30999 For some reason the crotch reminds me of a robosapien... I think it is the slight angle where the codpiece meets the first hip joint.

Some Pulley adjacent actuators https://www.youtube.com/watch?v=6QEgNbaDCjw https://www.youtube.com/watch?v=utDagouxM5U https://www.youtube.com/watch?v=Bzsw-3sJ-I8

>>31012 >For some reason the crotch reminds me of a robosapien... I think it is the slight angle where the codpiece meets the first hip joint. IMHO, this literally the most critical design spot on the entire robowaifu. Everything else depends on the pelvis/hip confluence being correct to function well as an overall bipedal locomotive system (not to mention snu-snu working properly).

>>31013 Excellent resources, Kiwi. Thanks! :^)

Tesla Bot Motors: https://youtu.be/5MNz0YyqgLY https://youtu.be/1xChD-gv_pc

Capstan actuators are cool, quiet, and efficient! https://www.youtube.com/watch?v=MwIBTbumd1Q

>>31403 >Capstan actuators I came online just to link this. You beat me too it. I second that this is a really great video and has a lot of big ideas. He talks about the mechanics of his drives in such a way that the info could be used for all sorts of drives. BTW the change in the look and operation of the website is...not good. It does not seem to function as well and is harder to read. When I go to references and hover over them, some show the reference, some don't. The old look and function was better...unless you can upload files now, in which case it's better to suffer. I notice this happens A LOT with software. They take something perfectly functional but has a sort of older look and spiff it up in the "new style", like Apples sort of flat candy look, and then it never works as good. It loses functionality and becomes harder to read and use. Don't get me started on Apple iphones. I have one, given to me, and it's a total usability piece of shit. There's all these beautiful icons that, have no explanation. There's all sorts of functions that if you get into them, there's no clear way to back up out of them and you can't read anything. What use is a device with no instructions but hieroglyphics that you have to go online in order to figure how to use it? Apple is the peacock tail of computers, the Thorstein Veblen of conspicuous consumption.

This is a first class link. It's a way to use low cost materials to make very powerful linear actuators. Something I've been talking about. I talked about using belts for windlasses but had not figured out how to recirculate the belts, and/or rope yet. I've been doing drawings of these but can't post drawings here so I gave up trying to describe them...Here's where I mentioned them >>31081 >>31087 My pitiful efforts have not come up with near, or even close, to the tidy package these guys have. I think there's some ways to optimize what they have done though their work is impressive . Don't miss this link. it's a great leap in linear actuator thinking that could lead to a good deal of progress. https://www.linearmotiontips.com/differential-windlass-drives-how-new-designs-work-for-linear-motion/ Why is this so good? I have talked before about using cycloids for bearings. Normal bearings are made of metal because the area they connect is so small. Therefore you need a super strong well machined surface. Big $$$$$. But if you can spread the forces out plenty of plastics and especially fiber reinforced plastics work not only as well but better because of some of the springiness of the material allows for some give. Meaning less likely to break. Manufacturing these sort of things could be easily done with a simple vacuum infusion set up with an accurate, but easily made die/mold for the form. Big time DIY.

>>31576 They have a great video here. It reiterates why I complain about hydraulics and the losses(though I have some changes in mind about this because I think hydraulics might be easier to mass produce.) https://youtu.be/Rp_kmhphpyk It kills me that I was so close to something like what they have, but too late.

>>31576 >https://www.linearmotiontips.com/differential-windlass-drives-how-new-designs-work-for-linear-motion/ Excellent information Anon. I've tried designing low-friction systems that follow similar concepts -- yet remain low-cost. (Faux) silk ribbons are a reasonably-good cording choice IMO; strong, light, resilent, and slippery as an eel ! :D Thanks Grommet. Another great post. Cheers. :^) >=== -minor edit

>>31577 You can just use Chinese/differential windlasses. They only own patents for over complicated versions. Vidrel has all the information you need. https://www.youtube.com/watch?v=zfaeVPQO2ek

>>31624 Neat! This gentleman is a wellspring of old lore. Thanks, Anon. Also, be sure to check out the next, related video of that playlist: https://www.youtube.com/watch?v=OXzGSB4i1UU&list=PL6HIFled82YVpPbRjhcfF7z5R-tmTAJiE&index=8

>>31403 >>31432 >Capstan actuator/drive Seems we all are in the recommendation algorithm loop for this. It seems pretty useful but I wonder what the ideal "rope" would be whether it is steel cable or nylon rope or maybe even steel chains.

>>31707 I was looking at 3D metal printing and ran across this. https://www.youtube.com/watch?v=Js3bJ1B8ySM It seems that "per weight( or specific strength), regular plastic is not so damn weak. A 4gm test piece of PLA+ held 154 lbs. before it broke. And most of the weight was the ends of the test piece. >I wonder what the ideal "rope" would be whether it is steel cable or nylon rope or maybe even steel chains. Strongest would be 4mm dyneema or UHMWPE Braided Cord(same thing). WOW! Black 1/8"(Dia.) 2200Lb $37.95 100 ft. Nylon would likely be WAY cheaper and good enough. The chain I reference below is less strong and more expensive but...using metal chain like this can be used as part of a motor, #35 x 100 ft. Zinc Plated Steel $48.47 Working load limit of 106 lb.[actually price per strength is not so bad. This is "working load" which is way less than breaking load]. The other ropes are shown, breaking load. https://www.homedepot.com/p/Everbilt-35-x-100-ft-Zinc-Plated-Steel-Sash-Chain-806600/203958837 Another picture of copper coated, https://cdn11.bigcommerce.com/s-e7e21/images/stencil/1200x1800/products/127986/328788/apii9rkiv__89322.1592325935.jpg?c=2 I was surprised by the low cost of UHMWPE. In the past I thought that the price for UHMWPE was really high. It must have come down or I didn't closely compare. Maybe the Chinese are driving the price down. I think the cheap price for UHMWPE is because it's actually polyethylene. The same stuff everything is made of. Plastic bags etc. But if I understand correctly, they make the poly chains much when they make the new plastic. They stretch them to align with each other. Now the long chain plastic molecular links are not extremely strong individually but they are so long that there are a huge amount of them and all the little molecular bonds add up to a really strong plastic. The big problem with this stuff is it's hard to tie knots or to connect it because it's so slippery but I'm sure that can be over come. Another problem is this stuff has NO stretch at all. So it would make poor tendons which need some stretch. They have nylon 1000 lbs. paracord [they say 1000 but...]for less which might be fine and stretches. $8.99 100 ft.

>>31403 >Capstan actuators are cool, quiet, and efficient! Github: https://github.com/aaedmusa/Capstan-Drive Yes, very interesting. But I guess this will only be usable in some places, since it is big. I also found this here. Since I always had the idea of springs or stretching rope for muscles in the back of my mind, I think this idea here is something to also keep in mind. Maybe for making the waifu jump in some cases, or for some fast movement. A loaded spring could be blocked with a solenoid for a little while. >Grasshopper leg explained https://youtu.be/MwIBTbumd1Q >>31576 >Chinese/differential windlasses >>31629 > Old lore, The differential chain hoist. Very interesting. Thanks.

Just as a thought exercise or a benchmark on how much force you can get with electromagnets, which also means EMF motors. I was looking at high power ceramic magnets and they had electromagnets mixed up with them in the site. Here's a couple with prices, Electromagnet with 1/4 in. Mounting Hole - 12V DC/3.3W - 1-9/16 in. dia x 1-9/16 in. thick - 59.54lbs Pull EM4441M6 Now From $20.99 Was: $25.20 | Save: $4.21 Electromagnet with M5 Mounting Hole - 24V/10W) - 110.25lbs Pull - 1-7/8 in. dia x 1-1/16 in. thick EM502027M5 From $35.28 https://www.first4magnets.com/us/44lbs-110lbs-t49 10W - 110.25lbs Pull That's a hell of a lot of force from a small amount of power. Now to be realistic this force is the attraction, likely, to a thick steel plate but a properly designed motor would have some of the same attributes in terms of force per watts, per weight, per size. Makes a good rule of thumb comparison. Say you shave off 20% and you could likely get these figures with a well designed motor, actuator. So with say...40W we could get somewhere around 400 lbs,. of force. That's what I'm looking for. At that level it could easily pick up most humans. We can use this to plan wire size. 40W/48 volts = 0.83 Amps Not a lot. Chart. https://2.bp.blogspot.com/-9TdTKLNaH_0/WBsh3YchzjI/AAAAAAAAEfU/BQZ9Nqc8IU80GFdBVoQgy0zwOEzWyI7tQCLcB/s640/wire-gauge-ampacity-chart_460534.jpg 26 gauge wire will handle 1.3A enclosed. So it would be better than needed. I see 250 feet with silicon jacket for $18.48 USD. I also see 1.000 foot, surely enough for more than one waifu, 28 gauge for $39.98. Could cost you a little, maybe even twice as much if you buy from US or European sources. (don't know just speculating) Chinese wire, maybe good, maybe not. Cheap but a risk. Figure in troubleshooting time if they send you bad wire and it may be worth twice the cost for Europe or US wire. Make sure, silicon jacket AND machine wire spec. Other jackets may be just as good but I do know silicon would be up to the job. More research may find stuff cheaper that will work but silicon jacket is the safe bet. Machine wire is specifically made to handle movement and vibration. Otherwise the wire will break through work hardening as it moves. I did these for myself to nail down some rough cost guidelines and thought I would comment it here. Maybe it will help others.

>>31961 >High pull force magnets It's important to remember you are given the force at contact with steel. Which is pretty close to your guess. >400+ pounds of force at 40W You can indeed get tremendous forces at low currents. The problem is this requires the steel to be touching the electromagnet. Which, would prohibit rotation. Using many of them to create a rotating magnetic field would be more expensive, heavier, and larger than a comparable purpose built motor. What I think you're going towards are solenoids. They will provide the forces at low power you are looking for. This provides you with 2 problems to solve. 1. How will you control the stroke to fit into your needs? 2. How will you retain position without the solenoid becoming a heater? Some helpful links; https://science.howstuffworks.com/solenoid.htm https://audioxpress.com/article/voice-coils-a-tutorial https://www.machinedesign.com/mechanical-motion-systems/article/21836669/what-is-a-voice-coil-actuator

>>31961 >>31978 Thanks, Anons.

>>31978 Great graph. Very informative. >What I think you're going towards are solenoids No I was just showing a general rule of thumb of what sort of forces we could get for what power. I do realize the force they are quoting is directly connected to a thick steel plate. A thin one would not show this sort of force. it would have to be thick. But I have brainstormed solenoid type actuators here. >>9984 >>10002 It's a terribly retarded idea, and likely noisy, but...really cheap and simple.

> (actuator/joint-braking convo-related : >>32321, ...) >=== -minor edit

Potentially a breakthrough increase in motor-efficiency design. https://www.youtube.com/watch?v=JGdIZF5gE7I

>>32828 Interesting, thanks. I kinda thought this was just some AI generated BS, after looking at the thumbnail, so I hesitated watching the video. There's too much BS on YouTube and I on the other hand tend to fall too much for drama. Anyways, there's also some guy providing special software for making motors, I saw him on Reddit or Twitter and YouTube, but don't remember the name. I thought this would just be another layer of difficulty and rabbit hole, but maybe we'll get to this one day. He exists, and some people are interested in it. One thing that interested me before was the alternative of motors without rare earth magnets, or perma magnets in general, since these could get rare or would be a possible vector to ban people from building robots. Good to see that there might now even be alternatives to these induction motors. But this is also just something to keep in mind, doesn't help us right now to get things done. I don't see anyone of us doing this anytime soon. These seem to be build for electric vehicles for now, and the one shown as a possible product weights 39kg (86 pounds). We certainly should keep an eye on it, maybe it will be available as a product for a scooter or paid printing services could deliver something like it one day. Though, then again, the video also says that this is "competitive" engineering and might not be economically feasible. And it's a prototype and hasn't really been tested. So, not really BS, but most likely also not useful anytime soon. Just want to throw in some keywords here for later search: copper printing, 3D printed copper coils, laser powder bed 3D printing, multi-material powder printing

>>32910 >since these could get rare or would be a possible vector to ban people from building robots Great point, NoidoDev. Yes, I've thought about this exact topic many times, trying to find a way around it. Of course, I'm far less concerned about the Chinese resisting robowaifus (also the biggest refiner of rare-earth metals, obvs), and far more concerned about the Globohomo we here in the West are all assaulted by. The troon-thug cycle is far more a personal pressing threat to us all, than w/e Dr. Lee or Mr. Chang think about us. <---> But REM aside, just the cost of the motors assemblies themselves is prohibitive, thus why I was exploring devising our own, opensauce, motor winding systems.

Came across an interesting actuator design on youtube: A 2-stage gearset with a 21.3/1 reduction ratio. Aside from the power he's able to generate with it, he mentions something that I never thought of, that being backdrivability being used to sense external forces applied to the robot. The only glaring issue I can see is it's large size. On a completely function oriented machine like what he's making that's no issue. But on a robowaifu? I'm not sure how easily they can be hidden. https://www.youtube.com/watch?v=v3g2VdduawI

>>33233 Excellent actuator, backdrivability is indeed important for man machine relations. In this case, it appears they are using current sensing to determine position and external torques being applied to the motor. I base this on a lack of light leakage, or dark spots indicating a magnet under the thin white plastic. The B-G431B-ESC1 (https://www.st.com/en/evaluation-tools/b-g431b-esc1.html) would be an ideal ESC to use for this due to its low cost and having plenty of support online. As for the gears, replacing them with eccentic cylcoidal mechanisms would allow for a smaller, more efficient system. Still would recommend adding some sort of grease or lubricant to reduce wear, heat, and drastically increase usable lifespan for the actuator.

Anon-designed, custom servo motor control board. Very cool! :^) https://www.youtube.com/watch?v=DDDD7iGMljw

Remarkably good linear actuator. I'm contemplating reworking my thigh/knee/shin complex to use these instead now. https://www.youtube.com/watch?v=bQl6gj_6oa8

>>33282 Update to the actual, working, prototype Jugglebot. Very entertaining Anon, and impressive project progress. https://www.youtube.com/watch?v=CiorGYCZOgk&list=PLe8shy-tVWtqXVAZ4kuTlRkpZlsaZRHAK&index=70 This guy should get to know /hover/ !! :^) https://trashchan.xyz/server/thread/4.html >=== -sp edit

>>33282 >>33283 What's especially-intriguing to me about this Anon's project is that from the outset, his goals for it (ie, womping teh heckin' heck out of those jugglin' balls!111) quite naturally channeled his design decisions into much of the same solutionspace that we here ourselves also need to manage arriving at (cf. >>33356, et al). Key above all others, was his (apparent) innate recognition that mass is everything : ( >>4313 ). His choices based on this single constraint (ie; REDUCE INERTIAL MOMENTS AT ALL COSTS -- aka 'thrown weight') led directly to his distinctly-successful linear actuator design. This success, in turn, directly enabled the success of his agile Stewart Platform approach. While this complex arrangement design won't likely be of much use to us for robowaifus beyond possibly the torso-actuation + collarbone/neck/skull -actuation complexes -- the basic linear actuator's low-mass & responsiveness & (relatively) low-cost & reasonable-manufacturability make it quite-well-suited to robowaifu skeletal actuation in general. Further, the fact that probably >95% of the actuator's mass is concentrated on one end (and therefore able to be arranged proximally to the robowaifu's main-torso monocoques -- thereby further reducing thrown weight+helping to coalesce dynamic central-mass around the pelvic volume geometric pivot-point) is just a big, energy-saving cherry on the top! :^) <---> >tl;dr This must be a kind of joke being played on me rn that I only just now discovered this project. It would have shortened much of the time for my design explorations thus far. Maybe I just needed a little """reminder""" to remember that 'The real prize was the friends we made along the wayduring this long & arduous, monumental climb' , hehe! Actually, I honestly do thank you for everything you've done for me, dearest Lord Jesus!! :D <insert: some_days_you_just_feel_like_getting_out_of_bed.jpg> >=== -prose edit -add crosslink

>>33283 That is one impressive actuator.

>>33315 I'm planning to totally steal his stuff for our waifus. Exploring the options now. Mine should be lighter than his. I wonder what he would think of robowaifus haha. :D

>>33320 BTW, look at this actuator detail I dreamed up a while ago. >>12941 See the motor like his. A little different. There's a cone on the end of the actuator. The pulley to the right has a spring pulling it down. "If" a large force is needed then the pulley is pulled towards the smaller end of the cone on the motor. Giving you a better force, but the motor has to spin more. Auto transmission based on spring constant. His is much better for linear but I think the one I dreamed up would be good for waifus. If I was going to improve his for linear actuators, I would make it like a scissor jack. I think it would be much stronger. Possibly the sprung weight would be higher but maybe not because the scissor mechanism would be lighter and far less cost as a truss wouldn't need carbon fiber which is outrageous in cost. You could use fiberglass. Carbon way more expensive but it is super stiff.

>>33364 >this actuator detail I dreamed up a while ago. Nice work! You're really ahead of the curve with that one, Grommet. :^) >this was my response back then, BTW : ( >>12943 ). I may say I loved the 'CVT power-pull' on that little bike -- smooth as silk. :) I like your ideas about further refinements to your design. We here could certainly all use such a device. To my own thinking, Jugglebot Anon's design is a great (& simple) start, and it's linear-actuation certainly works well with my tiny-mass/high-leverage-coefficient skeletal limb designs. Regardless, keep those ideas flowing Anon. Together, We're All Gonna Make It. Cheers. :^) >=== -prose edit

I was looking at gears and files I had saved and ran across an interesting very high reduction gear system that I had forgot about. A bit difficult to understand. It uses the teeth of gears very finely offset with a separate track to wedge them into place. So two tracks have teeth(gear teeth) slightly offset, and a third forces a flexible wedge into them to align them. Moving the track forward. It's primarily used in motors. The Moon rover used one to drive the wheels. They supposedly are very efficient, have very little to no backlash and compact with high gear ratios. Here's a link to video that shows how they work. Notice the linear example he shows first. Maybe, not sure, this could be set up as an actuator with the "flex spline" being activated or pushed into the two active geared splines by a roller that moves up the linear two splines. For testing these could be 3D printed with two solid gears(splines) and one flexible made with TPU or whatever. https://www.youtube.com/watch?v=7QidXf9pFYo https://www.engineeringclicks.com/harmonic-drive/ In some ways, kinds, sorts, this is like the idea I had to use a ball screw and one solenoid to scoot the balls up one at a time. >>9984 I wonder if you could do the same with these two wedge shaped gears and have a single edge move the parts just like the "flex spline" moves the rotary motion of the gear train in a harmonic drive?? Another way of looking at designs somewhat similar is the action of a farm jack.

>>33531 Uh oh, I'm repeating myself. Sigh...

Fridman interview, Electric vs. Hydraulic actuation: https://www.youtube.com/watch?v=6r6KPuJ689o

>>33990 Nice video. I hate hydraulics but...I could see with some sort of new valve design how it could be super cheap and easy to manufacture.

I'm finally writing down the plan for each movement, using the pattern from here. While listening to some soundtracks to prevent me from getting bored and going back to do something else. This plan might need several iterations, and this is just work in progress. I kept thinking about these things, but then forgetting it again, I really have to write it down. I think it's quite general, so it can fit into the thread here, but maybe it's time to make my own. - release and nudge means having a (soft) lock holding something in place, releasing it and just have some actuator the part into falling into one direction till it is stopped. - I think each movement will need more than one actuator. We need something for slow and something for fast movements. The latter might be limited in range. The precision might also vary. - loaded springs mean springs that are loaded and blocked by a solenoid and can be released to support or carry out a movement. > Head and Neck: > 1. Neck flexion and extension (forward and backward movement) - fast short movement: - forward: release and nudge, maybe multiple stages - solenoid, with a spring? - backward: twisted flexible string in the neck - maybe additional servo somewhere in the neck - slow full movement: - forward: flexible string in the neck slowly released - maybe one servo to pull forward at the end - backward: string in the neck fully twisted and maybe blocked > 1. Neck lateral flexion (side-to-side movement) - one servo for side rolling, two solenoids to lock - fast short movement: nudge and fall down, spring support back up - slow full movement: more in the shoulders > 2. Neck rotation (turning left and right) - fast full movement: maybe at least one loaded spring - slow full movement: more in the shoulders - regular fast movement: at least one fast spine servo, bones connected

>>34130 POTD I'll look into following your lead with my own software functions in my code, NoidoDev. Keep up the great work! Cheers. :^)

I'm sure everyone knows about those carbon nanotube muscles that were ten years away fifteen years ago, but has anyone tried just building the same thing using commercial carbon fiber? It's almost as strong and almost as conductive, so I don't see why it couldn't be used in the same way. https://www.youtube.com/watch?v=n-zXKrBoJGs

>>34133 >carbon nanotube muscles Interesting. Well, how do they work? Has anyone made them outside of a big lab? I looked a bit into it for a moment: >Rotary motors of conventional design can be rather complex and are therefore difficult to miniaturize; previous carbon nanotube artificial muscles provide contraction and bending, but not rotation. We show that an electrolyte-filled twist-spun carbon nanotube yarn, much thinner than a human hair, functions as a torsional artificial muscle in a simple three-electrode electrochemical system, providing a reversible 15,000° rotation and 590 revolutions per minute. A hydrostatic actuation mechanism, as seen in muscular hydrostats in nature, explains the simultaneous occurrence of lengthwise contraction and torsional rotation during the yarn volume increase caused by electrochemical double-layer charge injection. The use of a torsional yarn muscle as a mixer for a fluidic chip is demonstrated. https://www.science.org/doi/10.1126/science.1211220 >Improved electrically powered artificial muscles are needed for generating force, moving objects, and accomplishing work. Carbon nanotube aerogel sheets are the sole component of new artificial muscles that provide giant elongations and elongation rates of 220% and (3.7 × 104)% per second, respectively, at operating temperatures from 80 to 1900 kelvin. These solid-state–fabricated sheets are enthalpic rubbers having gaslike density and specific strength in one direction higher than those of steel plate. Actuation decreases nanotube aerogel density and can be permanently frozen for such device applications as transparent electrodes. Poisson's ratios reach 15, a factor of 30 higher than for conventional rubbers. These giant Poisson's ratios explain the observed opposite sign of width and length actuation and result in rare properties: negative linear compressibility and stretch densification. https://www.science.org/doi/10.1126/science.1168312 >Carbon aerogels are elastic, mechanically robust and fatigue resistant and are known for their promising applications in the fields of soft robotics, pressure sensors etc. However, these aerogels are generally fragile and/or easily deformable, which limits their applications. Here, we report a synthesis strategy for fabricating highly compressible and fatigue-resistant aerogels by assembling interconnected carbon tubes. The carbon tube aerogels demonstrate near-zero Poisson’s ratio, exhibit a maximum strength over 20 MPa and a completely recoverable strain up to 99%. They show high fatigue resistance (less than 1.5% permanent degradation after 1000 cycles at 99% strain) and are thermally stable up to 2500 °C in an Ar atmosphere. Additionally, they possess tunable conductivity and electromagnetic shielding. The combined mechanical and multi-functional properties offer an attractive material for the use in harsh environments. https://www.nature.com/articles/s41467-023-38664-6 Buying the ingredients might be possible: https://www.acsmaterial.com/carbon-nanotube-sponges.html - But it's around US$200.- per gram. Single-Walled Carbon Nanotube Paper: $300.00

>>34139 My understanding is that it contracts because the fibers repel each other when they're charged to a high voltage. Back when it was announced they were talking about it being 20000 times stronger than natural muscles, but after some initial excitement there were no further announcements and people just forgot about it. While I don't know why it was dropped, I'm guessing it's because they never found a cheap way to produce the fibers, which is why I think carbon fiber might be the answer: it's not as strong as carbon nanotubes but it's probably strong enough and $200 will buy you 4lb of it. https://acpcomposites.com/shop/fabrics/tow-cord/carbon-fiber-tow The high voltage may pose a problem but I think it could be safely contained using a PTFE sleeve wrapped in foil. PTFE can comfortably handle at least 10kV/mm, and if it wears too thin in one spot then the fibers would touch the foil and short to ground.

>>34141 >expensive and high voltage Too bad. If it's expensive but fast it would be interesting for small movements, like in the face or maybe the hands. But I wouldn't like high voltage there. >I think carbon fiber might be the answer Well, I'd like to see someone experimenting with it. Maybe try to convince people outside of this board. We simply aren't that many people. Or add a task to our "Stop Lurking" thread here: >>20037

>>34132 >I'll look into following your lead with my own software functions in my code Thanks for working on this. I hope you make this into modules which can be easily adapted to something. I could also imagine that I'd like to run the code for the actual movement decentralized on Arduinos. The realization and the command to do a set of movements, the motion planning, and the execution of the commands to the servo might be on different hardware (SBCs and controllers). Some notes from expanding >>34130 >Problems to solve - linear actuator, but hold in place by solenoid blockage - twisted string, freezing with solenoid blockage - solenoid holder, but with a limiter - adjustable spring mechanisms >General: - Any movement that happens automatically through gravity should have a spring that can be loaded and locked. Then the reverse movement is supported by a release of the spring(s). Compliance can be somewhat archived by making these movements not too strong or by using a brake mechanism. Precision isn't important. - Repetitive movement like some dancing moves should also be covered with spring mechanisms. Adjustable by moving the spring linearly a bit.

>>34130 >>34152 Seems that you're interested in dynamic motion with brakes for control. Check out passive dynamic robotics. There's much that can be done using gravity and precise control. https://journals.sagepub.com/doi/full/10.1177/1687814015620967

>>34157 >passive dynamic robotics Thanks. That's certainly interesting. Though, I'm not very focused on walking, and I'm not sure how much this will add to make something walk. I meant it more general. Anyways, it's probably good to keep these ideas about walking in mind right from the start.

>>34152 I might draw and scan something tomorrow. But here are the descriptions. We have to keep such concepts in mind, to design a good system. Not just relying on regular servos. > linear actuator, but hold in place by solenoid blockage Linear actuator in a tube. It being up is holding something. It can move down again relatively slowly (and noisy). But can release faster by "falling down" the tube after a solenoid blocking shaft is released sideways. Of course, another mechanism would need to get the linear actuator up to the neutral position again. > twisted string, freezing with solenoid blockage Some holding mechanism holding the twisted string in that position. The solenoid opened during the twisting of the string with a servo, it goes back to the unpowered position which holds the twisted string in place. > solenoid holder, but with a limiter Just adding a more flexible material in front of the shaft. This softer part is being used to block something. But it can give in, if a certain amount of force is used. The solenoid shaft itself is way stronger. > adjustable spring mechanisms Idk. I'm sure this is a solved issue. Some mountain-bikes have adjustable shock absorbers. I want something like that and smaller variants, but with a small servo that does the adjustment.

>>34130 These are the notes for now. Just posting this as a backup and if somebody wants to think about it himself. I have to think more about the details over time. > Head and Neck, 7 disks: 1. Neck flexion and extension (forward and backward movement) - fast short movement: - forward: release (5) and nudge (1), maybe multiple stages (6) - solenoid, with a spring? Adjustable (3) - backward: twisted flexible string in the neck (2) - maybe additional servo somewhere in the neck (1) - slow full movement: - forward: flexible string in the neck slowly released (2) - maybe one servo to pull forward at the end (1) - backward: string in the neck fully twisted and maybe blocked (2) (4) 2. Neck lateral flexion (side-to-side movement) - servos for side rolling, two solenoids to lock - fast short movement: nudge and fall down, spring support back up - slow full movement: maybe servos in the disks 3. Neck rotation (turning left and right) - fast full movement: maybe at least one loaded spring - slow full movement: flexible twisted strings in the shoulders - regular fast movement: at least one fast spine servo, bones connected > Upper Body: 1. Shoulder flexion and extension (raising and lowering the arm) - Maybe a fast linear actuators - Loaded spring actuators - Twisted string actuators for regular movement - Servos for precision movement 2. Shoulder abduction and adduction (arm away from or towards the body) - servos in the shoulder and chest (breast) - maybe solenoid under the armpits - maybe loaded spring actuator under the armpits 3. Shoulder internal and external rotation (rotating the arm inward or outward) - one servo per shoulder 4. Elbow flexion and extension (bending and straightening the arm) - fast movement: small servo in the elbow - maybe additional twisted string actuator - additional bending spring actuator >>13711 - slow full movement: windlass pulley in the arm 5. Forearm pronation and supination (rotating the forearm to face down or up) - one servo per shoulder - rotation within the (lower) arms with servos >>14474 - maybe also electro-magnets in the two bones (fast, no load) 6. Wrist flexion and extension (bending the wrist up or down) - actuators in the lower arm ... 7. Wrist radial and ulnar deviation (bending the wrist towards the thumb or little finger) - actuators in the lower arm ... 8. Hand and finger movements, including individual finger flexion and extension, abduction and adduction, and thumb opposition >>4577 This is going to be a topic on it's own. Mostly actuators in the lower arm and some small ones in the palm. - palm bending: ... - group of finger bending, and opposite direction: ... - individual finger bending: ... - finger abduction and adduction: ... > Spine and Torso: 1. Spinal flexion and extension (forward and backward bending), 5 disks - small servos in some spinal bones - servos with (twisted, flexible) strings in the hip area and lower torso 2. Spinal lateral flexion (side bending), 5 disks - small servos in some spinal bones - servos with (twisted, flexible) strings in the hip area and lower torso - the water- or airbag holding mechanism 3. Spinal rotation (twisting), 12 - small servos in some spinal bones - pulling the other bones with it if it rotates to a certain extent - servos with (twisted, flexible) strings in the hip area and lower torso 4. Pelvic tilt (tilting the pelvis forward or backward) - maybe one servo > Lower Body >>237: 1. Hip flexion and extension (lifting the leg forward or backward) - huge servos in the hip areas, maybe Capstan actuators - one servo in the back of the hip - to string-pull each leg in series just enough to walk - should probably also bend the knee a bit - string based actuators in the thighs, switching string idea 2. Hip abduction and adduction (moving the leg away from or towards the midline) - rotating uneven disk to push it outwards while normally moving inwards - maybe additional pull outwards for faster movement 3. Hip internal and external rotation (rotating the leg inward or outward) - servos in the thighs 4. Knee flexion and extension (bending and straightening the leg) - servos in the knees - one servo in the back of the hip - to string-pull each leg in series just enough to walk - same as for hip flexion and extension - string based actuators in the thighs 5. Ankle dorsiflexion and plantarflexion (pointing the foot up or down) - servos in the ankle - maybe additional loaded spring and solenoid in the lower legs - ideally a way to load a spring mechanism when landing with the heel 6. Ankle inversion and eversion (turning the foot inward or outward) - servos in the ankle and lower leg - some string based actuators - maybe additional loaded spring and solenoid in the lower legs 7. Toe movements, including flexion and extension, abduction and adduction - mostly all moving together - servos for slow and solenoids for fast movement - maybe additional spring and release mechanism for running

>>34152 >Thanks for working on this. >I hope you make this into modules which can be easily adapted to something. I could also imagine that I'd like to run the code for the actual movement decentralized on Arduinos. Y/w. I hope to expose every user-directable function as a pythonic API, suitable for scripting from, say, CyberPonk's Horsona library. Even Bash scripting isn't off the table ATP. >The realization and the command to do a set of movements, the motion planning, and the execution of the commands to the servo might be on different hardware (SBCs and controllers). Yes, very much so. Anon's IPCnet is the backbone to allow such distributed, decentralized computing to occur within our robowaifus. >>34167 >I might draw and scan something tomorrow Hope to see it soon, NoidoDev! Cheers. :^) >>34171 POTD This is growing into something important to us here, IMO. Keep it up, Anon!

Hear my schizo post, Anons. I think I have a good idea. We should use this train of connections to form an electric muscle. <Bone tether => Actuator Housing => Generic electric motor => magnetic sun-gear reduction => spool => tensile cable => bone tether. Don't forget a reset spring that . Turning on the motor puts torque through a double-acting gear-reducer, and clutch to wind up a spool, to pull a fiber, to pull the muscle tethers. Why this train? First off, you'll need to see this: https://www.youtube.com/watch?v=eaMD_9kOlTA&t=154s This video is the work of a genius on a magnetic gear train. Why do you care? Because... >We can swap out one of the magnetic gearwheels with *electromagnets*. >Electromagnets will vary strength based on the current running through them, and this allows us to essentially have a *frictionless clutch* between the electric motor, and the muscle fiber. A frictionless clutch! >The sun-gear set is concentric! We can fit them all into a tubular housing with a skinny aspect ratio! This takes a ton of complexity out of controlling the actuator. It will act like a muscle. You turn on the motor and activate the clutch, the muscle pulls. Turn up the power of the motor/clutch, it pulls harder. When you're done, it relaxes and resets to the spring settings. If the muscle is strained harder than it pulls, the clutch simply slips! It's cuddle-safe! This allows us to: >Spin up the motor and dump the clutch for explosive movement (can be dangerous). >Use the clutch slippage strength to vary or cap muscle tension for safety reasons. >Cap the power of the electric motor for safety reasons. >Simplify motion control to an easy pair of inputs that a neural network will easily understand. (clutch and motor) >Simplify the actuator to be easy to repair and replace. >Create a skinny actuator package which fits well onto a skeleton. >Quiet down motor whine because gear reduction. >Get sufficient torque because gear reduction. >Tune the neutral pose of the body by just pre-tensioning the springs (It would be nice if the roboGF could do this automatically somehow, would save her a lot of battery). >Have at MOST two pairs of wires per muscle. >Sink the actuator package into the bone itself if you so please (to save volume) >Use simple, off-the-shelf parts which you can get in highly variable sizes. Just build a good skeleton, and place a fucking ton of these things on it Have like 5-20 different sizes of e-muscles. Little skinny ones as thin as a pencil should be controlling the face. Big fat ones should be running the biceps. Excuse my drawings for being rough drafts, I just want you to get the picture.

>>34161 I meant that passive dynamic research has provided different mechanisms, algorithms, and equations which would benefit your designs in general. What works for walking, works for anything else that relies on dynamics. We all need to keep an open mind for using previous research in novel ways. For instance, you're working with breaks, and passive dynamic research involves using breaks to optimize energy use with gravity.

>>34319 POTD This is great stuff, Anon. I'd suggest you also have a look at the linear actuators for the Juggler Bot Anon's designs. They are fairly closely related to your approach as well, I think. I hope & pray you'll actually get the resources together to make these actuators real, and then share your experiences here with everyone. Please keep the great ideas coming, Anon! Cheers. :^)

Someone has taken that very-interesting 3-element 'U-Joint' design has been enhanced with what looks to be a cable/pulley system that probably acts as an actuation system, AFAICT. https://www.youtube.com/shorts/6q92GHuHhp8

Traditional hobby servos aren't dead yet. In fact they keep getting stronger and the price vs torque keeps getting lower. New on the market are a line of high-torque (as in 150kg/cm, reputedly) servos that look very promising for robotics projects. The product line includes lower strength servos in the standard size, but the stronger ones are in what we used to call 1/4 or giant scale, as in 1/4 scale model planes. Here's some specs lifted from the ebay listing: Specification: • Size:52*30*65(servo), 64*55.6*20(Assembly the metal brackets size) • Weight:165g • Wire length:320mm • Speed: 0.24sec/60 degree at(10v), 0.21sec/60 degree at(12v) 0.19sec/60 degree at(12.6v) • Torque:150kg.cm.at(10v), 165kg.cm.at(12v), 173kg.cm.at(12.6v) • Voltage: 10v-12.6v • Dead Zoon setting:3 Microseconds It looks like they also come with some useful hardware. Poppy anyone? They seem to be too new for many reviews on youtube yet, so here's an add that might bring up additional videos in time: https://www.youtube.com/watch?v=rmnZtFXuL0c And it looks like they are available at the usual suspects: https://www.ebay.com/itm/286032154221 https://www.aliexpress.us/item/3256806456032772.html Also at Temu but I can't get a clean url

>>35236 Thanks, Robophiliac! This type of spec information is deffo appreciated ITT. Cheers. :^) >Also at Temu but I can't get a clean url Just cutting off everything after the first part is usually sufficient (even if the site adds stuff back on afterwards again). Eg: https://www.temu.com/search_result.html?search_key=hobby%20servos

>>35236 i think weve moved past servos and agreed that bldc motors are the wsy to go since they make little noise. Hannahdev dropped them. For a worker robot it might not matter but robot waifus are held to a higher standard.

>>35236 A problem I've discovered with those servos is they suck up a lot of current. I haven't been able to take full advantage of 'em using a standard buck converter, but maybe a high-current power converter built especially for servos would do the trick... haven't tried it yet since SPUD got upgraded to the ASMC-0X types.

>>35254 I think it's important to keep things in perspective. Currently there are no replacements for servos to perform many of the actions we want our robowaifus to be able to execute. Oh, there are theoretical, and even experimental devices, but nothing else is as capable, as affordable, or has the decades of experience in use that servos do. So, unless you want to wait until the current vapor-ware comes to market in 5 to 10 years, like electro-elastomers or HASEL actuators, we are stuck making the best use we can of what is currently available. I think it may help to compare a robowaifu project to building a custom kit-car on a budget. Once you choose a body style of a car to build, the first purchase would be the windows and body panels that might become unavailable if the kit manufacturer went out of business. Then you would buy the chassis of the vehicle your kit was designed to be built on, and then as your budget allowed you buy the brakes, steering, electrical and engine components, etc. Even after buying and installing parts, even completing your car, you may replace parts with better parts as they become available or old ones wear out. But if instead you have a finished design in mind that you will not drive until it is exactly the way you envision it, it's possible, even likely that you never will. In the same way, if you approach a robowaifu project with the idea of only using the "perfect" parts to start with, it's possible you will never finish, or even start. Consider that all of the projects that anons have shared are not just works in progress, but also in development. Why not just start, using what is available, and if you decide the noise of servos is too annoying before better replacements are available, then investigate sound insulating materials. "Too noisy" is after all an opinion based on personal taste. Personally, I think a little mechanical noise would add to the charm of my ROBO-waifu. But honestly, after a couple of weeks of loving companionship, good conversation and fantastic sex i not only won't care about a little servo noise, I won't even notice it. And you wouldn't either.

>>35257 Good engineering advice, Mechnomancer. Thanks! >>35264 POTD Words of wisdom right there, Robophiliac. Every'non here please remember Gall's Law : ( >>33553 ). >tl;dr We'll always be able to incrementally-improve our Model-A robowaifu types, etc. But if we each wait until we've perfected the 'Flying Car' versions before enjoying our initial robowaifus, then we'll simply never get anywhere with this. And let's make sure we all get somewhere with this, alright? < Alway Rember: Start smol; Grow big. Cheers, Anons. :^) TWAGMI >=== -sp, fmt, prose edit

>>35264 >Currently there are no replacements for servos to perform many of the actions we want our robowaifus to be able to execute I'm trying to be mean but you are severely limiting yourself to think servos are the only answer. I ran some numbers on Tesla's Inductive reactance motor in their cars. Short answer is, Tesla 3 motor output = 258KW(345HP) at 27.99Kg(60lbs.) but a human rarely puts out more than 400 watts in peak. https://en.wikipedia.org/wiki/Human_power So the tesla motor puts out 4300 watts/pound and human puts out(at 250watts and 200lbs.), 1.25 watts/pound So there's a hell of lot of room for using cheaper less specialized materials and you can still get far better performance than a normal or even elite athlete human. The numbers don't lie. The path to performance is there. Here's a comment I made where I talked about the power of magnets and inductive reactance in comparison to presently built electric motors and the strength needed for humans to do work at a very high rate. There's LOTS of room for improvement over servos. Lots. >>12140 And if it pleases you, you could add mechanical noise like electric cars do to let people know they are going down the road. :)

>>35297 Oops. I'm NOT trying to be mean.

>>35297 >>35298 >Oops. I'm NOT trying to be mean. No Problem >>12140 So tell me, how long do you estimate before that technology is available in a form factor that will replace the 28 perfectly adequate servos I already have that I'm about to try to install in a head, and still leave room for electronics, and at a competitive price (<$2 each)?

>>35299 >install in a head I didn't get that part. They may be perfectly fine for the head but I read what you said and I did not comprehend that you were only talking about the head. My point was power over all. The ones you linked here >>35236 are $31.77. Not <$2.00 and they do not have near enough power for lifting things. Though that may not be important to you. To me it is. As for me, I plan to make them. It's iron laminations and coils. It's far less complicated than servos. I have some idea that instead of iron laminations you can use iron oxide in resin. Not sure about that. There are ferrites that can be used. At one time they were not used because controlling them was difficult but cheap microcontrollers has changed that.

>>35301 >I did not comprehend that you were only talking about the head No, I wasn't. >Currently there are no replacements for servos to perform _many_ of the actions we want our robowaifus to be able to execute. >The ones you linked here $31.77 $22.46 at Ali Express, $23.46 for the 160kgcm, until 1/8/25. >they do not have near enough power for lifting things. Let's check that with a real-world example. I want my robowaifu to be able to lift a gallon of milk, and these servos are a candidate for the job. Let's say her forearm (a modified inmoov design) from elbow to milk jug handle is 18", or 1.5 feet. 8 lbs at 1.5' = 12 ft/lbs = 144 lb/in = 165.9 kg/cm - So this servo is not strong enough to lift that jug of milk if it is mounted at the elbow, which only an idiot would do on something this size. Instead, I'm going to place the servo in the upper arm where God intended it and I'm going to have it bend the arm by using a 4cm diameter sprocket at the elbow, driven by a cable and chain from a 2cm diameter sprocket on the servo. 4 is twice the diameter of 2, so 2 x 150kg/cm = 300 kg/cm of torque applied to the forearm to lift the milk jug. Of course, the speed of motion has been halved. If that turns out to be slower than I would like, I can mount the servo so that it can slide on rails up and down the upper arm and attach a cable from the drive sprocket to some point near the shoulder. Now, as the forearm is lifted, the servo also pulls itself up the arm. In this way the speed of the moving servo is added to the motion of the arm, and the combined result is that the arm is faster than the servo alone without the torque reduction. I could go with a 3cm dia. sprocket if I prefer for style, and I would still get 225kg/cm of torque on the forearm. I could use a different configuration, double up on servos, and I'll definitely be hacking them. It's all about creativity and ingenuity to get what we need out what we have, and realizing that what we want is already there, if we know how to get it. That's why servos are probably a billion-dollar industry, if not more, world-wide, with millions of them in use daily, and they are constantly being improved. They work. >As for me, I plan to make them What? >It's iron laminations and coils. It's far less complicated than servos. I have some idea that instead of iron laminations you can use iron oxide in resin. Not sure about that. There are ferrites that can be used. >At one time they were not used because controlling them was difficult but cheap microcontrollers has changed that. I have no idea what you are talking about. Ask yourself what it is you are trying to achieve? Are you trying to build a robowaifu, or have you allowed yourself to wander off on a quest for a mythical better part? stay on target! Build your waifu now, with the materials you have collected and the parts you can use now STAY ON TARGET! . Then, if you still want to develop a new technology, do it with the help and support of your waifu.

>>35297 Grommet, the Tesla 3's motor is a servo. A servo is any system wherein a mechanism for sensing speed or position is implemented. Almost all brushless or AC motor applications involve some kind of servomechanism due to a lack of comutation to control rotary motion. What you're really saying is that brushless motors have a higher power density than their brushed counterparts. This is true, it's also true that you can buy AC servos that match Tesla's drive motor. They see extensive use in industrial robotics for die pressing metal sheets, moving large items on conveyor belts, and heaps of other things. >>35301 >Making your own motors Why bother? Ready made motors are cheaper, easier to use, and more efficient than anything we could build in our own labs. Just calculate your force and acceleration needs and either buy an appropriate servo or buy a brushless motor that exceeds your needs and use whatever sensors, reducers, and motor controllers to implement your custom servomechanism.

Really proud of the team here on /robowaifu/ , Lads! Such good ideas, such solid engineering advice, tackling such monumental complexity with an easy grace. TBH, I've never seen the like of it on any other IB in my experience. GG, Anons. Cheers. :^) Keep moving forward.

>>35303 >Ask yourself what it is you are trying to achieve? I know exactly what I want. I want a robowaifu strong enough to pick up a person. So stronger than most Men. I want it to be super cheap and have, a goal, not a command, the same number of major muscles as a human(300). I can not get there paying $30 for servos or $20 or $10. The only way I know to do this is some combination of a switched reluctance motor and/or solenoid which operates much the same. And the other way would be some sort of hydraulics, which I don't like, but, if a low cost enough valve can be made and controlled easily, then it would likely be even cheaper, so would be the path to take however it offends my sensibilities. What I don't want is something that is machine looking. >develop a new technology I made a mistake earlier and said "inductive reactance", I meant, "switched reluctance". I goofed. Switched reluctance electric motors were the very first electric motor made, (W. H. Taylor in 1838). They didn't catch on, as said, because they didn't have micro-controllers and they were jerky. A good article on them and their advantages. https://www.controleng.com/articles/resurgence-for-sr-motors-drives/

>>35304 > Tesla 3's motor No I wouldn't call it that at all. Though you may be "technically" correct but I would not say all motors that have control feedback are "servo motors". Would you call a pulse width modulated regular electric motor a "servo motor". Maybe but I never heard anyone call it that though technically that's true but people call it a PWM motor. If you want to get very specific Tesla's motor a "Permanent magnet assisted synchronous reluctance motor (PMA-SynRM) ". The magnet focuses the fields through the sheets of metal. Though I'm not sure, I believe it's to push the fields out in some areas and to concentrate them in others. This has the effect of smoothing out the pulsations of reluctance motors(I suspect). Knowing how anal Tesla is I'm sure the magnets give it a little boost too. The magnets are not very large and are not the primary force driving the motor. The primary force driving the motor is "reluctance". >What you're really saying is that brushless motors have a higher power density than their brushed counterparts. I'm not even remotely, even close to saying that. Most traditional motors have a coil or magnet in the rotor. switched-reluctance motors use the force of reactance, pulling the metal into alignment with the magnetic force field lines, to drive the motor. Totally different thing. When people say "servo motor"as a general rule they have been around a long time and people are generally not referring that as a switched-reluctance motors. >Making your own motors Why bother? I'm not going to talk about this any more. You folks buy all the $100, or more, motors you want as that is what it will cost you to get a real strong force. Lifting a bottle of milk will not get it for me. Maybe for you that's good enough.

>>35319 Grommet, you are being nonsensical. Your requirements are contradictory. >Doesn't look like a machine. This is a feat that no one has ever achieved. You are unlikely to be the first. Whereas I share in this ambition, the best you can get is equivalent to a theme park anamatronic. >Lifting you Assuming your mass is around ~100 kilo because you're a man. That is going to require significant power to lift yourself. Especially if you want her to lift you with her arms, that adds leverage, which multiplies the force needed from your actuator. She'll need to be very large to be able to do that. See picrel for a good idea of the minimum size you should expect. >Cheap Cheap would require her to be tiny. Consider material costs, power systems, reducers, etc... I say this not to discourage you. Rather, I want you to think critically and start doing the math. I see potential in you. But, you need to understand that you're working on something that's difficult and will be expensive. If you want to make something cheap, you need to start small and simple. >>35322 >Servos What matters is that there's some feedback to govern the behaviour of the motor. No, a PWM controller regulating a motor alone is not enough. For the same reason using a resistor isn't enough. Those systems lack sensors for feedback. Brushless motor controllers use the coils as inductive sensors (or other external sensors, like hall effect sensors) to determine positioning of the magnets to govern how the coils are energized. The higher end models use various other sensors >Reluctance motors They are a type of brushless motor. It's still magnetic fields interacting with each other. One is from permanent magnets, the other is from permeability. >Why bother? I do not understand how you think you can magically create better motors than established companies for less. They are complex machines requiring exacting tolerances.

Great introduction to how electric rotary motors work. http://solarbotics.net/starting/200111_dcmotor/200111_dcmotor.html How to make dirt cheap motor drivers along with great information on how they work. https://www.youtube.com/watch?v=xW2Nwg_RX84

>>35335 >http://solarbotics.net/starting/200111_dcmotor/200111_dcmotor.html Neat! Feels like a blast from the past museum tour. Great info Kiwi, thanks. :^) > Chobitsu's idol; the final, great, Natural Philosopher: Michael Faraday Exceptionally based! :D <---> >https://www.youtube.com/watch?v=xW2Nwg_RX84 Man this guy is a great technician & competent engineer + a very good teacher + a good video technician/producer. His cost-consciousness is a must-have for DIY'rs, too. Great combo! I hope he'll take the robowaifu-pill & find /robowaifu/ some day! :^) >=== -add'l cmnt -minor edit

>>35319 >Grommet, you are being nonsensical. >Your requirements are contradictory. They are not. I can prove it. What power to weight ratio is a tesla 3 motor in their cars. Tesla 3 motor output = 258KW(345HP) at 27.99Kg(60lbs.) 4300 watts/pound What is the total power in humans 200 lbs, 250 W(rough numbers) So the tesla motor puts out 4300 watts/pound and human puts out(at 250watts and 200lbs.), (It's actually much less) 1.25 watts/pound https://en.wikipedia.org/wiki/Human_power "...Normal human metabolism produces heat at a basal metabolic rate of around 80 watts..." (Note: Heat not work.) "...Over an 8-hour work shift, an average, healthy, well-fed and motivated manual laborer may sustain an output of around 75 watts of power...." "...During a bicycle race, an elite cyclist can produce close to 400 watts of mechanical power over an hour and in short bursts over double that—1000 to 1100 watts.... An adult of good fitness is more likely to average between 50 and 150 watts for an hour of vigorous exercise. Athlete human performance peak power, but only for seconds, 2,000 Watts..." For reference a good horse working at a good constant rate works at 746 Watts. Now look a these numbers and then look at the power to weight ratio of tesla's motor. Now maybe my actuators will not be as good as his but the far, far, far, far higher performance of these type motors means I can have something that operated at 5% of teslas motor and still beat the pants off of a normal human. The numbers are what they are. If you dispute them, well do so but show the math. Show how your numbers are different. > >Doesn't look like a machine I'll be more specific(because apparently this is necessary). I don't want to have for instance a leg with 3 motors. I want something that has all, or a good portion, of the muscles arranged like a humans. They contract just like human muscles. Now it will yes still look like a machine, but the "form" and "function" is not like a machine. It's humanoid. I want to note you are being "pedantic". It should go without saying, though apparently not for some, that I am not going to make a machine that can not be distinguished from a human. Same with "classification" of motors. When I say a switched reluctance it is not generally not thought of as a servo. For a long time servos have generally been not of switched reluctance type. Though "technically" you are correct, people don't refer to switched reluctance motors as servos. When I say like a human it's obvious that I'm not saying perfectly human. You are being "pedantic" and nitpicking. People complain I write too much, well this is why. Excessive nit picking over definitions when in general someone should know some generalities of things and not have to have every single point pointed out to them. But if that's what you need I can do it. >Lifting you I covered that above. The numbers are there. >Cheap would require her to be tiny No cheap means you have picked the right materials for the job. I've talked about this, obsessively. Continuously. Over and over. Some general ideas are using low cost fibers like cellulose bonded with glues. Possibly adding high quality carbon fiber in small strips for stiffness. The USSR made cars out of cellulose and glue. They were damn near indestructible. Canvas is cheap. You can buy cellulose fiber for insulation house super cheap. Talked about glue fiber composites here, >>24149 Titebond III glue (very strong) is $30 a gallon or so. I recently found a material "Plastisol" that sells really cheap and has good properties. Would make a good skin/muscle covering/whatever. Talked about it here. >>35259 Talked about glues here >>26603 Talked about using glue and concrete for bones here, >>23296 (BTW people goofed on me for mentioning concrete for bones but concrete is SUPER strong, as long as you add tension members to compress it. They build concrete slabs for apartments by embedding steel cables covered in plastic. They let the concrete dry, then they tighten up the cables (squeezing in from the sides)to compress it. Saves a fortune in concrete. All of these are, fairly obsessively, about cutting material cost and, more important, cutting assembly cost.. Cutting assembly cost may be the most important of all. >I do not understand how you think you can magically create better motors than established companies for less. They do not make what I want at a reasonable cost in the form function I think best. BTW how do you think they come up with these. Some guy figures it out or a couple guys. That mine might not be as efficient is not a problem as the numbers above I showed attest. And what's so unbelievable about some wire and possibly iron oxide that we can't make this? >Reluctance motors >They are a type of brushless motor No one calls them this or if they do I never heard of it. It's too broad of a definition. "pedantic".

>>35335 That's a great driver video. I need to learn more about this. I know some of the basics but not the details.

I took some Magnetite F3O4 and tried to infiltrate it with super glue. Big fail. It beaded up. I have a few pounds of this stuff and while attracted to magnets while still in the bag it doesn't seem to be attracted enough for the kind of force needed. I was hoping if it was immobilized in the superglue plastic it would be better. ehhh...not so good as the glue refuses to penetrate readily and if pushed into it just makes a big mess.

>>35341 >Power to weight of Tesla 3 motor I’m not sure why you think you can scale motors power by mass linearly. That simply isn’t true. As motors scale in size, the densities of the permeable materials, magnets, and copper coils all play different roles in the resultant power output. It’s also important to understand that motors are designed to meet certain requirements. A self balancing scooter motor weighs 6.8 pounds and maxes out at ~500 watts. A drone motor can weigh 0.11 pounds and output 800 watts. One will happily move you all day. The other would melt if it tried to move you. Yes, the drone motor can move you, and at similar electrical efficiency. They would require ~10 pounds of gears and cooling to achieve it. My point being, motors are built for a purpose. You cannot extrapolate from a car motor to an arm motor. It’s just not the same. >Human power Irrelevant to us. We are building robots. We have math based on physics to determine our requirements. You cannot evade mathematics by trying to imitate a person. This website provides a simple overview for calculating needed torque. Do remember your actuator will need to exceed this minimum required torque. https://www.wikihow.com/Calculate-Torque I’m working with torque instead of watts because you could have a billion watts of power and not be able to move a person. You’d just have a wheel which spins at ludicrous speeds. >Being pedantic This is because I’m an engineer. I hold you, and everyone else here, to engineer standards. Proper names, proper definitions, and using actual physics are all important. >Wanting all of the muscles a person has This is even worse and harder than simply making her look human. There is not much space in a robot's frame if you match the dimensions of a person. You are working with motors, not muscles. They are fundamentally different. I wanted to do the same years ago, Chobitsu likely remembers my crazy ideas for making winches from drone motors to match human dexterity. This was a terrible idea. Draw your ideas. Find motors and servos that match your needs and place them in the body. It’ll help you better understand what I learned. >Cheap means right materials This is somewhat true. Albeit, what I believe to be cheap for a robot is ~500 USD. What you define as cheap is subjective. Fabric composite materials are some of my favourites. They are a great idea for your robot, though I’ll say that a thin steel rod for a core to your bones would be better than envisioned steel ropes. This is because your bones will be under loads which are different from a building. Also, steel ropes release heaps of energy when the concrete shatters, best to avoid that possibility. You’ll also need a good sealant but, I’m sure you have plans for that already. I prefer urethane sealants for concrete myself. >Motors that match my requirements don’t exist It’s best if you change your requirements. Frankly, you should start small and grow from there. You’re trying to make something on the edge of what’s physically possible. Build a robot arm. Build a walking robot. Build something real to help ground you. Robotics is harder than anyone thinks it is.

>>35344 I have no idea what hell you are on about. I point out that there are very high power to weight ratios in some engines, well you can't dispute that so you make up from whole cloth things I did not say, >>I’m not sure why you think you can scale motors power by mass linearly Show me where I said that. You can not. I said that the power to weight ratio is SO MUCH higher that even if we, or me, no one else seems to believe they are capable of building electric motors of any value, that even if it's not as good it's still way better than organic muscle. Look at the ratios. Tesla 4300 watts/pound humans 200 lbs, 250 W(rough numbers) 1.25 watts/pound This is well over 3 orders of magnitude. If you feel that you can not get anywhere close to building something that has 3 orders of magnitude leeway then I don't know what to tell you. You say you can't do this well I looked up the power in drone motors and at a random site it said, "...sport planes are about 100w per lb..." That's a lot of power and it's NOT a big motor. I personally have no idea how the size of an electric motor scales with size. Just what magic happens in large motors that can not happen in small ones? Bear in mind were talking about over 3440% or more power than humans.Where are these 3440% loses? I would like to know. https://www.rcdronegood.com/calculate-quadcopter-power-consumption/ And if you are going to buy off the sheaf "servos" to run everything then your ass will go broke trying to sell them to people because people are not going to pay that much. And if you build something that looks like C-3PO, no one is going to buy that shit. Or no one like me. >Human power >Irrelevant to us Sigh... So you think it's inappropriate that we are supposedly making humanoid robots that, supposedly, look and operate like humans but equating what power humans have to do the same things we want our robots to do...is silly and somehow not valid??? >I’m working with torque instead of watts because you could have a billion watts of power and not be able to move a person. quadruple max sigh... Here's a Watt (W) to Torque (Nm) Conversion Calculator for you. https://electrical4u.net/calculator/watt-w-to-torque-nm-conversion-calculator/ >Being pedantic That's what I thought but now you say >I’m an engineer so I guess you just do not know what you are talking about. Babbling on about torque. Sigh, like that's something different. Some magical thing. No I use watts because it's a simple to visualize and EASILY equates to newton-meters. So if you have some sort of lever arm you can change to watts, or the opposite way around, to figure out what sort of wire, voltage, current etc. It far simpler. It is using the idea of starting with the power needed to do things. >I’m an engineer I'm extremely skeptical that this is true. If it is true then...fuck me, engineering has gone down the drain. Things you say like >you could have a billion watts of power and not be able to move a person. You’d just have a wheel which spins at ludicrous speeds. Only if you are setting up some sort of system that is so radically stupid that no one in a billion years would do such a stupid thing, then yeah I guess on some planet of the "zippy the pin heads" it could happen, but no real engineer would ever do such a thing. And how you can not imagine how to make a small motor that's any better than 3,500% worse. Either you are trolling me or, much more serious, you have a really poor grasp of engineering. >Wanting all of the muscles a >person has >This is even worse and harder >than simply making her look >human. If I can find a way to do without so many actuators then I would but to me, to get what I want it seems the best path is to find a way to make a shit load of super cheap actuators. Go back on this thread and you see I have babbled on incessantly about that. Some of it wrong, but I do thing I'm making some progress. BTW I've been looking at high permeability powders that have low magnetization memory, hysteresis, and ran across Carbonyl iron. It's been used a long time and was used a lot in WWII. Permalloy is another good one. It's difficult to find suppliers that openly list cost to buy small amounts of this to try or any of the high permeability powders. They ALL appear to want you to send them all sorts of data. I did find some carbonyl iron powder on ebay, from Ukraine,AAARRGGGHHH!!! I REALLY hate to sign up for a huge mass of spam that they will send. I may just call them on the phone. I have some iron oxide, but as I said I placed a bag of loose powder next to a fairly strong magnet. The magnet has 40 lbs. force on thick steel(ideal conditions). I was not impressed with the force. It could be that I need the particles immobilized in a resin or matrix and have them magnetically oriented in the resin to get some significant force. Or could be my iron powder sucks. I do know there are different kinds and supposedly what I have, or what they say they sold me, supposedly has good permeability. A big deal I believe, is to get something with really high permeability and that will lead to good things. And Kiwi I want no more of your advice. It is of no use to me.

>>35371 >Refusing advice Very well, I accept your dismissal. I hope you manage to benefit our community eventually.

>>35344 > Robotics is harder than anyone thinks it is. This. But I'm weird so most of my robotic problems have been aesthetics. Comparing my model 2022 workshop waifu to the 2024 robowaifu I can expect some serious developments in the next two years lol. SPUD's elbows have 70kg/cm servos in them so I really should get around to testing how strong she really is. Only way I was able to squeeze the ASMC04s (rated for 180kg/cm) into her chonky Alita-esque shoulders was making a motorcap to cover the protruding 550 motor. Looks delightfully robot-y. I'd love to use some ASMC04s in the elbow but I don't think I could squeeze 'em in without SPUD looking silly. I do have an idea for bipedal leg development but that is in future (I fiddled with it a bit and got better ideas, just no motors to do it). >>35371 >Looking human & stronk Well, a waifu is primarily an emotional connection to a fictional character, and I figure a robowaifu would be an avatar upon which one can express affection to the waifu: primarily a robowaifu should be capable of expressing romantic affection eg a robot capable of greeting you when you've been gone for a while with a cute little dance and say "welcome home, I missed you". Most ronery men (and a few wommins) just want an uwu plastic pal who is fun to be with. If you want a heavy-duty robowaifu with lots of lifting power you might want to consider the mecha musume aesthetic: build a heavy duty robot with a waifu doll-thingy "piloting" it. Big heavy arms to carry you like a princess and little waifu arms for giving you... I dunno... headpats? I'm not sure of the extent of your knowledge base so excuse me if I say something you already know. I think playing with some motors (and if you haven't done so already play with them some more, its fun!) would help you discover something very vital about robotics: mechanical disadvantage is a bitch. (Is mechanical disadvantage the word? I dunno, I only had 1 semester of engineering so 99% of my robotics/engineering knowledge is just from building stuff without the fancy terms and labels but experiencing the concepts firsthand) Lets say you have a 180kg servo... usually, that measurement 180kg of torque per cm. So a 100cm arm will really only be able to lift 1.8kg max at the hand, and that 1.8 kg will have to include the mass of the arm. Also bear in mind that most motors and actuators require a constant supply electricity to maintain position. The best way to figure stuff out is to do it. Buy a drone motor and try it out. Worst thing that happens is that you're right and make us look like dumb schmucks :D (Edited to carry the zero.)

>>35386 >Lets say you have a 180kg servo... usually, that measurement 180kg of torque per cm. So a 100cm arm will really only be able to lift 18kg max at the hand Umm, No. 180 kg/cm divided by 100cm = 1.8 kg. If the arm weighs more than 1.8kg, depending on where the center of mass is, the servo won't be able to lift just the arm. This is why putting servos right at the bend of a joint is almost always a bad idea for structures of this size. If you want to use that servo and get real use out of it, you need to change the installation so you can increase the torque being applied. See my example >>35303. I left out the part about also calculating the torque created by the weight of the arm to keep things simple. If you relocate the servo to the chest area behind the breast, you could use sprockets and chains to increase torque to rotate the arm at the shoulder. You could also use the sliding mount in the chest to recover speed. This relocation would mean the servo is also no-longer having to lift itself (we don't lift our shoulder muscles when we raise our arms) and you won't need the motor caps on spuds arms. You might want to consider also mounting the servos for upper arm rotation and lifting the arms straight out to the sides into the chest and using cables in sheaths for the movement. If you arrange the cable paths to emulate the human shoulder muscle arrangement, you could combine all three shoulder axis (vector math) for many motions, increasing the torque being applied.

>>35371 >Why do you think you scale motor power by mass lnearly? >Show me where I said that. You can not. Yes, I can, easily. "Tesla 4300 watts/pound humans 200 lbs, 250 W(rough numbers) 1.25 watts/pound" Assuming that power scales as watts/pound is exactly that. Once again, you cannot intuit the power of smaller or larger motors using another motor. Motors just do not scale by mass. >Humanoid robots aren't comparable to humans This is important for us all to understand. We aren't robots, and robots aren't us. This is for everyone, don't assume that a robot will work like you do. Design around the fact that you are building a robot. Research robots that are similar to what you want to achieve. Then, incorporate what you learn into your design. Look for motors and servos that match your requirements. I've personally sunk countless hours into understanding how robots from Boston Dynamics, LimX Dynamics, Tesla, and many more. It's worth the time and effort to learn from what exists. We all have room to grow and learn. >Ignoring torque for watts You cannot bypass physics. When you want to move, you need to understand the torques involved in your kinematics. If I want my robot to lift her arm with a Fosters, I need to do the math. Let's use my MaidCom model picking up the beer from my counter. Her largest torque requirement is going to be her shoulder, which is 35 cm away from her gripper. The Fosters has a mass of 350 grams. So, with a 100 gram hand, and 600 gram arm, 350g+100g+(600g/2)*35cm = 26.25kgf/cm (This is grossly simplified but, good enough to get a ballpark estimate). In other words, after accounting for a 20% safety factor, she needs at least a 35kg servo. I would give her a far stronger servo to accomplish the task because reality is never as forgiving as maths ideal world.

>>35389 >If you relocate the servo to the chest area behind the breast, you could use sprockets and chains to increase torque to rotate the arm at the shoulder. I like the way you think, Anon. >You could also use the sliding mount in the chest to recover speed. Care to spell this out more clearly for the uninitiate?

>>35392 >This is important for us all to understand. We aren't robots, and robots aren't us. This is for everyone, don't assume that a robot will work like you do. This. >We all have room to grow and learn. Also, this. Very this. >Fosters Green can or Blue? :D

>>35392 pretty much, just add the acceleration of gravity (kg x 10 for kN) since you want a force

>>35303 >>35389 Before I respond, could you please draw out your system? Frankly your description of how it looks and functions confuse me. >>35386 You continue to be one of my favourite posters. Keep doing what you do. Just remember your decimal points. I can edit them in if you'd like. (Unless you specifically declare you want the error to remain, I'm going to fix it. It really bothers me.) >>35394 >Green or blue? Buy heaps of blue's when sales hit :)

>>35389 Oops sorry I forgot to carry a zero lmao.

>>35303 >>35393 >You could also use the sliding mount in the chest to recover speed. >Care to spell this out more clearly for the uninitiate? This comes from old-school model airplane servo installations, before the days of mixing channels or computers in transmitters. Back in the day, if we wanted to have a non-standard control surface arrangement, elevons, tailerons, flaperons, ruddervators, etc., we did it by having two servos mounted on a sliding tray. For example, ruddervators in a v-tail. One servo would have a double-arm with drive-rods going to the actual control surfaces, one to each. The other servo would move the tray. The double arm servo acting alone provided rudder function, moving the surfaces in opposite directions, and the tray-moving servo provided elevator function, moving the surfaces in the same direction. Acting together the two functions were mixed. Sometimes a single servo would perform the operation and slide the tray it was mounted on at the same time. This might be necessary when operating landing gear that traveled fore-and-aft, shifting the center of gravity as big wheels swung up or down. We could have the servo move in the opposite direction as a balance weight. In this case we might want the servo to move itself at the same time as the motion it's causing to augment (or retard-unlikely) that motion. Depending on the linkage, or power train- in this case a chain drive- we need to have a system for adjusting the length of the chain as the servo moves, a tensioner, or possibly a double tensioner. We would want the chain to be "pinched" to resemble a figure eight or infinity symbol when the servo and driven sprockets are closest together, and a simple oval when they are farthest apart. It's basically bicycle and scooter parts applied to 60 year old tech. Most likely the servo will be hacked for continuous rotation, and the feedback pot mounted in the robot's shoulder joint. The system for moving the servo could be another chain and sprocket, or a control arm, it depends on your design. The installation would depend on the available space, so you might not be able to add simple brakes or clutches to maximize the performance of the chain for direction of travel, if you even wanted to. KISS. btw, that chain and sprocket system is still used in real aircraft, so it's very reliable. >>26046 Search the web for online copies of old model aircraft magazines and articles about servo installations, or plans. There was a time when designers had one servo with multiple control rods open the landing gear doors, raise the gear, and then close the doors on a scale P-38 Lightning. There are probably lots of forgotten gems we could use in that old tech. >>35396 Kiwi, is this enough? Right now I'm up to my eyeballs in eyeballs: https://www.thingiverse.com/search?q=inmoov+eyes&page=1&sort=newest

>>35396 >Just remember your decimal points. Yeah whoops! lol. Just make an edit in parenthesis stating I forgot to carry a zero. My point still stands tho: mechanical disadvantage sucks >:V >>35389 >servos, sprocket chains and bowden cables Quit the jaw flapping and build it, man! :)

>>35405 I get it.

>>35405 >>35406 And I made a mistake, corrected it, saw a typo in my post and fixed it, but reposted the mistake. >There was a time when designers had one servo with multiple control rods open the landing gear doors, raise the gear, and then close the doors on a scale P-38 Lightning. Should have been: There was a time when designers had one servo with multiple control rods unlock and raise the gear, and then close the doors, or reverse the process on a scale P-38 Lightning, with it's separate pod and booms. On the P-38 One of my favorite planes. How could I have forgotten? the doors stay open when the gear is down.

>>35389 The math ain't mathing. In order to get mechanical advantage you need r2 to be greater than r1, however you sacrifice speed/total movement for greater torque.

>>35410 Now extend your drawing to the left. Add a "cable" from the bottom of the servo wheel to an anchor point to the left. Assume the servo is not fixed in position. What happens when the servo wheel rotates counterclockwise? This illustrates the theory. NOTE- NOT "bowden cable", but steel, ie; bicycle cable and steel sheath.

>>35411 so this thing basically

>>35410 >>35411 >>35412 No. We are increasing the torque on the arm by increasing r2. This slows the arm movement. However, our servo is not fixed, but arranged to pull itself away from the shoulder at the same time. It is reeling in cable from opposite sides, reducing the length of the cable system twice as fast (if the cable pulleys are the same size). Since the shoulder is the "free" end of the system, the arm moves faster.

>>35411 Bicycle cables are called bowden cables :D What is the point of adding a second cable that does the opposite of the first? It doesn't multiply your forces or anything. >>35412 Even if one were to use a block and tackle the more pulleys you add the more you need to pull the cable for the load to traverse the same distance.

>>35414 so you want to reduce the load by increasing the distance but reduce the distance and increase the load? you cant have your cake and eat it

>>35415 >Even if one were to use a block and tackle the more pulleys you add the more you need to pull the cable for the load to traverse the same distance. So close.... If it was a snake, it would have bit him. Weekley live stream starting so goodnight all.

>>35416 The load is staying the same, the torque is changing because the load is being applied to the arm somewhere else. I'm out.

>>35389 >180 kg/cm divided by 100cm = 1.8 kg Yes. My thinking for the lowest cost, cheapest method so far is you can have a fast lower power motor and use cables that wind up on a pulley, as noted, or you can have an electric actuator with a really long set of smaller forces all acting at once like a Switched reluctance linear motor. https://en.wikipedia.org/wiki/Switched_reluctance_linear_motor "If" the SWLM can be made economically it would seem to me to be the best choice. It's simpler. No big mass of cables, pulleys etc. to route and deal with. I also keep coming back to the notion of scissor jacks. I have not figured this out. I have some ideas, which will likely not work, but the basic idea is a take on an electrical scissor jack type mechanism but all electric. The big idea of this is to take less powerful forces, line up a huge mass of them and use tiny leverage to move large forces. Speed would be reasonable for our task because it's electric. I don't like hydraulics but...I think that you could do them way cheaper if you count the labor to assemble. If you used large cast bladders it would beat everything else on cost and rapidity of assembly. Power output, would not be to difficult to get fairly large forces. I have an idea for valves, which is the cost problem in normal hydraulics. Imagine an oval feed line to a bladder type actuator. On the feed line is a clamp held down by a spring. The clamp is held down but could have either a small solenoid or better, another small control bladder actuator that would take off the spring pressure, opening the valve, which would then feed fluid to the bladder actuator. Since the force is the pressure times the area. The same pressure could be used for the valve opening bladder, but it would have a large area compared to the small area that squishes the line closed. The opening function could be a small bladder that pushes a pivoting arm that releases the spring pressure, leading to fluid filling your main actuator bladder. The small "control" bladder could be actuated by a very small tube since not much fluid would be need to fill it and take pressure off the main actuator bladder valve control spring. Yeah I know hard to garble this out. The point being you have one pressure so that minimizes cost. The originating control could be very small electromagnetic solenoids, simplifying the electronics. So the solenoid opens a smaller feed line to control the bladder that opens the big flow line to the main muscle actuator bladder. Turn one on to turn another, etc. until you have a large flow into the main actuator bladder. The other, smaller, actuating bladders have very little flow so need less fluid, so are faster to act. All of your muscles, no matter how, many could be fed with one large supply line. The supply line would be connected to a bladder tank that is filled and pressurized by an electric motor. I could see you being able to have a large mold. Hang some sort of reinforcement net in it (nylon?polypropylene?)then fill with some sort of flexible resin, cure and then you have the large majority of all actuators in one big ass pour. Hard to beat that for time. Certainly soldiering thousands of connectors would take more time. But...I hate hydraulics. I have a picture, a little off, slightly different actuation and not drawn very well in this comment, >>28447

>>35392 >>Why do you think you scale motor power by mass linearly? >Show me where I said that. >You can not. >Yes, I can, easily I don't want to sound like a dick but your fucked up, you talk like a fag and your shits all retarded. One is electric the other is human. I only used these numbers to show the difference in power to weight ratio. It says nothing about scaling. Nothing at all. Stop saying stupid shit that means nothing and is wrong. The more you say the sillier you look. 1. You confuse the power to weight ratio of electric motors and humans. Declare that this mean scaling. No it's a comparison of force. Not scaling. Scaling would be if I said that the power of an electric motor would be the same "per weight" in a smaller size. I did not say that. You are ether trolling me or have extremely poor comprehensions. Like really, really, really bad. 2. >Humanoid robots aren't comparable to humans This is important for us all to understand. We aren't robots, and robots aren't us Don't lecture me when you are so brain dead that you "can not comprehend" how to make an electric motor that is anywhere near 3 orders of magnitude in power to a Tesla motor. I might as well spell it because you may actually not know what this means. It's a 1,000 times difference. You can't possibly be so stupid as to believe that I do not know robots are not humans, or are you??? And since you should know this, why are telling me? You just look foolish and trollish. If that's is what you want people to think about you, fine but you're doing it to yourself. >Ignoring torque for watts >You cannot bypass physics You have no idea how foolish you sound. I suggest you use the torque-watts calculator I linked. I got a big laugh out of it. I got a flash back from some guy arguing about vaccines with me. You sound like him telling me,"vaccines are science", totally missing the point.

>>35420 >The more you say the sillier you look. Lol. I'm all for the le ebin trole on the occasion, Grommet. But please remember we're all on the same team here. The more ad hominems you use, the more like an Eglin AFB jew you look. <---> IMO Kiwi (and the others here, including yourself) are all acting on good faith. Debate the numbers all you want, but please restrain yourselves from (in effect) degenerating into calling each other niggers, heh. Cheers! :^) >=== -minor edit

>>35421 >IMO Kiwi (and the others here, including yourself) are all acting on good faith Is he??? Notice I explained myself, yet he would not leave well enough alone and kept lecturing me, making up things I did not say all the while getting the whole thing wrong and pretending he was talking down to me. I'm only dismissive and short with people that are that way to me. And Jews act the way they do to demean people who didn't ask for it. I only demean people who do so to me and will not let it well enough go. I asked him to stop giving me advice. I don't need his advice. It's wrong and he doesn't understand what I'm talking about and refuses to even entertain the notion that he could be wrong. "If" he would have asked me questions, like real ones, to clarify what I was talking about I would have readily done so but I have no time for dismissive people who do not know what they are talking about. The way he argues his point also really annoys me. It's Womanish. Passive aggressive. Like, kiwi,"I'm so offended, I was only trying to help". Please, don't piss on my leg and tell me it;s raining. (said in a whiny voice) >oh you need to learn physics >robots aren't humans > maybe some day you can be of use Give me a damn break. Fuck that, I'm a Man and my response is going to straight up tell him he is acting foolish.

A good idea of some of the basics I'm striving for is this part I saw in a video looking at solenoids. This part 874F Push/Pull Solenoid Has 2,000 Newtons of force (449.61 lb. force). It has a small stroke (14mm) but can do so at 20 to 40 ms duty cycle. Now they of course are not doing this continuously but it does show what sort of forces "can" be had. So if you are able to spread these out a little, for heat, and have a lot of small ones the forces add up very fast. And because some people are so picky, I'm not saying "use this solenoid, I'm only using it as a benchmark as to what might be achieved. In real life what we would have would be far less but using many small magnetic attraction elements that would not matter. https://www.geeplus.com/push-pull-solenoids/

>>35426 >And Jews act the way they do to demean people who didn't ask for it. I only demean people who do so to me and will not let it well enough go Fair enough. And please forgive me for the misunderstanding, Grommet. I'm not in any way implying you are one. No one so generous-hearted & giving of yourself towards our common goals here could be presumed to be so by any right-thinking man. One kind gives freely, and the other kind only takes. But rather simply that such behaviour looks similar on the surface. However, you have my full trust, Anon. :^) <---> Personal differences aside please, I'm basically asking everyone here to extend much grace towards one another, and to walk humbly with one another. Christ gave such an admonition to all of us (the so-called "Golden Rule", cf. >>3 ). If each of us fails to do so this year, then the laughter of our common enemies will be our only reward. Let it not be so! Cheers, Anons. :^) Onward! >=== -fmt, minor edit

>>35418 Just because the servo is further away from the joint doesn't mean the force decreases, energy cannot be lost from a system. If the servo reels in a cable from the shoulder and moves itself the sum of the forces on the servo still add up to the original load.

>>35449 >Just because the servo is further away from the joint doesn't mean the force decreases, energy cannot be lost from a system. If the servo reels in a cable from the shoulder and moves itself the sum of the forces on the servo still add up to the original load. That's what I said I just used fewer words. : >>35418 >"The load is staying the same" Did you mean this to reply to RT at >>35416 ?

>>35405 I found a video showing a form of the 'sliding servo mixer' that should make the concept clear. It may be a useful idea for some anon's robot. Maybe in the neck or spine, feet or toes? YMMV https://www.youtube.com/watch?v=h6GrhfN52lQ

>>35454 so its about combining x,y motion i think he already did this with the head

>>35405 >>35454 Got it! Thanks Robophiliac. I love elegant solutions tbh, whatever the domain. Cheers. :^)

The whole bldc vs hobby servo conversation is somewhat dumb or maybe rather misguided, in my opinion. You're not going to be able to afford all off the shelf bldc motors in every joint. Even if you can afford it, not everyone can. Not only that, using all bldc motors is going to require much more custom power electronics (3 phase sinusoidal) and still requires some type of gearing, direct drive bldc is difficult because you need lots of power per motor (adding lots of battery weight) and they are very costly. Most will have to use off the shelf bldc sparingly where it makes sense and use hobby servos everywhere else you can. It's more of a using the right tool for the job argument. Also, hobby servos are noisy because of the high-speed rotation of their small internal motor and the gears meshing within the gearbox. High speed motor + friction + impacts between the gear teeth == audible noise. Ultra high torque brushless servo exist, but are just as expensive as a bldc motor and still have friction + gearing impact noise. My point of view is to create custom servos and iron-less axial flux motors for everywhere you can. More work yes, but significantly cheaper than good hobby servos or off the shelf bldc, and far better because you can tune the motor parameters, motor housing shape, and mounting surface to your specific robowaifu design. I personally think this is a much better use of time than trying to shoehorn in existing servos/bldc into a robowaifu design. Using existing servos/bldc seems to cause an extreme 'ulgy-ification' of robowaifu and more generally robots. Look at robo-one stuff to see what I mean in a general sense. These are designed with cost and function over form. A true robowaifu imo, should be prioritizing form, cost, and function in that order. Function's last because robowaifus should be modular and you can always modify your robowaifu for more function over time. Axial Flux > Radial Flux. Don't bother with radial motors for robowaifu if you can. See https://www.emworks.com/blog/electromechanical/axial-and-radial-flux-permanent-magnet-machines-what-is-the-difference for more information on why axial is the best choice. Gearing stages with either custom servos or bldc should be targeting toothed belts to drastically reduce noise over meshed gears. Maybe use planetary or harmonic, or cycloidal gears if you have spacing requirements. A bldc with encoder and gearing is just another phrase for a high torque brushless servo btw. I have developed a shoulder/hip motor that only costs 8.50 usd dollars per motor, cost varies based on filament, bearings, magnets, and hardware used. All you need are a bearing, few stainless nuts/screws, 3d prints, magnets, and double coated enamel wire. This is ignoring the encoder stage but that can be added on later in the gearing or on the motor itself. Also, once you have a well designed motor/servo, you can get the 3d printed parts machined out of metal which generally still works out to be cheaper than buying off the shelf stuff. I got quoted for 300 usd on a set of 8 machined motor's parts, it gets even cheaper if you go the cnc router/cnc mill/lathe route and has an added benefit of being able to sell motors to others or quickly recoup a lost motor due to negligence/accident. Quality bldcs are outrageously expensive, really good ones can cost 200 usd or more a piece, a few of them are in the thousands. If you take the cost of 8 of my motors with machined parts, it works out to be only slightly less then 400 usd. This is a drastic cost reduction. That's also why i went the cnc router/mill/lathe route. If you don't know how to make your own coreless axial flux bldc motor, refer to https://cadenkraft.com/designing-a-coreless-axial-flux-motor-part-1/ for a complete walkthrough. Hope this helps.

>>35498 >See https://www.emworks.com/blog/electromechanical/axial-and-radial-flux-permanent-magnet-machines-what-is-the-difference for more information on why axial is the best choice. Thanks for the information, Anon! I'm glad there's a wide variety of somewhat differing opinions of robowaifu engineering topics going on here. This type of honest & healthy debate can only lead to a better outcome in the end for everyone involved here. <---> Please keep the good ideas coming, Anons! Cheers. :^)

Do you mean just using this motor? Or, do you have your own design? Need to know more before I can respond with anything of substance on your custom motor. When it comes down to the BLDC vs hobby servo argument, I don't think there is one. Both have their purpose as you noted. As for axial vs radial flux, they both have merits and demerits. It depends on use case. Radial flux are more popular because they are cheaper to mass produce. This is mostly due to material properties of the rotor. rotors go through uneven loading. A cylinder is more mechanically resistant to that uneven loading causing problems. Axial flux rotors are flat disks, they form waves more easily, meaning the material must have far greater rigidity to reach the same power output safely. Their high torque to volume/material at larger scales does lend them an advantage in mobility platforms. This is why when you look for motors, you'll mostly come across radial flux motors. I personally like them both equally, honestly wish that they were easier to mass produce so they would be easier to get at smaller scales.

Great introduction to the physics of counterbalancing. https://www.youtube.com/watch?v=OdmpYBSMWzM

>>35498 That certainly is a neat way to get your own motors. Definitely turn that speed down into torque tho. Then you just need to steal a control board off an ASMB-04b and you could control it just like a servo (not sure how much voltage it needs but the ASMB-04b does up to 24V). You can see the control board has separate connectors for the motor, motor power, and signal/board power. But being 3d printed it is probably going to be quite noisy. I might try building one eventually, but I'm focusing on just getting SPUD built fast n cheap as possible (without going mad, of course), and I think I'm doing pretty well at minimizing the uglification :) >>35454 Okay, now the concept is clear. Sorry about that :) Inmoov actually uses something kinda similar for the neck -which I modified for SPUD at one point, see vid.

>>35504 >This type of honest & healthy debate can only lead to a better outcome in the end for everyone involved here. Indeed! >>35549 >Do you mean just using this motor? Or, do you have your own design? I've designed my own. Though his is pretty good. I have designed three different sized axial motors (shoulder/hip/waist motor, knee/elbow motor, wrist/ankle motor), and two 'micro-ish' radial motors for neck and spine. The two radials are being redesigned as space is a premium and i can layout/fit some stuff better if they were larger diameter with thinner thickness and axial is well suited. I provided a link to that tutorial because it's well done. Largely my argument for using axial is because they can be easily diy'd at home since you can build them as air core, you can wind coils yourself easily with spools, you don't need costly arc magnets and you can pick and chose the kind of magnet you want, circular or bar. The axial flux efficiency/torque is largely just a bonus. The weight savings with axial is not insignificant either. >A cylinder is more mechanically resistant to that uneven loading causing problems. Axial flux rotors are flat disks, they form waves more easily, meaning the material must have far greater rigidity to reach the same power output safely. It largely depends on the construction, materials obviously, but the number of bearings and their placement, the coil/pole numbers, and your capabilities on making the airgap as small as possible. It's possible to remove a large amount of uneven loading from axial, it just requires some forethought. Like i said it's also better to get the parts machined out of some metal and use the 3dprints as a prototype. You can also stack coils and linked-stators in rotationally offset layers to further reduce uneven coil loading. I helped design a stacked and rotated layered axial pcb motor for school which uses large number of layers and it has low uneven loading and it has no bearings at all, just nylon spacers. There's a published paper on it, im sure you all can find it easily, I don't want to link it because I'd be doxing myself. >honestly wish that they were easier to mass produce so they would be easier to get at smaller scales. Personally after creating two different sized 'micro-ish' radials, using http://jystator.com/ I'd just stick with nema size motors if you can fit/afford them. It would be great if it radial were constructible at home, but it's not really feasible for most if anyone. Nema 8/11/17 are quite small and they can be gotten from China extremely cheaply for what they are. Sadly radial motors are quite heavy, for the same output you can see a 2 to 4 times weight reduction with axial, and you can use the weight savings for a better construction or live with uneven loading. The weight savings are extremely beneficial. Lower weight means less rotational mass in limbs and lower weight overall meaning smaller battery capacity can be used or exploit the savings for longer battery life. There is a reason most commercial humanoid robots today are using axial. >>35552 >Definitely turn that speed down into torque tho. Yeah toothed belt gearing on axial is the best way imo. my axials have the toothed belt gearing directly embedded on the outside of the rotor and the housing has spots for belt gears for torque stacking. >Then you just need to steal a control board off an ASMB-04b and you could control it just like a servo Interesting, that's a good idea. Does it support 3 phase? All my axial motors are 3 phase y configuration. i like doing the electronics from scratch as I can select components that will match the motor, I can select higher quality chips/components with the savings from diy-ing the motor, and I can move the high heat components to specific places for better cooling. plus a motor controller is pretty easy to make with a pico/pico2 with pio and the extra components i noted before. But i know not everyone wants to make everything from scratch. Personally everything from scratch is the fun part for me.

>>35551 THANKS! Achieving reasonably-pleasing humanoid gaits (with our feasible, available, hips/spine design choices available rn) will certainly need a number of these counterbalance design approaches (notably, for the central battery-mass located proximal to the robowaifu's pelvis/belly center-mass volume) to enable the nice, arc-based motions+balances needed. Great find, Kiwi. Cheers. :^) >=== -minor edit

>>35573 Walking gets exponentially harder when the robot gets bigger, because of things like the square-cube law. It's easy for the Femsapien (the toy robot on OP's post), but for any human sized robot, it's going to be difficult. Note that I'm not saying impossible.

>>35570 >Does it support 3 phase? I thought there was something wonky with that DIY motor, unfortunately the board only supports bog standard DC motors. I'm not sure about the official terminology (I am orkpilled after all) but it seems bog standard dc motors are technically 3 phase motors, they got a funny little dingus called a "communicator" which takes the standard dc electricity via brushes that bush against multiple metal contacts along the shaft so it mechanically handles sending current to coils 120 degrees out of synch as opposed to using an external controller. Or something like that. Me ork. Gib motah elecktrisity it zoggin' zoomz. Simpul az. Amusingly, brushes are typically the first things to go on motors /generators, the most notable I've experienced being a car's alternator (at approx 120-140k miles). Replacing them aka "rebuilding" an alternator can cost between $5-$35 (turns out a pontiac 6 shares the same alternator brushes as certain models of Hummer!), as opposed to buying a new one at $100+. A little DIY can save you lots of money and eliminating brushes can prolong the life of your robot.

>>35570 >Axial flux actually makes sense from a DIY perspective I’m in manufacturing, so my biased understanding can obscure some potentiality. I’ll take your word for it that axial flux is generally a superior design for mostly 3D printed motors. I’d like to know more. Read a few papers but, overall I still feel as if a definitive design that is easily reproducible remains elusive. Partially because coil winding requires special jigs to maintain proper tension and alignment. I’d like for you to create a dedicated thread on home fabrication of axial flux motors. You seem uniquely qualified for such a task and I’d like to work with you on the subject in a thread where it’ll be easy for others to find and follow along. >Nylon spacers as bearings You could also use glass or ceramic spheres/beads to make thrust bearings. Works great with PETG. I’d run a track near the center and near the outer extremes of the coils. 3D printing small indents to place them in would be easy and provide great support. Perhaps the nylon spacers are better, I’d say both methods are worth comparison. >Shoulder and hip motors You speak as if these are direct drive implementations. Care to elaborate their dimensions/volume, mass, torque speed curves, and cost/estimated time to fabricate? Deeply curious about the potential. >>35573 >Humanoid gates Walking like a person is sadly far away still. Still, Good to have eyes on the prize. >>35576 >Easy for Femisapien to walk That same walking gate can easily be scaled up well beyond human sizes. It’s just a simple mass shifting statically stable design. She moves some mass to adjust center of gravity from her midpoint to being over a foot. The design has some aspect that allows the rest of her to tilt, this can be done with slight slants or springs in her feet, springs or loose joints in her hips, or a more complex internal mechanism, which causes her other foot to be off the ground, enabling a step. Femisapien uses springs in her feet and hips, so when she twists her trunk with her arms splaying in and out to adjust her center of gravity, her legs naturally tilt due to gravity, with one foot hovering a few mm off the ground. Her thighs then move her lower legs back and forth, causing her to step. What makes Femisapien especially clever is that she distributes her mass via twisting her trunk and arms which puts less load on her motor relative to leaning. When scaled up to human scale, I’d recommend having an eccentric mass, (perhaps her breasts?) to cause her to move her center of gravity. This would reduce how much mass the arms would need to move. Though, the issue would then be giving her gigantic feet to maintain stability. Something like Nana in picrel. I’m a tad obsessed with mass shifting mechanisms, gravity is the most reliable power source. The most reliable actuator is the one that’s not there. No one has ever had a problem with Femisapiens hip and ankle tilt actuators.

>>35584 >Though, the issue would then be giving her gigantic feet to maintain stability. Something like Nana in picrel. That's where the square-cube law will strike HARD. Imagine walking in lead-filled snowshoes.

This thread is really paying off for it's general conceptual approach rn. >>35549 >honestly wish that they were easier to mass produce so they would be easier to get at smaller scales. This. >>35552 Good luck with your neck actuation assembly, Mechnomancer. It's looking good so far. >but I'm focusing on just getting SPUD built fast n cheap as possible (without going mad, of course) Lol. >and I think I'm doing pretty well at minimizing the uglification :) She's charming! One day you'll look back and enjoy the 'family album/growth chart' of dear SPUD and see just how far you've actually come with her, Anon! Keep moving forward. >>35570 >The two radials are being redesigned as space is a premium and i can layout/fit some stuff better if they were larger diameter with thinner thickness and axial is well suited. Exactly so, and this set of priorities is in play for several actuation sites around the robowaifu. Her hips, for example. >The weight savings with axial is not insignificant either. Low mass is arguably our single-most important design philosophy bullet point at this phase of our history (cf. >>4313, et al). Thanks for all the great information in your post, Anon. >>35576 >Walking gets exponentially harder when the robot gets bigger, because of things like the square-cube law. The basic reason we need to focus on the square-cube law is generally because of the (usually) concommitant increase in mass, as the volume cubes (see link above why this is a critical issue rn). If we use every trick in the book to keep our robowaifu's mass down, then it's simply a matter of the increased mechanical forces required for the longer limb 'levers'. A more-delicate balancing act, if you will. :^) >>35583 >(I am orkpilled after all) Lolwut? :D >>35584 >I’d like for you to create a dedicated thread on home fabrication of axial flux motors. You seem uniquely qualified for such a task and I’d like to work with you on the subject in a thread where it’ll be easy for others to find and follow along. This. All in favor of this idea, personally. >Still, Good to have eyes on the prize. Yup. IIRC, you yourself were the one who posted images here on the board of the prototype that had a mass located inside the torso volume that swung back and forth to provide the needed counterbalancing? My position is simply that we design a suitable system to use the high-mass batteries to swing back and forth instead of deadweight. Make sense? >What makes Femisapien especially clever is that she distributes her mass via twisting her trunk and arms which puts less load on her motor relative to leaning. Her designer is quite brilliant, IMO. >The most reliable actuator is the one that’s not there. This. The same is quite true of software as well, btw. :^) >>35590 >That's where the square-cube law will strike HARD. Imagine walking in lead-filled snowshoes. I think there's some subtleties there, GreerTech. The Square-Cube Law [1][2][3] is formulated regarding volume. For meatspace things like elephants, blue whales, and birbs, this roughly-translates directly into increased mass -- which is our general nemesis. But notice the last example, birbs. God designed them with a very low mass bone structure, and same for their primary lifting mechanism: feathers. As long as special tricks are used to keep the strength up and mass low (as with birbs), then the square-cube isn't quite the problem it would be otherwise. In Kiwi's example case, I'd suggest the concept is kind of like this instead: imagine you're walking on a pair of stilts, with 2'/60cm wide hoops affixed to the bottoms. While the results may be somewhat awkward (perhaps even comical), they wouldn't really require a significant increase in force required, but rather in the finesse involved. With the vase-printing approach (particularly if the 'mesh shell' paradigm [4] is used), her extra wide 'calves/feet' shouldn't really represent a significant increase in mass. Make sense? --- 1. https://worldbuilding.stackexchange.com/questions/215851/consequences-of-an-altered-square-cube-law 2. https://en.wikipedia.org/wiki/Square%E2%80%93cube_law 3. https://www.wolframalpha.com/input?i=cube+calculator 4. (cf. >>16170, >>16171, >>16525, et al) >=== -minor edit -add 'vase-printing' cmnt -add 'mesh shell' cmnt/crosslinks

>>35584 I kinda tried similar to the femisapien walking mechanism least year, not easy as it looks (especially when manually controlling the servos and some of them are the wrong way around!). That's why I'm gonna beef up a smol robodog with some ASMC-04bs and try to mount SPUD on the front of that. ASMC-04bs are pretty good in they have considerable torque (180kg/cm) and they control just like a servo except for an extra set of terminals for actual motor power. >>35604 >Good luck with your neck actuation assembly, Mechnomancer. It's looking good so far. Oh that is from like 6 month ago and I don't advise using that mechanism. The screw-drive takes too much current.

>>35590 >Square cube law More like square cube "guideline." That "law" assumes the part is 100% solid and scaled with complete uniformity. Reality can be whatever you want it to be, I can make the same part scale linearly with respect to volume and mass. Just need to get creative. In our case, we actually need to get creative because we are focusing on a large volume, low mass machine. >>35604 >Shifting internal mass That takes me back. Posted about that in 2017, the year I first got into this endeavor. Back then I was considering a wheel with bolts that'd work similar to old gyro trains. I'd say having a battery pack where the heart is, eccentrically connected to a shaft would be a far more clever, efficient, and practical. I have an itch to test this now. Got the ol' noggin' joggin'. >Femisapiens walking scales in processing "finesse" more than power/force required. Eloquently put, couldn't of elaborated better myself. >>35618 >Ork Femisapien legs You'd need an Ork psychic gestalt field for that to work. To put it seriously, your ankles and hip tilt shouldn't be powered. Your PVC has far too much play, cancelling out any potential height differential the legs could have. Most importantly, you're missing her entire body. The one part that is supposed to impart a height differential via manipulating her center of gravity. Your swing servos and parallel mechanism are correct. Your feet are also well thought out in terms of size and placement. It's a clever design, just needs a body, move her servos around, and find a way to reduce the play in her legs. Currently your design is similar to TecFoot from David Buckley. You may enjoy his work. http://davidbuckley.net/DB/A_Minimalist_Approach_to_Biped_Walking.htm

>>35604 >>35644 I see. Ironic that I failed to account for hollow designs, considering my design is mostly hollow. Most of my theorizing was based on the RoboSapien toy I had, there the walking and balancing was achieved by having giant feet with the relatively heavy batteries in the feet, hence the "lead-filled snowshoes", plus my own experimentation with building tall robots.

>>35618 Reminds me of the first Honda robot prototypes

>>35644 >You'd need an Ork psychic gestalt field for that to work. You also didn't notice the ankles were tilting the opposite way they should be :D I was only using 3 servo channels to control 6 of 'em. Frankly I'm amazed it even woobled. Springs might work well for shifting the mass of a smol robot, but a large robot's walk cycle could be easily upset if it was carrying a load or accidentally bumped which is why I tried powered. But once again why reinvent the wheel when you can yoink from other products lol

Here's some videos of the Femsapien and Robosapien walking https://youtu.be/IfyB558ItD8?si=0idXfxuLnUVGrjsu https://www.youtube.com/watch?v=4N0pERu9gQ4

>>35644 >gyro trains Lol, that's very cool-looking Anon! :D >I have an itch to test this now. Got the ol' noggin' joggin'. Please do! I think a single actuator that drives a pushrod in either direction should probably suffice. It's what I'll plan the control interface for. Cheers, Kiwi. Maidcom Mini's coming right along! :^)

>>35647 We're all learning, I've made far worse assumptions. Wish I had someone to correct me in the early days. >>35654 >Powering extra DOF for safety Sensible, I can understand how that makes sense to ensure she doesn't topple over when dealing with large inertial disturbances. Circumferential Flux Motors A new kind of motor that is essentially a rotary solenoid. Coils produce a flux field that imparts kinetic motion within a magnet, see gif for gist. This would potentially allow for a motor that has nearly perfect efficiency with the dirt simple control of brushed DC motors. The only site I found with information on them; https://www.motion-robotics.co.uk/cf-motor How do they interface with magnets inside the coils? Their coils are wound as a C-shape. I'd make one by having magnets with teeth which snap together to form an annulus/ring gear for a planetary gear train. Sets of coils would be wrapped around the annulus with space for planet gear meshing. The planets are held in place on the carrier to prevent clashing with. the coils. (Carrier not in picrel because it looked weird.) Sun gear providing the final output for the system. I put a hex connector because I like them. How would you design or implement a circumferential flux motor?

>>35662 >gif related Wow that's wild. What's the little 'tube car', some kind of battery or other? >How would you design or implement a circumferential flux motor? Could we co-locate 3 of these together into a tight volume to create a universal direction ball-actuator in the hip joints, ala Sukabu's designs, et al : (cf. >>35663, etc.)?

>>35664 It's an AA battery with neodymium magnets at the front and back. Those magnets function as both brushes to induce and electromagnetic field in the copper coil, and they act as the motive element by "pushing" against that induced field. An ingenious machine, essentially a flexible solenoid with power held within the armature. >Using three for her hip I very much intend to make that happen! Though I want to make it by end of year, it'll likely take at least 2 considering MaidCom 1 is primary focus.

>>35662 That was something I saw long time ago, but I overlooked that the "train" was a battery, and I thought I could use it to pull "muscle" strings to move robot parts. But it turned out that it was too weak: >>4429

>>35673 Just need to calculate based on your needs for your solenoid. https://calculatorultra.com/en/tool/solenoid-force-calculator.html#gsc.tab=0

>>35498 Great comment. I've thought about Halbach arrays but didn't think about them in a coreless usage. The paper got me to thinking.

Responding to >>36430 from the diy motor thread. >Maybe we should consider a more unconventional design thats more similar to how muscles work I just want to say upfront that I’m not trying to be critical of your idea, I really appreciate outside the box thinking, since it’s exactly what we need to push robowaifu development forward. On the topic of your design, If the magnet is fixed and the coil moves, the coil will experience a force, but it may not move efficiently because electromagnetic forces generally require a closed magnetic circuit for strong actuation. In your design, the permanent magnet at the bottom has a magnetic field extending outward. The coil generates its own magnetic field when current flows through it. However, the fields don’t have a clear return path, meaning most of the flux will escape into free space rather than interacting efficiently. The force might not be strong enough to move the coil significantly unless very high currents are used. The spring’s placement directly below the coil could also interfere with the electromagnetic forces. Solenoid-based actuators are notoriously difficult for precise movement due to their binary nature (typically fully on or off), high hysteresis, and nonlinear force response. But based on your drawing it's far more like a Linear Voice Coil Motor except your coil doesn't surround the return spring. Look into Linear Voice Coil Motor or Voice Coil Actuators. Typically these have no torque rating and instead have a linear force rating. VCA/LCVM generally produce a low linear force, but they excel in applications requiring high precision, fast response times. They are largely unsuitable for applications requiring high rotational torque. (I assume you're wanting to drive a linkage to rotate limbs via the linear force). The rest of my post will be about magnetic linear actuators and more specifically linear synchronous motors (LSAs). These LSAs are largely how maglev trains and modern roller coasters work. These systems rely on electromagnetic forces to generate linear motion directly, eliminating the need for mechanical conversion from rotary motion. Industrial automation often uses them for contactless linear transport. These are really unlike any other commonly used actuator, such as a ball screw, timing belt, or rack and pinion, they provide high precision, high velocity, high force and long travel. They are pretty complicated and extraordinary expensive devices. These systems are extremely useful in automation, but largely no one ever thinks of them or even knows about them outside of maglev trains. See https://www.youtube.com/watch?v=ICN2iO3nbiQ for the pic in action and see https://www.instructables.com/DIY-IRONLESS-LINEAR-SERVO-MOTOR/ for more information on the pic. Looks amazing right? While LSAs excel in high-speed automation (maglev trains, CNC automation), they face significant challenges in humanoid robotics due to power inefficiency, control complexity, and limited precision for small-range movements. Below is my experience with them. Using LSAs as linear actuators is a really intriguing idea, I played around with the idea many years ago as linear actuators in a CNC shop before there was really any large interest in humanoid robotics. I've been trying to build a good humanoid robot for many years (since the 90's) and the problem has always been good ways to translate motion with low power. At the shop I was tasked with building some automation and they had a few LSAs lying around unused and I thought they were pretty cool and thought they could be used as linear actuators, so I messed around with them on my lunches. I built a LSA bicep/forearm and wrote some pages of notes on my experiences with them. (I wish I had some pictures but this was before there were cameras on phones, lol.) Here's some of my observed limitations with using LSAs as linear actuators. The main problem with them for humanoid robotics is multi-faceted. LSAs require significant magnetic field strength and large electromagnets or permanent magnet arrays to function efficiently, making them bulky compared to other options. The amount of current required for high-torque, low-speed operation (needed for humanoid joints) is substantial, leading to high power consumption. The coils in the actuator generate heat due to resistive losses, requiring cooling systems, which further increases size and complexity. LSAs are often water-cooled in automation systems, at least the ones I used were. Unlike geared motors, an LSA requires continuous power to hold a position. If power is lost, the joint loses all resistance, making it impractical for static postures without a mechanical brake. Furthermore, Since LSAs operate on an electromagnetic field, they constantly draw power to maintain a position, unlike other kinds of motors which can maintain positions with mechanical resistance due to gearing. LSAs are great at high-speed travel over long distances, like maglev systems. But humanoid robots require fine, precise movements in small ranges (gripping, balancing, small joint corrections). Magnetic force distribution in LSAs doesn't favor short-range precision, making them inefficient for fine motor control. LSAs require precise real-time control of magnetic fields to achieve stable, smooth motion. Unlike traditional servos (which only need PID control), LSAs need a dynamic magnetic field generation system, leading to increased CPU load and additional sensor requirements. Limited Back-Drivability: Unlike geared motors, LSAs resist passive motion, which can make them problematic for compliance-based movements in humanoid robots (soft grasping, adapting to external forces). The lack of small precise movements, limited back-drivability, and high power requirements is a real killer here. Also unlike rotational motor-based humanoid robots (where actuators are largely modular), LSAs require custom electromagnetic configurations per joint, making them hard to scale for general use.

>>36432 POTD I truly appreciate the scientific basis around your engineering thinking process, Axial. Thanks for elevating the general discussion here on the board. Cheers.

>>36432 Great comment . I've been thinking about much the same sort of thing but weird tangents. Most automobiles are moving towards switched reluctance motors. Which, I believe, is sort of what you have pictured but linear. What I have been thinking of is a mash up of a linear and regular circular reluctance motor. Looking at the linear you showed some of the problems are, your moving the coils (heavy, needs wiring that flexes and takes space) and the actual coil (working) area is a very small portion of the track. As you mentioned this means high currents to pack all the forces you need into that small coil area. What if you turned the thing sort of inside out. What if instead you moved the track? But in this case the track turns in a circle. Much like a fan belt runs between two pulleys. So the coils are stationary and the inductive part is on the belt which can rotate linearly til it runs around the pulley and then comes back. It's a belt. You could put coils on both sides driving a very large linear area. This defeats some of the problems in the rotary inductive reactance motor. High heat build up because it's all enclosed in the cylinder motor with limited area for coils. With it spread out you don't have either problem, Another feature is you can make the belt, chain, moving part as long as you want with a large coil area over many links. This means the large majority of the whole of the motor is being driven. Not just some small section. For movement you could tie a line directly to the chain/belt/ driven part or you could wrap a line that is pulled around one of the axles giving you a step down transmission. I think a holding force, without power, is unnecessary as people don't have that so it's not needed in a waifu. In reference to Halbach arrays >>35695 You mentioned the return path for the magnetic field. I wonder what would that be for Halbach arrays? Could you put metal return paths on the sides and have that work?

>>36439 of course! >>36523 Hmm your idea is intriguing, I'll have to stew on that for a bit. The main thing I would be worried about would be complexity and the custom reluctance control algorithm. Interesting thought though! >switched reluctance motors These are very cool but very tricky to get right. In the future once I get the Coreless Axial Flux Motor stuff finished, I would like to check these out more as they are significantly cheaper to produce because there are no permanent magnets needed! Unlike permanent magnet motors, it relies solely on the interaction between the stator's electromagnets and the salient poles of the rotor. In general, with SRM you have more stator poles and fewer rotor poles, for example a 6/4 SRM has 6 stator poles and 4 rotor poles. When a stator coil is energized, it creates a strong magnetic field. The rotor aligns itself to minimize the magnetic reluctance (resistance to magnetic flux). This movement generates torque as the rotor seeks the lowest reluctance path. Once the rotor aligns with the first energized stator pole, that coil is turned off, and the next phase is activated. This switching sequence ensures continuous rotation. The rotor has no magnets, making SRMs highly robust, cheaper, and suitable for high-temperature environments. The main problem with these kinds of motors is the optimal drive waveform is not a pure sinusoid, making it harder to create electronics and program the driver. This is due to the non linear torque relative to rotor displacement and the windings highly position dependent inductance. The other problem is high torque ripple. Torque ripple is basically the periodic increase or decrease in output torque as the motor shaft rotates. Cogging torque is kind of a good example of torque ripple, if you ever rotated a stepper motor on a 3d printer and you feel the rotational detents, that's cogging torque. Technically speaking, cogging torque and torque ripple are different, but they both cause uneven motion in electric motors, as they come from different sources, the mental idea and picture is the same though. Cogging torque is a magnetic locking effect whereas torque ripple is a fluctuating output torque. When the electronics for an SRM are switching it must deliver power to the different windings and limit torque ripple, this is what causes the high complexity in electronic/programming design. (one of two)

>>36523 >You mentioned the return path for the magnetic field. I wonder what would that be for Halbach arrays? Could you put metal return paths on the sides and have that work? You can! I'll try and explain Halbach arrays and return paths for fields a bit further (this might be a better post for the DIY motor thread, but we can link it) So the interesting thing about Halbach arrays is the magnetic field is concentrated on one side while nearly canceling out on the opposite side. The Halbach Array works due to a carefully arranged sequence of magnetization directions that reinforce the magnetic field on one side while nearly canceling it on the opposite side. This is achieved through vector addition of magnetic fields, which results in constructive interference on one side and destructive interference on the other. If i remember correctly this is called the magnetic field superposition. Knowing this, we can tackle the return path problem in multiple ways. >Free Space Halbach In a typical linear or cylindrical Halbach array, the return flux is distributed in space around the array. This is common in applications where weight reduction is prioritized, such as maglev trains and brushless motors with air-core stators. One main problem with a Free Space Halbach is the lower efficiency due to leakage flux and weaker field strength on the active side. This is your general Halbach Array with no return path consideration, and is probably the most common. >Ferromagnetic Backing Plate (Steel or Soft Iron) If we place a high-permeability material (like steel or soft iron) on the "weak side" of the Halbach array, it provides a low-reluctance return path. This is generally used in rotor designs in axial flux motors and linear Halbach tracks. What this backing plate does is it increases field strength on the active side by directing more flux towards it. The main problem with backing plates is added weight and potential eddy current losses. >Yoke Based Return Path In cylindrical Halbach rotors (like a motor), a ferromagnetic yoke can be placed around the outer perimeter and can be used as a return path. This yoke guides the flux back to the magnets, increasing efficiency. Some axial flux motors place a steel backplate behind the magnets on each rotor. >Opposing Halbach Arrays (Self-Contained Return Path) Using two opposing Halbach arrays creates a closed-loop magnetic path. This is commonly used in linear Halbach motors and magnetic bearings. You can also have multiple Halbach Arrays in a radial flux motor where one array is within the coil and the other one surrounds the coil. The secondary Halbach array reduces flux leakage and increases field uniformity. The only downside is a more complex assembly and alignment. >Flux Return with Soft Magnetic Composites (SMCs) Instead of solid steel or soft iron, soft magnetic composites (SMCs) can be used to shape and direct the flux with reduced eddy currents. Useful in high frequency applications where laminated steel or soft iron would cause excessive losses. This is pretty advanced and costly, but you can diy fairly cheaply with iron powder and epoxy. Downside is the added weight and reduced eddy currents. >Non-Magnetic Spacers Interestingly using air gaps and non-magnetic spacers you can adjust field strength and direct return paths. It allows fine tuning without adding significant weight or increasing losses. >Hybrid Halbach and Conventional Designs This is also called a "Partial Halbach Array" where you have additional traditional magnet arrangements to provide a better return path. Some radial flux motors use a Halbach configuration on one side and traditional radial magnets on the other. I've actually never seen this configuration in use but one of the books I have talks about it, so I figured I'd mention it. The main problem with magnetics in general is the math is pretty heavy and the fields are invisible. What's worse is magnetic simulation software is extremely expensive, but some software people use have simpler models which are pretty useful (like freecad). You can drastically change field strength and return paths by subtle changes of magnet placement and air gaps alone.

>>36607 >>Non-Magnetic Spacers Interestingly using air gaps and non-magnetic spacers you can adjust field strength and direct return paths. It allows fine tuning without adding significant weight or increasing losses. I've been reading quite a bit about ferrites and soft magnetic materials as I'm interested in motors and magnetic amplifiers. I'm really sure I get the "air gap" thing. My reading, and likely I have this wrong, the air gaps help keep you from going into saturation in the field focusing material or maybe better termed as where the magnetic field can more easily go through. You said that the powdered materiel is more costly. I think for our needs that would be somewhat incorrect "if" you consider the processing of the material and not just raw, per pound cost. It's my understanding that the powdered material, or some of it, has the advantage of the powder being separated by air or glue of some sort gives it the advantages of air gaps built in. It's also my understanding that the air gaps have the ability of raising the inductance of your magnetic material temporarily lowering it's permeability (meaning it's ability to channel magnetic fields). With higher inductance it lowers the inrush of current. Or this is what I assume the purpose is. I;m not 100% sure of this but keep looking. It;s remarkably hard to find scientific information that gives you a broad overall view. They all cover you up with equations and graphs without much telling you the over all purpose. Could it be?? that using the air gaps you can have a large core with higher current wiring BUT by raising the inductance and lowering inrush current you can still get a large magnetic field while not melting down your coil or turning your actuator into a oven??? As I say I'm not exactly 100% sure what air gaps really do. I think I sort of do but I wonder is there not some more satisfactory method of doing this without essentially choking the field down? Like why not use a material with less permeability in the first place??? A weird factoid. The air gaps in the transformers in guitar amplifiers I read are what gives them their great sounding even harmonic distortion sound. Apparently they saturate and clip the output, but in a way that sounds full tilt rock and roll. https://en.wikipedia.org/wiki/Sendust I've been trying to buy some sendust magnetic material. https://en.wikipedia.org/wiki/Sendust I found good deals in China but the shipping is a killer. I may do it anyways. US suppliers are mostly interested in selling not the powder but the whole core already made. I have no interest in that. And US supplies always want you to write in, get quotes, blah, blah, its difficult to find a straight, buy this many kilos for this many dollars. Deepseek AI says that higher permeability cores can be controlled with smaller DC control currents in magnetic amplifiers. This seems contradictory to me but that's what it said. I will have to ask grok what it thinks. If I could understand the air gap thing it could save me from needing higher permeability (more expensive) cores.

>>36663 >still get a large magnetic field while not melting down your coil or turning your actuator into a oven??? When I say this maybe I should clarify so people know what I'm talking about. I know a decent amount about AC electric motors. In many of them they have a separate circuit or some sort of arrangement to limit the current when they start. Some have big resistors or external cast iron resistors in really big motors. Some have relays that run the coils in series until the motor gets up to speed then switches them to parallel. All sorts. I'm thinking that these air gaps do much the same. The total magnetic field is based on the current but starting inrushes of current can be really high. The idea of the air gap is to slow down the inrush or so I surmise, but of course I'm not really sure.

>>36663 >the air gaps help keep you from going into saturation in the field focusing material or maybe better termed as where the magnetic field can more easily go through. Yes, exactly! The air gap helps prevent saturation in the field focusing material (like iron or ferrite) by acting as a flux bottleneck, ensuring that the magnetic field does not become too concentrated in one area. A high-permeability material (like laminated iron or ferrite) naturally 'funnels' magnetic flux efficiently, but it has a limit. Once it reaches magnetic saturation, adding more current does not increase the magnetic field proportionally, this leads to inefficiencies and heat buildup. If you add an air gap, you can increase the total reluctance (resistance to magnetic flux), which spreads the field more evenly. You can lower the effective permeability of the core, meaning it takes more current before reaching saturation. and the magnetic energy is partially stored in the air gap itself, reducing localized core saturation. The best way to think of an air gap is it acts like a pressure relief valve for the magnetic field, preventing the material from becoming overwhelmed and ensuring a more controlled magnetic response. To recap, without an air gap, a high permeability core saturates quickly, and any extra current turns into wasted heat instead of increasing the field. By adding an air gap, the field is distributed more evenly, preventing localized hot spots and allowing you to push more current through the coil without immediate saturation. This reduces hysteresis losses and eddy currents, both of which contribute to heating. >You said that the powdered materiel is more costly. I think for our needs that would be somewhat incorrect "if" you consider the processing of the material and not just raw, per pound cost. I said that SMCs are expensive, and that you can diy them yourself cheaply with iron powder and epoxy. Well designed SMCs are ridiculously expensive. >It's my understanding that the powdered material, or some of it, has the advantage of the powder being separated by air or glue of some sort gives it the advantages of air gaps built in. Kind of, it all depends on the material properties of the binder (glue/epoxy/etc) and interaction with the iron powder itself. It can act like an airgap but really it's not an air gap at all but rather a nonmagnetic material so it will influence the system differently than just air. Plus some non-magnetic binder materials (glue/epoxy/etc) can interact with the iron powder causing funky unexpected results. There's a bunch of papers on this. I remember one paper where a non-magnetic binder material turned slightly magnetic in the presence of iron. If I remember correctly, there was a chemical reaction and caused the binder to turn slightly magnetic itself. The electrical and chemical properties of the binder are probably the most significant part in the equation, since iron powder itself is largely a constant. Also there is paramagnetism in materials, a kind of magnetism which only occurs in the presence of an externally applied magnetic field. Material science is a very complex subject that I don't know that much about tbh. >It's also my understanding that the air gaps have the ability of raising the inductance of your magnetic material temporarily lowering it's permeability (meaning it's ability to channel magnetic fields). With higher inductance it lowers the inrush of current. Or this is what I assume the purpose is. I;m not 100% sure of this but keep looking. Yeah you're on the right track! It seems like you already got the tricky part down. The tricky part is that an air gap reduces overall permeability, which you might assume would reduce inductance. However, because the gap stores more magnetic energy, in many practical applications, the inductance ends up increasing or stabilizing instead of dropping. When you power on a motor or transformer, an air gapped core slows down the rate at which current can rise, reducing inrush current. This prevents excessive current spikes that could damage components or cause overheating. (1of2)

>>36663 >Could it be?? that using the air gaps you can have a large core with higher current wiring BUT by raising the inductance and lowering inrush current you can still get a large magnetic field while not melting down your coil or turning your actuator into a oven??? You are absolutely correct in thinking that air gaps allow you to use a large core with higher current wiring while managing inductance and preventing excessive heating. The key here is that an air gap regulates flux density, allowing you to generate strong magnetic fields without saturating the core and overheating the coil. >As I say I'm not exactly 100% sure what air gaps really do. I think I sort of do but I wonder is there not some more satisfactory method of doing this without essentially choking the field down? Like why not use a material with less permeability in the first place??? The reason is that low-permeability materials do not concentrate flux well. You need much more current to generate the same field strength, leading to higher electrical losses. Magnetic circuits become inefficient because the flux spreads out rather than staying focused where you need it. Energy storage is less effective, which can cause performance problems in motors, transformers, and inductors. By contrast, a high-permeability material with an air gap gives you the best of both worlds. The core still focuses most of the flux, but the air gap acts as a buffer, preventing saturation and excessive heating. You get better energy storage and less wasted power than if you just used a low-permeability material. Air gaps let you fine-tune performance without completely "choking" the field, making them the best practical option. >It;s remarkably hard to find scientific information that gives you a broad overall view. They all cover you up with equations and graphs without much telling you the over all purpose. Good books help but largely have the same problem. >Sendust Way to expensive for DIY. Iron powder is the most cost effective of all the powder cores. Don't bother with trying to source sendust. If you really wanted sendust, it's composition is typically 85% iron, 9% silicon and 6% aluminum. You can always mix your own batch up. I've seen this done before, though personally I wouldn't do it, I prefer coreless designs. >US suppliers are mostly interested in selling not the powder but the whole core already made. Well yeah, that's how they make money. They generally are selling a service and not really selling the end product. You're paying for their service to do a whole lot more than just make a core. They do all sorts of things to optimize your core design and provide you with a end product. >And US supplies always want you to write in, get quotes, blah, blah, its difficult to find a straight, buy this many kilos for this many dollars. Yep, the old "if you have to ask the price, you can't afford it". >If I could understand the air gap thing it could save me from needing higher permeability (more expensive) cores. Coreless motors are far better for robowaifu, imo. (2of2)

>>36673 >Good books help but largely have the same problem. The extraordinarily-eminent Carver Mead said that (my own very crude paraphrase here): >"Much of our difficulties in physics today comes from having to deal with Maxwell's equations, because they are all contrived around the idea of the Aether. You gotta get rid of that stuff! Then things get much simpler. :)" (cf. : >>33943, @ ~27mins ). I'm very interested to hear your take on that notion, Axial. >=== -fmt, minor edit

>>36672 In an earlier post I made a mistake an left out the I "DON'T" really understand air gaps. I'm getting much closer. >air gap ... acting as a flux bottleneck, ensuring that the magnetic field does not become too concentrated in one area. A high-permeability material (like laminated iron or ferrite) naturally 'funnels' magnetic flux efficiently, but it has a limit. Once it reaches magnetic saturation, adding more current does not increase the magnetic field proportionally, this leads to inefficiencies and heat buildup That I really get. Makes good sense. >To recap, without an air gap, a high permeability core saturates quickly, and any extra current turns into wasted heat instead of increasing the field. By adding an air gap, the field is distributed more evenly, preventing localized hot spots and allowing you to push more current through the coil without immediate saturation. This reduces hysteresis losses and eddy currents, both of which contribute to heating. I get this too. The problem I have is "IF" the air gap causes lower permeability then what's the point of high permeability material? Why not use something cheaper with less? Now "IF" what I said about electric motors here >>36665 and avoiding large inrushes of current BUT allowing large magnetic fields to go through with the higher permeability material, then I get that. I guess it depends on "IF" it's like AC motors. (An assumption)The killer is the "rapid change in current" not necessarily the current volume or amount. (Speculation)Rapid changes mean saturation but more gradual rises of current are less saturating. It's a bit confusing because according to Deepseek high permeability materials can be saturated with lower levels of DC current when used in magnetic amplifier. But it would seem that the low permeability would allow MORE magnetic field, meaning the level of DC bias would have to be higher before it choked. Some of the data I got from a paper referenced by Duckduckgo AI. This one below. I can not find a direct link. "A Comparison of Molybdenum Permalloy Powder and Sendust Cores for Energy Storage Inductors" Tim Slattery, Applications Engineer, The Arnold Engineering Company, Marengo, Illinois, USA May, 2000i "Molybdenum permalloy powder and sendust cores are described and compared." Now it says, "...cores in power conversion applications. Energy storage and release takes place in the gaps between magnetic metal powder particles. An insulation material that is applied to the powder before compaction maintains these gaps. Particle-to-particle insulation also reduces eddy currents in the core. The magnetizing force (magnetic field of the winding) works with the easily magnetized metal to achieve high magnetic flux density (induction) in free space within the core. The free space is divided into the many gaps that are uniformly distributed along the entire length of the magnetic path. This distributed gap is one of the most important aspects of powder cores. Fringing fields and associated winding eddy current losses caused by one or two discrete gaps in the core path length are avoided.." It seems almost as if the air gaps store more energy than the high perm material itself. Is it, as you said, the real reason this is, is because the air gaps make the magnetic field EVENLY"spread around the material? So that all of it is used. And is the energy stored in the coil alone or is some of it stored in the air or a combination of the two? The paper seems to posit that the field is stored in the "Air gaps" and the permeable materials is only to channel it and focus on these gaps. This of course brings up the question of why have the permeable material in the first place if the energy is stored in the air gaps?? It's all so confusing. My interest in this mostly lies in magnetic amplifiers but of course switch mode power supplies, inductive reactance motors and forces created. BTW I found Sendust powder suppliers in China that would sell sendust powder for $12 a Kg in 3Kg quantities but the shipping was like $90. One company told me they would send me 3Kg for free if I paid shipping. So $90 USD. I may end up doing that. Get a FedEx account, which seems to be what they want. But not until I understand the parameters I need. May be sendust is a waste of money for what I want.

Ok you answered stuff I didn't see before my last comment. Thanks a heap for the tutoring. It's VERY MUCH appreciated. And in turn I will tell you why I'm so interested in this. If you want to recreate all the muscles in a human then you need about 300 actuators. This means 300 MOSFETS or equivalent. They are cheap but good ones, not Chinese, will run you 50 cents to a dollar. Then of course you have driver circuits to drive the MOSFETS completely so you don't burn them up, more money, more complexity, then diodes and resistors to control the back EMF from the coils. It gets complicated fast. And pricey. You also have, with separate MODFETS like this, a mass of wires of fairly high size to go to each individual muscle. All of this of course requires boards, big flow soldier equipment to place it. It adds up real fast. So "what if" instead you use magnetic amplifiers made from powdered cores . Compare that to buying 3Kg of sendust for $12 a Kg, some silver(alloy) solder for maybe $30 a pound, a little epoxy and manufacturing this stuff in a oven. Mag amps are seriously robust. And the permeable material, if metal based, will allow very high short term amperages because they conduct heat really well. Mag amps is how V-2 rockets and WWII battleships were controlled and they hardly ever broke. Possibly you could integrate the amplifier as part of the actuator. I see no reason you couldn't. The power would be one long bus of AC power. The control wires for the various muscles would only need to be very small DC control level circuits. I have an idea for the coils in the permeable material. Make molds with hollows in them. The hollows would be the coils. Pack in epoxy and your permeable material. Now you have the core with a "nothing" area for the coils. Pack the coil area with very fine copper powder, and use a good silver alloy soldier that can be sucked up into the copper powder spaces along with the copper. Put in a oven, solder runs into copper powder and now you have a mass produced magnetic amplifier/actuator. All this greatly depends on 3D printers to make the molds. Another method is to print the actual coils in a resin 3D printer. The resin printers have outstanding resolution. Make a mold of the coil. Pour low temperature metal into the mold. Pack your permeable material around the metal coil then melt it out. You only need boiling water temp or less to do so. Then add copper powder, soldier. The key here once you set this up you can easily mass produce all these molds and rapidly squeegee in all the materials. The other alternative is hydraulics which might be ok if you made bladder type hydraulics but...I don't like hydraulics, but even if I don't like them it may be the cheapest, best solution.

>>36677 What I understand is that quote was in reference to special relativity and how light needed a medium to travel through in Maxwell's original equations. Originally Maxell's equations were formulated in the context of the luminiferous aether. An assumed medium through which light waves propagated, similar to how sound waves propagate through air. Once special relativity hit and showed that no such medium was necessary, Maxwell's equations were reinterpreted in a relativistic, aether free framework. The removal of aether made Maxwell's equations simpler, symmetric, and consistent with relativity. Instead of relying on a fictitious medium, modern physics understands electromagnetism as a manifestation of relativistic field theory, where the electric and magnetic fields transform into each other depending on the observer's motion. (First Picture) Interestingly if you look at Maxwell's equations through the lens of Geometric Algebra/Geometric Calculus and specifically in the SpaceTime Algebra (STA) formulation you can make it even simpler without tensors. This is the second picture. Instead of dealing with four separate equations (or the complicated tensor formulations in differential geometry), Maxwell's equations reduce to a single geometric equation. This approach unifies the equations into a single, compact form, showing that electromagnetism is inherently geometric. Which is pretty cool and intuitive! Though this requires esoteric GA/GC knowledge. Side note: Is it possible to include math like this picture on here in text instead of me using github with markdown and pictures?

>>36672 >When you power on a motor or transformer, an air gapped core slows down the rate at which current can rise, reducing inrush current. This prevents excessive current spikes that could damage components or cause overheating. Ok I got this now. Just like a motor when sitting still has a very low inductance, meaning the current flowing into it will melt it down, but after it gets going the inductance rises. This means all the power, like a pure load resister, will be dumped into the motor coils BUT with higher resistance, lowering the current so as to not cook the motor. A question. How does the air gap effect the attraction of high permeable materials to magnetic fields? Will it lower the force or only for a moment? Basically the the air gap acting as a dampener? I'm going to read over, and over, what you wrote, then formulate some really specific questions and hit Deepseek and grok up. I I bet I can really understand this. I'm one of those weird individuals that likes to understand the whole package instead of just cook booking things. "If" i can come to this understanding I think I could make something very cost effective.

>>36672 > I remember one paper where a non-magnetic binder material turned slightly magnetic in the presence of iron. My present plan is to use high temperature epoxy. I've read you can get it to 600F but I know of one product that gets to 450F but it's air, room temperature cured. Some of the very high temp stuff you have to bake cure it for many hours at high temperatures of 400F or more. Non starter as it will be a huge time killer. Maybe ok for fighter planes but it would slow down production. I want to note that I did some rough figures on efficiency of humans compared to Tesla's motor and we have a HUGE amount of room to work with. We can make something FAR, FAR less efficient and power to weight without much penalty or basically none at all. Tesla 3 motor output = 258KW(345HP) at 27.99Kg(60lbs.) but a human rarely puts out more than 400 watts in peak.(this is athlete level) https://en.wikipedia.org/wiki/Human_power So the Tesla motor puts out 4300 watts/pound and human puts out(at 250watts and 200lbs.), 1.25 watts/pound My goal is a robowaifu at least as strong as a very good athlete, so I get it to carry stuff. Very extreme peak performance in humans is 1,000 watts but only for a few seconds.

Axil I REALLY APPRECIATE all the tutoring you have done. It's super helpful. I think have a really good handle on things now. I now need some sendust alloy. There;'s a HUGE heap of materials but I settled, so far on sendust because it's not too expensive, has really good permeability and very low to no magnetostriction. Meaning it does not expand and contract in magnetic fields. Materials that do have a hum. Like the 60Hz hum you get from large iron transformers. Don't need a humming buzzing waifu. It also being metallic will conduct heat fairly well. My basic idea is a buss of AC at higher frequency, to be determined. For giggles, 50KHz with magnetic amplifiers driving actuators.

>>36677 >>36708 >Interestingly if you look at Maxwell's equations through the lens of Geometric Algebra/Geometric Calculus and specifically in the SpaceTime Algebra (STA) formulation you can make it even simpler without tensors. This is the second picture. Instead of dealing with four separate equations (or the complicated tensor formulations in differential geometry), Maxwell's equations reduce to a single geometric equation. This approach unifies the equations into a single, compact form, showing that electromagnetism is inherently geometric. Which is pretty cool and intuitive! Though this requires esoteric GA/GC knowledge. I should probably add some more context for digestion. When I said electromagnetism is inherently geometric, I mean that Maxwell's equations describe the structure of spacetime itself and that the electromagnetic field is a geometric object, not just a set of arbitrary vector equations. Traditionally, electromagnetism is described using electric (E) and magnetic (B) fields, which are treated as separate three-dimensional vector fields. However, in spacetime algebra (STA), these are naturally unified into a single object, a bivector F: F=E+IB. Where I=e1e2e3 is the pseudoscalar (volume element of space). E is a vector (directional quantity). B is a pseudovector (axial vector related to rotations). This means that electromagnetism is not just a set of field values floating in space; it's actually a structure embedded in spacetime itself. Furthermore, electromagnetic waves are really just geometric rotations in Spacetime. When an electromagnetic wave propagates, what is actually happening is a rotational oscillation of the electromagnetic field bivector F in spacetime. >In 3D, we think of electric and magnetic fields oscillating perpendicular to each other. >In 4D spacetime, this is better understood as a single rotating bivector, much like a complex number or spinor. This interpretation makes electromagnetism look structurally similar to spinor rotations in quantum mechanics, hinting at a deep connection between electromagnetism and fundamental spacetime geometry. Maxwell’s equations are really a single geometric constraint, instead of being four separate equations, Maxwell’s equations collapse into a single equation provided in the second image. This is structurally similar to the Clifford-Dirac equation for spinors in quantum mechanics. In other words, electromagnetism is not just a force; it is a geometric feature of spacetime itself. The Lorentz Force is a natural geometric action, instead of defining the Lorentz force as an arbitrary rule: F=q(E+v*B). Geometric Algebra shows that the force law is actually a consequence of spacetime structure: Fqv. where v is the four-velocity of a charged particle. This tells us that charged particles follow geodesic-like paths through spacetime, modified by the structure of the electromagnetic field. The unification with relativity, in standard vector calculus, we awkwardly separate E and B and then recombine them in relativity into the Faraday tensor F_muν (Faraday tensor F_muv represents the electromagnetic field in a relativistic framework, and it’s a rank-2 antisymmetric tensor, meaning it has two indices [mu and v]). But in Geometric Algebra, we never need to separate them to begin with! The bivector field F is already a naturally four-dimensional object. This means electromagnetism is not something separate from spacetime; it is a direct consequence of spacetime geometry. >Maybe this is all too esoteric, but hopefully it's somewhat informative or at least thought provoking. If anyone is interested further and wants to know more, https://bivector.net/ is the best place online about GA/GC, while Alan Macdonald has the best books to learn about GA/GC (Linear and Geometric Algebra, Vector and Geometric Calculus). Hopefully this has been somewhat mind-opening. I know when I learned that electromagnetism was a structure embedded in spacetime itself. It kind of blew my mind for awhile. >>36711 You're very welcome. You are definitely getting it, in no time you'll understand it as much as anyone else does. Would you like me to continue to going over your posts or are you good? I don't mind.

>>36712 Anything, anything, you wish to add would only add to my thrill and joy. >>36677 >>36708 >Geometric Algebra Yep yep. I've been babbling about Geometric algebra a good deal. Not that I can work with it competently but I recognize the significance and I have some rudimentary understanding of what it's good for...like everything in physics. Chobitsu you didn't ask me but I have something to add that is very pertinent. The experiment that all the textbooks say proves that there is no aether is NOT represented correctly THEY LIE. Directly and bold faced lie. Here's what the Duckduckgo AI says, "...Michelson-Morley experiment was designed to detect the presence of the luminiferous ether, a medium thought to carry light waves. However, the experiment found no difference in the speed of light in different directions..." This is not true. I have personally seen an actual 100% original issue of the report myself in the univ. library. It does not say zero. I read it myself. I was clued into this by G. Harry Stine. They DID NOT find zero difference as is always said. I can't remember the exact numbers but these are probably close. They were looking for the speed of the Earth in orbit like 25Km/s but they found something like 8/Km/s. The statement that they found zero is a lie. And furthermore other scientist have repeated the experiment hundreds of times with more accurate equipment and they, also, did not get zero.(These used to be listed in Wikipedia but they've obscured them) One scientist did hundreds and hundreds of test with all sorts of conditions. He made sheds to cover the light path, all sorts of stuff. He found that the difference was also tied to the movement of the planets. Now you ask why would they do this...well if there is a difference, and there is, then Einsteins's theory is in trouble and you know who prints the textbooks. There's your answer.

>>36712 >Furthermore, electromagnetic waves are really just geometric rotations in Spacetime. When an electromagnetic wave propagates, what is actually happening is a rotational oscillation of the electromagnetic field bivector F in spacetime This reminds me very much of Dewey B. Larsons Reciprocal System of physical theory He says everything is motion and that twist of spacetime create everything (I think Have that right). Now this is of course called stupid and here, https://rationalwiki.org/wiki/Reciprocal_Theory they tell you how they have completely disproved it. But...I think they take some things out of context They also use earlier works of his which are more fleshed out in later books and papers. They even use equations of other people to damn Larson. Larson did some very significant things. He predicted certain astronomical type stars before they were found. (I think I have this somewhat right)His theory appears to explain the properties of materials based on their twist?. Like specific heat and lots of other properties. The site, meant to complain about him, says quantum physics explains all of chemical matter combinations HHAHAHHA yeah if you have the life of the universe to do the math. but Larson lays this all out. Even if it's wrong he explains, and can make predictions, of a lot of stuff, in a very easy manner that fits together with a few postulates. https://reciprocalsystem.org/dewey-b-larson I'm not so sure Larson is right(and I'm not smart enough to say) but, his formulation of how to attack the problem seems to be of value. An interesting idea he has, matter is in our universe sucked into another in black holes and blasted out in the mirror universe, same here, and these large jets of matter are where matter comes from. Star formations do seem to bear this out or so I read somewhere long ago.

>>36708 Thanks! You've filled in the details that Mead simply glossed over in that informal chat (+ my clear misinterpretations of his words). Yes, much simpler! Thanks, Axial. >Side note: Is it possible to include math like this picture on here in text instead of me using github with markdown and pictures? Apologies but not as far as I know, Anon. MathJax isn't supported on IBs that I'm aware of. Codeblocks and simple formating is about all you get here. https://alogs.space/.static/pages/posting.html <---> Let's move this to /meta pls, Anons Haha, this is extremely interesting to me, but we are all clearly-derailing our Actuators thread. I should have seen that coming a mile off with that question, so my fault entirely. :D >=== -minor edit

>>36712 >Furthermore, electromagnetic waves are really just geometric rotations in Spacetime. When an electromagnetic wave propagates, what is actually happening is a rotational oscillation of the electromagnetic field bivector F in spacetime. Where did you learn that? Because it just answered a question I've had since High School

>>36673 >Coreless motors are far better for robowaifu, imo. Back to actuators. Axil you said this but I'm not sure this is true nor can I say it is not. But I can game the logic why this might not be so. My confusion is the return path problem. I do understand that Halbach arrays focus magnetic fields but...I see this as changing the magnetic field into a sort of ellipsis or egg shape. Yes there is a very strong concentration in front of it but, does it not have a return path for the magnetic fields? To the best of my knowledge Halbach arrays do not create monopole magnets. I can't imagine it does. I see the magnetic field exactly analogous to electric circuits. The current (magnetic field strength) being alike and if you have a high resistance in the path (low permeable materials..air) then you are lowering the magnetic field strength throughout the whole circuit. Even if you focus the magnetic field you still are going through air in the return path, and air's low permeability means you are choking down your magnetic field. "If" you have a permeable material providing the return path it will increase the field strength over the WHOLE path. I also take issue, for cost reasons, on your use of neodymium magnets( and yes I fully acknowledge you brought that up, clearly and directly). I'm not in any way faulting the ease, performance, feasibility, just cost. In my mind switched reluctance is the only way to go because of the dirt cheap cost. They make the majority of motors and transformers with electrical steel for a reason. It's cheap. For me, one of the major overriding things to watch is cost. A waifu no one can afford is wonderful but of little use. My dollar figure I would like is $2,000 but more likely it will end up at $3,000. But that is affordable. I suspect the compute/electronics will cost $1,500 and more likely $2,000 so that doesn't leave a lot of coin for all the rest. You also said that the compute for inductive reactance motors was troublesome. Now they've been saying that for a long time and I think it's become a catch phrase that was once correct but I don't believe this is true any more, exactly Because now you can get a cheap micro-controllers that will do all the compute very fast for less than $9(or cheaper depending on form/function). With the ESP-32 you'd need about 20 to control 300 actuators equivalent to human muscles. Further about permeable materials. Example-You have two magnets and stick them together such that they are attracted. "If" you add a permeable material around the outside of them connecting the end of one and the beginning of the other...will not the attraction between them be much higher due to the completion of the magnetic circuit? An addendum that might be of great use to all actuator and motor driver schemes. Don Lancaster's Magic Sinewaves This is some really crafty stuff. Don has managed to find a way to use spikes of on-off MOSFET's. The sequence of spikes in actuality, with harmonics, add up to a very good sine wave even though it's made up of a bunch of sharp on off signals. The only harmonics in the end result are high frequency so can be easily filtered. Here's a link asking about them where they also provide links to Lancaster's work. https://electronics.stackexchange.com/questions/11844/don-lancasters-magic-sinewaves#12502 Magic Sinewaves Library https://www.tinaja.com/magsn01.shtml Lancaster is a very smart guy who has written a whole lot of dead tree books and articles on electronics, computing, etc.

>>36607 >ne main problem with a Free Space Halbach is the lower efficiency due to leakage flux and weaker field strength on the active side. This is your general Halbach Array with no return path consideration My apologies. What I wrote I did so before seeing this. I don't know how I missed it, but I did.

I don't know if anyone has heard of this guy he's old school dead tree publishing but I expect a lot of older guys who were interested in electronics have. Don Lancaster. I haven't looked at his stuff in a really long time but used to religiously read all his work. He did a stupendous amount of writing in electronics magazines. An absolute golds mine of all sorts of electronics, robotics stuff. Well I found he passed away by looking for his work on a"magic sinewaves" I commented on above. While looking at some of his other articles I found something I had mentioned here "somewhere" before. Grays code or binary codes that can use a few lines on around wheel to provide very accurate position sensing on rotating machines, actuators, etc. The link is on this page https://www.tinaja.com/hack01.shtml go down to Hardware Hacker Columns #73-87 HACKAR4.PDF and get the link. The article is on page 80.2. If you are into this sort of thing, massive intelligent broad range info on radio wavers, electronic, computing and all other sorts of interesting stuff, I recommend you download his work before it's gone. I have no idea whose hosting this, his kids maybe? I "thought" I had copies of this but searched and, I can't find them so I need to copy all his work before it disappears.

>>36941 F Have no idea who that is, but I'm glad there is interest in preserving his stuff. I dream of our robowaifu's AIs one day being able to sift everything available to find hidden gems for our current needs, engineering or otherwise. Cheers, Grommet. :^) >tl;dr Save.everything.

>>36944 >Have no idea who that is SERIOUS LEGEND I've been going though and saving everything. So much good stuff. Very relevant. All sorts parts interface tricks, Driver tricks. I'm going to post a "field's" super easy calculate link in the motors from him. He put out articles for 50??? years. A long time. He wrote for all sorts of what were major magazines before the internet. I don't even think you can imagine before the internet. Nut & Volt, Electronic Design, Electronics Magazine, there was one that was almost all ads but had a few articles that the ads paid for called Computer Shopper. It was like 3/4 to 1 inch thick in large format like Shotgun News (you may not even know what that is)of all sorts of computer parts and electronics. Huge. It was cheap because the ads paid for the magazine. Like a buck. It used to be fairly difficult to get what you can get with a couple keystrokes now. And with AI that will become even easier. To find out stuff I used to wander through huge stacks of bound older magazines in the Univ. library. Looking through print indexes to find stuff.

>>36949 He sounds amazing. What a career this Anon must've had. I'm sure I must have seen some of his stuff in a bookstore or something. Also, I found this page on his site: https://www.tinaja.com/ebksamp1.shtml Glad you're saving his stuff, Anon. Anything in specific you think would be the most-important for Anons here on /robowaifu/ to know about his stuff? Say a 'Top 5' list or something? Cheers. <---> BTW: Can we please continue this in /meta : ( >>36952, ... )? We're derailing the Actuators bread, IMO. --- https://hackaday.com/2023/07/02/saying-goodbye-to-don-lancaster/ https://www.youtube.com/watch?v=GiI2ayNVDQg >=== -add'l hotlinks -/meta note

I found something that I didn't know while looking at some of Robert Murray's stuff. This is one of those things you might want to keep the back of your mind as it may come in handy. I found that Silicon Carbide diodes have very high temperature ratings, they have almost no time to switch from on to off and are more efficient with less leakage. Silicon needs a finite amount of time that can cause problems. They appear to be close to indestructible compared to silicon. The reason this appeals to me is I'm interested in switch reactance actuators and all other sorts and you are going to have to have diodes to stop high reverse currents AND, while I do not know this for sure I expect that we can build our own. They are in a class called Schottky diodes and these were made a really long time ago with a semiconductor like silicon carbide and a metal. "if" you have 300 muscle circuits you are going to need a lot of them and you can buy silicon carbide super cheap by the pound for abrasives. Maybe you could use this stuff??? I'll look around and see. It beats trying to solder 600 or so diodes. These you might could silk screen them on. https://en.wikipedia.org/wiki/Schottky_diode https://www.electronicdesign.com/technologies/power/power-supply/discrete-power-semis/article/21193085/schottky-diodes-the-old-ones-are-good-the-new-ones-are-better I "think", not sure at all, that you can also make transistors from SiC...hmmm.... not sure. Something to think about.

>>37425 Any idea how well they handle high voltages?

>>37433 >high voltages Yes noted for it's high voltage handling. They are using them in auto motors because of this. Far better than typical silicon. Not that I completely understand this but normal P-N type silicon diodes have a built up charge barrier when reversed. It takes energy and time to reverse this and allow current to flow. Schottky diodes do not have this and are therefore faster and if I remember correctly more efficient. Note there is Schottky diodes made of silicon. I'm mostly talking about the SiCarbide ones though to be specific but I of course do not rule out others that I may not have heard of yet. SiC has very high temperature ratings. I wonder. One of the metals used to make Schottky diodes is tungsten which of course has very high melting point. I wonder if you took some SiC powder and electroplated it with tungsten would that work????

>>37425 > Schottky diodes...silicon carbide and a metal....you can buy silicon carbide super cheap by the pound for abrasives. Maybe you could use this stuff??? The Llama 3.3 70B AI is very skeptical of this approach. It doesn't say you categorically couldn't do it, but it seems to discourage the idea this can be done. I'm still not so sure that some sort of functionally useful Schottky diode can not be made economically. After all these were made in the early days. This is likely one of those things I'll have to keep n the back of my mind and slowly research and maybe something will pop up.

>>37623 Interesting, Grommet. Do you have a good feel yet for why it's spitting that negative response back?

I've been hunting for this paper. I found it when the site was down and thought it very good but when I went to find...arrrggh. It's about using black sand, basically iron oxide you dig up in a yard or find in a steam to make electrical parts, including motors and actuators. I told Chobitsu you could do this but didn't have a paper that backed it up like this one. The reason I couldn't find it was I forgot the name and the title didn't have iron or F2O3 or whatever. So searching all over didn't help. But I finally did. It's great and a way to cheaply get material to make your own motors and actuators using "dirt" as the permeable material. Like the electrical steel in motors and transformers. Here's the link and on that page is a download for, "Digging Dirt: on Inductors: Experiments with Custom Magnetics Made from "Black Sand."" https://hpfriedrichs.com/radioroom/inductors/inductors.htm One thing he found was that while iron oxide is not the best permeable material he found that as long as it was not over driven with too much magnetic field strength it had very low Magnetic reluctance which like resistance in circuits but for magnetics. So good efficiency at low field levels. This is fine for us because you can get fairly strong forces with even this material. This is well worth reading even if you don't plan to use iron oxide as a material.

>>37627 >spitting that negative response back I think purity and maybe not crystalline enough. Basically rough material. I'm guessing so don't count on that as definitive. It could be wrong though. People made crystal radios in the old days from rough stuff. I've caught AI's saying wrong things more than once. But they are sooooo useful, mostly.

>>37754 That's really pretty cool, Grommet. Stuff like this gives me hope that humans will be able to recover from catastrophe should the need arise. Cheers. >>37755 Yeah, that seems to make sense to me.

>"Digging Dirt: on Inductors: Experiments with Custom Magnetics Made from "Black Sand."" I want to say one more thing about this article. If you look at the price of motors, actuators, etc. they are high as hell (for the numbers we need for real looking waifu's)and with the recent tariff troubles they are only going to get higher. I think that there is no way possible to economically make a robowaifu if you buy all these actuators. So the article above may not be exactly what you need but it's a great good start and shows you can do this with found materials and some methods. I've been looking at synthesizing iron oxide nano-particles. There's a LOT of work on this. Some of it doesn't look extraordinarily complicated though I just started looking. I don't know all the numbers yet but the reason its so popular as a research subject is these have great permeability and a very high level of magnetic flux before they choke. This paper looks REALLY GOOD for making these. A quick preview I see no super odd chemicals needed or specialized equipment. "A Milestone in the Chemical Synthesis of Fe3O4 Nanoparticles Unreported Bulklike Properties Lead to a Remarkable Magnetic Hyperthermia" bulklike It uses PEG which can be bought as a laxative Polyethylene Glycol Powder for Solution (MiraLAX) and Iron oleate I don't know for sure yet but oleate is a surfactant so I expect you make the above from it and iron I don't think this will be a huge problem. None the less if you can start with regular black sand from the ground or streams and then try more powerful stuff. I only started looking to make this stuff when I priced permeable powders and they were high. China had them reasonable but the shipping cost alone was like $90 for a couple kilos. Now, way higher.

>>37780 >I think that there is no way possible to economically make a robowaifu if you buy all these actuators. Now you're getting it, Anon: actuators have always been the highest-cost single factor category of constructing pleasing & effective robowaifus (solely b/c the software coming from here is FREE [as in beer & speech: MIT-licensed (so, actually free speech! :D ]). This is one of the highly-important reasons why Kiwi, myself, and other cadres are so pleased with the ideas of Tensegrity : ( >>37672 ) for humanoid robotics. Amongst the many, many benefits this approach brings to us here on /robowaifu/ , significantly-reduced actuation costs (primarily due to significantly-reduced mass : >>4313 ) is a BIG one. <---> I like this guy's stuff, Grommet. Thanks for introducing us here to him! Cheers, Anon. :^)

Artificial Muscles from Fishing Line and Sewing Thread haines2014 abstract https://www.science.org/doi/10.1126/science.1246906 paper https://sci-hub.ru/10.1126/science.1246906

>>38110 Seems like we've been over this before, and it didn't seem to turn out positive. Didn't one of our guys actually do a fair bit of realworld experiments on this idea? Regardless, I'm sure the potential there is real. It's simply going to take a professionally-manufactured set of components to pull off well, AFAICT. The >tl;dr being this isn't well-suited to typical Anon's garage lab approaches.

Crystal Structure and Magnetic Properties of Fe3O4 Nanoparticles Using Iron Sand as a Raw Material for Mercury Removal https://jonuns.com/index.php/journal/article/view/1240/1234 The key here is not mercury removal but making nano magnetite (Fe3O4) from black sand you can dig up. Nano magnetite (Fe3O4) has very high superparamagnetism. Means that magnetic fields are channeled through it at a very high rate and the material has high saturation figures to these fields and it makes for very high power solenoids or electromagnets. An it's totally non toxic to boot. They use this for contrast in medical scans. People swallow it. Here's a paper that goes over, roughly, the main methods of making this stuff. Magnetite nanoparticles Synthesis methods – A comparative review IJCRT22A6681 https://ijcrt.org/papers/IJCRT22A6681.pdf It's a good brief review and if you are at all interested in making magnetic actuators, from scratch, you should download it. That and the paper using black sands directly above. It also has the references in it to learn the methods. There are many. A paper needed is a way to make a flexible rubbery like conductive material. Combine these and I think you would really have something.

I found a really, really good video on this, Making ferrofluid from scratch https://www.youtube.com/watch?v=6L8yUY-doNc

>>38166 Great information, Grommet! Thanks & cheers. :^)

>>37780 What application are you trying to synthesize them for? Are there any requirements for size or crystal structure of the nano particles? If so what are the tolerances?

>>38187 >What application are you trying to synthesize them for? Magnetic amplifiers, solenoids and electric motors Are there any requirements for size or crystal structure of the nano particles? If so what are the tolerances? They are typically looking at 5-20nm. For our purposes I suppose we could do with larger. A lot of research is for medical use. The magnetic particles provide high contrast for medical imaging equipment. Me, I want to tag along on their research. I will not say that I can write a symposium on all the effects of this but I get generally that they have very high permeability, they are able to sustain high magnetic fields without them choking and becoming a high resistance. As a general rule higher permeability means higher frequency and much smaller equipment with larger forces. Concentration and force per magnet field force is the big one. They also have a benefit that they can operate as one large mass, though I'm not exactly sure about all of the relevance of that but I believe it's directly related to the other traits I just listed and, I think it lowers eddy currents which are wasted energy. If you are making magnetic amplifiers from them you can use a much smaller (very small)DC voltage to control a very large AC voltage and current. This could come in very handy. They used these to control theater lights and battleships turrets in WWII. They are VERY robust. There's, I think, two different forms they're looking for. One is better than the other but I think both "not bad". I believe the one with less oxygen F3O2, a guess, is better, but I could be wrong. I have briefly looked at the papers but I can't remember all of the details off hand. The data is in the papers and in the video that I linked. The two forms have different crystal forms.

I may have it wrong. I think this is the form you want 2 Fe3+ + Fe2+ + 8 OH− → Fe3O4↓ + 4 H2O So more oxygen??? I note that the synthesis in the video he worried about too much oxygen as he was stirring upsetting the reaction to a less desirable state. I think it might produce side reactions that are none of the above. FeO maybe??? I;m not a chemist but took a couple chemistry courses so I know just enough to know, that I know very little. Like Barbie says,"Chemistry is hard". Here's links on these, https://en.wikipedia.org/wiki/Iron_oxide https://en.wikipedia.org/wiki/Iron_oxide_nanoparticle

If I could easily get powders online I wouldn't waste my time with this but the people supplying this are typically only interested in selling you tons and you have to sign up, talk to reps, it's a huge annoyance and on their part they don't care. They want to sell the already made cores not the powders. The ones who list just powders, with price, want an ASTRONOMICAL fee for them. It may very well end up that using black sands or some chemically made version of them is the best path even if the material is inferior. I wish I could get a couple kilos of Sendust alloy powder. That would be good. China will sell it to you but the shipping is a killer and now, with tariffs and trade disputes...not going to waste time on that.

>>38224 Nile red's video looks like your best bet for making them at home. If you are really interested in the biomedical applications you could try cultivating magnetotactic bacteria for their magnetosomes. You would have more control over chemical composition, size, and shape, but it is a slow and energy intense process and it is not a mature process. If you are making them on a small scale those might not be huge issues. https://microbialcellfactories.biomedcentral.com/articles/10.1186/s12934-020-01455-5

>>38232 >interested in the biomedical applications I have zero interest in biomed. Only noted it as this is the a lot of why they are researching this.

I found these brief articles. It's a great summary of why to use permeable materials for actuators, motors, etc. Very concise and short with a lot of info using graphs. https://www.horizontechnology.biz/blog/what-is-sintered-soft-magnetic-material https://www.horizontechnology.biz/blog/iron-powder-types-permeability Earlier I noted that one of the advantages of nano-materials is that they acted like one solid mass. I didn't really understand thta til I saw this i the article, "...However, what if your design requires higher perm and saturation? A logical transition is to the iron phosphorus material now with the benefit of advanced compaction technologies. The higher densities - up to 7.3 g/cm³ will promote higher perm - up to 4000 coupled with higher saturation induction..." So "maybe" the nano materials act like a compressed solid without actually sintering it into a solid, or so I read it that way. Maybe I seem to be blathering away a lot about this but in my opinion it;s at the very root of making efficient actuators. Now if a way can be found to make coils using these powdered materials without actually winding them up... Lots of resources here, https://www.horizontechnology.biz/soft-magnetic-composite-resources?hsCtaTracking=bf1123d2-dd65-4967-a84e-b0297046fa01%7C22279883-1ca0-4188-85a0-15819ac3a283

One more I find myself looking at. It's a coil gun link. Directly applies because it;s all about getting the most force from magnetic systems. https://coilgun.info/theorymath/saturation.htm

>>38277 >Maybe I seem to be blathering away a lot about this but in my opinion it;s at the very root of making efficient actuators. Nah, it's fine Anon. We all get it, I think. I'm hopeful that @Axial will return soon to pick up his thread again (cf. >>36271 ). Your investigations into these things may have bearing there. >Now if a way can be found to make coils using these powdered materials without actually winding them up... This. Cheers, Grommet. :^)

>>38294 >Axial I personally believe that any non return path, no permeable material, motor/actuator will always be inferior to the one with a return path. My understanding is it's like an electrical circuit and the permeable material is like a less resistive wire and a return path that has none, only air, is like a big ass resistor in a circuit. You will have way less magnetic force. If it;s such a good idea then why does every single electric car makers motors have permeable material to concentrate the magnetic field? Now it "may" be possible to use rare earth magnets and use purely attraction to the coils but REM cost a damn fortune and kill the price/performance level. Interesting factoid https://www.nextbigfuture.com/2025/05/nvidia-james-fan-proposes-physical-turing-test.html "1.5 million parameters is all that is needed to capture the sophistication of the human body for humanoid robot training." Anyone here know what that means exactly? A guess, 1.5 million bytes of data to monitor??? Or could it be nodes in an AI and if so what sort of compute would be needed for this. Anyone with a rough guess? Not a perfect guess but maybe one order of magnitude. I'll throw some numbers out there for fun. (not to be used for insurance claims nor do I accept responsibility for inaccuracies) A ESP32 microcontroller has 600 DMIPS. So at 1.5 millon parameters we have 400 operations per parameter and that's just a regular microcontroller. Looks as if one microcontroller could handle the whole entire waifu. Note that the whole body is not going to be moving so it could be less compute. Hell lets be generous and make one for gross body movement and one each for legs, head. You're still easily in range.

>>38312 >I personally believe that any non return path, no permeable material, motor/actuator will always be inferior to the one with a return path. I don't have sufficient training in EM/Field-theory (yet) to know any better one way or other, but the instincts of my mind's imagination tell me that this is the correct take. >why does every single electric car makers motors have permeable material to concentrate the magnetic field? We have a video linked here on the board somewhere about the remarkable stator, etc., design of the newer Tesla motors. It would be nice to know more about the manufacturing techniques used to produce them! There's also another video about the 'print-in-place' method used behind an even more-sophisticated, smol'r design motor (in response to your 'make coils:no winding' comment posted above). <---> >"1.5 million parameters is all that is needed to capture the sophistication of the human body for humanoid robot training." <sauce: bull's a*rse With all due respect to those esteemed researchers, I believe this is yet to be determined! :D We simply can't say something like this definitively yet, I deem. But what we can say definitively is that the eminent Carver Mead has done extensive research on so-called Neuromorphics (being the founder of the field back in the day) [1]. And what Dr. Mead, et al, says is that the "intelligence" of creatures gets 'pushed out' to the periphery of their body plans. In other words, for us humans, a very large part (by far the majority) of our own sensorimotor / kinesthetic "understanding" happens not inside our brains -- but within our peripheral, spinal-column + limbs neural tissue! Certainly such a model of reality seems to me (at least with merely a surface, cursory view on these matters; as yet) to warrant against using just a simple, "LLM-focused" discourse as the correct method to approach understanding these robotics C4 systems [2] topics with. /$0.02 . <---> As to your back-of-the-napkin estimate of the h/w power needed for our robowaifus to be safe & successful, effective & pleasing waifus... again, hard to say yet. OTOH, I'm rather-confident that we can, in fact, solve the basic kinematic needs part with a reasonably modest array of smol SBCs & MCUs + sensors, etc. [3] Its the very language/emotions & world-modelling/judgment problems themselves that may need an inordinate amount of compute to manage well, I think. Stay tuned. My understanding is that most of us plan to use some sort of home server setup for offloading robowaifu compute. Personally, I mean rather to have it all onboard in the nominal case; I'll be satisfied (at least at first) with having conversations not much more extensive than her saying "Chii?", "Chii!", "Hideki?", "Hideki!!" After all, teaching her about life & things is where all the fun is! :DD Cheers, Grommet. :^) --- 1. (cf. >>12828, >>12857, >>12861 ) 2. Command, Control, Communications, & Compute. While our needs here as robowaifuists go much-deeper still **, these are the commonplace basics we'll need to solve just to 'get off the starting block' with advanced, realworld robowaifus. ** (cf. >>10000, >>24783, >>17125, et al) 3. With the fundamental proviso that the software itself running on said hardware is highly-efficient, realtime-compatible, C++ & C code!

>>38312 >>38314 >Motors without permeable material to control flux paths They're fine, there's trade offs. Motors without flux guiding material (called coreless) can respond to changes in electromagnetic fields faster. This is useful in certain fields, such as RC cars and animatronics where high reaction speeds are crucial and budgets are small. They have far lower torque without flux focusing. Usually, this is overcome via having a larger armature diameter and/or gears. Efficiency isn't a problem, the flux from the coils is still used to push against the flux of magnets effectively. They can still have 98% efficiency if designed correctly. It's best to just think of it as, coreless has higher speeds and reactions, flux focusing motors have higher torque, relative to each other. They're both equally useful, just for different purposes. This train demonstrates it well. https://www.youtube.com/watch?v=G2r_Ucd2Fdw >Why do electric vehicles use permeable materials to control flux paths? Min-maxing torque per cost and mass. The physics governing them is different from what we are dealing with. (Unless you're building a waifu that's gigantic and needs to move tons of mass affordably.) Many cars such as modern Tesla's, use SynRM motors. They're a half step using flux paths so they can get away with smaller, cheaper, lighter magnets. It's all about being cheap while having enough torque to throw tons up a hill. For us, brushless is far superior and is cheap due to mass production. >How are they made Dies pressing metal sheets into shapes. High silicone steel pressed into many layers to direct flux fields. Shove wound coils and magnets where needed. Spice it up as you want depending on the car. https://www.youtube.com/watch?v=oVge8I6kxPY https://www.youtube.com/watch?v=-Uw0v6YaIc0

>>38322 >It's best to just think of it as, coreless has higher speeds and reactions, flux focusing motors have higher torque, relative to each other. Thanks! Makes sense. Sounds like -- as with all engineering -- its a matter of tradeoffs, and choosing the best in-situ approach.

>>38322 >Motors without flux guiding material (called coreless) can respond to changes in electromagnetic fields faster I do not believe that to be true at all. Some of the very fastest reacting motors are switched reluctance motors and all the rotor is "is" a permeable material.. .That's all it has in it. "...SR technology promises an impressive set of benefits over its competition. Among these are high efficiency over a wide speed range (and partial loads), high-speed capability (>100 krpm, with the proper drive), easy cooling with heat source only in the stator, ruggedness for high-temperature or vibration environments, and relatively simple mechanical construction (see “SR motor anatomy” online). ..." "...SR systems deliver high efficiency across the entire load range..." "...Cool-running SR motors also allow temporary high-current operation for peak torque output for severe applications..." I don't have the link offhand, it's somewhere, but I've also seen that SR motors can accelerate and stop super, super fast. Like 50,000 RPM and stop in milliseconds. . https://www.controleng.com/articles/resurgence-for-sr-motors-drives Here's a link I linked here, >12014 Tesla motor. Sandy Monroe associates who tear down and cost cars. He has a great electric car company comparison on their motors and drive electronics. The link again direct, https://insideevs.com/features/443775/tesla-motors-invertors-compared-competition-sandy-munro/ I've seen these taken apart. The picture kiwi put here, >38322 shows the general layout from the side. These are a bunch of thin iron, likely, high silicon plates, that are electrically separated by some sort of paint. Without separation they induce currents in each plate and heat up. The thin plates guide the magnetic fields. The spaces are stamped out holes. In some cases they have rare earth magnets in them or near by. A little speculation here. I know what the spaces are for. Since air has terrible permeability the spaces "guide" the magnetic fields to go around the spaces. I believe the reason they do this is the longer path is much like a lever arm. If it were too short less of the material would be attracted but the slots make sure it goes through a larger portion. Tesla has magnets in them too towards the center. More speculation. I say that the magnet also operate like the slots but serve dual purpose. It's a fact that DC magnetic fields, like in magnetic amplifiers, block magnetic flux. So I'm guessing the magnets extend the field like slots but also switched reluctance motors have a bit of cogging or jerkiness to them. I suspect the magnets help smooth this out so they can have less complicated electronics. The jerkiness or cogging can be eliminated by electronics but it gets tricky. I think the picture above is wrong. I see fields going through the slots... that doesn't look right. Maybe the area with slots have magnets in them and they are showing the magnetic fields lining up with the magnets??? If so they should have used different colors for those fields attracting the magnets and those going through the steel plates. It would have been more clear.

>>38314 >>"1.5 million parameters is all that is needed to capture the sophistication of the human body for humanoid robot training." <sauce: bull's a*rse With all due respect, I think this is yet to be determined! :^) Maybe I should have made it clear that "I" personally am not declaring this to be so. Only that I saw this link and the people proclaiming that are not necessarily idiots. Note I did ask just what these "parameters" might be. What "I" said might be needed I covered here, >22109 I came up with 240 joints and three muscles for each joint so I guess you could call it 720 parameters. Not covering the face so call it a rough 800 for a total.

I found this really good chart and it shows the type of electrical motors and the range of action based on "interaction torque", which I suppose is the coreless and straight magnet types, and the "inductive reluctance" torque with supposing the Switched reluctance all the way towards the reluctance side. https://assets.rebelmouse.io/eyJ0eXAiOiJKV1QiLCJhbGciOiJIUzI1NiJ9.eyJleHBpcmVzX2F0IjoxNzQ2OTEyMjYyLCJpbWFnZSI6Imh0dHBzOi8vYXNzZXRzLnJibC5tcy81MjUxMzg2Mi9vcmlnaW4ucG5nIn0.LiOyiRB1nxkTyB16vypaiM6TVXHBq-SkXYAu5Cz8uwY/img.png

>>38329 No worries mate, I knew that. <---> >Note I did ask just what these "parameters" might be. Which kind of feeds into my main point: treating the manifold physical aspects of a full-featured, natural-motion robowaifu as 'language model tokens' is fundamentally the wrong discussion to have, IMO. >I came up with 240 joints and three muscles for each joint so I guess you could call it 720 parameters. Not covering the face so call it a rough 800 for a total. All the interrelated finite-element-analysis + dynamical, floating center-mass, inertial-moment imparts of all these multi-nodal lever complexes (alone & in coordinated groups) would also need to be accounted for, to have asymptotically-complete descriptions of all the mechanics involved. >tl;dr In the millions at least, actually. <---> Thankfully, we don't have to have perfect simulations in advance. In the realworld, we can simply test + record results => adjust => lather, rinse, repeat. >ttl;dr HUMAN INSTINCT will take us 90% of the way there; science & good engineering will run final leg across the finish line. If it was good enough for my Grandpappy, it's good enough for me!! :DD

>>38331 >treating the manifold aspects of a full-featured, natural-motion robowaifu as 'language model tokens' is fundamentally the wrong discussion to have Yes, I agree. I had a thought. I took statics in college. It's mostly about adding all the forces up all over a structure and translating them to a fixed point or some cases a dynamic moving point. I see no reason the same can not be done. So each and every muscle/joint if rotated all about with varying force, in all directions, then stopping, recording the acceleration, deceleration and position. You get a force and a vector. Basically a calibrated value for how much it takes to move and to stop, with which momentum can be calculated. When moving, all these can be added up and if done correctly you can get the absolute mass force vector for any spot on the waifu. So as it's moving it can take, what's the last force vector(every joint, muscle, everything), add in the new force vector for all the parts(what you need to get to the next point) and you get the balance and position and you can calculate where and how much force to go where you need to be or if you are getting out of whack and need to correct before you fall over or hit something or etc. Now if you take my 240 joints and you have already run calibration on the waifu. You come up with something like a measurement of Kg-m/watt or how many watts to move the waifu where you want to go and where you have been so momentum is take account of. Each of these 240 joints would have a force and a X-Y-Z vector to show which way it was going(this would be a force-velocity vector in 3D). So 240*3=720 parameters + the same for the last reading to get your already moving speed-vector-force. This could also be calibrated by vision and force if it's walking when it hits the ground to see that the various inaccuracies will not add up to being completely discombobulated. While there's more to it than that it's not far off from actually describing what, I think, needs done. With such a small number of values and the blistering speed of today's microcontrollers I think you could check and correct these at a very high rate. So 600 DMIPS/(720 now + 720 last)=416,666 instructions for each second of movement. So every miliSec you could do 416 computations per joint including velocity vectors. That's a lot. I thnk you could set these force vector numbers up in such a way that you could just add them. I can;t remeber how we did it in statics but it seems like we did much the same, but with beams, buildings, etc. and just added up the forces and the X-Y-Z vectors. Maybe we took the sin of the force then added? Can't remember. I wonder how many instructions adding a vector would take.

>>38333 >I wonder how many instructions adding a vector would take. Once the data is moved into the registers, one.

>>38328 >Doubting coreless response to flux changes are faster than switched reluctance motors. I understand why you'd think that given your example. Coreless motors rely on torque interactions between permanent magnets and electromagnets. Reluctance motors rely on torque from reluctance, inducing magnetic fields into the rotor that then react to electromagnets. The material naturally has some reluctance, which is beneficial for controlling rotor position, but also reduces the speed at which the rotor can react to changes in the stators flux. It's a difference in speed that is small but, it exists due to the inherent physics of the respective systems. As for why race cars don't use coreless motors, the difference in costs between a coreless motors magnets and a switched reluctance motors electrical steel sheet rotor are enormous. The difference in performance at that scale using modern control algorithms is small. It's a simple cost benefit analysis lead decision. A coreless motor race car would perform just as well, but it would cost tens of thousands of dollars more, for barely any difference. >50,000 RPM and stop in milliseconds. These numbers aren't impressive for electric motors. That RPM is easily achieved by bog standard motors of all kinds. Almost all motors can also stop within milliseconds in a system designed for it. I understand how these numbers can seem impressive, they seem large compared to what we interact with normally. >Motor plates Those are made of electrical steel, a type of steel you correctly guessed is high in silicon. They have a strong dielectric separator. Which, could be a type of paint, resin, or other material. Paper can be used to separate them, I've seen it in big old motors used as pumps. >Speculation on flux guides They're actually really simple, they're there to keep the flux going straight from stator pole to stator pole. Consider a disc, without slots, the flux would permeate it evenly. This would prevent the rotor from having external torques rotate it. With slots, the disc now acts like it has many independent poles for the flux to travel through, providing a way for the torques to rotate. Consider picrel, because of the flux paths, both motors have the poles. The discs poles are physically larger, allowing for more flux and thus, more torque. The space between the poles is also smaller, reducing torque ripple and allowing for smoother operation. >Tesla The magnets in the slots are just there to provide magnetic flux like any permanent magnet motor. They're just small to save on costs utilize the designs of switches reluctance motors to get more power than the magnets would allow on their own. It's a really clever way to save heaps, because magnets at that scale are expensive. It's a cost saving half step between a switched reluctance motor and a permanent magnet motor. You are correct in that this kind of motors inherently has less cogging than a standard reluctance motor. The major problem for us is the lack of mass produced reluctance motors at the scale a waifu would use. They would cost less but, be larger and heavier to achieve the same power output as standard brushless motors. Which is why those brushless motors are common instead, the added cost is worth the higher power density at smaller scales. At large scales, the costs of materials adds up, making reluctance motors attractive. A larger motor and a few extra pounds, are worth saving tens or hundreds of thousands of dollars in manufacturing.

Magnetic shock-absorbers? Kind of 'anti-actuators' AFAICT. May be just the ticket for smooth robowaifus gaits & movements. https://www.tiktok.com/@super.accelerator/video/7487984992623004950

>>38336 POTD >At large scales, the costs of materials adds up, making reluctance motors attractive. You encourage us all, fren Kiwi. Cheers. :^)

>>38336 >The material naturally has some reluctance, which is beneficial for controlling rotor position, but also reduces the speed at which the rotor can react to changes in the stators flux Ok I do get that. Makes sense. >difference in costs between a coreless motors magnets and a switched reluctance motors electrical steel sheet rotor are enormous ehhhh...I don't buy that. Yes this used to be true. Every time you see any article on pure reluctance motors they always bleet on about the control system but you can get microprocessors for 50 cents now or ones with more outputs that are less per drive and the transistors to drive them cost much the same for either. I'm not saying it's not more difficult to set up but once you do, I think the cost is way less. If I'm not mistaken most appliances the major motors have all moved to some form of reluctance drive because...cost. I think there's just way fewer people that can easily design reluctance drives. I never said it was easy, so therefore, there are less of them. Most car manufacturers are,likely, moving towards reluctance drives with some, a magnet here and there. >50,000 RPM and stop in milliseconds. I am impressed by this. Most of the electric motors I have dealt with were larger and couldn't dream of doing that on their best day. >a type of steel you correctly guessed is high in silicon Not a guess. Some are made of different alloys but the cheapest and most common is high silicon iron. I "think" there may be a way to use the high inductance in the Stator. The idea is to have not coils but sheets of aluminium. By using pulses of electricity into them combined with high inductance (the sheets are mixed up with the iron sheets/powders) you could have a lost cost system easy to put together. Winding coils is time consuming, costly and tedious. If you notice I went on and on about Don Lancasters "magic sine waves". Basically pulses adding up in a circuit to give you sinusoidal waves.Or maybe I didn't. I thought I did but I did provide a link. >>27823

>>38336 >coreless motors magnets and a switched reluctance motors electrical steel sheet rotor are enormous I read that wrong you are correct.