>>34877 i got lucky and came across 1kg of iton powder for $3 i dont know what im going to do with all that iron though. i am aware that mixing iron with resin lowers its permiability however as long as its more permeable than steel its okay. The guy on the youtube video said he got good results but he used polyutharine i think not polyester and normally epoxy is used for inductors however i ended spending $15 for the materials so good overall. Since i got a lot of iron i can try dirt or clay too later on if this doesnt work and since nobody is selling ieon rods for electromagnets i might idk.

>>34877 That's a very neat idea, Grommet! I'm located in mountains, and I'd guess there's lots and lots of soluble iron content. I may try this the next time I'm panning for gold. Thanks, Anon. :^)

>>34881 taking diy to the next level are we lol. id say the most straightforward way is by collecting pyrite rocks as those are common and easy to spot. Theyre made up of iron and sulfur so you might be able to get both.

>>34882 Yeah good thinking peteblank. I actually had some Fool's Gold lying around at some point.

>>34878 It may be that the polyester is not so bad. I'm, slowly, looking into this. Iron transformers have a coating between the sheets. The reason is that as the magnetic field goes through the iron core currents are introduced, I think, perpendicular, to the magnetic field. I do know that currents are introduced that cause waste heat and does not help you actively. It's wasted. Newer ceramic ferrites have less of this because they do not conduct across the particles (I think). Here's a good paper on this and a quote, "...Several soft magnetic materials show promise for high-frequency operation. As oxides, soft ferrites stand out from other magnetic materials because they are insulating and therefore excel at reducing losses from eddy currents. However, soft ferrites suffer from a saturation magnetization (Ms) that is approximately one-fourth that of silicon steel. This substantially limits the power density in devices designed using soft ferrites and therefore limits their application..." https://www.science.org/doi/10.1126/science.aao0195 Now this saturation may not be a problem. I guess it depends. Is the current waste problem with iron worse than the magnetic saturation problem with the ceramic? I do not know the answer to this as it can get rather complicated. I suspect some guy whose been at this for a long time could give a quick answer. My plan is the same as yours, use ferrites (iron oxide) and mix that with resin. I'm not sure what the loss of permability with resin coating the iron oxide is. It maybe better to have the losses from eddy currents. I guess you could make a core for a solenoid and find out. I looked into some of these new wonder ceramic ferrites (Manganese-zinc ferrite (MnZn))but there's a problem. getting them. I can find them and they are not prohibitively expensive "if" you buy them 25 pounds at a time. Not cheap because who needs 25 pounds for prototyping. Otherwise they are difficult to get or too expensive in small lots. Likely the best plan of action, for me anyways, is to use iron oxide for prototyping which is super cheap and "if" it works out test some of this super whiz bang ceramic stuff for mass production where buying in quantity would be ok. Believe me it's extremely likely that ANY dirt you dig up will have a lot of iron oxide in it. One idea I had for the iron from dirt is to separate it with a magnet then grind between a couple bricks, it so that the spacing between the iron particles will be smaller when I add resin. It could be, but I don't know the answer, that aligning the iron particles with a magnet as the resin sets up might make it better for actuators. One more link. https://en.wikipedia.org/wiki/Ferrite_core

Found a chart looking for iron-silicon powder CHART: Permeability of Popular Iron Powder Types https://www.horizontechnology.biz/hs-fs/hubfs/iron%20powder%20chart.jpg?width=2063&name=iron%20powder%20chart.jpg An interesting thing on the chart. Pure iron powder is not as good as silicon powder. Pure iron permeability (meaning how much magnetic field it will allow with less resistance) is REALLY high for pure iron but this must only be for continuous sheet. Interesting.

A link and its a good one for understanding. Look down at the chart for permeability vs saturation. Regular iron is not so bad at all. I'm not sure this can be said for iron oxide but I do believe "roughly" that it's not that bad and will be perfectly suitable for testing. https://www.e-magnetica.pl/doku.php/coercivity So I looked it up and "may" have an answeron the iron oxide. It's from the paper, "Fabrication and comparative study of magnetic Fe and α-Fe 2 O 3 nanoparticles dispersed hybrid polymer (PVA+Chitosan) novel nanocomposite film" The chart https://www.researchgate.net/figure/Frequency-dependent-real-permeability-of-iron-and-iron-oxide-nanoparticles_fig6_325786171 This looks good. Looking at this I could see a finely ground iron oxide in some sort of glue matrix could be a very successful material for us.. It does very well at low frequency, which is fine for us. A lot of papers are more concerned with high frequency stuff for microwaves. So if you grind your oxide you get from dirt repeatedly then wash in a plastic bottle, only collecting the magnetic particles that stick, you could easily end up with nano-particles of the stuff you need.

>>34916 Two items arrived. all thats missing is the polyester silicone oil and the acetone. The game plan as far as the skin is mixing a little bit of acetone, a little bit of silicon/e oil and for the pigmemt i also got a little of part b silicone rubber, the part a bucket spilled if the skin works that could be a big deal because even if its not entirely a skin doll surface there might still need for the bottom, er chest snd hands. with ecoflex or dragon skin thatd be cost prohibitive. 1 gallon of ecoflex is ~$260 one galon is 0.134 cubic feet. one gallon sounds like a lot. i said before id divide the volume by five to guesstimate the amount if skin. 1/5=0.2 say we could get away with 1 gallon per foot. if the robot us 5 feet tall*1foot depth*1foot length ao roughly 5 gallons. 260*5=$1300 right there. All in all is not great but not terrible if you consider the amount people are prepared to spend on their nvidia cards ff. silicon caulk is much cheaper by comparison at 280 ml itd be $27 per gallon. 5 gallons is $135 so 1/10th the price As far as the eyes it could be more convinient to make screen eyes really. but maybe for the lips it could be more useful. Not that it wouldnt be useful if the eyes work with the electromagnets. might as well finish them since i printed it... The polyester will also be useful for making the negative part of the mold. i got 1kg so 1 liter. The mask positive print i have right now is 14*12*4 cm=672 cm^3 1 liter is a thousand cm^3 so it should be enpugh to make the negative part of the mold barely. Anyways using electromagnets for the lips... It could steel plates sorrounding the lips. rough estimates since from shoulder to shoulder itd be really 2 feet but you redo the geometry.

>>34918 Here is my idea for the lips For the letter O which none have solved as far i can tell youd turn on the electromagnet in the middle.

>>34919 a blown out O maybe, are they all magnets? i dont think thats feasible, it will have to extremely strong and the points of attraction are too close together to have any specific control over, each of those magnets will do the exact same thing

>>34919 That's probably a good idea, peteblank. And for the entire smooshy facemask, not just the lips. Cheers. :^)

>>34924 yes i underestimated how force required to move the lips. on my current mask... i just measured with the dynometer moving the lip down and it took 4 netwons of force. The better appeoach woukd be to use neodyum magnets. Neodyum magnets generate 1.5 tesla which is more than 4 newtons according to my guesstimemates. Theres also premade electromagnets but those consume 12 volts and might be overkill. Copper is not normally magnetic but it becomes magnetic if electricity i applied to it. the electromagnets are also pretty bulky.

>>34934 oh that makes more sense, like you would not only be able to turn on the magnets individually but also the points of the mouth that would be pulled by the outer magnets, thats smart but still really complicated, copper doesnt become magnetic it becomes a magnet, a really weak one, you would have to make them into coils like the electromagnet but without the core so that theyre not attracted when not on otherwise the whole mouth will move, and make sure the polarity is correct with the outer magnets otherwise it will repeal instead of attract

>>34936 actually you can just reverse the polarity if you reverse the current so they can push/pull the points on the mouth, you just have to make sure the poles of the inner and outer magnets are facing each other

>>34936 i could use a plastic core i think

>>34942 there must be alloys that are good cores that arent magnetic, i know magnesium and aluminium arent magnetic so maybe try a pencil sharpener lol

>>34943 okay then aluminum core. i dont like magnesium because it might wear down with the heat and aluminum screws are cheap and easy to find either way.

>>34944 i was thinking about alloys, the benefit with pure aluminum is too small to care about for a core, stainless steel isnt magnetic and way better its like 500 but i dont know what kind, some of them are apparently magnetic

>>34946 so yeah like most people these days i asked chatgpt and chatgpt said it coukdnt be done. But i came across this https://hackaday.com/2024/03/27/sort-of-electromagnet-attracts-copper-aluminum/ so is a little more tricky than i thought. The two aluminum electromagnets need to generate opposite alternating currents. However that implies itd also need an oscillator circuit to generate the ac. So this idea is turning out to be more complicated than i thought. 4 newtons of force is not an incredible amount of force. However if the aluminum electromagnet were used itd have the advatange of being able to be pulled and pushed...

>>34947 well no an arduino can generate ac with pwm too i guess.

>>34947 thats clever, its the same thing youre already going to do though, its just two electromagnets but one is made by inducing a current in the one that doesnt have any current in it by making the magnetic field go back and forth to induce a current in it (which makes it create a magnetic field that can be attracted), thats the only reason you would need an ac current to cause induction, but you can already just make both of them electromagnets without having to induce a current, you just need to make them strong

>>34949 yeah too itd be too much effort to use tge aluminum idea. Here are two calculators related to electromagnets okay so my phone is not letting me copy paste okay. Anyways. There are two calculators that are relevant two this in omnicalcukator.com the solenoid formula which outputs tesla and lorentz formula which outputs newtons i know my lithium batteries are 4 volts and 9900 mah however i dont know the current i think i broke my multimeter trying to measure a live wire on the wall switch So im just going to go ahead and assume 100 turns and my batteries ought to be enough to generate the required force

>>34863 Not sure if this belongs in the thread here or if it should be in >>154 anyways. Though, since this is about making a electromagnet, I guess it's right here, but maybe I will add a crosslink later. However, I asked Claude on the topic. According to that, first it claimed that pure or not pure iron would not make such a huge difference. Good iron cores would best be bought from a scientific supply shop and also it would be good to have these iron cores laminated to reduce eddy current losses. Low carbon steel would be a more accessible alternative. Pure iron would be 200-300% better. Cobalt alloy would be in between. However, coil winding, core geometry and other factors are also important: U-/E-shape or better toroidal, lamination between layers, length and diameter, number of turns and pattern, ... >>34866 Everything you do is always so wild that people don't know if this is trolling or not. >>34867 Okay, maybe it is doable, but ... >>34873 Yeah, especially for small ones it's probably better to just buy the magnets on Ali-Express? Or just make simple ones out of bend nails? >>34877 Okay, Doctor Stone. Or we just buy it? Unless we live in a slum in Pakistan or India. >>34942 I somehow lost you here. You also want something that's not magnetic?! >>34918 >ecoflex or dragon skin thatd be cost prohibitive This is probably fine for prototyping, but medical silicone is more safe. The one for bathrooms contains anti-fungal ingredients which might be unhealthy. Especially zinc and silver compounds.

>>34955 ty ty ideally the skin would be food grade however youre not going to be coming into contact with the skin all day unless youre a giga coomer. If it were mass produced i think injection molds and tpe would be the best way i think. Unrelated but for the ai i like the current ai i dont think it needs to be scrapped. chatgpt is a language model(llm) which is a subset of machine learning. Machine learning can adapt and learn. My plans for the movement is mostly preprogramned and correcting utself with sensors. Ill use ai for visual recognition and llm. i said id make a small scale robot next year and take a concious break from this for the rest of december but idk.

Yikes what happemed here? Could be theyre getting scared. im not scared because all my files are on microsoft onedrive anyways so if they p0wn me i can just reinstall. Anyways here is the diy silicone rubber https://youtu.be/xoFGHcsb1_Q

>>35026 >>34962

>>35027 finals? did chobitsu start this site as a teenager? This site has been up for 9 years chobitsu if youve been in college for 9 years drop out

>>35028 >chobitsu if youve been in college for 9 years drop out A MAN CAN DREAM, CAN'T HE!? :^)

the fifth time is the charm. couldnt show the screw on lid cause its hard to hold the phone camera and do this at the same time.

>>35156 okay im just going to scale this up to 3 cm diameter by ~6 cm length and trust my eye as far as the ratios. Here goes the 6th and probably last attempt before i go do something else.

>>35159 I thought I had seen the idea of magnetically positioned eyes before- NSFW https://dollforum.com/forum/viewtopic.php?p=1749667#p1749667

If you want eyes, just use something like this with silent actuators. https://www.thingiverse.com/thing:266765/makes

>>35165 >>35163 in the last video i mentioned that id skip the robot eyes and that the iron core would be a problem. i came this far though so itd be interesting to see it in action. like how fast would the eyes move uf pulled by an electromagnet. i also mentiomed using electromagnets for mouth movements. But yeah this is mostly a waste of time tbh. ill be neglecting the aesthetic side of the small scale robot. i just want to test to see if it walks ir if it falls etc... but unlike the result the process is boring and i set the deadline to late january so im not in a rush... the thicc wiifit trainer is my reference. So yeah that has to be chopped and turned into robot parts. boring. Also have black and yellow petg filament.

>>35165 >related https://www.youtube.com/watch?v=htWWpMQlaSY

>>35156 >the fifth time is the charm. Nice work, peteblank. Persistance is the key here. Cheers, Anon. :^) >>35165 Nice choice Anon! Looks really compact/inexpensive/manufacturable. Total win, Kiwi! Cheers. :^)

>>35186 yeah i gave up on making an iron core oh well

>>35200 Heh, that could be a good idea to IMO. OTOH, a) that was your choice (again probably a good one). And, b) doesn't in any way negate what I said, tbh. >tl;dr Find something more productive to pursue next time, Anon. Good luck, peteblank. Cheers. :^)

>>35204 it technically worked. its so powdery it should be mostly iron. im surprised at how good the steel screws turned out...

Galatea prototype. More proportional arms, and new legs and processing unit are coming soon.

>>37375 I love these smol, practical, incremental improvents you're using GreerTech. This will prove to be the key to your eventual success in a big way, I predict (cf. Gall's Law, >>33553 ). P.S. I'm really glad you chose blue-white for dear Galatea's eyes, Anon. They look cool. Cheers. :^)

++reposting bunker content++ Here's a free body for everyone. https://files.catbox.moe/cfzf8f.obj

>>37661 Wow! Those are hot enough that maybe you should spoiler them here Anon? :DD Sauce on these please, Kiwi?

>>37662 I am the sauce That's how I can guarantee they're free

>>37663 I AM THE SAUCE

>>37663 >I am the sauce Excellent! I was hoping you'd say that, Anon. I figured you were. :^) Then I'll plan to try scaling/fitting my design mocks for hips & shoulders + actuators into it's volumes to see where I stand with your shell sculpt.

From [Unusable board link - edit later] A good lesson for the prototypers here, if you intend for your design to be replicated, make sure the third-party devices have multiple sellers, the more the better, so that if one goes down, your design isn't instantly impossible to build. Even then, you should have backup designs in case the exact model is not being sold anymore. I learned that the hard way with a design in 2023, and with Galatea I have had it happen twice, once before Galatea v1 with the speaker, and recently with the Pleasure Unit, the latter because it had some other sellers, but not too many. TL:DR, with third party devices in open source designs, make sure there are several sellers easily found, across multiple websites.

>>37666 My words turned out to be prophetic, since these new Tariffs will raise the price of certain sources. I myself am slightly modifying Galatea to have a "Tariff-proof" design.

MaidCom0 is coming along well. The plan is for her shell to be 5 pieces that fit within 200mm^2 build plate area. Though, I could get it to 175mm^2 without her ass and tiddies. Everything else fits well within 175mm^2, so I'm considering making them add-ons. To keep myself on track, she will be limited to the feature set of Femisapien scaled to 150cm with skates rather than walking. I find the limitation both frustrating and inspiring. My inner engineer always wants to push boundaries, but they're also why MaidCom is years behind schedule.

>>37668 >MaidCom0 Oh cool!! I personally think there's a lot of wisdom in this approach, Kiwi. The old adage: >"Start smol, grow big." is just as true today! :^) Keep moving forward. Cheers.

++reposting end++ >>37668 Nice!