The knot is known for never coming "loose"...

>>39047 I decided to do some testing with PE line on the zeppelin knot and the grapple hitch out of curiosity. These performed very poorly, much more so than I anticipated. I tied both knots with the heaviest braided line I had on hand, 40 lbs test YGK XBRAID UPGRADE X-8. The zeppelin knot pulled apart under maybe 8-10 lbs of pull ( I did tie small overhand stopping knots). The friction of the knot pulling apart melted the surface of the line. This was impossible for me to capture on camera, but the line feels crunchy and the outside surface has fused a bit. A very bulky and complex stopper may have held it together, but that would be a bit self defeating in terms of effort. The grapple hitch was a complete failure. The knot, no mater how tightly cinched, will slide up and down the line under strain. As it did so, it frayed and destroyed the line.

>>39049 I tied another zeppelin knot with large dropper loops as stops on the tag ends. I made sure to moisten and properly set the knot before pulling on it. The knot slipped under light loading again, but the larger stoppers functioned. Unfortunately, the knot slipping still melted the surface of the line and caused it to fail at what I would guess was about 15 lbs of pull.

This thread -- and particularly @Fishing anon's inputs -- have encouraged me all the more that pulley-and-cord based approaches to distal actuation (elbows/wrists, knees/ankles) is the correct course for my design work [1][2]. I hope to soon begin some new prototypes with UHMWPE cords & low-friction plates, combined with aluminum tubing & struts/etc., assembled with 3D-printed fixtures/joints/etc., into a basic humanoid skellington prototype frame system. In the later stages of assembly for it, tension everything down, tightly-knit-together using tensegrity-'alike' approaches (using more PE cording). Its not "pure" tensegrity maybe, but many good benefits will still be there for us, I think. (Plus, it will be much simpler -- assembly-wise -- to just 'strap down' everything together using tensioning lines, after the fact. Throw a few [4-6, maybe?] strategically-located aluminum turnbuckles into the mix for simple re-tensioning maintenance further after the fact, and Bob's your Uncle! :^) <---> Should that go swimmingly, then I'll probably move from thence onto using so-called LARP-foam (EVA) to 'bulk out' her volumetric appearance(s) in very low-mass, comfy- feeling/looking ways. --- 1. Hey, it was good enough for Leonardo da Vinci!! (cf. >>37689, >>37679, et al) :DD 2. Because of the 'Thrown Weight' problem. >tl;dr Keep the actuator/battery masses located as close to the pelvic/thorax volume as possible.

>>39054 If you don't already have UHMWPE line for the pulley system, I would suggest going with something like the pictured 200 lbs test AFW Downrigger Braid. It should be more than strong enough to resist any realistic stresses. There are definitely stronger and/or cheaper cords not marketed as fishing lines, but they appear to be much less tightly weaved. I have not had any in hand for examination though. I assume (very likely incorrectly) a loose weave could cause durability issues by staining portions of cord unevenly in a moving system. As an alternative to using distal pullies, maybe look at incorporating fishing rod guides. Small aluminum oxide ones for casting rods would probably work well and are very cheap. The double footed varieties would be more sturdy. You could even use some of the line and super glue, epoxy, etc. to fix their feet in place on the skellington.

>>39060 POTD Thanks Fishing anon! I'll buy some of these things soon-ish to begin prototyping investigations with them. Cheers, Anon. :^)

After looking at random products for ideas, I believe there is potential in providing adjustable tension to a PE support system with an A-Just-A Bubble or an imitation of it. I believe you could pass a loop of line through it and adjust the length of the loop to set your tension. It might be possible to tension multiple line with one device as well. This would be both cheap and extremely light. https://www.youtube.com/watch?v=wxLITFtpPmI

>>39050 I stand corrected. I really love this knot it's done me great service and due to it's construction, the way it circles and clinches I had no doubt it would be magnificent but...apparently I'm wrong. Maybe the turn is too tight. Out of curiosity could you do the same test, with the same line, with a FG knot? So I look up PE and find that "supposedly" PE 40 pounds should take 400 pounds force. I notice the line you show is 50 pounds???? Was it another? At some time in the future I wish to tie a short line onto a longer line so that I have several short lines at intervals along the longer line. What knots do you suggest for this. FG knot? Couple others I like because they're easy to remember. Blake's hitch (for one way pull only)(you could use many loops to spread the stress)(I like this one too) https://en.wikipedia.org/wiki/Blake's_hitch https://en.wikipedia.org/wiki/Prusik_knot For some reason I look at the FG hitch and think,"I'll never remember that". Solely my failing. Some knots or hitches just seem natural to me and easy to remember. Blake's and Zeppelin, adjustable grip hitch are some that just easily stick in my head and "I assumed", and we know how dangerous that can be, that they worked equally well on all types of rope.

>>39054 >Throw a few [4-6, maybe?] strategically-located aluminum turnbuckles into the mix for simple re-tensioning A idea. If you can monitor the angle of the limbs, whether the cable is loose or tight makes no difference. A small amount of stretch means you just pull the actuator a little further til the angle is correct. Deletes a part.

I got the info on PE rating from this page https://www.oceanbluefishing.com/magazine/pe-rating-fishing-line-diameter-chart/ I think a fail written by an AI that confused things as their charts don't line up with what they are saying. I hate all these AI pages and pages and pages that show up first in search engines that often have faulty info. William Casey (CIA Director), "We'll know our disinformation program is complete when everything the American public believes is false." This was supposedly verified by a member who was in the cabinet meeting when he said it. It's a vile crime that US government organizations should have disinformation fed to the US population as one of their goals. Though I think we need intelligence services I sometimes think it would be better to not have them at all, even with all the deficits of this action, if they are going to constantly try to discombobulate the populations minds.

>>39049 Thanks for testing this it's very helpful.

>>39066 I was also looking at corsets and hoop skirts for squish-able body design ideas. I imagine a soft torso, or maybe a thigh, could be based on the hoop skirt. PE lines anchored at the top and bottom could suspend springy plastic bands of Curv, Tegris, or some similar plastic product. >>37700 I'm not sure what the best way to bring the ends of the hoop together would be. My fist thought was feeding the ends trough some kind of keeper and anchoring them with shock cord so the hoop could expand and compress. >>39067 >>39069 >>39070 When knots in PE fishing line slip, the friction destroys/melts the surface of the line, causing it weaken and fail. To form most fishing knots correctly, you actually need lubrication (spit or something) to keep the line from destroying itself. >PE 40 pounds should take 400 pounds force >I notice the line you show is 50 pounds???? Was it another? For fishing line ratings, it would defiantly be only the listed 40 lbs. The 50 lbs spool pictured was the only image I could find. The line I used came off of a reel and I no longer have the original spool. >At some time in the future I wish to tie a short line onto a longer line so that I have several short lines at intervals along the longer line. What knots do you suggest for this. FG knot? If you are talking about adjoining the short lines to the main line at a perpendicular angle, I would suggest not doing that at all and using only one line. Try looking at a dropper loop knot. You could just make VERY long loops: https://www.netknots.com/fishing_knots/dropper-loop . If that is not appropriate, I don't have any meaningful suggestions. The two knots you listed should both function. Adding more wraps, maybe 10 or so, to the Blake's hitch would help. You would need some additional knot to create the loop for the Prusik knot. I would try a King Sling (skip the threading it through something part) : https://www.netknots.com/fishing_knots/king-sling . You could also loop the Prusik knot around the main line additional times to keep it from sliding up and down the main line. >Out of curiosity could you do the same test, with the same line, with a FG knot? I have no heavy leader lines available to appropriately test an FG knot. The knot is for really only for combining a PE main line to a leader of fluorocarbon or monofilament. I'll try one of braid to braid in the morning, but I expect it will pull apart easily. >>39068 I assume he means in the static structure?

>>39072 >>39071 Whoops, no .png files huh?

>>39067 >>39071 When knots in PE fishing line slip, the friction destroys/melts the surface of the line, causing it weaken and fail. Sorry, that was unclear. The Zeppelin knot failed because it allowed the super slick line to slip in the knot. If you modified it to multiple have extra passes in the pictured step, it might do better.

I tested a braid to braid FG knot and a modified zeppelin knot with 6 twist per tag end. The FG knot eventually pinched the leader end off at maybe 70-80% breaking strength. The end of the knot that broke fused together from the heat. This was better than I expected. The modified zeppelin knot failed almost instantly. It might have held 3-4 lbs of pull.

>>39071 >>39073 >hoop-skirts Great idea, Anon! This has been discussed on /robowaifu/ before, and I'm still convinced its a good idea today. I personally see a kind of 'tensegrity cage' (probably also including your hoops at the outer perimeter) underlying an intentionally-designed shell of LARP foam to fill out the silhouette of our robowaifus. Again, good thinking Anon. Cheers. :^)

I love these hand-drawn graphics ITT! GG, Anon.

>>39066 This is excellent, but the outer 'bobber' shell of the pictured device is far too bulky for our purposes here on /robowaifu/ , I think. Any chance we could ditch that shell, and just use a 'spindle' (or even narrower) mechanism to turn for tightening up the slack? (Preferably, this process would simply be a hex tool inserted right through the shell, then rotated for re-tensioning maintenance...no shell disassembly required.)(or even better : fully-automated retensioning, completely internally! But let's all just take one step at a time here, eheh. :^) <---> The issue is the extremely-tight volumes around the hips & shoulders where several of these (w/ at least 4 for each leg & arm) need to be wedged into place... given the distal actuation requirements -- eg; both of the knees and further of the ankles (w/ 4 PE cords [2 cords per tightener] for each of these joints [extensor + retractor; x2 ]). Combined with the already-tightly packed volume of 3 major actuation systems for, eg, the hip itself (x2 ), plus the further 4 smol'r actuators (x2 ) for the knee/ankle also nearby -- then these big bobber shells are deffo a nogo right in that same complex, proximal volume.

>>39082 I was thinking something like that could be used in the torso to tension lines for "shell" parts. Something like >>39071 might be able to be able to have independent adjustable tension for all the lines with one device. I don't think it would be useful in the pulley system lines as Grommet suggested >>39068 . As far as I can tell,the bobber is just using a twisted latex rubber tube. Anything that could hold one end of the tube stationary while allowing you to rotate the other and lock it in place should work. A "U" of tube fixed at both ends might be more practical to make. You could feed the lines through and then twist the "U" and hook it over some kind of peg.

>>39085 >>39082 Couldn't the same cord and motor be used to extension and retraction if you used a rubbery pulley and wrapped the line around a it a few times? In any case, I think you could adjust tension by using bolts as the anchors and twisting them. It would be almost like tuning a guitar. I would suggest getting partial threaded socket head cap screws, wrapping the non-threaded part in electrical tape, tying the knot form the video, and gluing the knot all the way around the bolt with super glue. Thread looker would probably be needed. This would allow you to tighten the lines with a hex key. https://www.mcmaster.com/products/socket-head-cap-screws/socket-head-screws-2~/threading~partially-threaded/drive-style~hex/ https://www.youtube.com/watch?v=cghRdGdkLPg&t=216s Using something like a small eye bolt for the anchor bolt would easier to do for experimenting.

>>39082 So many typos. : (

>>39087 >So many typos. : ( LOL! Its said that people edit their documents to fix the errors. By all appearances, I, Chobitsu, edit my posts to add them!! :D >tl;dr I've always needed a good editor, Anon. If you're willing to post a correction of my post for me ITT, then I'm willing to go back in and fix some things with it. Cheers, Anon. :^)

>>39085 Excellent! This is a great concept about the 'U-tube' locking peg. Yes, I consider it will be needed to (re-)tension not only the shell, but the primary torso framework itself. Like @Kiwi, my low-mass goals are pretty significant (mine in particular are likely even more austere -- think: the 'human-powered flight' planes [1][2] ). Having components that lightweight will benefit from being 'strapped together' internally (much like the functions fascia [3] serves within humans' musculoskeletal systems). >issues w/ pulleys Yeah, the tightener mechanisms will need to be situated in such a way they don't interact with the guide-eyes/pulleys at all. Nor with the actuation drive-rotors. My presumption here is that we can devise some sort of 'guide rail' system, that keeps the tensioners both out of the way, and allows them to slide along with the PE cords during actuation. Maybe the rails could be situated within, eg, the thigh volumes? >redacted : (cf. >>39095, >>39101 ) Good thinking, Anon...nice graphic BTW! Thanks & cheers. :^) --- 1. https://en.wikipedia.org/wiki/Human-powered_aircraft 2. https://www.youtube.com/watch?v=Nug0Oho-b1k 3. https://en.wikipedia.org/wiki/Fascia

>>39086 >Couldn't the same cord and motor be used to extension and retraction if you used a rubbery pulley and wrapped the line around a it a few times? Ahh, the ol' 'double winding spool' trick, ehh (think: two spindle spools rigidly-attached axially...one wind goes one way, the other wind in the opposite)? :^) Yes, I think it's likely a feasible approach. There will be 'winding-rates going out of sync' issues if the changes in 'spool lay diameter' of the wound filaments isn't properly managed somehow. (OTOH, with the relatively-short throw distances involved with a humanoid's limb systems, this is likely negligible and easily-resolved in software.) This dual-purpose 'extensor/retractor' on a single actuator drive rotor is certainly something on my plate to investigate when time & resources allow me to. Good thinking, Anon. >In any case, I think you could adjust tension by using bolts as the anchors and twisting them. It would be almost like tuning a guitar. That is a great analogy! Do they make anything like a 'torque-wrench hex key', do you know? Guitars can use sound as the feedback mechanism for judging proper tension. We'd need a useful way to do the same for this scenario as well. >I would suggest getting partial threaded socket head cap screws, wrapping the non-threaded part in electrical tape, tying the knot form the video, and gluing the knot all the way around the bolt with super glue. Thread looker would probably be needed. >This would allow you to tighten the lines with a hex key. Yeah, that would be perfect. I presume we could devise some sort of 'neutral pose maintenance access panel' that would allow us to go in and retension all the lines at the same time (one panel per limb)? >Using something like a small eye bolt for the anchor bolt would easier to do for experimenting. Good thinking. Anything we can do to simplify testing/manufacture will make the entire set of processes easier on everyone! Great ideas, Anon. Your graphic really helps, too! Cheers. :^)

>>39091 >Ahh, the ol' 'double winding spool' trick, ehh Just a normal pulley or roller of some kind. The blue direction would be extension and the red would be contraction for a particular joint. Some kind of guide would be needed to keep cord spaced out properly on the pulley/roller. The line would never wind in multiple layers on the spool in this layout. >Do they make anything like a 'torque-wrench hex key', do you know? There are small torque wrenches like picrel that use 1/4" bits. There are also small, in-line, torque drives and limiters also based around 1/4" bits. I'm sure there is more variety on these, but that is all I am aware of. >Yeah, that would be perfect. I presume we could devise some sort of 'neutral pose maintenance access panel' that would allow us to go in and retension all the lines at the same time (one panel per limb)? For this to work, the screws would probably need to be at the extreme end of each cord system. A tiny access hole for each screw could be made directly above or below them in the outer shell. If you used an appropriately long screw, It could stick out away from the skeleton and be flush with the exterior. >>39089 >the tightener mechanisms will need to be situated in such a way they don't interact with the guide-eyes/pulleys at all. Nor with the actuation drive-rotors. My presumption here is that we can devise some sort of 'guide rail' system, that keeps the tensioners both out of the way, and allows them to slide along with the PE cords during actuation. Maybe the rails could be situated within, eg, the thigh volumes? Using fixed adjusters as anchors at the distal end of the cord would be much easier to set up and it would cut down on the number of parts and weight.

>>39071 >If you are talking about adjoining the short lines to the main line at a perpendicular angle, I would suggest not doing that at all and using only one line A little off topic but to make it clear I'll tell you what the point was. I like sailing videos. I saw this one group that caught next to no fish sailing across the Atlantic while another group caught all sorts of fish doing the same BUT the first group only had one hook and artificial bait on the line. The second, who caught all the fish, had several in a row attached to along line. My guess. The first was going to fast for the fish so they missed it but the second could see one go by then see another coming up and they would bite. The dropper loop looks good but, for me anyways, I think the Blake, even if not as good, suits me because I can easily remember it and the knot I can remember is better than, no knot at all, no matter how good it is. I'm not sure why this is but some knots, for me, just stick in my head while others I constantly struggle how to tie them. Don't know why. Thanks for your reply.

>>39074 I don;t get that picture. Looks like a regular zeppelin to me BUT if you are talking about making extra loops for the B and Q, then I actually thought of that also but didn't say anything about it.. Clarify. You make a "b" and the "O" part of the "b" . and the "q", is two or more loops before you run the line through the center.

>>39076 >FG knot...70-80% breaking strength....zeppelin knot failed almost instantly. It might have held 3-4 lbs I;m guessing the sharp turn, with small line, put more stress on it. The FG with more loops and straight pull, far better. I wonder. Could a zeppelin be better with larger ropes, like used to hold an actual airship, because that is the knot they used for that task.

>>39089 >tension not only the shell, but the primary torso framework itself I see, said the blind man. I was only thinking about the limb actuation, not body parts.

>>39095 Ahh, OK, now I've got it. Excellent! Heh, I think my obsession surrounding thrown weight was keeping me from seeing the idea of putting the tensioning screws at the far end of the rig. Makes sense now. Maybe we can find some kind of specialty very-lightweight screws for this. Isn't magnesium really lite but strong? Also, glad to hear about the smol torque wrenches/limiters. Thanks again, Anon. Your diagrams really help! Cheers. :^)

>>39099 Heh, when you're shooting for <25Kgs of mass, you need to get inventive!! :^)

>>39097 >I don;t get that picture It's just from the website you shared. I added more passes of the tag end through the "O" of the "b" and 'q" in an attempt to add enough friction to keep it from moving. >>39098 The zeppelin bend is made to be untied for mooring lines, etc. and simply does not have enough static friction to keep the knot from sliding apart with the very slippery PE fishing line. It's not exactly breaking when it fails: it's either untying or melting. If the knot does not simply pull apart, the fibbers sliding across each other under tension heat up enough to start melting, resulting in a snap. Rope would not have the same issue as it has a much higher friction coefficient. It wouldn't start sliding, so it wouldn't damage itself.

>>39107 >It wouldn't start sliding, so it wouldn't damage itself. From what I understood of the chemical/physical discussions of UHMWPE, the reaaaally long chains create a tremendous number of covalent bonding sites -- thus the strength. But that the polymer chains themselves were particularly prone to breaking down at a relatively low temperature -- thus the melting issue. Please correct me if I'm wrong here.

>>39101 >magnesium Magnesium screws like the one pictured are used in some orthopedic surgeries as an alternative to titanium. Magnesium embedded in bone is actually bioabsorbable. I'm not seeing much out there for other applications, but Maruemu Works and Melfast appear to make magnesium socket head cap screws. I can't find any for sale though. Maruemu's stuff might be available though some Japanese retailer and Melfast appears to only do bulk commercial sales. Aluminum would probably be the way to go, if you think the higher cost is worth it. https://www.mcmaster.com/products/~/material~7075-aluminum/material~2024-aluminum/?s=aluminum+socket+head+cap+screws >>39108 >thus the melting issue. This is very much the case. There is a relatively large amount of surface area on the braided line, and the friction of dragging it over itself can quickly bring it to the 270 °F or so point it starts melting at. >It wouldn't start sliding, so it wouldn't damage itself. I didn't intend to imply rope would fail in the same way. If a natural fiber rope slid in a knot, It believe would start peeling itself apart and shedding fibers until what was left it was week enough to snap.

>>39124 >Aluminum would probably be the way to go, if you think the higher cost is worth it. Certainly that's fine for prototyping work ATM. Bookmarked. <---> Now let's move on to the rods/struts. I'm planning on aluminum tubes, simply b/c that easily-obtainable at 'big-box' hardware stores here in Burgerland (Lowe's, Home Depot). But what I'd really like is carbon fiber tubes that are very strong & slightly compliant. Any leads there? (Please remember I'd like our lifesize [~150cm] robowaifus to be able to lift & carry 100Kgs of weight for 15Km [Fireman's Carry or backpack] across level terrain.) For a loose analogy of the use case; as a rock-climber I can tell you that its your bones, not your muscles that give you your primary strength during climbs. Thus the importance of high- compression/tensile strength rods (+ low-friction UHMWPE pads inside joints, ala replacement knees, etc. [1] ) for our robowaifu's skellington. <---> >There is a relatively large amount of surface area on the braided line, and the friction of dragging it over itself can quickly bring it to the 270 °F or so point it starts melting at. It's remarkable stuff, and I'm really glad it exists for us robowaifuists! I don't think this temperature is going to be a frequent problem for our usages here -- as long as we keep it it well clear of hot actuator/etc. cases! :^) --- 1. ( >>39006 )

>>39126 https://www.mcmaster.com/products/carbon-fiber-tubing/ >Please remember I'd like our lifesize [~150cm] robowaifus to be able to lift & carry 100Kgs of weight for 15Km. What's going to be the power source? All the muscles/actuators in the world don't matter if you don't have to energy to use them.

>>39128 Thanks! (Lol, they kicked me out with 'Please Log In' before I could see it.) Any rough ranges on the prices for tubes suited for building robowaifu's skellingtons, GreerTech? >power source? We've had this discussion at length in our Power & Energy bread : ( >>37810 ). Our primary ME here (& I wholeheartedly agree with him on this) say its lithium ion batteries. Its the only practical solution for us still yet. BTW, this kind of derails our Tensegrity bread. If we need much discussion on this, then let's please have it there thanks. :^)

>>39132 >Any rough ranges on the prices for tubes suited for building robowaifu's skellingtons, GreerTech? What thickness and total length do you need? >power source? I don't have any new discussion on it, my only concern is that it would take an e-bike sized battery to carry a 220 lb (100 kg) weight for 9 miles (15 km), not to mention any computer systems.

>>39133 >What thickness and total length do you need? It varies, depending on the part in question. The thigh 'bone' will need to be the thickest & strongest, so let's start there. Maybe 18" long X 1.25" diameter (unsure about the wall thickness, but let's guesstimate about 1/8")? >e-bike sized battery Given. In fact I'd estimate two of them. Somehow that's got to all be fitted inside the torso as well. Hey, if this was easy everyone would be doing it!! :^)

>>39134 Thigh bones alone would be $104.30 I would just go with aluminum

>>39135 Thanks Anon! Now we have a tent-peg in the ground for it. Yeah, this will probably need to go in the higher performance/higher cost robowaifus. I'll go with aluminum for now. Cheers, GreerTech. :^)

>>39136 Might it be better to do some prototyping in wooden dowels?

>>39136 However, I did find these. I don't know why McMaster Carr is more expensive, but I'm more inclined to believe them https://www.amazon.com/ARRIS-500mm-Wrapped-Carbon-Surface/dp/B00TF8V9KI/ https://www.amazon.com/FANCYWING-Surface-Wrapped-Quadcopter-Multicopter/dp/B0B99PG6W7/

>>39137 Yes. And in fact I've actually already done some efforts with various quality dowel rods (White Ash, I think maybe?) Wood should definitely give sufficient compression strength for cheap robowaifus around the home (doing light, menial tasks [laundry, dishes, etc.]) It doesn't stand up very well to outdoors environments AFAICT (maybe they just need a good treatment to help with this). Good thinking, Anon! :^) >>39139 Thanks!! This is actually inexpensive enough to give some consideration now. Probably best to do the prototyping with aluminum in the meantime, then test out replacements with the CF rods after the designs are all working well, for better performance. Need to be frugal with costs during this stage as well. >I don't know why McMaster Carr is more expensive, but I'm more inclined to believe them As am I. :^)

>>39141 Speaking of Futurama...

>>39142 Lol. I don't recall seeing that one! :D

>>39139 The cheaper tubes might use a poorer quality epoxy. There are no also listed msi ratings, so they are probably a made with a week low modulus fiber. The orientation and alternation of the wraps would also be a factor. The Amazon ones look to be made from tube a of loose twill weave. The three McMaster Car options have different materials. One has all the fibbers running in a single direction, one with a tighter weave, and the last has crisscrossed alternating wraps with a much high modulus fiber. >>39142 >>39143 Curse you merciful Poseidon!

>>39145 Their pictures are terrible, but the last McMaster Car's last option is probably similar in construction to a fancy fishing rod blank. The first option probably would be a bad choice as its outer layer could potentially splinter into sharp needle-like fragments.

>>39146 This is a good point. And thanks for the reminder! I'll plan to use a safety box around when doing the skellington failure tests! Cheers. :^)