Way overly complicated

You had a think? How do you have a think, oh that's what you do when you're sitting on the toilet? You all over there on the other side of the pond need to fix your English in grammar. I would believe thinking of something that you do not something that you have.

You could probably save weight by swapping the bearings for some linear bushings like IGUS drylin ones. They do have carbon fiber shafting as well. Be sure to get a decent spacing between your bushings.

10:58 maybe a clam type system? 2 screws and one plate that those screws press down? or pass through aka hole and then screw bites down from the side. or thats at least how most rebuildable vape tanks works. hole and one 3mm scrub screw that bites down on the heating wire from the side

I just discovered your actuator and I find if awesome ! Regarding the string lock at the top of the actuator, you can use blocking rings such as those used in RC airplanes to hold the wheels. Regarding the strings, there are high quality/stiffness/endurance 2mm (maybe less too) cables used in light aircrafts to link flight surfaces to the control rod and pedals. Regarding the general structure, I wonder why you don't print the negative shapes and mould everything in carbon fiber+epoxy. Stifness will increase precision. Finally, Regarding the brake stage overheating, I strongly recommend that you check how it's done (and which mosfets used) in F5B competition ESC. These devices are able to brake a 8kw motor from full power to full stop in a fraction of a second.

11:20 Would Urethane rubber like the type on skateborad wheels work for the bearing covers.

1:51 🥵

Super cool project! Immediately got a couple ideas as well. Sorry if these are already named, but here we go: - Route the extend and retract spring via the same wire, so kind of get a capstan drive. - Make the sleeves for the inner bearings slightly concave. to make them self align to the rod? Maybe also make a groove in the middle of them to make a channel for the wire, to turn that issue into a feature. It might remove an extra bearing for routing the wire. - Make the moving bearings three sets at a 90deg angle, to get a kind of rollercoaster setup. - Add a fan to the output shaft of the motor, that spins a cooling fan over the brake resistor. Well now that that's out in the world so I don't think about it again, good luck!

Hi @harrisonlow - great work! I wanted to point your attention to a type of string that has much higher strength called "spectra" or "dyneema". used most commonly in sailing, paragliding and kitesurfing, such strings can support 100s of kilograms in rather thin gauges. The fibers are also cut-resistant but you might have a heat problem with them and might consider lubrication to control that.

how far you've come on this project! Thanks for being so detailed and sharing it with us.

This is awesome!! I've been needing some scalable linear actuators for a project. These look Great!

11:40 for this there is ha easy fix, just add other two (a total of three, une per carbon tube)

Pornstars love these.

Ähm the runtime is without load. I would suggest test that with the nominal load otherwise you get wrong confidence.

Just remember before you terminate it you tell it “ I’ll Be Back” this is a very important step or your actuator a.k.a. “ Connor” yes I named your actuator 😝 will not survive 🫣

How are you controlling the motor and what type of motor is it? The best way to brake a BLDC or synchronous AC motor is to control it using Field Oriented Control (FOC). This will give you much more braking torque than other control methods. However, it will also result in regen current going back to the power supply. That is where you might need a brake resistor. Another solution is to use a battery instead of a power supply. The battery, unlike the power supply, can absorb the regen current. Then you probably will not need a brake resistor.

To get rid of the torsion along the central CF tube, instead of using six bearings and increasing the cost. Maybe consider using bushings around all three outer tubes. My thinking was PTFE but, another material could perform better

What about attaching a small spring to a small rubber o-ring then attach string to the o-ring available in those plumbing kits to fix a leaky tap. Might be smoother and not require that spiral string. Would be interesting to see which lasts longer. Just a thought...

could you crimp a steel ring eye terminal to the synco cable,thenm put the bolt throught it,youd need a steel crimp not alluminium or brass or copper

wouldnt using some sort of encoder allow you firstly be able to know the position at any time and match it with tiny pwm module on each motor

a good technique for getting the threaded inserts flush and square and clean is to ALMOST push it in the whole way, and finish with a hard edge flushing it to the surface and sucking in any slop.

9 months late, but could you use a pinch bolt for the syncromesh cable? The sort of pinch bolt that you might use for the braided cable on your bicycle brakes.

I think you should consider using Dyneema/Spectra for your ropes. It is made from high strength polyethylene fibers, and it's very resistant to wear (and thus quite hard to cut, even with a proper knife). It's used a lot in sailing and fishing, and I've seen some only 1mm in diameter, which I guess should be thin enough for your application, claiming several hundred kilos of working load.

странно что я не могу найти ни одного сексоробота сделанного по этой схеме... пожалуй сам захвачу такой сегмент

why do the bellows need to resist torsion? and how much?

I saw weird spikes in the position data like that before and what the issue was for me was that the SPI bus between the encoder buffer and the arduino(brain) was getting disconnected during vibration. Took me a while to figure that one out. Also for the bellows, this is a blind guess, but you could try adding fillets to the edges(outside) as well as giving it a bit less thickness to the parts you want to bend. Kind of creating a designed "weak point" so they bend correctly.

Keep up the great work.

If you still dont know how to terminate the syncro mesh cable you can crimp electral wire terminals onto the end. Youd want to do some strength testing for your needs, but in my experience, it works great for a mounting point.

Why not have the motor on it's side so you don't need so many pulleys.

If you fix the moving rod at the bottom, you might be able to get away with scrapping one of the top bearing rings. That will lighten up the actuator a lot

you can make a 3d sketch by creating your own plane (likely a 3 point plane) then creating a sketch on that plane coincident to the joint points. (using mate connectors in a sketch can be a little difficult, so perhaps the "use" took would be more beneficial.)

10:59 "guitar tuning machines" straight up made for tensioning cable. Might be a good idea to make your own to some degree though. No need for a huge key poking out the side should you try doing that. It's a worm and wheel gearset with a cylinder with a hole in the side attached to the wheel. You could turn the worm gear with a screwdriver, and the wheel would rotate slowly, allowing you to tension the cable. Also for the resistors all you really need to do is ensure super low resistance and dissipate the heat. You do need to know however just how much current that motor can handle continuously while torque stalled (rotor locked in place) as this will determine the minimum resistance for the long-term use of braking. Then it's just a case of full bridge rectifying the 3-phase the motor output, and having a MOSFET closing the gap on the output. The voltage to the MOSFET would change the resistance and so you can alter the dissipation that way. You will need to heatsink it though. No need for an actual resistor as long as you have a chunky enough MOSFET to handle it. Braking resistors are only really needed when you're working with super high power stuff that has a lot of innertial energy to dissipate, and the motor is going to be your weakest link there, second is the string. I've worked on megawatt scale robots in factories before and the braking resistors be crazy. Like bigger than me crazy. Not really needed for this. If you use MOSFETs you have another advantage, and that is you can vector the voltage being sent to it based on stator temperature. So say that motor is at 95C already. You could decrease the braking force so the motor doesn't start melting its enameled wire by sending 2 volts to the gate instead of 10. You will however need to make a logic gate based interlock system such that if you're telling the motor to move, that braking system should never activate while the motor is running and vice-versa. If you have trouble, I can elaborate but I think it'll be more fun for you, now you know it's possible to do it with simple components, to figure it out.

For your cable clamp, you'll need 2 metal plates with a hole at each end (10x5mm, with M3 holes should suffice) - fix the plates with screws at each end and clamp the cable between the plates. Might work better if you set a shallow channel in it to keep the cable from moving about. If you can't make the plates, plates from a bicycle chain would work.

I only have a 3d printer and i am wondering if a stepper motor could be used instead. And use a belt not string, so you do not need to spool the string. Dependimg on how it spools, you would gain linearity errors

Don't take measurements of travel by measuring string travel. That doesn't compensate for any wear. Your durability test should have measured how far the actuator actually is from where it thinks it is.

String is very stretchy. Under loads the braid tightens, eventually that is permanent. You had no way to compensate for that. For terminating the cable, which may also require stretch compensation, look at how it is done on steel cables. Crimp on a terminal to prevent fraying, but not for a load, or for loads, slip on a bigger terminal than loop the cable and put it back into the terminal before crimping. in electrical applications sometimes soldering is used, instead of crimping. I don't know what that cable is, but perhaps epoxy would work, maybe with crimping.

Hello stumbled into the video and the topic pulled me in. Retired Engineer in Medical, Automation and Robotics Systems. You do a great job of scientific thinking and setting goals. We used a Small Stewart Platform to shake test medical Data collectors in Scanners. It was amazing as we rolled through the frequency scale how bits would actually start to vibrate so much we could only see a blur where the part was. We added braces and stiffeners and in the end it improved our end result significantly. The advantage of having access to Aerospace folks to help. We also used Carbon Fiber for patient support to keep them nice and still. Later i used Carbon Fiber to build robotic arms, Large ones for Defense manufacturing, we added active and passive vibration dampening to get long arms with heavy grippers to move fast and stop without rebound. I appreciate what you doing and how you are doing it but still unclear what gizmo is doing. oh and this work i mention was done in the late 70s and early 80s. Maybe you were born after that time. Keep it going troop you are educating as well as fixing your machine. We had to know position as well so we added a rotary counter on the motor to feed a location calculator. Have fun and keep up the good work. Dennis another thought on the center Round Shaft. I used a square Shaft on one device, machined two corners opposite each other and a three V shaped rollers to capture the shaft and maintain alignment with lowest drag. Those were all metal. I was surprised to see you running metal bearings on six sides of resin cured Carbon Fiber tube. The resin will require some compliance as you have learned. We found Delrin worked well it is machinable and robust with some compliance to prevent damage to the resin surface of the composite tubes. D

Just lying here, thinking that there are a couple of things that i might be able to usefully comment on... a bit off the wall, but i wonder if using two motors, although more complex, might give you control over things like stretch in a string, and providing some kind of dynamic control over the string action and loading... with encoders on the motors, and optical "zeroing" sensors, you get calibration on every cycle... i wonder if some kind of silicone 'oil' impregnating the string might improve lifetime ... indeed, steel cable with a 'dry' lubricant might work well, even on the smallish radius pulleys? I guess that a lubricant could be 'wet', if the actuator 'tube' was enclosed....anyway, sick in bed, just the ramblings of an amateur....

Maybe look into stringing machines for tennis or badminton rackets. They grab onto the strings with a clamp, maybe you can find something useful to design a tensioning mechanism

Please Please remove the music when you are talking, it is very distracting

Terminate the same way as they do on bikes with a bowden cable a threaded pin with a hole in it and a nut, very nice project thanks for the video

Use dyneema rope

First and foremost: You are doing God´s work, if you didn´t know that yet...!!! (and I say that, while being an atheist, mind you...!!!). Second: I know little about such sort of things, but I guess, stepper-motors would distibute the "breaking-heat-energy" much better in their thermal mass, compared to the brake-resistors... (and no brake-resistors would then be needed) So, if my layman-assumptions are correct, why wasn´t any stepper-motors used instead...??? Thanks á priori for any answer...!!!

10:45 use the brake cable bolt from a set of bicycle callipers, they have a hole in the middle. Edit, nm, more sophisticated ideas have been suggested

You should be able achieve any string axis with two pulleys (not counting the main spool), your first pulley set does not need to rotate on the same axis as the spool, the only requirement is for the string to be tangent to the spool.

Hmm.. there are super strong, super soft (curve ability), miniaturised cable (multi-strand woven cable) jewelry wires out there. They have a plastic cover standard, but are fantastically strong. I used Accu-flex 49 strand. As another alternative, many specialty and competitive fishing lines face these wear challenges.

For the slowing down of the actuator, you can put a coil on the ends of the main frame and have magnets attached at the ends of sliding rod. By using hall sensors you can determine the proximity of the magnets and switch on the coils accordingly and hence decreasing the speed effectively. This can moreover be used in a way that woks like a coil gun and help accelerate the moving arm after a full stop.

Maximum hair today

Don't let the of girls find this shit

use dyneema for the cables no fatigue like metal and insanely strong, just don't leave it in the sun