Thank you everyone for the votes and comments. Really quickly, I did want to briefly address what Ms. Zimmerman said because I should have said something about this earlier. Regarding schools providing crowd funding and local grants, this is completely fair and actually my university does offer a crowd funding platform however it is a very lengthy process lasting at least 4 months, requiring the applicant(s) to continuously campaign, and charges a fee. Just to clarify (cut/pasted this here instead of under the vote).

Thank you Mr. Bender for your comments, I will try to respond to your points,

1) Regarding partnering up with someone who has prosthetic experience, I actually was able to do this last year in April/May where I worked with a prosthetist and discussed my design, what focuses amputees often look for from his experience, and how my design should balance those priorities. I got a lot of feedback from him about things like optimizing weight, center of gravity, simplifying certain functions, etc. which did have a large impact on my design as I went from my previous (frankly impractical) prototype to this one.

2) The weight of the device will be suspended on the arm using a set of outer braces mounted on a set of hinges such that there is some room for the outer braces to pivot to expand apart to allow some degree of growth or expansion. These outer braces are meant to tighten around the inner brace which will fit around the residual limb sleeve. This sort of loosening/tightening function was something that was inspired by the way the Open Bionics Hero Arm works if you wanted to get a better idea of this, but I also tried to keep the CG as close to the residual limb as possible to reduce the amount of torque the prosthetic arm is applying. Also as I stated in my proposal, my end goal here for the purposes of setting a more attainable goal is to make this as developed of a proof of concept as possible because of the amount of difficulty I expect to experience in getting to a stage where I could test this design on an amputee. Copied from the proposal, "This is not to say exactly that I am ready to give up on the idea of testing this design with an amputee, however, I think there is value in this design and by moving in an open source direction this would allow people who are more familiar and experienced in the clinical application of prosthetics to maybe take this design to that step. Also with many members having such a strong understanding of socket design and fitting and this being a field I am not that knowledgeable in, I think that an open source direction allowing someone to kind of take over that part of design might be better - especially if that someone already has experience with for example fitting amputees using methods like 3D scanning." I mention this here because of the way the device's weight will be supported in this design has a lot to do with how the arm is fitted to the amputee. Again, because this section of the arm is a separate module attached via 4 bolts, this can easily be revised should whoever looking to change this desires.

3) Regarding preventing servo backdrive/whine, I do not think this will be too much of an issue considering that I am using continuous rotation servos as opposed to standard servos. Going back to how these servos are being used in my design, the servos rotate as for example the fingers close around the object until a certain resistance is met which is in the form of a current draw value. When that current draw threshold is met, the servos stop turning or depending on what my tests show, the servos might be directed to continue turning at a very low speed to keep the amount of force being applied as constant as possible. Back to the main point about backdrive, because continuous rotation servos do not have a potentiometer telling the servo to keep applying force to hold at a certain position, backdriving the servos is possible. This is a good thing here because even if the servos are backdriven this will not result in servo whine (if the servo can stop rotating), or very little (if the servo has to rotate at a very low and constant speed). This video testing the servo I am planning to use might also offer some insight: https://www.youtube.com/watch?v=hSgzR5Ga-lU

4) For how long the wrist module is/how short the residual limb needs to be, with the whole arm being 17.87 in long, the the residual limb is about 4.59 in long. A longer/shorter residual limb will force the total length of the prosthetic arm to of course be longer/shorter, however this total length can be adjusted in the longer direction by adding space between the limb and wrist region but will shift the center of gravity by an amount depending on where that space is added.

5) For self-funding smaller proof of concepts, I think this relates to Mr. Schull's question about whether the project could be divided into sub-components/systems. For this, I answered with a long reasoning about why I thought it would be more efficient to divide the project into a 1) mechanical assembly stage, and 2) the integration of electronic components. Assuming you haven't already fully read that response, I would be glad to address any specific questions you might have about my method/reasoning there. Regarding self-funding, as I talked about the mechanical stage being what I thought to be a good way of validating my design and the concepts encompassed before moving forward from there, the amount of money in hardware required for the mechanical stage would be $82.73. At the moment this is not really an amount I can afford to self-fund without having to ask my parents; something I am really trying to avoid as I have already asked them for help with previous prototypes.

Hopefully these answered your questions and maybe changed your mind? Anyways, I would be happy to discuss any more concerns you have further.

Hey thanks for your reply, my intentions is not to be discouraging, but just understand that Kickstarter/Youtube/Hackaday is a graveyard of student and/or maker projects that where going to do "ABC better than product XYZ for a fraction of the cost," but they don't get off the ground because 3d printing something that moves like a hand and making a wearable product are vastly different things. I appreciate your honesty about achieving the 2nd part though. As a prosthetist, I think your gear-based adaptive grip is the most novel out of anything you have proposed here, so I would focus on that mechanism primary if I could encourage you in any particular direction--but figure out that backdrive problem or it will be all for nothing! :)

A few thoughts:

1) I'm not sure you fully understand the problem of backdrive. If you don't use some kind of mechanical backdrive prevention (worm gear or leadscrew) then you have to continue to apply current to maintain grip force if the user just wants to say, carry something heavy across the room. This means ++ heat, ++ noise, - - battery life, and has been the fatal flaw in so many maker projects I've lost count. The cool thing about worm drive or leadscrew is you can completely shut the motor off once grip is achieved and you don't waste battery, don't make heat, and don't make noise.

2) The problem with the long wrist module is that it limits the amount of users that can use it. Notice almost every commercial design has all the electronics either inside the hand or external to the socket, that way no matter how long your residual limb is, you can still use the socket. Is it possible to do what you're doing but moving more parts inside the hand and/or external to the socket?

You've clearly given this a lot of thought so don't let me discourage you! Happy to change my vote (the money asked for is marginal) if you can consider some of the things I've said above.

Got it; for the gear based adaptive grip I will try to demonstrate this first if I am given approval so that I can better show this concept to the community. Also about your distinction between how difficult 3D printing something and making that device wearable, I definitely agree with that. Again if I am given approval to start building, I will probably save my efforts to really optimize the "wearability" of this arm for last after getting the rest of the building process out of the way as my proposal goal implies. (Also, excuse my long reply)

1) Okay I understand what you mean now, previously I thought you were simply talking about the ability of the servo to rotate backwards given resistance. If I were to incorporate a backdrive option from the ones you listed, I think the worm drive would make the most sense as I am going from rotary to rotary instead of a lead screw going from rotary to linear. There are some drawbacks to this however that I am having trouble considering such as with speed. If I do go with a worm drive system, this will decrease the speed of the fingers considerably unless I adjust the angle of the worm wheel and worm gear threads drastically enough to warrant a 1:1 ratio, however the more this is adjusted the easier it is for the worm gear to backdrive against the worm wheel. With the high torque ratio the servos I am using are putting out already, I start to get the idea that users might become frustrated or see the slow speed of the fingers as tedious as opposed to a more instantaneous grasp. Currently, I was planning to rely on the gear train friction and current of the servo which work against rotational resistance especially given the high torque ratio of the servo putting out 20 kg-cm to hold the force applied with maybe some motor assistance depending on the task/preset (even though this trade-off will use some battery and produce some heat and noise). The other issue I was thinking about was compliance because if sudden forces are applied to the fingers in either direction, it would be a good thing if the fingers are able to accommodate this which a robust design should do. With an anti-backdrive system like a worm drive or lead screw system that will most likely have to be 3D printed to get the right ratio and be fitted onto the servo, if sudden forces are applied the worm wheel/lead screw would not permit any movement making parts like the worm wheel/lead screw threads susceptible to snapping assuming they are the weakest points. I think I have a work around to this instead though, by using a sort of locking mechanism like a ratchet, but in a manually locking way. By having 4 manual push-locks for each of the 4 fingers (not thumb because thumb is cable, not gear driven), this would allow the push locks to engage the lowest gear on the finger which drives the rest of the gear train, such that the lowest gear and the rest of the train get locked in place. For example when holding a heavy grocery bag, I think this would work well instead of a ratchet which is difficult to implement because of space and noise. As for actually implementing this into my design, this will be easy to accomplish because I don't need to alter anything inside the hand, but need to add mounts to the back cover for the push locks. This can then be retrofitted with the rest of the design. As far as when I am going to implement this however, I do think I am going to wait because I want to see how much force it actually takes to backdrive the 20 kg-cm servos because if this is sufficient to hold force for most applications with/without minimal servo motor assistance, this change might not be worth it. If it is relatively easy to backdrive, then the all parts for the change can be 3D printed and no extra hardware is needed so I will not have to request anything.

2) About the length of the wrist module, I agree that this is a limiting factor. Actually in my old prototype, I tried to keep the ability to accommodate as many different length residual limbs as possible by housing the DC motors used for actuation in the fingers themselves. After talking with the prosthetist I mentioned however, he reinforced the idea of keeping the arm as lightweight as possible as a major factor amputees consider because although the motors used were fairly small and lightweight, the combined weight but especially the placement further away from the socket placed a lot of torque on the residual limb. I tried to use this same idea here also because the servos are too large to fit in the hand area. By placing them right in front of the socket this was my way to minimizing the amount of torque exerted on the residual limb. I do understand that the trade-off because of this is a narrower range of residual limb lengths but I thought this tradeoff might be worth it also if the servos and rest of the electronic components are close in proximity. As far as making these parts external to the socket, I am currently doing this with battery pack with that rectangular extension on one side of the outer braces in the picture, however I don't think I can really do this with the servos because this would add volume around the socket area. I was trying to avoid this after I was advised by Mr. Liu to debulk the that area where I previously had a tightening mechanism but removed it in place of a more compact method. Also for custom fitting this to an amputee, this would be difficult because if the length of the outer braces/inner brace is lengthened/shortened depending on the length of the residual limb, this would mean that if the servos were mounted externally around the socket then the length between the servos and hand would change length, altering the length of the flexible shaft that needs to be ordered. From a sourcing perspective, I see this as being problematic because of how long it takes the Chinese manufacturers I am currently looking into to produce and ship custom flexible shafts (around a month), whereas if that length is fixed, this sacrifices a bit more torque on the residual limb in exchange for an easier custom fitting process and a less bulky socket/brace region.

I think maybe you can see why some of the designs you mentioned in your proposal make some of the compromises they have. While the wish list is certainly fully-articulated, lightweight, low-cost, in reality those things all become difficult to fit together in a package that is cosmetically appealing and acceptable to a large variety of users due to the ALL the additional factors (like battery life, weight, where to put all the electronics, etc).

FWIW I'm willing to change my vote as you appear open to addressing issues raised by the community.

A few additional thoughts:

1) Your manual lock solution is a possibility, but do understand it does go against your design principle of minimizing bi-manual operations.

2) If you're not using a potentiometer to control the servos, have you looked at using standard DC motors or gearmotors instead? That way you could choose the output speeds you want and could potentially work in a worm-drive setup. Just a thought.

3) Don't underestimate the compromise of decreasing your battery life and adding heat through constant-powered grip to fight backdrive. Like I said, this issue killed the Hackberry, and EXTREMELY elegant and highly-funded design, not to mention countless other Youtube/Kickstarter/Hackaday designs.

1. You're right that adding some of fail-safe for the fingers is a nice design feature. Lot of ways to do this besides allowing backdrive. Lot of commercial designs use a breakaway pin or knuckle joints that "pop out" under too much force. That way they can snap back in place and carry on. Again, not saying you need to do this, just adding some information from my experience.

You've got a HUGE task ahead of you, but I think if you can break it up into bite-sized chunks to minimize funding risk from the community I won't stand in your way!

Firstly, thank you Mr. Bender and Ms. Zimmerman for changing your votes, I appreciate it a lot! To respond to your comments Mr. Bender:

1) You're correct, I think I am being a bit of a hypocrite here by including a function that would require the off-hand when I described my goal to partially focus on minimizing use of the off-hand. In total then, this would mean the off hand would be used for adjusting the position of the thumb, and locking the fingers. Despite my focus on minimizing this, I think this would still be a worthwhile compromise whenever situations to lift something heavy in that way arise.

2) Actually with the servos I am using being continuous rotation types, they do have the ability to alter speed/direction but even at the highest speed this is not very fast due to the high torque ratio. Based on a past prototype that used DC motors, I was also trying to avoid this because using DC motors with an Arduino means having to use motor driver carriers for every 1 or 2 motors which I found to be a hassle in terms of the volume taken up by the board and extra wires too.

3) This is definitely an area of concern for me, especially with you mentioning this being the fatal flaw of the Hackberry which I was not previously aware of. I will take this into account as I am testing the arm and will make the changes I specified if necessary--thanks for emphasizing this.

4) Actually now that I think about this, adding a fail safe in addition to the current design would not be that difficult considering that the I need to create a "weakest point" somewhere in that gear train/servo flex shaft system. Because these components are going to be 3D printed anyways, I could just have the gear train's driving gear be 3D printed using maybe 5% or less infill such that if enough stress is applied, then the gear would break before any other parts do. The only drawback with this of course is that unlike the popping knuckle joints/breakaway pins you mentioned, this would not be re-settable without having to print that gear again.

Hi everyone, it has been a bit since I have posted an update after my proposal was approved so I wanted to mention a couple things:

• As stated in my proposal (or in a reply, I can't remember which), I will be documenting my progress for anyone to follow at: https://bhargavp225.wixsite.com/mysite I know the name "The HAND-icap Project" sounds kind of corny so I might change that once I think of something better.
• After getting approved for funding and submitting an Open Collective expense request, the $452.00 was payed out to me and I have that money now so I will begin ordering parts and taking further action soon (see update on website for more info)

Again thank you all for the comments and support, and be sure to check the website link for all further updates.

How about Handy-Hand

BR/Yoav

Hi everyone, I know it has been a while since I last posted here on the E-nablio forum and in the updates section of my project website. I wanted to provide an update though on how things have been going, and I recently emailed Mr. Schull about this too and so I'll paste that email here because I think it covers my thoughts:

"Hi Mr. Schull,

Firstly, this is going to be a somewhat long email. I know it has been a while since I last spoke to you, but I wanted to talk to you about the status of my project and some steps I want to take. You may recall that about a year ago, I received a mini-grant through the e-NABLE foundation for my project to build a novel transradial prosthetic arm. At this point, however, having not yet reached the goals I set out to accomplish, I wanted to talk to you about returning the grant money I was given.

I am saying this because, if I can speak honestly, the urge I keep feeling to finish this project and how it has conflicted with the commitments I have to school and extracurriculars has been very frustrating for me, and I have felt increasingly guilty about accepting the grant money but not yet delivering what I promised to the community. I apologize for this.

To give a recap and status update on my progress, originally with the backing of the community and the time to work on the project I progressed for a while and reached a decent level of completion with the mechanical stage of the project completed. Since then, however, the weight of my college courses and other extracurriculars has been very heavy on me to the point that my progress with this project became very slow.

To date, since completing the mechanical assembly I have done a very small amount of testing concerning control of the hand using an Arduino Uno, but I wouldn't really say I have ventured into the electronics integration stage yet. There is one other objective that I was not able to fulfill as well: my goal to document my progress in an open source format. Again, this was due to my inability to dedicate a sufficient amount of time to do this correctly. Although, I did briefly share my CAD and STL files directly from my google drive for anyone interested to use, but this was careless of me to do without any creative commons license protection. Even with the updates to my website, at some point these became very infrequent as I had felt there was very little to update on without repeating that I had yet another setback.

As far as the future of this project is concerned, my other commitments unfortunately have not slowed in terms of my involvement in them and so it is still difficult for me to promise a certain amount of progress in a certain amount of time. However, I know that at some point I am going to finish this project just because of how much time I have already invested in this, and because I am still motivated to reach the goals I have set. Again though, I understand that I cannot just take my sweet time because I am using money that was given to me with the expectation of completion by a certain time. I hope you can understand why it has been difficult for me to complete this project in the timeline I originally promised. This is why I think the right decision for me should be to return the grant money.

Hopefully sometime in the future after I have completed my project, I hope you and/or the community may be open to the idea of possibly allowing me to reapply for the grant money again after I have proven results and have shared the project on an open source platform. If you and/or the community do not feel the same way about this though, I would understand completely. I was also planning on sharing this in the E-nablio forum for the users who originally backed my project to see, but I wanted to hear your thoughts first. Thank you for your trust.

Best regards,

Bhargav Parthasarathy"

Anyways, I wanted to let you all know that I just returned the grant money ($438.76) I was given via a donation to the e-Nable fund and that Mr. Schull was very understanding of my situation so I appreciate that as well. As the email above mentions, I will be sure to get back to you all when I have some results to share that I hope will be useful for the community's progress. Thank you all for your support.