Announcing the E-Nable Eagle Hand, currently in alpha

For discussion of hand designs.
Jason M Bryant
Posts: 198
Joined: Mon Feb 16, 2015 1:59 am

Announcing the E-Nable Eagle Hand, currently in alpha

Postby Jason M Bryant » Mon Jun 01, 2015 1:46 pm

Evolved from the Raptor, sharing DNA with the Falcon, I'd like to announce the Eagle Hand.

Eagle Hand.jpg
Eagle Hand.jpg (69.47 KiB) Viewed 3267 times


First off, this isn't ready for testing. It doesn't even have screw holes for the velcro yet. This is just the very first test of the hand. I already see stuff I want to change, like how the thumb is set up. I'd also like some feedback on certain aspects which I'll list at the end of this post. After these thing have been addressed, I'll put STLs up for people to try out.

However, the functionality is already quite good. I took the Raptor and redid a ton of stuff, not leaving a single polygon of the palm. (Although the gauntlet is 100% the Raptor Reloaded gauntlet. Feel free to use your thermoformed RR gauntlets with the Eagle Hand.) Here's a quick overview of the changes from the Raptor reloaded.


Dental Elastic Bands Instead of Elastic Cords

dental bands.jpg
dental bands.jpg (80.2 KiB) Viewed 3267 times


This is the biggest change, and it has already been tested a little in an RR mod. The elastic bands are much easier to put on and don't require as much force to close the fingers. In addition, the lack of tie-off bars on the palm gives it a much more streamlined, human look. I may even be able to hide some more of those wire routes.


Better Gripping

pinch grip.jpg
pinch grip.jpg (50.79 KiB) Viewed 3267 times


With the dental bands set up right and some wire routing changes, the fingers bend in a better way now. They always bend at the palm first, then bend at the fingertips only when the palms joints meet resistance. This makes the fingers wrap around objects a little easier.


More Opposed Thumb Position and a Metacarpal Arch

thumb position.jpg
thumb position.jpg (66.76 KiB) Viewed 3267 times


The thumb has been rotated back a bit. Walking around my home and picking things up with this hand is a joy. Cups just stay in the hand better. The thumb does a much better job of pushing things against the fingers.

The metacarpal arch is the offset of the fingers, so that the ring finger and the pinky are lower than the other fingers. This also aids in grip, since the closed fingers now form a diagonal line that is more perpendicular to the direction of the thumb. On the Raptor Reloaded, these fingers are often too far away from the grabbed object to provide any help.


Gripper Box

gripper box.jpg
gripper box.jpg (68.7 KiB) Viewed 3267 times


I've already written a ton about this. The short version is that it adds some adaptive grip to the hand, so that some fingers will bend more than others, depending on the shape of the thing being grabbed.


Naturally Shaped Fingers

finger shapes.jpg
finger shapes.jpg (69.15 KiB) Viewed 3267 times


The thumb look like a thumb. Fingers are round. Fingertips have a wedge shape. This doesn't just look pretty, these things aid slightly in gripping and picking up items.

They're also really pretty.

Several of these features have already received various amounts of testing. I've already incorporated some of the feedback into the studs on the dental bands.

Issues

Thumb position: I made it so that the thumb does not lie flat. I did that because having it lean all the way back with this new angle would create an inhuman shape. The resting position is a little strained for a human hand, but it is possible. That's an aesthetic choice.

The downside is that the hand dies not lie as flat on a table as the Raptor Reloaded. Is this going to be an issue? Will it be uncomfortable, or will recipients casually lay their hands on the side and not worry about it? The concern is that they will forget and lay their hands on their tables or desks too roughly, causing damage.


The thumb joint on the palm is a mess: I plan to work on this a lot, hopefully making it have a rounded joint with the palm.


Adult sizes: Will they have trouble grabbing medium sized things? There is a lot of space between the thumb and the fingers. Technically it's about the same as a human hand when the fingers are outstretched. However, when a human hand closes, the palm bends, moving the base of the thumb very close to the pinky, if needed. That doesn't happen with a plastic hand, so this hand doesn't have the same size flexibility. Will hands scaled up for adults be able to palm a basketball, but unable to hold a cup comfortably?


Inner palm space is slightly smaller: It's actually just one corner, where the pinky came down a bit. The rest of the space is slightly longer, making the palm a few millimeters longer than the Raptor Reloaded. Will the corner that I cut off be a problem? This will require some testing and feedback from users.


In addition to these issues, I have a big question for the community. Does this design lose anything that the Raptor hands do well? I don't want to unintentionally sacrifice a design advantage of the other hands. It's important that as we design new hands, we understand what worked in the previous hands.

I could write a whole book on this, but that's enough for now. Take a look and tell me what you think.

Attribution
The Falcon Hand Team at Seattle Pacific University – dental band system
Skip Meetze – whippletree for adaptive grip
Trouble Sturm and the R&D Hangout regulars – sounding board and idea givers
The Raptor Reloaded designers – The basis for the hand, including the hand shape, the gauntlet and many other details.

bosco2003a
Posts: 2
Joined: Mon Jun 01, 2015 4:04 pm

Re: Announcing the E-Nable Eagle Hand, currently in alpha

Postby bosco2003a » Mon Jun 01, 2015 4:09 pm

hi Jason;
the hand looks great cant wait to print, the fingers need to be set to a normal length a person all are different :D

bob

Jason M Bryant
Posts: 198
Joined: Mon Feb 16, 2015 1:59 am

Re: Announcing the E-Nable Eagle Hand, currently in alpha

Postby Jason M Bryant » Mon Jun 01, 2015 4:27 pm

I did consider that. The Raptor Classic had different length fingers. I'm not sure why that was not done in the Raptor Reloaded. I do know that occasionally people would not notice the different lengths and put the fingers in the wrong places.

The easiest thing would be to have two lengths of fingers, normal and a little shorter. The pinky and the index finger would use the shorter fingers, which would get us pretty close to normal finger lengths.

Is this just about looks? Is there any functional reason for changing the finger lengths? One thing that I notice is that if I make a fist, my fingertips line up evenly on my palm. That's because of several little movements in the hand, so we're not going to mimic that easily. I'm not sure it's necessary, either.

I'm not sure if this would make the pinky contribute less to the grip. It already doesn't reach the gripped object as often as the other fingers, so I wouldn't want to make that worse.

What do people think? Making short and long fingers would be a simple change. Is it worth doing?

User avatar
andreas
Posts: 31
Joined: Thu Feb 12, 2015 4:55 pm

Re: Announcing the E-Nable Eagle Hand, currently in alpha

Postby andreas » Tue Jun 02, 2015 7:04 am

Jason, amazing work!

Huge fan of the improvements you've made. The dental elastics, metacarpal arch, gripper box, and new fingers are all big steps towards a more functional and more accessible design. What software are you designing this hand in?

I think users will have to answer some of your questions about the thumb positioning and palm fit. In terms of any potential sacrifices of the Raptor Reloaded's capabilities, I'd have to sit down with my beta tester and do some A/B testing with the Eagle Hand his Raptor Reloaded. I don't see anything from a design perspective, unless the changes to the palm cavity are less accommodating to residual palms. I'd argue for moving away from the velcro in the palm in favor of thermomesh or riveted leather, but we probably need a little more user feedback before we can whether establish thermomesh as definitively better than velcro. My beta tester and Luke Dennison are fans.

A couple suggestions for the next rev (not to suggest that they're present in this rev) to improve printability across different machines:
  • Consider making the flexsor line channels a square profile rather than a round one. This will help with printability across many different machines as concave curved surfaces can be trouble points for a system that isn't 100% dialed in. We made the channels in the Raptor Reloaded open to address the problem of the channels closing up at smaller scales and on certain machines, but a square-profile channel would ensure a clean roof to the tube composed of small bridges, which nearly all machines can accomplish.
  • To eliminate the need for support material in the cavity of the palm, consider lofting a horizontal roof to the cavity to again enable relatively short (20-30mm) bridges across the span. I've had excellent results with the thermopalms I've been working on (see below).
Screenshot 2015-06-01 23.13.00.png
Screenshot 2015-06-01 23.13.00.png (107.71 KiB) Viewed 3239 times

Screenshot 2015-06-01 23.13.22.png
Screenshot 2015-06-01 23.13.22.png (413.76 KiB) Viewed 3239 times

IMG_2538.JPG
IMG_2538.JPG (80.57 KiB) Viewed 3239 times

Jason M Bryant
Posts: 198
Joined: Mon Feb 16, 2015 1:59 am

Re: Announcing the E-Nable Eagle Hand, currently in alpha

Postby Jason M Bryant » Tue Jun 02, 2015 12:07 pm

Andreas, thanks a lot for the feedback.

I'm using Blender 3D. It's a pain sometimes, but it's done okay. I'm going to install Fusion this summer and take a serious look at it.

Squaring off the flexor line channels: How about square on top and round on the bottom? Square channels will make sharp corners, which I'm worried will cause wear. Making the tops of the channels square would not be hard, while allowing the bottom to be nice and smooth.

Flattening out the roof: I'm not sure about this one. It seems like it isn't likely to fit people's hands as well, although the variation in hand shapes might make that a moot point. Still, it seems like it will make the hand more bulky. Sleekness is important to me, I want to try getting a latex glove over this thing.

Partly, I'm a little emotional over the shape of the hand. I put a lot of time into it, and I'm a little attached to it. I will think about it.

Finger length: Do you have an opinion? The Raptor Classic had different length fingers, although the difference was hard to see. Do you know why the Raptor Reloaded lost that? I'm strongly considering making a shorter fingertip for the pinky and index finger. It's a fairly quick modification for a nice aesthetic gain.

Jason M Bryant
Posts: 198
Joined: Mon Feb 16, 2015 1:59 am

Re: Announcing the E-Nable Eagle Hand, currently in alpha

Postby Jason M Bryant » Sat Jun 06, 2015 7:22 am

The testing process:

10 Look at thing and make list of problems to fix
20 Fix problems
30 Goto 10

User avatar
andreas
Posts: 31
Joined: Thu Feb 12, 2015 4:55 pm

Re: Announcing the E-Nable Eagle Hand, currently in alpha

Postby andreas » Sat Jun 13, 2015 5:09 am

Nice modeling! I was just curious if there were compatible files that I could take a look at in Fusion/Inventor. As long as we document all the interfaces of the joints as well as the overall dimensions (as with the Raptor Reloaded), we should be in good shape for people to build compatible pieces.

I love your suggestion about the channel profiles-- I think that sounds like the best of both worlds.

I'd advocate for at least a moderate flat to reduce the need for a support structure. What kind of supports are you using now? To help put the flattened roof into context, I only added 2mm of material to get the flat that removes the need for support, so there's not a huge loss of space. I think a lot of our existing palms could stand to be deepened to make more room for residual palms too-- will have to see how this new palm fits with some scan data.

Regarding finger length, it couldn't hurt to make the option available. The Raptor Reloaded team chose not to do different length fingers because it increases the assembly complexity, but as this design is making big steps towards a more anthropomorphic form, maybe different sized fingers would contribute to the aesthetic (not sure of their impact in functionality/gripping).

Super excited to print this out and take it for a spin. Will try to print this out over the weekend.

Screenshot 2015-06-12 22.07.54.png
Screenshot 2015-06-12 22.07.54.png (167.75 KiB) Viewed 3157 times

Jason M Bryant
Posts: 198
Joined: Mon Feb 16, 2015 1:59 am

Re: Announcing the E-Nable Eagle Hand, currently in alpha

Postby Jason M Bryant » Sat Jun 13, 2015 6:18 am

Thanks!

I did flatten out the roof a little bit. After printing a couple of hands, I've flattened it a tad more. It's not as much flat area as your suggestion, but I'm trying to get it to work better with the built-in supports. I'm also a little unhappy with a couple other areas of the support, but it is getting there. They're a grid structure, with a little extra in spots that need it. It uses less material and is much faster to print than the RR supports. I think I can get comparable print results. It doesn't have to be pretty, since these areas will be covered with padding, but I want it to be solid.

The files are all .blend files from Blender 3D. I think the best I can do is export them as 3DS files. Even that might be a little weird. When I export them as 3DS and import those files into Blender again, they're 1/100th the scale. Blender handles units weirdly.

I'd like to hold off a bit longer on posting the files, since I'm planning to make more changes as testing continues. I might put them up in a few days.


Return to “Hands”

Who is online

Users browsing this forum: No registered users and 1 guest