"Waffle" East Angelbots 1339 Offseason Project

[End-Game]

The Angelbots are proud to present our current offseason project, fondly named Waffle.

http://i.imgur.com/DyIJD68.png

We are currently building this robot to play our version of capture the flag with another team here in colorado. We have to acquire the flag from the top of a 3'x3'x3' cube, and get it across the field to the safe zone as many times as possible in a match.

http://imgur.com/a/ppAcy

http://imgur.com/a/mzQJ8

We decided to push our boundaries this offseason, and tried a great many new things on this robot. We are using two different styles of drive, a #25 Chain drive with a WCD Bearing Block, and an exact CC belt in tube drive. The chain uses a 18t sprocket, and the 9mm belt uses a 24t pulley. We are using a WCP 2 Cim DS, with a spread of 2.92, for a high gear of 16ft/sec, and a low gear of 5.5ft/sec.

We also decided to try something new with the claw mechanism, and used an over-center linkage to lock the claw in place when actuated shut.

The Reduction on the Claw roller is 50-1 with a BAG motor, and the arm has a reduction of 200-1 with two BAGs.

Another this that we are rather excited about is the bellypan. We routed it in house out of a sheet of 1/16in aluminum, and thus far has worked beautifully.

Any Comments, Questions, and Critiques are Welcome.

[asid61]

Looking good! Make sure to post pics of the final robot and it in action, it looks very well designed.
Did you add any extra for your C-C on the chain or did you go exact?

[End-Game]

We went with plus 3 thousandths for the 9mm Belt in tube, and a good old fashioned WCP cam tensioner for the chain.

As a broader question, What sort of experience have people had with an exact CC beltwork? Any tips or tricks that may make our experience easier?

[kyle_hamblett]

Looks nice! As far as bumper mounts go on what looks to be a West Coast-esque drivetrain, how are you mounting them?

[mrnoble]

We discovered several years ago that small sections of andymark's C-channel could serve as a wrap around 1" square tube. We combine that wrap with these:

http://www.mcmaster.com/#standard-quick-release-pins/=zyayty

With this combination, bumpers are easy and quick to attach and detach firmly and safely.

[philso]

That looks great! It is nice to see that you have taken the initiative to anticipate elements of next year's game.

You may want to delete the pocketing in the bottom plate where the electrical components go so that you have something to put screws or zipties into or stick double sided tape or Velcro on.

[s_forbes]

Cool stuff! Keep us updated with pics on the build, it looks great. I like that you're using two different drive approaches in the same build, you'll get double the experience.

There are two parts of the design that I would be slightly worried about:

The end segment of your arm is floating in the middle of two very long shafts. If you put much load on the end of the arm, the shafts will deflect quite a bit and allow the arm to sag. It will probably work fine if the game piece isn't very heavy, but you generally don't want to have a long shaft supporting a load like that. Snug the bearings from each segment up as close to each other as possible!
It looks like you are transferring all of the load to move the arm from the big sprocket, through a shaft, and connected to the arm segment. You're relying on the strength of the shaft and hub that it's connected to to withstand all of the torque of the arm, and the shaft/hub is generally the first part to break in that situation. If you are able to design an arm so that the big sprocket is bolted directly to the arm itself, it will be much stronger. This is one of the more common issues I've seen at competitions during years where teams make robots with arms, and results in broken parts.


As this is an off season project and a chance to learn stuff, I'd say build it as is and abuse it til you break it! You learn a ton from finding the limits of your designs, and it helps you design them better in the future.

[TheBoulderite]

Wow! Looks pretty good!

[End-Game]

An Update to our Progress.

With Two Weeks Left in our build season, We have completed the Frame and the majority of our Superstructure. We are working toward having a fully functioning Robot by the end of next week.

Thank you all for the feedback, it was incorporated into changes to the design.

http://i.imgur.com/sO9ohvH.jpg

http://i.imgur.com/1ltmEZq.jpg

http://i.imgur.com/KGDla6b.jpg

[Cash4587]

This looks really great. One thing I noticed is that it looks like you are using a hex shaft to move that arm only with the sprocket further down the hex shaft. I think it would be better if you did everything you could to bolt that sprocket to that versa block on the arm. I think it would put a whole lot less stress on the hex shaft. It might be fine as long as something not to heavy is being lifted with the arm but in the future it would be better to do so. Over all though I really like the use of all of the vex parts. Vex is one of my favorite places to order parts and I almost exclusively do so. Great work.

[mrnoble]

The design has been modified a little to drive the linkage arm instead of the shaft, sorry we didn't load an updated CAD image yet. Thanks, it's really fun watching the team work on this. I don't know that I've ever seen an off season project like this before, and it's exciting. I agree that the vex stuff is fantastic.

[s_forbes]

Nice progress! Good looking modifications, too.

Are you planning to use the AndyMark churro tubes for shafts? They'll probably work fine as pivoting shafts like you're using them there, but not so much if you try to transfer torque through them. They make for a good lighweight, inexpensive shaft material, so it will be good to see if they hold up.

[mrnoble]

No, we were doing assembly at another school and just used what was sitting around for the rough build. We've got shaft at our school that we will cut to length tomorrow

[End-Game]

Nice progress! Good looking modifications, too.

Are you planning to use the AndyMark churro tubes for shafts? They'll probably work fine as pivoting shafts like you're using them there, but not so much if you try to transfer torque through them. They make for a good lightweight, inexpensive shaft material, so it will be good to see if they hold up.

We may end up using the Churro on the front supporting axle, just to test exactly that.

[BBray_T1296]

Churro tubes are a Solid Mechanics example of one of the worst cross-sectional profiles imaginable to transmit torque.

Internal shear stress distribution in a pure torsion model is linear with radius. That is to say the further-out parts of the shaft (from the neutral axis) carry more rotational shear stress than the further-in parts.

In a circular shaft, from the center, the 0 to r/2 part of the shaft handles (1/2)^4 of the torsion (6.25%) while the outer half (r/2 to r) of the radius carries the remaining 93.75%.
This means a shaft with a hole in the center of half the diameter is only 93.25% as stiff, but with %75 the weight, making it a reasonable trade-off.
This also means a solid shaft with half the radius will deflect the same amount (radians/unit length) as the full shaft with just 6.25% the load and 25% the weight.

The churro tube with its "snowflake" cross section gains its weight savings by carving out the outer sections of the shaft. This is removing much of the primary torsion carrying material, leading to a horrendous torsional rigidity. I don't have the time to run numbers unfortunately, but it has to be really pitiful.

Note that in tension/compression or beam shear proximity to the neutral axis is irrelevant, making a churro tube no different than any other shaft with identical x-sectional area in these properties.
Beam bending moments are probably pretty decent due to the I-beam-like sections so churro tube may be a reasonable choice there

tl;dr Churro tube has an exceptionally poor torsional stiffness per unit weight, for weight savings use a hex axle with a hole drilled in it instead

[End-Game]

Another update on our progress. We are in the midst of wiring and plumbing the Robot, and should have the structure, with the exception of a few parts, done by the end of tomorrow.

http://imgur.com/a/634HB

http://i.imgur.com/DPo0rv1.jpg

[jkelleyrtp]

tl;dr Churro tube has an exceptionally poor torsional stiffness per unit weight, for weight savings use a hex axle with a hole drilled in it instead

Does that mean ThunderHex by Vex is the best COTS option for weight savings and torque distribution?
By your numbers:

ThunderHex is 1/2" in size
.2" central bore
15% less weight by cross sectional area. ((.5^2)pi-(.2^2)pi))/((.5^2)pi)
The central .2" carries less than 2% of the total load?

[s_forbes]

Does that mean ThunderHex by Vex is the best COTS option for weight savings and torque distribution?


I would assume that to be the case, but it would be interesting to see what torque is required to make a ThunderHex shaft fail vs a standard Vex Hex shaft... the rounded corners on the ThunderHex and bored out center might let it round off inside of a hub with enough torque. Or not... it sounds like a neat test.

Keep in mind the weight savings of the hole in the shaft are pretty minuscule, something you might be able to offset with several strategically placed helium balloons.

[End-Game]

An image album of the completed structure, with most of the wiring and pneumatics finished.

This also shows our strategy for acquiring the flag.

http://imgur.com/a/4Uoq6

http://i.imgur.com/YIeaaPph.jpg

[Sl3dman]

http://i.imgur.com/oEQhHYf.jpg
http://i.imgur.com/HTy7lDw.jpg
Waffle bot update

[BBray_T1296]

Does that mean ThunderHex by Vex is the best COTS option for weight savings and torque distribution?
By your numbers:

ThunderHex is 1/2" in size
.2" central bore
15% less weight by cross sectional area. ((.5^2)pi-(.2^2)pi))/((.5^2)pi)
The central .2" carries less than 2% of the total load?

Those should be powers of 4, but the result is correct. A hollow round shaft with those parameters is just 2.56% less torsionally rigid than a solid shaft.
The existence of the hexagon-generating surface features complicates things if you care for precise numbers, but from this simplified situation, ThunderHex is the most efficient choice.

What is happening on a molecular level is, the atomic bonds on the outside edges of the shaft are being stretched in plane shear more than the atoms on the inside of the shaft for the same rotational displacement. In addition, there are more atoms on the outside edge than the inside near the center that are being stretched, and therefore resisting the displacement. Coring out the shaft removes the lightly-stretched central atoms while leaving the atoms actually seeing the most stress. Also, consider the stretched bonds on the edge are further from the axis of rotation and thus generate a higher moment.


Out of curiosity I wanted to find the internal radius of a hollow shaft that would maximize rigidity/weight. I calculated the moment of inertia ratio of the hollow shaft compared to a solid one, then calculated the mass ratio of the hollow shaft compared to a solid one. I then calculated inertia ratio/mass ratio. By setting the second derivative with respect to the inner radius equal to zero (maximizing), I found that the internal hole should have a radius equal to sqrt(3)/3 of the shaft radius (or 57.7%), which for a shaft of 0.5" diameter is 0.29" This shaft would have 66.6% the weight of a solid shaft, but maintain 88.9% the rigidity.

[mrnoble]

http://youtu.be/ymrUYVZW844

Waffle's first run.

[asid61]

http://youtu.be/ymrUYVZW844

Waffle's first run.

I love that quick snap on the PVC, and the flawless lifting and lowering of it.

[Bluman56]

http://youtu.be/ymrUYVZW844

Waffle's first run.

This is very cool. Reminds me of 179s can manipulator that creatively utilized mecanum wheels.

I noticed it takes a bit of adjustment to get the grabber to snap onto the PVC tube properly. In a game where there may be many of these tubes, I would try to aim for a quick lock on the tube after your robot aligns with the edge of the box. I can imagine putting an encoder or (probably even better) a potentiometer to the extending mechanism to program it to stop going forward when the grabber is aligned with the pipe.

I'm curious what the team is doing (if anything) to address that.

[Sl3dman]

I can imagine putting an encoder or (probably even better) a potentiometer to the extending mechanism to program it to stop going forward when the grabber is aligned with the pipe.

I'm curious what the team is doing (if anything) to address that.

Actually, we are indeed using an encoder (which is reset by some limit switches that the linkage hits when in its upright position) and a PID loop to track how far forward the linkage is from the robot. The only problem is that the pipe isn't always at the same position in the box which proves to be a challenge for automation. So we ended up deciding that it would be better to overshoot the game piece by a little bit rather than coming up short to it.

[mrnoble]

Gotta drop this here. The team is now competing to make the best off-season build video, and since this is the first one, I figured it should be shared.

https://m.youtube.com/watch?v=OSH0sN4_U5k

We had our "competition" with 2240 last Saturday. It was fun, and gratifying to see how the team's work paid off. The robot was as good as any practice bot during build season, in spite of limited hours and resources. It would have been nice to see additional sensors used, including more limit switches and a gyro, and of course drive practice is something we will always need more of. Nevertheless, I'm very proud of the team.

[End-Game]

We are releasing the CAD for Waffle, Available on our GrabCAD account.

Link to this here
https://workbench.grabcad.com/workbench/projects/gcMSq3bUSTTpUPqgkfWA6ybFKtkb0lv1T0XfrrJkEzhyO1#/space/gcKnvKJtNSNDBOtFyf3aHZ5W_NByL3zjH0_vBO4C75WvaU

There are two new Reveal Videos on our Youtube account.

https://youtu.be/3aahoGNRPow
https://youtu.be/E5dKcjiZz3s

I hope everybody enjoys them.
Can't wait till Kickoff, good luck to everyone!