Re: Crab Drive
 

Oh, 60 RPM free speed! That makes a difference-- I'd say you turned at less than that under load, right? And with two sides/ends, it's a bit different than with, say, the V6 system. (For example, it's possible to drive it like a skid-steer...)

Re: Crab Drive
 

Here's my two bits:

2008 was 68's first year experimenting with crab drive. I'd say we were very successful. In fact, I'm almost entirely sure we had zero downtime. That's really good for an experimental system!

The mechanical side: (keep in mind I'm the programmer, so I don't know a huge amount about the mechanical side, but I'll try) Don't think you necessarily need to use chain. We used cable. It went around a spool in the middle of the two wheels, and then out to each of the two front or back wheels. (the two front were tied together, and the two back were tied together) We mechanically limited our drive to +/-90 degrees so that if there was anything wrong with software, nothing got destroyed.

The programming side: We used straight-up potentiometers, gearing them so we got almost the full travel of the pot per full travel of the mechanism. I broke the control software into two pieces. The first piece read the pot values, scaled them to degrees, and drove the PWM outputs to keep the mechanisms at the stored degree values. The second piece read the joystick and set the degree values for the first piece to use. This setup was really good because it was easy to write auton--stop executing the piece that reads the joysticks, and set the degree and velocity values for the first piece to use.

Let me know if you'd like to see some code--I am willing to release fragments of the crab modules I used this year.

As a side note, as a programmer having an interface to the robot through serial was HUGE. I could poll the values of all the pots to figure out why software was pegging the mechanism at +90. I could re-calibrate easily without re-compiling code. I am also willing to release pieces of that code.

I can also get pictures of our setup.

Ahh...crab drive. This was a good year--that was really fun to program! :) I'm not a nerd or anything...when pigs fly...

JBot

Re: Crab Drive
 

the modules could rotate forever, so that should not be an issue. As for programming, our team needs a bit of help, we have a mentor who did it all last year, so we might take you up on that. Thanks a lot. :D

Re: Crab Drive
 

*hits self over forehead* Belts! Same deal as cable, with the sure grip of chain, or close to it. Either way, keep them tight or you'll get some slipping modules.

Re: Crab Drive
 

Unless you are going to be using slip rings, your modules will be limited by the wires to the CIMs. They can only move and twist so far. Therefore, most crab designs only allow for 180-360 degrees of motion.

here's a good example of what I think he was describing with a cable system.

Re: Crab Drive
 

Which would you recommend?:D

Re: Crab Drive
 

Actually, our earlier prototype used belts. They were screwed to the modules in one place per pod, so they did not slip. We only ditched them because the design we used took up more space than cable did. Honestly, I think belts would've been a workable solution this year.

I'm not so sure you want to do that. I've seen cases in the shop where a pot has come unplugged or some other disaster has happened in which the mechanism got pinned to one of the limits. Are you absolutely certain your device can withstand two minutes of spinning full circles? That's worst-case, of course, but it is something that needs to be thought of. We had limit switches planned, but alas, time constraints kept me from implementing them...luckily we didn't need them.

That looks to be the same basic design I was describing.

I will try to scrape together a good list of things to check out and a basic programming guide for crab. It may not happen today or tomorrow, though, as I have my open house to prepare for.

JBot

Re: Crab Drive
 

Anytime is good, I will be out of town for 2 weeks. You could send it to davidallen@westminster.net whenever you get around to it. I really appreciate it. :D

Re: Crab Drive
 

I'd say use a timing belt, properly tensioned; some teams have used polypropylene belts (I think) in their drivetrains. A quick CD-Media search could help.

For a cable, you would probably need a crimper. JBotAlan, am I right about that? If so, that's more money to spend...

Re: Crab Drive
 

I don't think we used any kind of crimper, though I would have to ask the mechanical guys. I'm pretty sure it was just screwed down, possibly under a washer? I'll see if I can get you more info...sorry for the lack thereof.

JBot

Re: Crab Drive
 

What about the plastic belting like that used for car windows? I've seen it used on 71's bot for swerve steering and 27 used it to elevate thier shooter this year. I've been told several times that it is available from McMaster-Carr, but my memory doesn't seem to be long enough to remember what it is called.

Re: Crab Drive
 

Window tape is what that is called, I think.

Re: Crab Drive
 

well rick all i can say is wow we (your mentors) are really proud of you for taking a bit of independence in this thin.
I would look into belt also but as you know we have had issues in the past with belt.
something i discussed earlier with the thread makes was weight. does anyone think this looks at all bulky? I have never done crab before so that is why i am asking.

Also yes, we are building this crab over the summer so that we can get a feel for it in the event that we wish to pursue it in the next build season. And this crab may also serve the purpose of helping the newer members get involved with building as well as learning about the different systems involved in FIRST, and yes of course, the main thing, outreach to get others in the community excited about FIRST.

Re: Crab Drive
 

Here is a picture of our first module, which we have (finally) gotten to build. Everything works well so far. We have ordered parts for the other 4 modules, and we already have those plates cut. More pictures are at http://s296.photobucket.com/albums/m..._tab_my_albums
It is now about 7 lbs, we are hoping we can take off at least 1.5 pounds for the other modules. :D

Re: Crab Drive
 

I am by no means a gearbox expert, but those gears look quite thin compared to many other gearboxes and Crab drives I have seen.

Any particular reason for that, other than weight? Sure they will work?