View Single Post
  #8   Spotlight this post!  
Unread 23-04-2003, 18:33
Jnadke Jnadke is offline
Go Badgers!
#0093
Team Role: Alumni
 
Join Date: Jan 2002
Location: Appleton, WI
Posts: 775
Jnadke is on a distinguished road
Send a message via ICQ to Jnadke Send a message via AIM to Jnadke Send a message via Yahoo to Jnadke
Quote:
Originally posted by wysiswyg
Stupid question but did you have a differntial on you drive shaft or did you just mount a chain to it directly???? I was wondering this.
We did originally have a differential on it, but we had many problems at the Midwest Regional, so we ended up dumping it and going with a drive system developed by one of our sister teams (with the help of our engineers). The other option was to replace the gears on the differential with a belt system. The problem wasn't in the differential itself (that worked flawlessly), but in transferring power to the differential. I noticed they were using 1/4" shafts and set screws...

The rear (newer) wheels had less traction than the steering (front) wheels (which we were originally going to use with the diff), so there were no problems in turning even though both rear weels were operating at the same speed. Unfortunately we were pushed a little easier than we would have hoped.


I'm just a college student/alumni that was helping to mentor them. I wasn't at nationals but I was at the Midwest Regional.

Quote:
Ahhh thanks by the way do you know where they got it??? The only differentials I have seen look like they would not work in a robot.
The differential was actually a free-floating differential (also called an open differential - there are many different designs). It consisted of a nylon housing to which a large gear was attached (the drive gear). The housing was basically 4 bevel gears of the same size, to which 2 shafts were attached (the 2 that are perpendicular to the drive gear).

You can see a good picture here:
Differential
Steering

Unfortunately, I don't have any information as to whether they made it or bought it. If you look at the design, it wouldn't be that hard to make with a mill and drill press. All you need is a high-strength plastic like Delrin or Polyethylene. Take a square of it and mill out a perfect square inside. Drill 4 symmetrical holes on each side and mount the bevel gears (preferrably steel). They should all be mounted solid (can't move), 2 of which should be attached to the wheel shafts. Then attach a drive gear (or sprocket) to the housing (not to the wheel shaft). A bit more complex, but that's the basic idea.

At MR they replaced the nylon gears with steel one's. The differential worked nicely for a match, but then it broke again when a nut came out, so the decision was made to switch to the alternate, alternate drive system permanently (Plan A was nylon gears and hollow driveshaft, Plan B was steel gears and solid steel driveshaft, Plan C was belts, and Plan D was the drive system you see now). If you can get past all the kinks (using 3/8" steel shafts, keyways, and getting rid of those nuts), it'd be perfect.


There are many different kinds of differentials. There are differentials with bevel gears (open differentials). There are differentials with worm gears (Torsen differentials). There are even differentials with planetary gears (TrueTrac differentials). Open differentials are the simplest and most efficient, but if one wheel loses traction you're screwed. Torsen are the most complex and least efficient, but they have the ability of transferring torque to the wheel with traction. The TrueTracs are somewhere in the middle, but they require a little bit of initial friction to transfer the power to the other wheel (the slippage can't be instantaneous).

Quote:
Were you ever worried that if one of your wheels were up in the air that the robot would not move since all of the power would be going to that wheel?
Yeah, that was a concern, but if you look at the steering picture, you can see that our it pivots on the vertical. It's semi-articulated so that both rear and front wheels are on the ground no matter what position the robot is in relation to the ramp. The only way we would have that problem is if a robot got under us.
__________________
The best moments of our lives fall in two categories: those that did happen and those that did not.

Last edited by Jnadke : 23-04-2003 at 19:32.