Chief Delphi - Timing Belt Drive System

Quote:

Originally Posted by KenWittlief

Something has been bothering me about the drawings that were posted by a few people for proposed belt drive systems. I just figured out what it is.

The serpentine belt on the AC and alternator on a car only spins in one direction. The drive belt (or chain) on a motorcycle wheel normally only supplies force in one direction (unless you let off the throttle to slow down)

those systems are ok with spring tensioners to take up the slack. But for a robot drive train that goes both ways, the side of the belt that is under tension would pull the slack out of the tensioner. When you reverse directions it will pull the slack to the other side.

This is not good. That would introduce slop into the system. You want your wheels to turn together, to supply force together. As the spring tensioner is pulled tight on one side, then on the other that will make the wheel on the other side of the tensioner turn unsynchronized to the rest of the drive train.

So Im thinking you dont want a spring tension system - you want something that is adjusted manually, a lead screw sort of thing, that wont flop around when the robot reverses direction.

Excellent point Ken

One important thing to keep in mind is that you normally see spring tensioners in friction drive systems and in situations where the belt/chain length might change. It a timing belt application you have a positive drive system, not a friction system, and a steel reinforced are even a nylon reinforce belt will have basically no stretch. There is no need for a spring loaded tensioner. All you need is to be able to move one of the pulleys to take up any slack that comes from manufacturing variation. If your drive system is simple enough you may even be able to design it with no adjustment. If you have ever torn apart the from end of an overhead cam engine you would see what I mean. They often use a timing belt with no provision for tensioning.

Matt B