We’re looking at using timing belts this year for the first time. I have a desired center to center distance of 4.5 inches.
When I punch that in with a 24 tooth and 48 tooth pulley using a 3mm belt, it gives me a lower belt of 110 and an upper of 115 for a GT2 3mm. However, I know people in high speed applications tend to run the belts pretty loose. Does anyone have a rule of thumb on just how loose timing belts should be run?
If I go with the 115 tooth belt, I get an extra .14" center to center distance (4.6442"). So if I run that at 4.5, is that enough to minimize friction?
I’ve seen it on a number of FIRST robots. The one that stuck in my mind is Hot’s 2013 frisbee shooter because I talked with them and asked one of their team members about it. Their belts were quite loose. I also remember a short discussion about it here once upon a time, but the search function isn’t helping.
As long as the belt won’t slip, it should be ok. Knowing the application could make a difference. If it’s for a spinning wheel that needs very little torque to start spinning, then you can probably get away with quite a bit of looseness. I think that loading is probably more important than speed, for determining how tight the belt needs to be.
Our team uses timing belts quite often, usually for drivetrain stuff. When we use it for mechanisms and such, if we have enough slack where the belts are slipping on the pulley, or we need it tight for its purpose, we add a tensioner. The tensioner should take out slack, but the belt and pulleys should still be able to free spin. If you can’t free spin or it’s too tight, then your tensioner is too strong. Tensioners don’t need to always be spring loaded either. It can be a couple bearings on a bolt and the belt goes around it, just as long it takes out most of the slack out.
The timing belts on our 2012 and 2013 shooters had the motor pinon slide to adjust the tension. They were always at what I would consider normal tension for belts; meaning the teeth don’t slip, but you can deflect the belt slightly with a finger. Now that we have machining capabilities, our belts are set to the exact center-to-center distance. We have done this for drive trains, intakes, conveyors, and shooters with no issues. I agree with MrForbes that loading is more important than speed for determining belt tension.
With many belt runs stuck on top of each other, like with 254’s 2017 hopper, if they’re all tensioned you’d get fairly large losses. But if you just have 1 run, then it doesn’t matter too much, I believe you can get >90% efficiency with timing belts.
For a single run I would do the calculated CC distance, whereas if you have multiple runs stacked up on top of each other in a situation where it doesn’t matter if the belts skip (like an intake or hopper), then I’d go -0.060 or less.