paper: Belt vs. Chain Drive Evaluation - Team 234

frc234
chain
belt
cyberblue
#1

Thread created automatically to discuss a document in CD-Media.

Belt vs. Chain Drive Evaluation - Team 234
by: Chris Fultz

Team 234 completed a performance evaluation of belt drives and chain drives for use in FRC robots. This paper contains the results of the testing, how we defined our tests, and the test processes.

This is a formal test report of belt drive vs. chain drive for FRC robots. We compared several parameters in “side by side” testing to determine the “best” solution for FRC robot drive systems. We compared weight, performance, efficiency, noise levels and other parameters. We used some basic design of experiments processes, ccreated test plans and processes, and then completed the testing and analyzed the data.

The design, fabrication, testing, analysis and reporting were completed by Team 234, Cyber Blue, in the fall of 2008.

FIRST Robot Drive System Analysis - Team 234.pdf (1.17 MB)

This paper is a test report of work completed by Team 234 during the fall of 2008.

We compared Belt Drive vs. Chain Drive on several parameters, working with some basic Design of Experiments principles and working to make sure we would obtain objective, unbiased results from the testing.

Our object was to determine if there was a “best” drive type for FRC robots.

#2

Amazing test and report, well done! I expect the entire project was a great learning experience for everyone on your team, it’s way more exposure to ‘real engineering’ than most high-schoolers ever come close to seeing.

I’m not very surprised by the results, but it is good to have a solid reference. Unfortunately the gain in performance from using belts over chains isn’t worth the extra design effort and hassle for most teams, but it’s good to see it being pursued (and in such a professional way!).

#3

It is not only a great learning experience, but it is a great way to keep the team involved all year around and mentor new members before the season begins. Plus, with no strict time constant like during the build season,it gives us all time to fully understand the all aspects of the project.

I personally didn’t realize that how many variables you have to keep the same or similar so we are analyzing just the belt and chain.

#4

Thank You 234 for this great paper! Team 612 has been using belt’s on our drive train since 2007 and have been trying to convince other teams about the improvements they provide but we have never had any hard facts until now.

#5

Has either 234 or 612 (or any other team using belts on their drive train) used non-aluminum timing belt sprockets? Or more specifically, has anyone used the SDP/SI (or similar) timing belt sprockets that have an aluminum Fairloc/T-Insert hub embedded inside a polycarbonate timing belt sprocket on a FRC drive train?

I’ve been using some of these with 5mm pitch belt on the shooter (as shown below) and ball intake on 228’s robot this year, and was considering them for the drive train in the future. They are a lot lighter than the aluminum ones (freeing up machining resources for more important things), and at high RPMs are noticeably quieter than roller chain (compared to our Aim High robot).

http://www.team228.org/gallery/103/slideshow/build-season-week-six_b03b0-17e10.jpg](http://www.team228.org/media/pictures/view/4959)

#6

We have not used them before. I personally have used plastic sprockets for FRC and other applications from various sources with good results. Perhaps you have given us a Fall 2009 test project…

#7

I havent used them in a drivetrain, but what seems to fail in drivetrains is the belt. I cannot verify that a plastic one will definitely work, but my feeling is that it probably would depending on size, mounting style, etc.

#8

So, having conducted this experiment, what does 234 plan to use in the future? I see this year you guys went with chains on the drive and belts for everything else.
Another interesting experiment would be to see how chains & belts respond to rapid reversals of direction, as often seen in FRC drive systems. Is one more durable than the other in this respect?

#9

During the 2007 and 2008 seasons Team 612 used plastic pulleys with the inserted aluminum hub. Both systems experienced 2 regionals, Championship and multiple off season events and still have yet to show signs of failure. Currently my teams photo gallery is under construction so I don’t have any better photos of our 2007 or 2008 drive train, this is all I could find.

http://www.chiefdelphi.com/media/photos/32192

What we recommend using is 15mm wide, 5mm HTD belts. In 2007 we used a different kind of belt, that didn’t have any problems, but after testing before 2008 season we determined these belts have the past teeth pattern for the FIRST application. If your team uses Inventor 2009, the loop tool helps calculate the pitch length for the exact belts you need. Also through SDP/SI and BBman they provide belt length calculators on their website to help determine the exact belt you need.

This year we have further improved our design by eliminating the use of a transmission. Not only do we save weight by using belts, we save weight by not using transmissions. We currently are running a 12tooth pulley on our cim motors directly to a 90 tooth pulley on our wheels that were custom made on our CNC.

http://www.chiefdelphi.com/media/photos/32600

If your at New Jersey, VCU or Championship and have any questions about belt drives, stop by 612’s pit and will be happy to answer or explain how we use belts on our drive train.

#10

If you don’t mind, how did you exactly make the pulley on the cnc? We are interested in doing this.

#11

We did not use belts for our 2009 robot for a few reasons -

With this game, and the minimal torque requirements, we did not feel #35 chain or 15mm belt was required. Since we have no experience with smaller belt (9mm), we decided to use #25 chain on our drive. (We are direct driving our rear wheels, and chain to the front ones). We have our practice robot and may switch it to belts to get the experience.

We have been using our “test chassis” to pull a trailer while we practiced.THis chassis has chain on one side and belt on the other. So far, the chain has come off two or three times from rapid direction changes or wall impacts, and the belts have not come off once. This is not really valid testing, but is making us more and more comfortable with the belt drive / pulley option.