Chief Delphi

Chief Delphi (http://www.chiefdelphi.com/forums/index.php)
-   Technical Discussion (http://www.chiefdelphi.com/forums/forumdisplay.php?f=22)
-   -   Timing belt in drive success (http://www.chiefdelphi.com/forums/showthread.php?t=108091)

Brandon Holley 05-09-2012 13:53

Re: Timing belt in drive success
 
Quote:

Originally Posted by AdamHeard (Post 1184284)
Our dream is to run 5mm pitch gt2 9mm wide in 24T with exact c-c; per the math this is marginal, so our expectations is that our prototype will fail. We would then run with tensioners.

I think you guys will be pleasantly surprised ;).

-Brando

steelerborn 05-09-2012 14:03

Re: Timing belt in drive success
 
Yeah I was looking at the strengths of the different belts and it seemed like 15mm was just kinda overkill, I really think the thinner belts look a lot cleaner too.

sdcantrell56 05-09-2012 14:21

Re: Timing belt in drive success
 
Quote:

Originally Posted by steelerborn (Post 1184292)
Yeah I was looking at the strengths of the different belts and it seemed like 15mm was just kinda overkill, I really think the thinner belts look a lot cleaner too.

As someone who has designed and run belt drivetrains the past 3 years I will say 15mm width is not overkill. We tried 9mm width GT2 in 2010 with bigger pulleys than most people are talking about and had belt ratcheting issues every time we crossed the bumps. Swapping to 15mm width for 2011 and 2012 resulted in the belts never ratcheting even with larger wheels. Stick to 15mm width

Rob Stehlik 05-09-2012 14:37

Re: Timing belt in drive success
 
All of this discussion on timing belts is great. I have wanted to use them for a couple years, but availability and lead times has been an issue. Chain and sprockets are just so much easier to get.

That being said, we are close to finishing a 6wd drivetrain that uses a single 15mm wide 5mm pitch gt2 belt on each side. While the student was putting together the design, we kind of guessed at the belt width and pulley size. Actually, he wanted the pulleys to fit inside a 2" tube, so that dictated their size.

Recently I have been looking through the gates design guide, and the proper way to select your belt and pulley size is to base it on the power input. A CIM motor's peak power is 340 watts, so the average gearbox is putting out 680 watts, or 0.91 hp. In the design guide there is a table that shows you the power rating of the belt for a given rpm and pulley size. In our case, we use a single belt, so it should be able to handle the full 0.91 hp. The chart says we should have used 30 tooth pulleys instead of 20 tooth. Oops! I'm sure it will run fine as a prototype, but we'll have to be more careful when designing an actual robot. I'm curious what design methodology others use?

For those who are interested, you can download the Gates design guide here:
http://www.gates.com/brochure.cfm?br...ation_id=11539

Also, for those making their own pulleys, you can buy flanges separately from SDP.

sdcantrell56 05-09-2012 16:11

Re: Timing belt in drive success
 
Quote:

Originally Posted by Rob Stehlik (Post 1184294)
All of this discussion on timing belts is great. I have wanted to use them for a couple years, but availability and lead times has been an issue. Chain and sprockets are just so much easier to get.

That being said, we are close to finishing a 6wd drivetrain that uses a single 15mm wide 5mm pitch gt2 belt on each side. While the student was putting together the design, we kind of guessed at the belt width and pulley size. Actually, he wanted the pulleys to fit inside a 2" tube, so that dictated their size.

Recently I have been looking through the gates design guide, and the proper way to select your belt and pulley size is to base it on the power input. A CIM motor's peak power is 340 watts, so the average gearbox is putting out 680 watts, or 0.91 hp. In the design guide there is a table that shows you the power rating of the belt for a given rpm and pulley size. In our case, we use a single belt, so it should be able to handle the full 0.91 hp. The chart says we should have used 30 tooth pulleys instead of 20 tooth. Oops! I'm sure it will run fine as a prototype, but we'll have to be more careful when designing an actual robot. I'm curious what design methodology others use?

For those who are interested, you can download the Gates design guide here:
http://www.gates.com/brochure.cfm?br...ation_id=11539

Also, for those making their own pulleys, you can buy flanges separately from SDP.


I might be wrong but I think the Gates guide does not really account for the rapid starting and stopping that a FRC robot encounters either. Belt width is directly correlated to the ability to resist skipping/ratcheting.

Akash Rastogi 05-09-2012 17:30

Re: Timing belt in drive success
 
Quote:

Originally Posted by sdcantrell56 (Post 1184293)
As someone who has designed and run belt drivetrains the past 3 years I will say 15mm width is not overkill. We tried 9mm width GT2 in 2010 with bigger pulleys than most people are talking about and had belt ratcheting issues every time we crossed the bumps. Swapping to 15mm width for 2011 and 2012 resulted in the belts never ratcheting even with larger wheels. Stick to 15mm width

Did you happen to have any ratcheting balancing on the bridges this season Sean?

Adrian Clark 06-09-2012 22:10

Re: Timing belt in drive success
 
Quote:

Originally Posted by AdamHeard (Post 1184284)
Our dream is to run 5mm pitch gt2 9mm wide in 24T

I might be a little late here, but could you elaborate on how you chose your width and tooth#?

Without knowing your gear ratios I can't really tell much about the forces that are going to be on your belts. But if I were to assume that your wheels are between 3.5" and 4" and your top speed was going to be 15-17fps (fairly standard WCD numbers), then according to the gates catalog the smallest pulley you should run would be 30T, but that's just for high gear. On low gear (assuming you have a 256% spread, just for comparisons sake) the gates catalog says the minimum pulley size would be 56T. Also, gates has a minimum recommended sprocket OD chart which says that in my above hypothetical drivetrain the minimum sprocket OD in high gear would be 2.2", and low gear is off the chart, literally.

I apologize in advance if I come off as bashing you here. All my above information comes from gates charts, not from experience, so chances are that you know something I don't when it comes to implementing a belt drive.

sdcantrell56 06-09-2012 23:22

Re: Timing belt in drive success
 
Quote:

Originally Posted by Akash Rastogi (Post 1184316)
Did you happen to have any ratcheting balancing on the bridges this season Sean?

No ratcheting that I ever heard and that even included driving onto the bridge and engaging mechanical brakes which locked the drive transmissions.

AdamHeard 07-09-2012 01:07

Re: Timing belt in drive success
 
Quote:

Originally Posted by Adrian Clark (Post 1184491)
I might be a little late here, but could you elaborate on how you chose your width and tooth#?

Without knowing your gear ratios I can't really tell much about the forces that are going to be on your belts. But if I were to assume that your wheels are between 3.5" and 4" and your top speed was going to be 15-17fps (fairly standard WCD numbers), then according to the gates catalog the smallest pulley you should run would be 30T, but that's just for high gear. On low gear (assuming you have a 256% spread, just for comparisons sake) the gates catalog says the minimum pulley size would be 56T. Also, gates has a minimum recommended sprocket OD chart which says that in my above hypothetical drivetrain the minimum sprocket OD in high gear would be 2.2", and low gear is off the chart, literally.

I apologize in advance if I come off as bashing you here. All my above information comes from gates charts, not from experience, so chances are that you know something I don't when it comes to implementing a belt drive.

We would be running 3.5-4" wheel with the stock AM shifter ratio (11T pinions instead of 12T for 4" wheel).

I've used belts a lot on the team and in personal experience, and we know that for the reduced runtimes we experience in frc, you can get away with higher loading than gates recommends by a good margin. We know that design as posted doesn't pass the Gates' documentation, but that doesn't mean it won't work. It's a pretty marginal design, we don't expect it to work in an exact c-c implementation, but that's why we test.

The sizes were determined as an easy size to integrate into our existing WCD.

Adrian Clark 07-09-2012 03:18

Re: Timing belt in drive success
 
Quote:

Originally Posted by AdamHeard (Post 1184506)
We would be running 3.5-4" wheel with the stock AM shifter ratio (11T pinions instead of 12T for 4" wheel).

I've used belts a lot on the team and in personal experience, and we know that for the reduced runtimes we experience in frc, you can get away with higher loading than gates recommends by a good margin. We know that design as posted doesn't pass the Gates' documentation, but that doesn't mean it won't work. It's a pretty marginal design, we don't expect it to work in an exact c-c implementation, but that's why we test.

The sizes were determined as an easy size to integrate into our existing WCD.

I'm sure the gates charts have a sizable safety margin, but without knowing the exact size of that margin its hard to tell if a belt drive will have ratcheting problems or not. But I guess that's what testings for. My biggest worry about using pulley below suggested size is that weather you're using exact c-c or using a tensioning system if the pulley is too small there's nothing that can be done to prevent it ratcheting, even with massive amounts of tension.

Also, why did you choose to use 9mm belt over 15mm?

DampRobot 07-09-2012 09:56

Re: Timing belt in drive success
 
Quote:

Originally Posted by Adrian Clark (Post 1184518)
I'm sure the gates charts have a sizable safety margin, but without knowing the exact size of that margin its hard to tell if a belt drive will have ratcheting problems or not.

Actually, I would guess not. I was doing some calculations for bevel gear max HP and torque, and my results were within rounding error of the published values. Perhaps Gates divides by a safety factor, but I would suspect not.

Also, keep in mind that the amount of torque that the pullys will start slipping at is highly dependent on the tension on the belts. Perhaps all these stories of belts slipping is less because they exceeded their rated HP and torque, but because they were in a center to center design and couldn't be properly tensioned.

Brandon Holley 07-09-2012 10:07

Re: Timing belt in drive success
 
Quote:

Originally Posted by AdamHeard (Post 1184506)
We know that design as posted doesn't pass the Gates' documentation, but that doesn't mean it won't work.

For our drivetrain this past year, the Gates documentation vehemently showed our design would fail. We rolled the dice and decided to go for it anyway, based purely on intuition and previous experience.

After the Suffield Shakedown scrimmage, New York City Regional, Boston Regional, Championships, Battlecry, Beantown Blitz, and IRI we've haven't had a single slip, ratchet, thrown belt or any other failure related to our belt selection. This was with an exact c-c setup.

Not that I condone ignoring manufacturers data sheets and calculators, I just wanted to point out that its not always black and white.

-Brando

Chris is me 07-09-2012 10:27

Re: Timing belt in drive success
 
Quote:

Originally Posted by AdamHeard (Post 1184506)
It's a pretty marginal design, we don't expect it to work in an exact c-c implementation, but that's why we test.

The sizes were determined as an easy size to integrate into our existing WCD.

I'm a little confused. Are you saying the design won't work because of the exact center spacing? I don't see how that would make the belt any weaker than a tensioned system - if anything a tensioned system would be less foolproof and more finicky.

Quote:

Originally Posted by DampRobot (Post 1184538)
Also, keep in mind that the amount of torque that the pullys will start slipping at is highly dependent on the tension on the belts. Perhaps all these stories of belts slipping is less because they exceeded their rated HP and torque, but because they were in a center to center design and couldn't be properly tensioned.

Of the two belt designs we've run, our exact center design didn't ratchet / slip and our tensioned design did. It has a lot more to do with drive sizing than method of holding tension.

JamesCH95 07-09-2012 10:44

Re: Timing belt in drive success
 
My limited experience with belt drives tells me that exact C-to-C is fantastic.

I designed and built a CNC plasma cutter drive (two identical units) when I was in HS, back in 2005, using exact C-to-C spacing belt drives. When the plasma cutter table was disassembled in 2011 in favor of a more modern setup the belts seemed as taut as the day I installed them. Granted they didn't see production-level usage, but they did suffer HS students leanring how to use CNC machinery for six years ;)

AdamHeard 07-09-2012 14:28

Re: Timing belt in drive success
 
Quote:

Originally Posted by Chris is me (Post 1184542)
I'm a little confused. Are you saying the design won't work because of the exact center spacing? I don't see how that would make the belt any weaker than a tensioned system - if anything a tensioned system would be less foolproof and more finicky.

You have to assume that an exact c-c will run at a reduced strength.

For a given toothcount, beltwidth, etc... combination the strength is proportional to tension in a bell curve shape; too little or too much tension is a strength loss.

With exact c-c you are assuming you're merely high enough on the peak, it would be foolish to believe you both picked the EXACT best c-c value and had zero error machining.

The lack of a tensioner is certainly attractive with less work to fab, assemble and maintain. The cost of this however is reduced strength compared to a properly tensioned setup.


Quote:

Originally Posted by Adrian Clark (Post 1184518)
I'm sure the gates charts have a sizable safety margin, but without knowing the exact size of that margin its hard to tell if a belt drive will have ratcheting problems or not. But I guess that's what testings for. My biggest worry about using pulley below suggested size is that weather you're using exact c-c or using a tensioning system if the pulley is too small there's nothing that can be done to prevent it ratcheting, even with massive amounts of tension.

Also, why did you choose to use 9mm belt over 15mm?

Like i mentioned before, the toothcount, width, etc... were all picked so that they could integrate into our existing WCD (with zero modifications to the WCD). If substantial modifications are required, it diminishes the appeal as we have years of standardized drive parts useful for prototyping, practice bot, etc...


All times are GMT -5. The time now is 03:55.

Powered by vBulletin® Version 3.6.4
Copyright ©2000 - 2017, Jelsoft Enterprises Ltd.
Copyright © Chief Delphi