Position idlers to get better chain wrap -- a question about good practices.
 

So, I guess I'm looking for words of advice and consolation as I deal with trying to elegantly tension chain and achieve good wrap around my drive sprockets.


Assume that tensioning is achieved by moving the outer drive wheel and sprocket away from all of the others -- toward the end of the robot. The idler is attached to the frame rail and is fixed in place.

How close can I reasonably put the idler to the gearbox output sprocket (shown floating in space) to achieve increased chain wrap without it being so close that the sprockets are acting as spur gears, too. :)

Are there any situations that may arise from creating serpentine chain wrap, speaking generally? I've never designed such a complex chain run before; and I realize it's not even that complex.

Re: Position idlers to get better chain wrap -- a question about good practices.
 

Maybe I don't understand what you are saying, but why not try this? I made it in Paint in about 10 seconds, so don't kill me...

Re: Position idlers to get better chain wrap -- a question about good practices.
 

In 2007 we had an two 10 tooth sprockets stacked vertically with about 3/8" clearance between them. We had no problems with it at all. The chain was coming in about 30 degrees off horizontal from each side (30 degrees towards the side that gave it more wrap).

Hope that helps. The mentor who designed it with us was from 687 and he had build a lot chain driven robots in the past and I believe he said as long as there is room for the chain to go trough, there won't be problems.

Re: Position idlers to get better chain wrap -- a question about good practices.
 

Well, I guess I'm trying to find out what sort of advice or experience folks have with such an arrangement -- which is really just an extreme case for what I have on my model. It seems to me that, as you increase chain wrap with such a setup, you correspondingly increase the chain's desire to tear the idler off its mount entirely. :) I'm just curious to learn what folks have found to be a happy medium or that I'm concerned for nothing.

Re: Position idlers to get better chain wrap -- a question about good practices.
 

Bicycle idler sprockets routinely have a serpentine chain wrap--more than you seem to want. You should have no problem.

Re: Position idlers to get better chain wrap -- a question about good practices.
 

Well, this is true. We broke the shaft holding on our idler last year, but it was really a bad design flaw because the metal was so thin where it broke. The amount of force on the idler is a lot (I think our engineer was saying that we had over 400lbs of force on our idler) We ended up replacing our lathe built idler bolt assembly with just a plain 'ole bolt that worked for the rest of the regional.

Re: Position idlers to get better chain wrap -- a question about good practices.
 

We used a serpentine chain-wrap system in 2006 and it was a disaster. The chains fell off, they even broke! We had a six motor drive train and so much of the power was lost to friction that we barely had any pushing power at all!

I'll never use tensioners of any kind again. Not in the drive train, on the arm, or anywhere else. Moving the axels is a far more elegant and efficient solution.

And yes, I realize hundreds of teams use tensioners just fine. Thats just my opinion.

Re: Position idlers to get better chain wrap -- a question about good practices.
 

That's weird.... we had a serpentine wrap in a six motor drive and there we no problems. Never had a chain derail, ever. We still had great pushing power as well, no more power loss than a regular setup.

Re: Position idlers to get better chain wrap -- a question about good practices.
 

true, but they cheat and make sure there is very little chain tension on the part of the chain with the derailleur (the part with all that wrap). The loaded (top) part of the chain is a straight shot from crank sprocket to wheel hub sprocket.

One thing we learned in 2006 is that if you use the "elegant" solution of making the axle moveable to adjust chain tension, that you really do need to make it elegant, not just a bolt thru a slot.

One thing we learned in 2007 is that the idlers/tensioners need to be part of the initial design concept, it's not easy to just add them in later and have it all work out.

In other words, I don't have an easy answer for Madison!

Re: Position idlers to get better chain wrap -- a question about good practices.
 

You won't have trouble with chain if you have the correct setup. Little sprockets save weight, but they put more tension on the chains, allowing them to snap. The constant forward and reverse of our robots makes the chains slap around, often derailing them, so use guides (we make our own nylon guides, and haven't had a chain derail since 2005. Just my advice...

Re: Position idlers to get better chain wrap -- a question about good practices.
 

There's a nice picture of the setup on page 45 of FIRST Robots: Behind the Designs.

Re: Position idlers to get better chain wrap -- a question about good practices.
 

From the pictures in FIRST Robots: Behind the Design, it appears that the idlers that 100 had in 2006 consisted of cylinders of plastic (looks to be polyethylene or maybe nylon) so it isn't really a big surprise to me that it had high frictional losses. On the other hand, we have used ball bearing idler sprockets for the last two years, and they spin really easily even with a significant amount of force applied to the chain.

Re: Position idlers to get better chain wrap -- a question about good practices.
 

oh man.... dont have my book with me, but that much wrap on plastic rod seems like a lot of friction. No wonder.

Re: Position idlers to get better chain wrap -- a question about good practices.
 

Actually if they did it right there might not have been much friction at all. The trick is to install the rod in such a way as to make sure the chain is tight and then let it "wear in" a slot on the plastic rod. You migh want to pre-wear a groove in the rod to make sure the chain stays put, but it probably is not necessary. Once the tight chain wears into the rod a little bit it will go slightly slack. It will then cease to wear and the friction will be negligible.

Another approach is to not use tensioners at all. We have done this the past few years. In this case you need to be able to locate the sprocket axles precisely. You need to design the chain loop to have an integer number of links and preferably an even integer.

If there's interest I can do a white paper on how to do this.

Chris

Re: Position idlers to get better chain wrap -- a question about good practices.
 

If you can mount the drive motor/trans assembly midway between the wheels AND have it mount such that it can be moved vertically slightly, you could use a double-row drive sprocket and two independent chains. Each chain run is shorter, which seems to reduce tensioning problems. To tension it, just loosen the drive assembly and shim it up slightly. Team 975 has done this a couple of years. Tensioning with #35 chain was rarely required, and slipping a thin washer or two between the drive mounting plate and the frame seemed to take up more slack than you'd imagine.

http://www.chiefdelphi.com/forums/at...d=119696522 0

Re: Position idlers to get better chain wrap -- a question about good practices.
 

http://www.chiefdelphi.com/media/photos/27587

Here's a picture of our drive last year. I think it's similar to what you are looking at doing. We had zero problems with this and it worked very well. So well in fact, that we are using a version of it this year. Hope this helps.

Re: Position idlers to get better chain wrap -- a question about good practices.
 

We used the little things that you take off of the gear end of the big CIMs that come in the kit. They are sort of a spring loaded idler. We have used lots of other hand-made items and also moved the wheels in and out in years past. Last year we had no problem with tension. They kept the chain tight even in extreme situations of flex due to the constant pressure of the spring load. I am sure they can even be used here to get your desired serpentine effect you want which would mimic the serpentine belt system on the alternator on your car. I am not with Cyber Blue (234) anymore but if you contacted one of them I bet they would send you a picture.

Re: Position idlers to get better chain wrap -- a question about good practices.
 

I'm a fan of having 2 separate chain loops, one for front one for rear, it gives a lot of engagement on all the sprockets, and provides redundancy in the event of losing a chain. We had another robot run on top of us in 2005, and bend our front axles enough that one of our chains came off. But as long as we were flat on the ground (and not tipping forward :p like it liked to so so much) control was not hindered much at all.

If you have a metal idler sprocket, then you still have 4 sprockets for each side, so any weight savings from having the single loop should be minimum. We used sliding axle mounting blocks for tensioning, each one with at least 6 socket head cap screws to hold it in place in the frame rail

Re: Position idlers to get better chain wrap -- a question about good practices.
 

Madison,
We've built serpentine #35 - 6 wheel drive systems that incorporated idler / chain tensioners mechanisms in the past, along with many other types of chain drive systems.

The distance apart and the amount of chain wrap is directly related in a design like the one you are showing. We've had them as close as "the teeth almost touching", to a few inches apart with 3 teeth engaged.

So, it can be done - but, there are other issues to also consider when routing the chain this way.

The chain wrap and chain tensioning will impact the drive motor output and temperature.
Too much tension is not good and too little tension is not good. Too much wrap is not good and too little wrap is not good.

The chain tensioning can also be achieved by placing shims or blocks beneath the idler or motor mount. Thus lifting one or the other upward, taking up the chain tension and increasing the chain engagement. This can be done either in conjunction with, or seperate from, sliding fore and aft on the rail.

Also watch that the longest unengaged side of the chain (in this case it is on the bottom side) is protected. Any object that can push against the side of the loosest side will act like a derailer.

I may not have provided a specific answer to your question, but I hope this helps,

Mike Aubry

Re: Position idlers to get better chain wrap -- a question about good practices.
 

ok so i had a quick question... with te tough boxes we recieved a sprocket that fits perfectly ont the box... however the output sprocket is at least 2x the size of the input... so... for every rotation in the tough box we get .5 on the actual wheel...for a 1st year team we are trying to get a drivable robot thatd be a decent accomplishment in out eyes... and we are working towards that but a faster driving robot would be nice...

and my second question is actually relevant to this forum... where do you get idlers... and any ideas on mounting them?

Re: Position idlers to get better chain wrap -- a question about good practices.
 

If you can vary the gear ratio provided in the Kit of Parts by replacing either of the sprockets. To go faster, use a smaller sprocket on your wheel and leave that on the Toughbox output unchanged -- or similarly, use a larger output sprocket on the Toughbox.

Keep in mind when sourcing sprockets that the Toughbox output is a 1/2" diameter, keyed shaft. It's very difficult to find COTS sprockets that have a 1/2" bore and keyway. You may have better luck buying #35 sprockets from www.andymark.biz or www.ifirobotics.com, as those contain a bolt pattern that will mate directly with the Kit of Parts wheels.

Idlers can be made in a variety of ways. Search these forums for "idlers" or "chain idlers" or similar and you'll find lots of information.

Re: Position idlers to get better chain wrap -- a question about good practices.
 

Another way to get a faster speed out of the kit transmissions is to take out the second reduction stage. You have to move some parts around, add a washer to space the gear on the output shaft, and add more holes to the aluminum plate to clear the ends of the motor shafts. Also if you want to use the gear tooth sensors, you'll have to relocate them (after you get them repaired). Then you can use about 3:1 chain reduction (with 6" wheels) to give you a reasonable "high gear" robot speed. The Andymark 12 tooth transmission sprocket, as well as 36 tooth wheel sprockets from Andymark or IFI, will do this.

Re: Position idlers to get better chain wrap -- a question about good practices.
 

Another option that we considered was to buy the wheel hub from AndyMark and a larger sprocket (I think the KOP wheel sprockets are 22 tooth - you could use a larger one on the output shaftO)- use this as the output gear of the toughbox instead of actually attaching it to the wheel.

for the chain tensioner question, we designed a custom spring loaded idler with bearings pressed into the 22 tooth sprockets. we are not concerned about the chain wrap issue, however - just tensioning.

I agree that a little serpentine alignment of the idler and output sprocket will work fine like the ones used in cycling (for a bike, the sprockets are directly above each other and spaced about 2" apart). they are usually 11-12 tooth sprockets so the serpentine path engages ~1/2 of the teeth. they are spring loaded derailures, however, so a stagnant idler like your design will take a lot of force. Your arrangement will not likely slip (which is why you have it in the first place) but will cause a lot of stress on the frame, idler, and gearbox. to get around this much stress just make sure that the chain is not too tight - you want to just barely take up the slack. On a fixed gear bike (no derailure and no coasting) the chain can move about 1/4" when the wheel is locked in place - this relieves all of the drag on the chain.

I hope this helps.

Re: Position idlers to get better chain wrap -- a question about good practices.
 

As another source: McMster-Carr http://www.mcmaster.com
"Finished Bore" is keyed - just as Madison posted, be sure to check bore size and key size....page 1012, 9 tooth to 25 tooth are available in 1/2" bore for #35 Chain

We're also experimenting with new idler designs for chain drive system tensioning.
Chris H - A white paper would be great! The "Milwrights and Mechanics Guide" is tough for our kids to follow...