Chain Tensioning
 

Okay, I was talking with someone about ways to tension a chain. And one of the things that was brought up was a spring. (Or some kind of potential energy device) Basically you would have a rod mounted on a pivot. you would have a spring attached to one side, and a free floating sprocket on the other. (Or you could use AM chain tensioner if you modified it a little.)

We loved the idea. But that's just it; it is still an idea. We do not know of anything like this every being used on a FIRST robot. We know it has been used before, but it's unknown if it's been used on FIRST robots. Is there a specific reason? Or is it just because no one has tried it?

Thanks
-Rion

Re: Chain Tensioning
 

I don't think that there is any reason why this wouldnt work. We were actually discussing doing something like this, but we found it much simpler to do it another way. But in truth, its a great idea and somebody should do it!!! Let us know how it goes!

Re: Chain Tensioning
 

A better way? Care to elaborate? I'm okay dieing after you tell me if I must. ;)

Re: Chain Tensioning
 

Well a big question we (me and Rion) had about using this kind of tensioner, was why don't teams use spring tensioners on their drive chains (chains that run drive wheels)? Ive seen one or two used on random chains that drove mechanisms on robots, but never a spring tensioner on a drivetrain. Could anyone explain why this is?

Re: Chain Tensioning
 

Last year 932 used this method for tensioning our gates belts for our ball gathering system
I'll try and get a picture but we had to scavenge some parts from the ball gatherer for this years design

As for why, in my opinion (my own experience with robot drive chains) it requires a structure around the chains to mount the bar and a 2nd structure for the spring, versus the andymark tensioners which do not require said structure

Re: Chain Tensioning
 

Bike's usually have the spring loaded tensioning setup, this is because the chain is only ever going one direction with force, meaning the spring takes up tension on the side that is always slack.

Robot drivetrains dont do this for the reason they run both directions and placing a spring loaded tensioner would yield situations in which the motor would be pulling in chain directly against the spring and when this happened slack would develop in the other areas of the chain run giving increased likelyhood of chain popping off.

Using a spring loaded idler to reduce harmonic bouncing of the chain over long spans though is a good use though.

Re: Chain Tensioning
 

Also I'll suggest that my first intuition regarding this is that you would need a very stiff spring.

if the spring is... anti-stiff (what's the word for that? :confused: )... then under load it could allow the chain to lose tension somewhere along it's length, which could cause the chain to slip off.

without having much experience with chain tensioners, my intuition tends to lean towards an idler gear that can be moved in order to tension the chain, and then can be locked into place. this could be designed to be fairly simple.

currently my team has one axle on each chain loop mounted on slots so it can be tensioned pulled away and tension the chain. this is somewhat problematic since the axle is moving and thus is less than perfectly secure (even when locked into place).

I'd continue working on this idea and developing ways to tension chains..

maybe try making a list of possible solutions, and then make a pro/con list for each solution...

-Leav

Re: Chain Tensioning
 

Page 1022 on McMaster's website.
devices similar to what you explained are used quite commonly in industry. they're great ways to keep tension, especially if the mechanism is subject to shock loads.

I personally haven't seen these used on robots, probably due to weight, but other then that I dont really see why they wouldnt be since they are great for keeping constant tension on the mechanism.

Re: Chain Tensioning
 

This is similar to the type of tensioner we are talking about.
http://www.cheapbikeparts360.com/images/b/13958.jpg

Re: Chain Tensioning
 

Even simpler than spring-loaded tensioners: floating idler sprockets.

Re: Chain Tensioning
 

Did you all lasercut that yourselves? How did you decide how big it needed to be and other design decisions? We use 25 chain so we cant use mcmasters. Thanks!

Re: Chain Tensioning
 

Our team loves to use simple aluminum spaces as a temporary tensioning method. We use a lathe to make a hole down the axis of a cylindrical piece of aluminum and bolt it in place. The chain runs over the aluminum spacer and the size and diameter of the spacer can be easily changed. This works fairly well as a temporary measure and is lightweight.

That being said, the spacers are damaged by the chain and eventually break. I'd use them only if you really need to, otherwise the idler gear would be a better bet as there is no risk of damaging the chain.

Another solution we found was half links. Until this year we'd never heard of them but when used with a chain puller, its easier to make chains that are of the proper length.

Re: Chain Tensioning
 

At Long Beach we had trouble with a slack chain on our lifter arm. We used a floating idler for a couple matches and it solved our problem. We switched to a a rigid idler sprocket on bearings which worked perfect.

I've tried spring loaded tensioners before. They only work well in one direction. I wouldn't recommend it. Rigid is best. Best way is probably to make one end of your mechanism sliding, to increase the center distance.

Re: Chain Tensioning
 

We used this for our 2010 Robot.

http://www.chiefdelphi.com/media/photos/34570


The hub of an AM' Kit wheel cut out with a KOP Sprocket bolted on and two eye bolts with a 3/8 all-thread through it.

Re: Chain Tensioning
 

we used one on this years robot and the inspectors almost didnt like it... but eventually they took it and we used it

Re: Chain Tensioning
 

Back in 2008 we had problems with the drive chain slipping for our elevator under load. It was in such a weird spot that any conventional tensioner would've either hit our frame or rubbed on the ground. So, I figured, what the heck, let's try a floating idler. And it worked, it make it the entire season without any issues. (Except once - we almost tipped the robot over pushing the limit in some pre-ship practice and the force must've popped the sprocket out...)

Also, we used these guys on our '05 robot: http://www.mcmaster.com/#6233k33/=6iqpvw

I really don't like them because I don't feel that they tensioned the chain all that well, but it also wasn't the best chain routing, so these may work too.

Something else we use on 816 is adjustable transmission mounts. Every robot we've built since 2006 has had one in some form or another. This year, the transmissions are mounted on 1/4-20 threaded rod which allows them to move up and down to tension if needed.

Re: Chain Tensioning
 

The best method of chain tensioning is to properly design your chain run so that there is no need for a tensioner. However, to be more realistic for FIRST, the most robust system of chain tensioning is to be able to vary the distance between two sprockets. Anything else adds unnecessary complexity and strain on what should be a rather simple system.

Once 114 implemented a proper chain tensioning system (Aim High, 06), we did not break a single chain, or have any chain related problems. We have since had zero problems with chain.

Re: Chain Tensioning
 

In 2007, 2008, and 2010 We've used small cylinders of HDPE to tension our chain. We simply bolt it on a piece of angle with the hole off-center, rotate to proper tension, then tighten it down .

(2009 We didn't use chain we did direct gears instead)

Here are pictures of it's implementation on those 3 robots mentioned above:
2007:
http://www.vandenrobotics.com/plogge...n/dsc07244.JPG
2008:
http://www.vandenrobotics.com/plogge...n/dsc07275.JPG
2010:
http://www.vandenrobotics.com/plogge...n/sn850302.JPG

Re: Chain Tensioning
 

We used the same thing, but with delrin.

Re: Chain Tensioning
 

It should also be noted that we had an issue with these tensioners rotating on our 2010 robot no matter how tight the bolts were cranked down.

Re: Chain Tensioning
 

Thanks for the advice guys, we really appreciate all the help.
I was wondering though, if spring tensioners only work in one direction, why not use two?
One oriented one way, and another oriented the opposite way.
Maybe something like this?

Re: Chain Tensioning
 

While we only had this problem about once if you can figure out exactly how much tension you need, then you can drill the hole in the center of the cylinder and mount it like that so even if it rotates you maintain your tension. Works like a charm.

Re: Chain Tensioning
 

Its true rick but wouldnt we need a plastic cutting bit?

Re: Chain Tensioning
 

They were milled on a CNC machine; you can't laser cut polycarbonate. ;)

As for design decisions, we just made a bunch of different sizes. To tension the chain, simply pop the floating idler sprocket out and move it closer toward one of the driving/driven sprockets. You may have to readjust the idler position somewhat frequently as the chain breaks in, but once the chain finishes stretching you can leave the floating sprockets in the same position for the rest of the season.

If you don't have [access to] a CNC machine, you can always just get plastic (nylon) sprockets and use them as floating idlers.

Re: Chain Tensioning
 

Art,

Did you guys put a chamfer on the sprockets and how thick are they?

Thanks,

-RC

Re: Chain Tensioning
 

We used metric 3mm polycarbonate (.118") sheet for these sprockets, as it reduced the amount of necessary chamfer. The chamfers were done manually with files, since cutting them out on a CNC machine takes long enough as it is.

Re: Chain Tensioning
 

I suspect that the best way to tension a chain will be a long-standing debate in FRC, with no clear winner. Every system has its own pros and cons.

Re: Chain Tensioning
 

Garrett, there is a patent for a dual tensioner very much like the one you illustrated.

Re: Chain Tensioning
 

Would you mind posting a link to the patent so I could check it out?

Re: Chain Tensioning
 

Here's a link to patent 4069719. There are additional illustrations in the patent doc.

http://www.google.com/patents?id=8LAwAAAAEBAJ

I often use the Google patent search for ideas for mechanisms. If you find one relevant patent, there are often others cited and you can check them out.

Then there are endless hours of entertainment reading patents such as the MIG welder attachment for a drill that converts it into a spool gun. Or the golf clubs that are powered by a shotgun shell.

Re: Chain Tensioning
 

This is very true. But I will probably be making a list of everything that is said in this thread. Just so that it will be easily accessible from here on out. ;)

Thank you everybody for all of your help. I never thought of looking through the patents... I'll definitely have to do that from now on.

Are there any other ways that you guys know of to tension chains?

-Rion

Re: Chain Tensioning
 

http://www.andymark.biz/am-0286.html

I've also seen teams with leadscrews attached to their drive bearings (which are attached to their frame in a slotted fashion). They tighten the screw to pull the bearing towards the end of the frame and tension the chain.

I've seen a number of teams use a hard-mounted idler sprocket (as mentioned a couple times in this thread), but also attach a shaft encoder to that sprocket (instead of to the gearbox or drive wheels).

Re: Chain Tensioning
 

Whenever I could get away with it, I slapped the gearboxes and the motor on the frame, put some chain in there and got everything snug. Then I would drill the mounting holes and throw a little shim in somewhere if the chain wasn't strung tight enough to almost sing. That was with decent #35 chain, so wear wasn't a problem. On a real long chain run then perhaps you have to have a bit of a tensioner.

There are really a couple of different issues. You don't want a lot of slack in a bidirectional drive mechanism. If the drive is in one direction, then you can make a chain guide to prevent excess slop so you don't throw a chain, but "tension" is not so much of a concern.

Re: Chain Tensioning
 

Did you get away with having wheels in the bumper zone on the 08 robot?

Re: Chain Tensioning
 

spring and an idle wheel, enough said

Re: Chain Tensioning
 

Only the very smallest wheels could avoid being in the bumper zone, since it began five inches from the floor.

Re: Chain Tensioning
 

2/3 perimeter is covered + min 6" segments = good bumpers. '08 was the first year bumpers were required but the rules were not as detailed. Abstract concepts like the BUMPER/FRAME PERIMETER (set of exterior vertices) and rule prohibiting extending past it that perimeter came the next year. Plenty of mechanisms had to reached far past their robot perimeter was to acquire/hurdle giant trackballs, so wheels are the least of your concern.

Re: Chain Tensioning
 

No, not enough said. Please re-read this thread, where you'll find the information on why this system is not adequate for a FIRST application.

Re: Chain Tensioning
 

The inverse of stiffness is compliance! :)

Re: Chain Tensioning
 

Thank you for getting to that before me. :)

Re: Chain Tensioning
 

For applications that don't require a continuous rotation (like an arm perhaps), we've used a turnbuckle to tension chain. Now obviously, the turnbuckle cannot pass over the sprocket.

Re: Chain Tensioning
 

Actually, Craig, an idle wheel and a spring would work. If you had the spring pulling up on the wheel. This would only apply force in an "upward" direction. This is in a completely different plane than the chain. Thus, no matter which way the chain was moving, it would still work. At least, I'm pretty sure.

-Rion

Re: Chain Tensioning
 

A third idler (What I assumed he was referring to) was addressed earlier. To reiterate, it will only tension in a single direction. If you switch directions, you will be directly pushing against the idler spring. Unless you're using a hefty spring, this will not be an adequate tensioning method, and you will have chain skipping and hopping in one direction.

We ran into this exact issue with a belt one year, which led us to develop the bolt pulled tensioner (You can search my CD Media submissions to find it). Once we implemented that, we had zero chain failures, and never needed to change tension to the system for entire regionals.

Re: Chain Tensioning
 

Team 975 once used an old skateboard wheel for a tensioner/idler. A groove was turned in the tire to prevent lateral movement.

Re: Chain Tensioning
 

Our team just slotted our wheels and then on each side of the wheel's axle we put small cams and then just turned them till the chain was tight, then we tighted down countersunk screws into them to keep them in place

Re: Chain Tensioning
 

Do you have any pictures of this? i cant visualize this.

Re: Chain Tensioning
 

Like Andre, I can't visualize how slotted wheels would work. I think you mean you used slots in your frame through which you mounted your axles, right?

Re: Chain Tensioning
 

Sorry, yes that is what I meant.
We cut slots into our frame and then placed our wheels in

Re: Chain Tensioning
 

I'm seen this done before. It really is a good idea. You typically see it on WCD chassis though don't you? And, at the moment; my team only has hand drills and a drill press. No mill in which to do slotting on.

Re: Chain Tensioning
 

I really dig this idea for a quick fix! But I suppose I'm biased *points to 14x24x24 box of skate wheels next to the door*.

Re: Chain Tensioning
 

Okay, I get it now.

Re: Chain Tensioning
 

lol sorry. By better way, i mean for what we were making it for. the chain for our gear box wasnt very long, and the cim was mounted pretty close. basically, we ended up with slits on one side and holes for bolts on the other. we could slide it in easily and bolt it down. before we realized this, though, we were going to do that. actually, im pretty sure it was sent to me with the caption "genius tensioner design" :)
Really, i think that the method of tensioning is gonna vary based on the situation. For ours, we would have liked to do this, but when we followed the KISS method we figured this might be better

Re: Chain Tensioning
 

I'm a big fan of the put the holes in the right spot method of chain tensioning. It does however get more complicated with weird chain runs like we had on our 2010 bot in Waterloo.

Re: Chain Tensioning
 

525 has used the Fenner Drives composite idler (Mfr #: CS3502) for many years. It is lightweight and has a high quality bearing. It's available from MSC Direct.
http://www1.mscdirect.com/CGI/GSDRVS...00000134143903

Re: Chain Tensioning
 

Hmm. I'm getting a bad link from the post above me for the idler...

Re: Chain Tensioning
 

Austin,

Just copy and paste the part number at the top, I had the same issue.

-RC