Chain Drive Question

[Adam Y.]

Just some friendly advice (for next time) go find the formula that will tell you the center to center distance of the sprockets so that the chain fits and no tensioner is needed

Err that is nice and all but I really doubt you would be able to get an answer that would really work nicely. You are probably still going to need tensioning. Anyway here goes the process.
Chain Length=L
Center Distance(Distance between sprokets center)=C
Pitch Diameter of Larger Sprocket=D
Pitch Diamter of Smaller Sprocket=d
Errr wow is that confusing or what. I hope someone can find a simpler form than what I just typed. Instead of typing it I scanned them. Unfortunately its sideways.

[sanddrag]

Originally Posted by Adam Y.

Err that is nice and all but I really doubt you would be able to get an answer that would really work nicely. You are probably still going to need tensioning. Anyway here goes the process.
Chain Length=L
Center Distance(Distance between sprokets center)=C
Pitch Diameter of Larger Sprocket=D
Pitch Diamter of Smaller Sprocket=d
Errr wow is that confusing or what. I hope someone can find a simpler form than what I just typed. Instead of typing it I scanned them. Unfortunately its sideways.

What type of units is L in? Where exactly do you measure the length of chain? Overall length or between pins or what? Or do you count the number of links?

[Matt Adams]

DANGER: Shameless Plug!

Click here for the link to the thread with my center distance calculator

Matt

[Adam Y.]

What type of units is L in? Where exactly do you measure the length of chain? Overall length or between pins or what? Or do you count the number of links?

Might as well borrow a bit more from my book.

These equations will provide physical lengths. To determine the correct chain length and center distance, use this procedure:
1)Estimate what the center distance (C) should be.
2)Chain lengths are based on an even number of pitches. To determine the number of pitches, divide the chain lenth by the pitch length. If the number of pitches of the chain is not an even number, round up or down for an even number.
3)Multiply this result by pitch length to get the final physical length of the chain.
4)Plug the pitch length into Equations 19 and 20 to calculate the center distance between sprokets.

The book was a whee bit vague as to what units L was in. Im guessing that it should be whatever untis the pitch is in. Metric for metric chain and the same for American chain.

[maclaren]

Oh yes I forgot to mention something FRICTION IS YOUR ENEMY. Gloified plastic chain gaurds are a large drain on efficeny. We used idler sprockets which have built in bearings. We took the idler sprockets and mounted them to a small piece of 80-20 and then bolted the slide to another piece of 80-20 and you simply slide the idler sprocket and then tighten the bolts.

It's simple and easy.

[Astronouth7303]

Oh, right, that.
One word: WEIGHT.

[Kevin Sevcik]

I'll toss in my two cents. In every chain length formula I've seen, L is is actually in number of pitches, as opposed to and measure of length. and the chain doesn't HAVE to be an even number of pitches, it's just good practice not to use offset links.

[maclaren]

Originally Posted by Astronouth7303

Oh, right, that.
One word: WEIGHT.

Yeah that's true weight is allows a problem in first however that is also why you follow good engineering practices.

Just for the person that posted the picture of their teams robot using short section of chains. AVOID CAST IRON LIKE THE PLAUGE and you will save at least 1-2 lbs per bearing.

The best way to remove weight is to go to aluminum axles and and fixed axles with live wheels. It takes a little effort but could give you at least 5-8 lbs back out of your drive train. Also the wheels will be easy to install, remove, and repair.

[Dick Linn]

A couple of times, we've mounted each drill motor and small sprocket shaft on an aluminum plate. This plate has slotted holes and can be moved in relation to the drive wheel to tension the chain when necessary. We can also use washers between the plate and the chassis rails to get additional range (yeah, the slots weren't long enough...). Keep the chains lubed, snug and aligned and you'll have less chance of breaking a chain when it tries to climb up and over the sprocket teeth. Also avoid tiny sprockets.

[Dan Richardson]

Another way to tension chain besides Slotted wheel mounts and tensioners is an adjustable gearbox.

This year we built our gear boxes on the frame with a sandwhich mount and shims. WHen the chain would get loose we could slide the gear box up or down ( depending on location of the 2nd sprocket ) and tension that way. It worked out really well for us. It ended up being much easier than sliding every wheel like we used to in old chain drive designs. Just loosen a few bolts, slide, tighten the bolts up again and walla Tight chains.

Its probably late to consider something like this but Its just an interesting idea that works really well.

[Grommit]

Yes, Travis is quite right, I neglected to mention that our wheels have always been mounted such that they could slide for us to tension them. What I intended to explain was that spring tensioning is not the only solution, and that slots are a convenient alternative.

Also, note that Bosch extrusion particularly has very easy, built in slots that hold well (if you are using 30mm X 30mm, that is). There are T-Blocks available (at great expense ) that make mounts very solid.

Even if you do not wish to use Bosch, and you don't like your chassis to have a big aluminum plate base, you can still use slots. Take your welded chassis and bold on a plate, almost as a gusset, but with a big slot and two smaller ones on each side to hold the wheel and two mounting blocks. The slots on the side allow you to adjust the mounting block position. But it gets better: many teams have trouble tensioning this, because it takes two people: one to pull on the wheel and one to tighten, and it just isn't fun. The clever plan is to use a custom tool. Mount a block with a threaded hole in front of the mounting block slots. Use a larger hold for greater control and ease of use. Then take a long bolt and put it in the thread, so that when you tighten the bolt, it will push on the mounting block. Then you can even use a drill (on very low clutch setting, maybe as low as 11 or so) to tighten the mounting blocks for you, and they won't bounce back. VERY useful for quick tightening jobs.

[ahecht]

I'll add in another plug for the adjustable sprocket-box that 190 has used the past few years:



The black line is the chain, the two larger sprockets are idlers, and the smaller sprocket is attached to the motor. On the bottom of the image, you can see the chain returning. By tightening or loosening the bolt, we can adjust the tension. Also, this design assures that we have a full 180 degrees of wrap around the motor's sprocket (this image isn's quite accurate).

The other nice thing is that by flipping this design around on the other side of the robot, and driving the lower part of the chain, we can have both motors driving in the same direction when we drive forward:

EDIT: Just to clarify, on each side, there are two identical lexan plates, one on either side of the sprocket. Therefore the adjustable sprocket is actually supported by two bolts, which need to be adjusted together.

[Grommit]

Originally Posted by ahecht

I'll add in another plug for the adjustable sprocket-box that 190 has used the past few years:

The black line is the chain, the two larger sprockets are idlers, and the smaller sprocket is attached to the motor. On the bottom of the image, you can see the chain returning. By tightening or loosening the bolt, we can adjust the tension. Also, this design assures that we have a full 180 degrees of wrap around the motor's sprocket (this image isn's quite accurate).

The other nice thing is that by flipping this design around on the other side of the robot, and driving the lower part of the chain, we can have both motors driving in the same direction when we drive forward:
EDIT: Just to clarify, on each side, there are two identical lexan plates, one on either side of the sprocket. Therefore the adjustable sprocket is actually supported by two bolts, which need to be adjusted together.

That's a great way to keep the chain tensioned! I was curious about your bolt: did you have trouble with the bolt loosening in the competition? Did you use some kind of solution inside to increase friction so the bolt wouldn't slip? Or was that not a problem anyways? Also, was the sprocket shaft on a sliding slot or was there some other way to move the sprocket along?

We might consider something along those lines next year, especially if we added another pair of wheels. It would avoid using three sprockets on a shaft, one for drive and one to each other wheel: we could drive off one of the other two without sacrificing the 180 degrees of chain contact.

Very well implemented idea, thanks for sharing it.

[ahecht]

Originally Posted by Grommit

That's a great way to keep the chain tensioned! I was curious about your bolt: did you have trouble with the bolt loosening in the competition? Did you use some kind of solution inside to increase friction so the bolt wouldn't slip? Or was that not a problem anyways? Also, was the sprocket shaft on a sliding slot or was there some other way to move the sprocket along?

We didn't have any trouble with the bolt loosening, but if it is a concern, you can coat the bolt with a very thin layer of locktite before inserting it (make sure the locktite is completely dry before inserting, as the solvent will melt the lexan). In 2k3 we had trouble with our chains popping off, but that was due to our flexible frame and flaws in our manually milled (not CNC) 80 tooth sprocket, not our chain tension.

The adjustable idler's shaft slid in a slot milled into the two lexan plates, relying on the tension in the chain to keep it pressed up againt the bolt. The fixed idler's shaft just fit into two holes.

Another thing to note is that if you don't need your motors spinning in the same direction (which we didn't this year because we could pneumatically lock the two sides of the driveline when we needed to go straight), it is possible to design the system without the stationary idler (thus having your chain make a 'z' shape).

[Biff]

Originally Posted by maclaren

Yeah that's true weight is allows a problem in first however that is also why you follow good engineering practices.

Just for the person that posted the picture of their teams robot using short section of chains. AVOID CAST IRON LIKE THE PLAUGE and you will save at least 1-2 lbs per bearing.

The best way to remove weight is to go to aluminum axles and and fixed axles with live wheels. It takes a little effort but could give you at least 5-8 lbs back out of your drive train. Also the wheels will be easy to install, remove, and repair.

Like that idea, we considered it but could not figure out a way to get the sprockets attached to the live wheels, Skyway had a bike wheel that was close but it had a coster break installed and was to large anyway. Had to go with the live axles. Last year we got cantiliverd (sp) axles bent, this year we over compensated. We just finished a design in inventor for a 45 tooth #35 chain Aluminum sprocket, With out the weld that holds the hub on inventor says weight is .317lbs or 5.072 oz our target was 6 oz or less.