Bearings vs Bearing Blocks

[Peyton Yeung]

Originally Posted by Munchskull

So from my understanding C to C chain is a bad idea?

If it is chain in tube c to c chain it usually is fine.

[AdamHeard]

Originally Posted by Munchskull

So from my understanding C to C chain is a bad idea?

Not necessarily.

C-C with chain and no understanding of what's going on is a bad idea.

If you plan for it and understand the variables involved it's totally doable.

[Dunngeon]

Originally Posted by AdamHeard

Not necessarily.

C-C with chain and no understanding of what's going on is a bad idea.

If you plan for it and understand the variables involved it's totally doable.

Yup! 955's ran variations with #25 and #35 chain with success over the last 5 years. If you're curious Anthony (Munchskull) I'm sure you could go across Corvallis and ask for a look.

[Chris is me]

Originally Posted by Munchskull

So from my understanding C to C chain is a bad idea?

The way I see it is: If you needed this CD thread to figure out if it was viable and what variables to consider, it's probably not a safe bet for you.

Regarding pin / bushing wear (the primary contributor in chain stretch) - load is a lot less of a factor on this wear than you might expect. Use in a low load or no load state can stretch chain as well. I suspect that the wear occurs as the chain rounds the bend of the sprocket. I'm just speculating here, but I imagine you could "break in" chain by just running it in alternating dimensions on the bench for a few hours. Once the initial wear happens, chain "stretches" a lot more slowly, so you may be able to get your re-tensioning out of the way early.

In 2015, my old team (2791) ran a 6 foot long chain run (so more than 12 feet of actual chain) at exact center distances. What should surprise no one is that the chain did indeed stretch, but we only had to tension it once. We did end up offsetting our mounting holes for the elevator by one half link after the initial stretch though. All of this stretch occurred before the relatively light load was placed on the chain (chain never lifted more than 1 tote ideally) .

[Michael Corsetto]

Originally Posted by Chris is me

The way I see it is: If you needed this CD thread to figure out if it was viable and what variables to consider, it's probably not a safe bet for you.

Can you expand on what you meant by this statement?

-Mike

[Chris is me]

Originally Posted by Michael Corsetto

Can you expand on what you meant by this statement?

-Mike

What I mean by that is that it takes a degree of engineering knowledge, design knowledge, and care to run exact center chain runs (outside of a tube), and if a designer wants to use exact center chain runs, they should do some thorough research to understand the problem as best they can. If their first consideration of this idea was a suggestion in a CD thread, if they don't have the background knowledge it's probably safest to just use a simple tensioner.

I think the statement I made was too broad and sweeping. In short: know what you're doing.

[Monochron]

Originally Posted by Chris is me

What I mean by that is that it takes a degree of engineering knowledge, design knowledge, and care to run exact center chain runs (outside of a tube)

I agree mostly that to build an efficient and high quality c-c chain system you should investigate the principles in depth.
I'll offer a counterpoint though in that this past year we ended up having to run our elevator c-c at the last minute when a couple of other things failed shortly before build season ended. It was always our intent to use a tensioner but when that failed, it was the quickest way forward was to just go straight c-c and it worked just fine without any design changes. We had to adjust the chain once in our two competitions (removed a link) to keep it ratchet free. While I'm sure it wasn't as efficient as it could have been (after removing the link it was certainly over tensioned), it worked well and didn't cause us any noticeable problems.

[Cory]

Originally Posted by Chris is me

What I mean by that is that it takes a degree of engineering knowledge, design knowledge, and care to run exact center chain runs (outside of a tube), and if a designer wants to use exact center chain runs, they should do some thorough research to understand the problem as best they can. If their first consideration of this idea was a suggestion in a CD thread, if they don't have the background knowledge it's probably safest to just use a simple tensioner.

I think the statement I made was too broad and sweeping. In short: know what you're doing.

I don't understand why I keep hearing references to "outside of a tube". What difference does it make? There should be no difference in how well the shaft is supported in either scenario and your shaft deflection is basically nonexistent in both scenarios.

[z_beeblebrox]

Originally Posted by Cory

I don't understand why I keep hearing references to "outside of a tube". What difference does it make? There should be no difference in how well the shaft is supported in either scenario and your shaft deflection is basically nonexistent in both scenarios.

I think the point is that the tube keeps the chain from coming off the sprockets, even if it's not sufficiently tensioned.

[Cory]

Originally Posted by z_beeblebrox

I think the point is that the tube keeps the chain from coming off the sprockets, even if it's not sufficiently tensioned.

I still think that will have very, very little effect on whether chain details or not. 233 has done exact center to center for ages in a tube, but a tube that is far larger than the sprocket, so there is nothing constraining the chain top-bottom from coming off.

Chain is pretty tolerant to tension. You can dramatically over tension or under tension it with no issues as long as you maintain good lateral alignment and you aren't right on the upper edge of its load rating. Teams should be focusing on good lateral alignment no matter what style chain drive they run.

There's nothing mystical about running exact c-c chain...

[Monochron]

Originally Posted by Cory

Chain is pretty tolerant to tension. You can dramatically over tension or under tension it with no issues as long as you maintain good lateral alignment and you aren't right on the upper edge of its load rating. Teams should be focusing on good lateral alignment no matter what style chain drive they run.

You're saying that ratcheting isn't a big issue even with "loose" chain as long as your lateral alignment is good?

[Cory]

Originally Posted by Monochron

You're saying that ratcheting isn't a big issue even with "loose" chain as long as your lateral alignment is good?

I don't know what you'd have to even be doing to see chain ratchet on a FRC drive.

I'm not saying it can be flopping all over the place. But chain stretch isn't going to cause your chains to derail if you do exact c-c with Paul's added distance.

[GeeTwo]

Originally Posted by Monochron

You're saying that ratcheting isn't a big issue even with "loose" chain as long as your lateral alignment is good?

That would be an overstatement. Consider the chain on the sprocket being driven by the chain. The chain is exerting a torque T on the sprocket. This is exerted as a force at the pitch radius r of the sprocket. The chain is exerting a net force along its length of T/r. This means that the tension on the "pull" side of the chain is T/r greater than the tension on the back side. If the "at rest" tension in the chain is less that T/2r, the tension on the back side when applying torque is less than zero (assuming the instantaneous chain stretch is linear). That's an almost sure recipe for ratcheting unless the chain is otherwise constrained to stay on the sprocket. If the sprockets aren't essentially in the same plane, it's a certain recipe for derailment.

Edit/added:

Originally Posted by Cory

I don't know what you'd have to even be doing to see chain ratchet on a FRC drive.

We had it happen this year, not on a drive, but on our lift. We mounted hex shaft bearings in holes bored directly in our lift frame (with a spade bit, no less), but those did not give us any trouble. In order to accommodate chain stretch, we did not make a full loop of the chain, but secured both ends to the back of the lift plate using end point mounts copied in concept from a garage door opener. We used thumb screws, and drilled holes through the thumb tabs an appropriate size for the pin in a master link. We then secured each end of the chain to these screws with a master link. The thumb screws passed through a hole, and were secured on the back side with a nylock nut. I know they were originally built that way; I inspected them myself. However, it appears that on one of our tournament repairs, the lift was put back together with regular hex nuts (in the interest of speed), which loosened, which caused ratcheting on one side and derailment on the other. (One was apparently lined up better.) We then re-tensioned and used lock nuts again, and haven't had the problem since. That reminds me - it's time to check the tension on that demo robot again.

[BrendanB]

Originally Posted by Cory

I don't understand why I keep hearing references to "outside of a tube". What difference does it make? There should be no difference in how well the shaft is supported in either scenario and your shaft deflection is basically nonexistent in both scenarios.

Its probably stemming from people seeing the walls of the tube acting as an insurance policy in that if the chain does stretch or it wasn't machined right there's a larger tolerance and the tube will keep the chains on the sprockets. Outside the tube, not so much and if the frame isn't that rigid it gets even better.

When we did C-C #35 in a plate drive in 2014 we got a lot of funny looks talking to people because for teams using the pre-2014 kitbot it was a norm to have tensioners integrated and most mentors work with machines with tensioned chains so therefore the robot needs it too. Some people talked to us like we found some type of black magic thinking every chain needed to have a built in tensioner.

You are also very correct that having proper alignment on chains is crucial as well and is often overlooked.

If you have the ability to build your drivebase with C-C (chains or belts) its a great opportunity to lose a few parts, headache, and possible failure points. There is plenty of time between now and kickoff to design and build one as a prototype.

[GeeTwo]

Originally Posted by Cory

I don't understand why I keep hearing references to "outside of a tube". What difference does it make? There should be no difference in how well the shaft is supported in either scenario and your shaft deflection is basically nonexistent in both scenarios.

Originally Posted by BrendanB

Its probably stemming from people seeing the walls of the tube acting as an insurance policy in that if the chain does stretch or it wasn't machined right there's a larger tolerance and the tube will keep the chains on the sprockets. Outside the tube, not so much and if the frame isn't that rigid it gets even better. ..

Yes, that's exactly right. The vast majority of "chain in tube" designs I've seen recently are designed so that the inside opening of the tube is greater than the pitch diameter of the sprocket plus the height of the link as viewed from the side (H in this image), and (much) less than the sprocket outer diameter plus H plus R, where R is the hub diameter (same image). This means that the chain is free to move when it engages the sprocket, but cannot jump from one tooth to another even when there's no tension on the "back side". Cory referred above to tube larger than this, where the "in-tube" advantage does not apply:

Originally Posted by Cory

I still think that will have very, very little effect on whether chain details or not. 233 has done exact center to center for ages in a tube, but a tube that is far larger than the sprocket, so there is nothing constraining the chain top-bottom from coming off.

While I cannot cite experience, I can see where C-C spacing can work for fairly short runs (where C-C is only ten or twenty full links), but I would absolutely shun C-C without a tensioning option for long runs, like a 6' lift. Maybe someone with more experience in this can better tune my gut "ten or twenty full links" to a more definite number.