Cause of chain ratcheting issues

[Monochron]

I am trying to determine the reason that the chain that lifts our elevator now skips on our sprocket when simply running the elevator stages up and down.

We are making some improvements to our robot in preparation for an off-season event and when we got to the point of re-testing our elevator, the chain started skipping. We physically can't lift a single RC more than a couple inches.
We have a two stage cascade elevator; the first stage is lifted by a chain, and the second stage is lifted by a run of steel cable. We are using #35 chain, 15 tooth sprockets, and the distance between sprockets is roughly 3.5'. Pictures of the chain can be seen in these albums:
http://imgur.com/a/uCo9b
http://imgur.com/a/34fYz
I tried to get it from as many angles as possible.

Things I have thought to look at:

Chain tention

From what I can tell, tension is just fine. Here is a picture (http://imgur.com/JrIR1vC) of the chain at rest, and here is a picture (http://imgur.com/I3SuvmP) of me flexing the chain as much as my hand was capable.

Chain misalignment

I can see some slight misalignment on the bottom sprocket, though (to me at least) it doesn't seem like enough to cause the constant ratcheting that we are seeing. You can see for yourself here (http://imgur.com/4baECMR).

Additional load

This is possible because we had to change out our top pulley holding the steel cable due to binding, and the new solution seem to be binding a bit as well. Even with this though, the physical force it takes to lift the elevator with your hands doesn't seem drastically more than it did in April. In April we never ratcheted the chain as long as our chain was tensioned.

Elevator stages "loosening"

The rectangular tubing making up the elevator stages has "loosened" a bit over time so the stage rocks inside it's parent stage a bit more than it used to.

Flexing sprocket mount (ie. c-c distance changing)

I don't believe that this is happening. The VersaPlanetary attached to the bottom sprocket is mounted very securely. I can look into this though.



That is what I have looked into so far, does anyone have any ideas as to causes for this? I am assuming that there are multiple causes for this, but some expertise in isolating those causes would really help us out. Any suggestions for things to look into or modifications to make would be great. As it is now we can't really compete :(

[Lil' Lavery]

I can't really tell from the pictures, but is the shaft on your Versa planetary supported on both ends? This essentially ties into point #5 you made. During build season, we experienced chain skipping due to the shaft and mounting on our versa planetary flexing under load. Considering how far away the load is from your gearbox face, I feel that may be the issue you're seeing.

[MrForbes]

might take some video of the chain skipping, but also get other parts of the system in the video. My guess is that something is moving, so the C-C distance becomes less as the chain skips. If you have video that you can analyze frame by frame, it might help you find it? also look to see if the beams supporting the things that hold the sprockets, can flex.

[Thad House]

We noticed when we were having ratcheting issues a few seasons ago after enough time the sprockets had just worn down and couldn't hold anymore. Could we get a good close up of the sprocket. They might just be so bad they need replacement.

[Madison]

Chain tention

From what I can tell, tension is just fine. Here is a picture (http://imgur.com/JrIR1vC) of the chain at rest, and here is a picture (http://imgur.com/I3SuvmP) of me flexing the chain as much as my hand was capable.




If there's slack in your chain, it's going to show up at the bottom where the chain doesn't wrap tightly around your sprocket.

[Monochron]

During build season, we experienced chain skipping due to the shaft and mounting on our versa planetary flexing under load. Considering how far away the load is from your gearbox face, I feel that may be the issue you're seeing.Last time I checked I didn't see any flexing of note, but I will check this again. Having that sprocket so far out definitely isn't ideal, but it never had issues during the season.

We noticed when we were having ratcheting issues a few seasons ago after enough time the sprockets had just worn down and couldn't hold anymore. Could we get a good close up of the sprocket. They might just be so bad they need replacement.Here is the skipping sprocket (http://imgur.com/yXLOiIV), and here is the upper sprocket (http://imgur.com/OZ2dknD). There is some wear, do you think that is enough to contribute to the skipping?

I also updated the OP with a couple more detailed pictures.

[D.Allred]

Brian,
Couldn't tell from the close up shots, but looking at your robot picture on the BlueAlliance it definitely looks like your frame could easily flex under load and change c-to-c distance.

The idler sprocket at the top of the lower stage appears to be mounted on u-frame bolted to the sides of the base stage. It only has to pivot a fraction of an inch under load to cause issues.

David

[Chris is me]

It looks like the sprocket on the bottom is cantilevered almost an inch away from the gearbox face. This is a big part of your problem - there is a lot of flex in the VP output shaft when you have an unsupported load that far from the output bearing.

No real way around it; you have to support the other end of the shaft or move the gearbox closer. If you didn't cut a big gash in the 1x1 tube right by the output shaft, you could have popped a half inch bushing in there and turned the end of the VP output shaft to round in order to ride in the bushing. A bit more friction than is ideal, but it'll work.

[Monochron]

No real way around it; you have to support the other end of the shaft or move the gearbox closer. If you didn't cut a big gash in the 1x1 tube right by the output shaft, you could have popped a half inch bushing in there and turned the end of the VP output shaft to round in order to ride in the bushing. A bit more friction than is ideal, but it'll work.
The VP output shaft actually doesn't quite reach far enough for that.

I can run it through some tests tomorrow but I guess I'm still a little skeptical that this issue would only crop up now, after our other changes. How much deflection in the end of the shaft are you expecting? All the mounts for the motor are still as solid as ever.

[Thad House]

The VP output shaft actually doesn't quite reach far enough for that.

I can run it through some tests tomorrow but I guess I'm still a little skeptical that this issue would only crop up now, after our other changes. How much deflection in the end of the shaft are you expecting? All the mounts for the motor are still as solid as ever.

Actually theres another trick you can do to extend a VP shaft. If you get some 1/2 round stock with a 1/4 inch hole, like the vex round tubing, you can screw that into the end of the VP shaft with a 1/4-20 bolt, and then put the round stock through some tubing and a bearing. As long as you get the bolt tight, it will have plenty of support.

[GDG 2337]

On our 2011 Bot arm, we experienced chain skipping using a long run of properly tensioned dual #25 chains with 16 tooth sprockets. We eliminated the issue by adding a roller that increased chain wrap around the sprockets. We made different diameter rollers to account for chain stretch and sprocket wear throughout the season.

[FrankJ]

Looking at your pictures.

If you are ratcheting, either the CD is changing or the chain is elastically stretching. CD changing seems more likely to me.

You definitely have some drive sprocket wear. Whether or not it is enough to be an issue, hard to say. The wear changes the pressure angle which can increase chain tension for a given torque. Maybe trade the idler sprocket for the drive sprocket?

Make provision to take up slack in the chain?

The motor support & motor sprocket has a lot of cantilever. Run a piece of flat bar to the frame rails or wheel gear boxes to add a bearing to help support the gear box shaft?

[Monochron]

We have mostly fixed the issue by doing two main things.

Sliding the drive sprocket further back on the drive shaft. Looking at it more closely (and very hard to see in the pictures) revealed that the point in the chain that was mounted to the elevator was directly above the middle of the drive shaft, not directly above the end of the drive shaft. I'm guessing that over time the shaft collar holding the sprocket in place loosened and inched forward.
We also removed some of the slack in the steel cable that drives the second stage of the elevator. I don't think this had a big influence, but it definitely reduced some of the shock loading that was going on when the cable was pulled taught.


After these two changes the ratcheting is gone. However, now when the elevator is under a lot of load I definitely see the motor mount bending upwards, thus changing the c-c distance. It doesn't ratchet anymore because the chain is better aligned with the drive sprocket, but it shows us the next thing to be fixed.
Thanks for the help everyone. All of your suggestions have been great.

[BJT]

It looks like your whole motor mount is sticking out quite a ways from where it attaches to the robot. As others have said, you maybe should have used that piece you cut out to support the motor mount and/or the shaft. Two sprockets need to be as solidly tied together as possible to eliminate that change in c-c distance you are experiencing.

[Monochron]

It looks like your whole motor mount is sticking out quite a ways from where it attaches to the robot. As others have said, you maybe should have used that piece you cut out to support the motor mount and/or the shaft. Two sprockets need to be as solidly tied together as possible to eliminate that change in c-c distance you are experiencing.

This isn't quite the problem, the position of the motor in relation to where it mounts to the robot hasn't caused skipping at all, rather it was the perpendicular alignment of the chain and the recent extra load that caused the skipping. As I said before, this setup ran without issue since February until now. I don't think we could have used the cut away piece of 2x1 because the shaft is hex, the shaft extended past the part we cut away, we had no way to turn the shaft down, and no bearings we found could fit in that 1" aluminum face.

What you say about tieing the sprockets tightly and supporting the motor shaft is all absolutely true, but it's important to know that none of that caused our ratcheting. I know this to be the case because we haven't changed any of that and our ratcheting is completely gone.