Chain tensioning in difficult location

We are struggling with designing a chain tensioning mechanism for the chain in the attached photo. It is under a great deal of strain as it Powers a large arm mechanism. The slack is causing problems with the encoder reading, not so much with the chain slipping.

We have a few ideas but are wanting advice. We cannot change the sprocket size, or at least not easily. We are worried that if we attach a spool or something to the 1x2 behind the chain that it will get bent due to the significant force on the chain.

We had a similiar issue with chain skipping- the set up looked almost identical to yours. We bought this tensioner from McMaster ( the floating version) and it got rid of the issue. You will have to make the chain much looser for this tensioner to work properly.


On the rear wheels of bikes they have chain tensions that are made of a 5-7 tooth sprockets that have a lever and a recoil spring to make sure the chain stays tight but not too tight. You could add one on the top or just reduce a link or 2


Thank you. There isn’t enough slack to remove a link. The encoder is very sensitive to slop so we eed the chain to be really tight.

We actually tried to make something like this but the short distance and the significant size difference made it difficult. We may try again.
Thank you

One thing we did in terms of fixing encoder readings on a shaft driven by a chain that slipped was putting the encoder directly on the shaft.


You may want to try popping a floating sprocket in there. It will take up the slack, preventing slipping, but doesn’t prevent all of the backlash. The backlash is the issue that your encoder is seeing and giving you fits.
The floating sprocket will help, but you really need a hard mounted chain tentioner.


Last year, my team had an arm for the Power Up season and we tensioned our belts my creating a part that basically squeezed around the belt making the gap more narrow. The two prongs had wheels on them to help the belts move through the part. I’ll attach a picture of what we did to make it easier to understand, but you could probably create something similar for chain.

Both sides of the belt ran through the two prongs.

1 Like

if you use a longer bolt on the top, I think you could easily add a piece of delrin or a bearing with a small sprocket depending on how much slack you have (may need to actually add a few links if you go with the 2nd option) but either way utilizing the top bolt seems like the simplest way with the least extra parts/mods. if its under considerable tension and you have the room, I would make that top bolt a quarter inch.

1 Like

I think that’s where we’re heading. Thank you all for your help

1 Like

Is there a way you could get an encoder directly on the shaft? Removing slop will still help for PID tuning, but feedback from the mechanism itself is extremely helpful for programming. (removes complexity, may eliminate the need for a limit switch, etc)


Yeah, this 1000%.

reducing slop from chain tension will help, but moving the encoder AND removing the slop will help significantly more.

1 Like

It’s possible, but the slop would still be problematic. Encoder is final VP stage right now.

If you have access to a 3D Printer, I’ve had some decent luck printing tensioners in the past. Granted, the ones I made were used in low-load situations, so you’re mileage may vary, but it might be something worth exploring.

1 Like

use a chain tentioner with a bolt and nut right next to the other bolts and drill a whole

This topic was automatically closed 365 days after the last reply. New replies are no longer allowed.