How to design efficient Tensioners for long or irregular Belt, Chain, or Cable paths?

We’ve struggled over the years to properly tension our chains, belts, and cables. I’m sure it was very entertaining for those not on our alliance when we used oversized, free-spinning sprockets between the lengths to take up the slack. They went flying in all directions upon each defensive impact.

My personal query is regarding longer runs where the sprocket or motor axles cannot sustain massive tension loads. I’m thinking that increasing the angle a chain/belt wraps around the sprocket reduces the required tension or stress.

I’m sure there are different tensioning challenges others will find informative and useful.

I am hoping some of you may have advice and examples of robust, clever, or compact solutions for long or irregular conditions where Center-to-Center calculation is not enough. When is a spring-loaded tensioner appropriate and when should a solid one be used?

One example by @gear caught my eye and triggered my post. This tensioner appears to be robustly mounted on sheet metal. I’m hoping he will describe the sliding block and if the sprocket axle and tension screw may intersect.

What sort of center-to-center distances do you typically use? What sort of structures are you typically using? Sheetmetal? 2 x 1 or 1 x 1 tube? Other? Do you have any photos you can post?

We’ve used these structure types and 1/4" plates too. But I am hoping that posts show different methods that have worked well in different conditions that anyone can learn from.
I guess we do have one simplistic example that worked well and will add an image when I can.
This friction tensioner is an aluminum boss bolted to a sideplate. The bolt was drilled off-center causing a cam action when twisted, taking up the slack. This was placed on the slack side of a winch when climbing for Steamworks. Definitely not appropriate for high usage or bi-directional loads. We only did it because we hadn’t planned for it and needed a quick solution.

The following are the types of tensioners that we have had the best luck with.

If it is a continuous run the WCP cams. These work well with the WCP sliding bearing blocks.

If it is something like an elevator that can have an inline tensioner we have used something like this.

In the past, we have also struggled to get chain tensioned correctly. (Just look at our 2018 robot drivetrain and arm chain here and here). Between the 2018 and 2019 seasons, we developed an offseason drivetrain, and used VexPro’s sliding bearing blocks with great success. Of course, these bearing blocks are intending for drivetrains rails, and are not easy to integrate into a more complex mechanism.

As you can see, we added a slot to our gearbox plates so the nuts towards the top could be turned to push the sprocket into the chain, thus tensioning it. All black parts except the sprocket were printed out of NylonX, while the sprocket was printed out of Nylon 645.

For a few printed parts, some ingenuity was required. Most notably:

  • Bolt Capture

In order to constrain the sprocket bolt to the perpendicular bolt holding the sprocket, we designed this part. First, the part was printed up to the first layer that would cover the top, paused, and then resumed one the bolt head was dropped in the slot.

  • Nylon Idler Sprocket

After considering idler sprockets from McMaster-Carr such as this one, we decided on making our own because of the price. This part was printed in a similar way to the bolt capture-the bottom of it was printed up till where it bridges over, then the this bearing was pushed into sprocket, allowing for the rest of the print to finish. In addition to the idler sprocket, two spacers were printed to go on either side to allow the sprocket to continue to rotate when the bolt it is mounted to was tightened.

Throughout our nearly 70 matches and countless hours of driver practice, this chain tensioner held up with little to no wear on all printed parts. In fact, the only time a part broke was when it was overtightened.

While this is certainly not a solution for all teams looking to fix loose chain, our resources allowed us to make it in a timely manner.


For the 2017 climber, we used a piece of 2 x 1 tube bolted to the gearbox. We used a VEXPRO VersaBlock clamping bearing block to support one end of the winch shaft. We just slid the VersaBlock out until the chain was tight. It just worked after the block was tightened onto the tube. I don’t think we even needed to put a screw into the tube to prevent the block from slipping.

I would actually say that that sort of tensioner is somewhat appropriate for high-use or bi-directional.

But you don’t want to use aluminum. Use Delrin instead. It might wear a bit, but eventually it’ll stop. I’ve seen something similar used in a drivetrain–delrin half-round placed to guard chain going over an edge and tension it incidentally.


We like the Delrin approach as well.

We were looking at similar issues during our 2018 season so we did some research in the offseason and actually came across 118’s 2018 cad and used that for inspiration.

This is our 2019 bot where we used long belts that weren’t loops but just one long line, and then we machined grooves that fit the teeth of the belt and a back plate that was then clamped to it. that was then attached to another similar block that held the other end of the belt and they had a bolt between them that we used to tension them. We rarely ever had to retension them and we could even adjust are positions of the cart/1st stage if we needed to.


For our elevator this year, we ran a custom tensioning system that served us pretty well. If you were looking to do something like this, hopefully this helps and feel free to contact me with any questions!

After learning about the AndyMark Inline Chain Tensioner through 1114’s elevator this year, I think that if we were to do it again, we would use those instead (lighter and simpler).


Just got a photo of our simpleton example tonight. As others mentioned, it would be better and lighter if we had made it from Delrin. At this point it is rotated to maximum tension.
Probably shouldn’t mention that we left my PLA “test fit” piece in place for 2 regionals, before failing half way through State competition.:disappointed: Somebody had simply placed it in the spare parts.

We’ve struggled with this on occasion on wrists so I went to check out how 2056 was doing it this year and really liked their solution (below). Unrelated but also cool, magnet and rubber stopper at end of the up travel.


Am I correct in thinking that the shaft for the big sprocket just fits into a slot in the thick aluminum plate that the rubber stopper is pressing against? Is it the socket cap screw that adjusts the chain tension?

Yup. I was very impressed with how simple and effective of a solution it was.

Disclaimer: It’s not my robot I just took a picture and asked questions so I may be wrong.

The socket cap screw adjusts the chain tension, but I don’t think the shaft is in the slot, just the bolt that’s in the end of the shaft. Looked to me like the fastener on the end still provided clamp load onto that plate, which it couldn’t do if the shaft was in the slot.

1 Like

Here’s an image from a little bit of a different angle that shows it a little better with the shaft.

I pretty sure you are correct about the clamping.


Thanks @Nessie and @Mike_Schreiber. Looking through pictures on my phone from 2018, I see they used a similar tensioning screw on their drive train and elevator.

Our Inspiration. You’d be surprised how many design ideas we’ve had come from the bicycle.



what are some other design ideas that came from bicycles?


That tensioning solution is super neat!
A few questions:
Was the screw directly contacting the shoulder bolt?
Did you have any issues with the screws coming loose/stripping under impact?
Did you have tensioners on both sides?
How was the hole in the plate for the screw manufactured?