Chain in tube tension WCD

Hello, I hope everyone is having a good day.
I am designing an offseason west coast drive for my team for the first time and I have several questions that were not answered in other threads. I appreciate any advice other teams can share.
Chain in tube:
Initially, I designed my WCD very close to 1678’s 2019 robot with WCP side bearing blocks and WCP cams. However, my team wanted to try a chain in tube design so I altered the design to have fixed wheel locations (chain dist calculator + 0.018") and press fit holes for bearings to be inserted directly into the tube, inspired by 319’s design. This method does not allow us to tension the chains, but many teams that have used chain-in-tube have demonstrated that with the correct distance between sprockets, this tensioning is unnecessary (since skipping teeth is impossible).

What are your thoughts on whether tensioning is required on chain-in-tube drivetrains and how to design them? In addition, if you have any other advice on how to design west coast drives, I would really appreciate it.

Close up of the 1678 tensioning set up

And 319’s set up
Both 1678 and 319 have generously shared their CAD so feel free to check them out for yourself. The links above are to their onshape documents.


Here is the video series for 2363’s CIT drivetrain.


One of the main advantages of chain-in-tube is that the top and bottom tube walls help constrain the chain so that it can’t jump off the sprocket even if it’s loose enough that it would jump on a normal WCD. You may get a bit more backlash, but that shouldn’t be a problem if you consistently pre-load your chain before auton.

We went chain-in-tube with exact C-C in 2018 and had no problems with it. The biggest problem is that if you throw a chain it’s a PITA to replace, but luckily that never happened to us. I do recommend using #35 chain to get an extra FoS in that area for that reason, though teams have definitely used #25 and been fine.

1 Like

@notmattlythgoe, thanks for the video! I should have also mentioned versablocks as an option for tensioning chain in tube. However, I prefer the WCP side bearing blocks because they are more compact. I was wondering whether I really needed that since almost all teams using chain in tube that I’ve seen haven’t used tensioners.

@AriMB I think we will use #25 chain since it packages well in 2x1 in combination with andymark’s 17 tooth double sprocket.

We’ve joined Brando’s camp on this, and have been happy ever since.


Another datapoint - if you decide you need a tensioner, the solution is to use an outside-tube tensioner. The fan favorite seems to be these. This allows you to be able to slide back and forth without the interference issues the in-tube bearing blocks have. The only issue is that it reduces the maximum possible wheel spacing, since bringing the wheels all the way to the edge interferes with the corner gussets.

1 Like

So do you recommend just using a fixed C-C distance with no 0.018" adder? Or do you mean I need a proper tensioner?

Many teams have tensioning systems inspired by 27 / the 221 SimpleTube SuperLight chassis, and that works fine too.


We prefer to use a tensioning system unless we actually measure the C-C distance needed for the chain run using all the same components (including the same batch of chain) we plan to use on our robot. We don’t use an adder that we didn’t find experimentally, anymore. We describe a measuring technique here:

1 Like

My team ran a chain in tube WCD in 2020 and from our experience I would advise against running without tensioners. We had to tighten our chain multiple times during the season. At one competition we had chains both fall off and break after running matches without re tensioning the chains.

1 Like

Clamping blocks prevent the slick no-cutouts in the bellypan look, but they’re too convenient to not use in many cases.

I’ve seen at least two teams that forgot about that as evidenced by some nice hacksaw cutouts.


@Nate_Laverdure I couldn’t find what you meant by the 118 / the 221 SimpleTube chassis tensioning system. I looked at 118’s 2019 robot but it appears that they do not have a chain tensioner either.

@troy_dietz I can see how the versablock would be useful, but if most teams aren’t using any tensioners for chain in tube, I don’t want to add unnecessary complexity.

Maybe I should rephrase my question. I’ve seen many robots that have chain in tube without tensioners. In fact, almost all of the chain in tube drive trains I’ve looked at have not had tensioners. I’ve calculated the c-c distance between the sprockets and add 0.018" for slop. Is that good enough for maintaining chain tension?

One more question, what is the point of the diamond shaped plate that the bearing sits on in the above image? I’ve seen it on another robot but I don’t know what its purpose is.

Since they’re using sheetmetal instead of 2x1 or similar tubing, I would assume that that’s just a thin plate for adding a second bearing so the shaft has two support points. The profile is similar to the WCP bearing block, which has a function to the plastic clamping bearing block mentioned above. It just has a smaller profile and is a way less viable tensioning block (I would assume), so teams use it for reinforcing shafts in places with thin-walled tubing.

I believe this is what Nate is talking about, sliding bearing blocks with a bolt in the end of the tubes to tension. Images publicly shared by team 33 in the frc discord and probably other places.

I designed and built an off season chassis with just the .018" adder. We used it for driver tryouts and some practice, after a while you could definitely see/hear the chain slapping around, so it doesn’t stay particularly will tensioned. However it kept working great and only had an increase in backlash, I was never worried that it was actually going to fail. I recommend the WCP double sprockets in 1/8" wall tubing, 17 or 18 teeth depending on if you want a center drop.


You typically want to make sure your bearings are well supported. Typical 1/2" round or hex bearings have .25" between the flange and the other end of the bearing, and you want to make sure that it’s well supported. This isn’t to say you always need the bearing in .25" of material, but I would say .125" minimum is a good rule of thumb. If you want to use thinner materials then to save weight, sometimes you need to add a diamond shaped plate like that to increase the thickness just around the bearing hole.

We used the setup in these pictures for the 2020 season and were very happy with the results. This resulted in the chain and the tensioner being inside the tube, so the screw head for tensioning was at the end of the drive tube, making it very easy to access and tension.

The T-block that the screw threads into to tension is counterbored into the bearing block plates on either side. We just machined a square counterbore pocket on the inner face of each bearing plate to allow for this (you can see the square shape in the second picture, there are little ears in each corner which allowed it to be made on our router).

The plate on the end was just a 1/4" plate that had grooves machined on either end to bottom out on the end of the drive tube, and then a counterbored hole for the 1/4-20 bolt to go through. I never knew that 221 sold the block they have to effectively do the same thing at the end of the tube, that would work the same way here.

I do think in the future it could just be a #10-32 bolt, but it’s not really costing much to do 1/4-20 so meh.

EDIT: I will also say I did run direct C-C #25 chain on DTs during my time with 2451 using the +0.018" method that Paul Copioli brought to the community, and iirc we never had a chain fail using this method, and was always happy with the tension we had throughout the season.

However, the years I was with 2451 we were playing 2-3 regionals and worlds, whereas in districts with 33 we played 3 district events, dcmp, and worlds, which was a good amount more matches, so I don’t have the personal experience to say that the direct C-C keeps a good tension over like 100 matches.

Edited my post above, I was pointing at the wrong thing, sorry!

1 Like

The screw tensioning method is brilliant, but my team has very limited machining capability so I’m not sure if we can do that. Maybe heat set inserts in a 3D print is an option.

So far my understanding is that chain in tube prevents the chain from coming off of the sprocket even if it becomes stretched, but a loose chain is noisier and less accurate. Is that true? Do any teams recommend using chain in tube with no tensioner?