pic: Floating Zip Tie Chain Tensioner

I think it’d be best for him to step back and view this from an economical perspective.

Think about how many of these you need to make, how quickly each one needs to be made, and how much you need to sell them for to break even.

Then think about how many you think you can possibly sell.

If you do the math, I don’t think it’s a viable investment to try to mass produce these, or even make them on a fairly large scale.

To actually make money you need to be injection molding, not milling these. Injection molding is orders of magnitude more expensive than even the most generous estimate of demand for these.

Given demand, and cost of contracting these out to a machine shop, or time involved in making them yourself, I don’t think you can come out on top. Especially given that anyone with a manual mill or even drill press with x+y table can make one of these in about 15 minutes.

Keep in mind that if you choose to make them or have someone else machine them, you have to be able to meet the criteria of a vendor, as well as be capable to potentially ship one to every team in FIRST, if they ordered one, within x time period (I think 5-7 days? can’t recall).

That is why I am talking to Andy Mark, but Cory. Do you think these will sell if they are priced just right???

Greg & Jason,

Thank you for setting me straight on the copyright issue, and for your thorough discussions.

I’ve mentioned this discussion to “those I mentioned in my post” who use this practice, but they swear it would hold up in court…oh well…

Last night I took this into the preliminary CADs my team has for our '09 prototype bases, and we noticed an inherent flaw. Perhaps this is a flaw with all floating tensioners though.

If the two sprockets in the chain run are significantly different relative to their distance apart (e.g. 22-tooth to a 32-tooth, 6" apart) or their axes of rotation are not level with each other (e.g. transmission output shaft is higher than the wheel axle), this tensioner will have a tendendency to “drift” towards the smaller or lower sprocket and stay there without putting any tension on the chain. Is this correct or are we seeing something that isn’t really true?

http://www.snapidle.com/chain_tensioners.htm We used these when I was working with team 507. they worked nicely however since then we have gone to other methods of chain tensioning and may or may not return to these in the future.

That is true but with this design it is not a problem. On the tops there are spots where you can drill holes and put springs in them so they don’t lean one way or the other too much. We had to do this with the original floating tensioners, but they didn’t have any spots to drill holes into them (a super hassle and a waste of time).

Installation of a snap idle : Enough tension need only be applied to take out the slack of the chain. When the drive is in operation the Snapidle® will move back and forth on the chain periodically.
This is a qoute from the snap idle page I referenced earlier. While we did not have such a large difference between our sprockets in use your application may still work due to the snap idle’s slight movement. It wouldn’t hurt to give it a try. If you do let us know.

I’ll agree with the importance of researching the market as part of a business plan. I would have a hard time seeing how someone could get rich manufacturing these tensioners, however I have no problem believing that they could be manufacutured profitably. Potentially the greatest profit to be had from marketing and manufacturing them is the experience of developing a business plan, refining the production techniques, setting up a business and running it. Even if the business runs at a small loss having that on a resume and that exeperience as part of a professional tool set will be worth thousands of dollars…

But I strongly disagree with the need to have them injection molded in order to turn a profit. The elegance of this particular design is that, as I have mentioned, it can be mass produced in a woodwork shop. The table saw is great for cutting long straight lines and long deep grooves. If you look at the jigs used for creating finger joints that might give you an idea how to make repeated, evenly spaced cuts. Do the cuts on one side of a sheet of HDPE, flip it over, turn it 90 degrees and do the cuts in the opposite direction, then slice it and cut to length to turn out the final product.

This will require a bit of R&D… talking to someone with some fine woodworking experience would be a big help. I would also suggest prototyping the production process using a less expensive material, such as MDF, until the jigs are worked out and acceptable tolerances are achieved.

For an investment of probably less than $200, plus hours and hours of work, it is quite possible to have hundreds of these produced and ready to go very quickly.

I don’t think anyone is going to become a millionaire off of this, but the beauty of it is that done properly any financial losses… even if sales are terrible… will be offset by the educational value of going through the entrepreneurial process.

Go for it!


Umm, so let me get this straight if you used this as a chain tension device than how long would it last due to the chain grinding against it. Also chain has that special pattern to it with that wouldn’t it get locked up on the edges of this device? might want some sort of bearingS to prevent friction but that is just me.

Its obviously not as efficient as something that rotates, but the plastic used is relatively low friction. I also imagine that the side walls do not hit the side of the chain unless the chain is moving side to side somewhat.

I would have to see some test data before i make any real judgment on things since the world is full surprises.

Go to mcmaster.com and type in 5973K1. My idea is based off of this but has a different way of tightening it and is superior in price and quality. Here is my final design of it:


We have been using Mcmaster Carr tensioners and chain does not get locked into the tensioner. UHMW is a very good plastic for this application and the side walls are usually never hit due to the smaller side walls. It is fit for both #25 and #35 chain.

what is the purpose of the holes on the front and back?

Say that you want to tension the chain on an arm robot. The chain is going vertically, so you add springs to each side of holes and the tensioner now just hangs: suspended by springs and it still tensions the chain. Also you can attach zip ties or springs to limit how far it floats or if you just want it to stay stationary. Example shown below:

Look at the white tensioner on the lift mechanism, it is suspended by springs and zip ties. The sprockets lifting the arm were of different sizes (a 60 tooth and a 15 tooth), so the tensioner had a tendency of diving down to the smaller sprocket.

While you can do a lot of engineering and material science related equations, there are a lot of good approximations and rules of thumb that high school students can use on FIRST robots. (Just don’t use that practice IRL ;)).

Delrin (and UHMW) are both engineering plastics that have very low friction, and both work great for chain tensioners of various kinds. I prefer Delrin where possible since it machines very nicely (you can mill it really fast too), but it does cost quite a bit.

But if you want proof of plastic chain tensioners, here’s a photo from Team 228’s 2007 robot.


The white plastic circles are either Delrin (back one) or UHMW (front one). You can also see a third Delrin one on the drivetrain through one of the chassis rail lightening holes in the back. These are really simple ones: plastic rod with an off-center hole. As you rotate it around, it varies the radius; and by cranking down the bolt, it will hold its position.

The Delrin ones held up the entire year (about five competitions and a lot of other random driving) with barely any noticeable surface scratches. The UHMW ones had small (about 1/16") grooves in them from the chain. (There was no particular reason why certain ones were Delrin and others UHMW; they just happened to be whichever our machinist mentor grabbed from the stockpile when making them.)

Saw this on 1902’s '06 bot and chuckled after remembering this thread… apparently they work :slight_smile:

I can see why you would want two different sized grooves for the chain to accommodate both #25 and #35 chain, but do you need the different sized grooves for the zip ties, too?

I ask only because it would seem to save a few manufacturing steps to put only a single groove in for each zip tie.

In a similar vein, would it, perhaps, be easier to drill the holes ALL the way through the tensioner? That way you would only need to do two drilling operations rather than four.

You may have a different production path in mind than I envision (as I’ve mentioned, I see this being an ideal shape to cut on a table saw). For instance with a CNC router, putting in extra grooves would require little in the way of additional manufacturing time. But if you are going to be kicking these out by the hundred using some manual production method, you probably want to include optimizing the production path as part of your design process.

Just a few thoughts that occurred to me without having “seen inside your head” to know why you made these design choices… perhaps I have overlooked something perfectly obvious to you. If so, my apologies.


I agree that the 2 grooves for zipties seems unneccesary, and that the holes could go all the way through.

Another thing to consider:
If you were to make 2 different models for two different sizes of chain, it would simplify your design immensely.
I doubt many teams will be switching their tensioners between different sizes of chain very often.
Just because there is demand for more than one chain size, doesn’t mean you need to incorporate all sizes into one product

Honestly, I liked your original design better. I think the most recent one you posted looks too complicated and time-consuming to machine.

Guys, the reason for one tensioner is that we used the #25 Tensioner for both 25 and 35 chain and it worked perfectly. There will be no machining, i am probably going to have a mold for this made. I will probably end up changing the final design after all the comments and after more test runs of this tensioner.

Thanks Cd

If you are molding it then I would suggest rounding the exterior edges and corners. You might also want to think about putting some draft in to your design, and possibly stamping a product name, website or trademark in to the side as part of the molding process.

I hope that the sales volume is sufficient to cover the up-front cost of the tooling.