Last season, we switched to chains on our robot partway through, and it caused us all sorts of problems because the wheel spacing was the worst possible (the chain was very loose, yet did not have enough slack to take off a half link) and we had problems with chains falling off. We would like to use chains next year, but avoid all those problems. From what I have heard, #35 chain is not very stretchable, and correctly calculating the center-center distance is sufficient. However, a mentor is still worried about stretching and recommended we add a chain tensioning system, which would add complexity to the chassis. What are your thoughts on adjustable tensioning with #35 chain? Worth it or not?
It depends on how long your chain runs are, and how well constructed your system is. Short distances in a very rigid part? Should be no problem. Long runs to a gearbox mounted such that it can flex over time? Add a tensioner.
Wait to see what others post on this topic, but I personally am a fan of tensioning. In a pinch, you don’t need an entire tensioning system, you can use beefy Zip ties to tension a chain.
We personally have done this in competition for belts, I have seen other teams use a similar approach for chains. It is very effective, cheap, fast, and simple, and because of that, its a no-brainer to add to any drive system.
Regards,
Kevin
We spaced our no. 35 chains during our design and the chains are still running on the robot although a little more stretched out after the brand new chain we used was stretched out.
I as little skeptical myself at first so we did have a backup chain tensioning system ready to use but the chains have been satisfactory.
Keep in mind that the longer your chains are the more links you have which is where the tiny bit of stretch is coming from. With the decreased frame perimeter sizes this is less of an issue compared to 2012 and before.
We are definitely going to try to use the same method in 2015. Our mechanical team loved not having to worry about chain tension this year for the first time.
The chain doesn’t really “stretch”…it wears. It will wear in a noticeable amount when you first use it, then will hold it’s length for quite a while. Proper lubrication helps it keep from wearing as quickly, too.
Having adjustment is nice, but not necessary if you do a good job on the design and fabrication.
One thing that is often over looked is the “springyness” or stiffness of the support system. If you are throwing chains, it might be because your sprockets are deflecting.
In any kind of chain or belt drive it’s a good idea to sue a tensioner. I don’t know if it matters as much for #35 chain, but if you’re using #35, you can afford the extra weight/space to add a simple tensioning block like in a WCD.
It doesn’t. For FRC use most 35 is fine C-C with no tensioning system. As long as your chain runs aren’t longer than 17" or so you’re set for the season.
Asid, I’d stray away from making assumptions like that that you’re not completely sure of. Misleading information like that is dangerous. I believe Anant usually made a point about misleading information to students each year. I remember he told me directly when I spoke to him at your workshops a few years back.
It depends on the OP’s application. If the team simply uses an arbitrary distance for the sprockets on a drive train, then the chain could be quite loose without half-links. Adding a tensioner allows for flexibility and accounts for unknown manufacturing tolerances. On the other hand, if they design properly they may not need one. Better safe than sorry (especially if almost unnoticable mistakes are made in calculating c-c distance).
I don’t think he’s misleading anyone Andrew.
And in away you are more so misleading, stating that “it doesn’t” and is “fine” for “most” use. And how did you arrive at the 17’ number?
And for this type of questions really depends. How precise can you make the C-C with your fabrication techniques, what length run, what kind of power is being put though the chain etcetcetc. The key as usual is making sure the sprockets are lined up **and ** proper tension is applied.
With the WCD type design adding in adjustable bearing blocks are easy and imho give lots of ease at mind for anything that could come up. Abit I personally used and seen C-C on chain work IF the machining tolerances are held up to spec (personally I did +.01 for a 15in run) and on a HAAS mill so it was dead on. But again it comes down to testing and experience.
My suggestion, always have some kind of tensioner and in many industrial applications this is an requirement to.
Edit* Forgot this, but C-C is a great thing and works great on theory and sometimes in real life… but I’ve seen it fail quite often in practice due to many reasons. And know teams that have regretted it more the once.
I’m referring to OP’s use in a drivetrain. There is no question about the application in question. OP asked a question, I answered it with the answer OP was looking for. No hypotheticals - just a straight answer. The 17" number was based off of the older frame dimensions where the longer lengths around that number would lengthen more over a season.
Precision for chain isn’t difficult to do, since you just need a multiple of the pitch, which is in eighths and easily measurable with a tape measure or ruler. It doesn’t take precision fabrication to get C-C distance with hand measurements and an electric drill (something I’ve done for the past 4 years as a student on 256). While power is indeed a variable to look at, this returns to the fact that OP was referring to a drive system, where judging by even the really extreme systems shows that 35 chain can consistently and reliably handle the loads thrown at it, as it has for years.
I agree that adding a tensioner is simple in a WCD. I’m not saying it’s not. OP asked if it’s needed. I’m saying it’s not.
You are correct in that there is much more that is being overlooked, but judging by OP’s question I do not doubt that this is the question they wanted answered and that - possibly - an excess of extra information like others have been posting may take away from the original point of this thread.
Yes the OP asked a simple question but with a complex answer.
Giving him a answer based on your experience without some context can be slightly misleading. There is a lot of “it worked for me so it has to work for you” type stuff floating around here, and often that is not the case. There are many types of drive trains out there and the OP didn’t specific anything about them. (He could be talking about swerve for all we know) and just many variables at play. As I said, C-C chain can work, but it doesn’t always usually based a team’s experience and testing. The drive base is one of the most important part of a FRC robot, and is something you want 100% confidence in. Planning some tensioners, even not used, can save a lot of headache down the road.
And drilling holes for C-C with a hand drill and a tape measure… I reckon that it needs to be within at least ±.04in for it to work effectively… probably you want something closer +/- to .01in. Sure this is possible to do on a drill press, but isn’t trivial. It isn’t “precision” machining but it isn’t just sharpie some marks and go for it and the OP should be aware of this and is a important part of getting C-C to work.
This is a question that comes up a lot “do we need tensioners” on the robot. And leaving a bit of information for the OP to think about and explore with off season prototyping and research is never a bad thing.
Although the topic of the thread implies it, I feel I should clarify that you should not fabricate parts for a chain run using this method for anything but #35 chain. And even for #35 chain I would not suggest suggest using a tape measure for your C-C distance. And I definitely would not make the statement that precision in a chain system is easily achievable with the methods you suggest. What you’re suggesting is somewhat worrying, using a ruler and hand drill is not a terribly accurate way of positioning holes. It may be true that #35 chain doesn’t need extremely accurate C-C distance, but only in the aspect that it can handle significant slack compared to belts of 25 chain. The point is that the tolerance for positioning the holes is a heavily lopsided bilateral tolerance and not a unilateral tolerance. This is extremely important, if you’re using a ruler or something similar to position your holes and attempting an exact C-C distance I wouldn’t be surprised if half the time the holes are just a hair too far apart. If you are going to drill your holes based off of ruler or tape measure measurements I suggest accounting for the error in the measurement and drilling of the holes and aiming for slightly less than the exact C-C distance.
-Adrian
Amen to that! https://twitter.com/team1306/status/447098680168820736
Thanks for all the responses. I think we will try center-center distance in the offseason to verify it works. Our chain runs well not be very long, about 9" center to center. We are using laser cutting to manufacture parts, so accuracy will not be an issue. What I am wondering now is the correct way to calculate sprocket diameter. Is tooth count*3/8" / pi sufficient, especially for small sprockets?
There are calculators online to find the correct number of links of chain in a sprocket drive. This is the one I like to use: