Chain C-C Add Amount

When looking back at a lot of topics on the C-C add for 25 chain, I see a lot of people saying that the center add for 25 chain should generally be around .018. I am currently trying to make a little bit more accurate chain calculator, that can factor in this add, but I have a few questions about it:

  1. Is this add for pre stretched chain, or regular 25 chain?

  2. And bridging off of an old post from Adam Heard:

The amount you add per distance doesn’t seem to be linear. is there a way to calculate this mathematically?

  1. Do any other teams have data on how much they add at certain distances?
1 Like

In reality, each run of chain, each set of sprockets, etc. is quite different, and the only way to get accurate numbers is to experimentally determine the proper spacing. This can explain the different numbers from multiple sources that don’t seem to scale.

1 Like

Here is a measuring procedure:

A lot of this stemmed from @Paul_Copioli’s presentation he gave at MSC in 2013 iirc. Someone else who has it could link the video of that presentation. A lot of what led him to that value he says in the presentation was testing they did and consulting a lot of engineering handbook information, and also consulting with chain manufacturers and suppliers. He said their findings were that the add distance did not need to change based on the length of the rub because the tolerancing maxes out. Again, this is what I remember from his presentation, if someone has the link they could post it here, or perhaps @Paul_Copioli could provide direct source information about it. Clearly @AdamHeard’s post above has different results on some of this.

1 Like

I have the pdf in my resources folder but I am pretty sure it was recorded. I’m still trying to find that.

Practical Design and Calculations by Paul Copioli.pdf (22.0 MB)

1 Like

Here is the spreadsheet I used in the presentation. It also includes a tab about pneumatic use since (before last year) we never used tanks so getting the air consumption correct was important. The pneumatic tab is based on the original sheet done by 1114 back in 2011 / 1012.
2019 Useful Calculations - working.xlsm (64.3 KB)

3 Likes

Interesting to know. Thanks!

Just curious, would this work regardless of if we are using pre stretched or regular chain?

Ive never used ore stretched.

3310 pretty much only uses the #25 HD chain from VEX nowadays.

1 Like

How would you install fixed c-c chain. I tried on a scrap piece and it was very hard to install it.

I don’t know if this is directed towards me, but the best way is to probably pre make the chain to the specified length, and slide it on when putting the sprockets on the assembly.

2 Likes

That is definitely the way we do it.

The tool I linked tells you how many chain links so we have students learn how to use the chain breakers and make enough chain for all robots plus spares.

We don’t do chain in tube so we can replace the chains quickly if a chain breaks unexpectedly.

My trend includes Paul’s point as an input, and I’d say it agrees.

To be really clear, the numbers we (and 973) saw agree that .018" is a good value, for what would be your common length chain run if you run chain from a center wheel to each outer wheel on a 6WD robot of typical frame dimensions

Our numbers do not agree that .018" is the right additional amount for all lengths of chain runs. We’re quite confident that as you get significantly longer or shorter than 10" C-C, .018" is not a good value to use, as we have empirically tested this by CNC’ing frame rails with 1.125 bores at different c-c distances, then installing bearings/shafts/sprockets/chain, leading to the numbers Adam posted. There is really no substitute for empirical testing. We still have to re-run our tests each time we make a drivetrain with non trivially different C-C distances, even after having done this multiple times.

1 Like

We too have done empirical testing on a frame rail with bearing holes at sightly different centers. What we have learned is different chain manufactures have slightly different lengths! Obviously, this shouldn’t be happening.
However, thinking about manufacturing tolerances for any given production run for the chain and sprockets, we might also see measurable differences from the same manufacturer.
The various components are subject to tolerances/manufacturing out of our control, and maybe we have chains made Friday afternoon as well as Wed morning. The only part we really could control was the bearing C-C. And even that may have contributed to stacking tolerances since the bearing holes were made by students.

Cory is correct that .018 isn’t perfect for every distance. For very short runs we tend to use .009” to .012”.

For long runs, we have still used .018. In 2017 our run was 23ish and we still used .018” with #25HD chain. The run was tight at first but was pretty loose near the end of the season. However, it was still fine for us for that game.

Again, we found something that worked for us and moved our efforts on to other things. We don’t like CNC machining stuff as much as 254 does. :slight_smile:

I totally trust Cory here, and it makes total sense that different chains are actually different.

With mcmaster and vexpro standard #25 chain, as well as a little 25H chain we’ve been happy enough with this equation derived from some data Cory posted I think in 2017, for runs from 10-40 inches long (mainly with 16t to 16t sprocket):

actual C-C distance= 1.0012*(calculated C-C distance) + 0.005

This equation gave us around 0.056" add for about a 40inch run that worked reasonably well, at least until the chain stretched. We have seen variation year to year in how well this works, but it’s always been good enough that I’m willing to post it.

It sounds like the answer is if you want to get a really good tension without tensioners you have to empirically determine it, but I think there are reasons it could be useful depending on your situation.

1 Like

Do these numbers significantly change with brand of chain? We have both the Vex 25H and Diamond 25. We will of course just test both to see what happens, but curious as to others’ results.