Gasman1440

How do you determine the correct chain length for a given drivetrain? I have seen several videos, tips and tricks that show pictures of the derailure and a drawn in line for example but I知 not smart enough to tell the difference visually. I知 an engineer not an artist so I would love a simple formulae or way to calculate the proper chasing length based on type of derailure and max number of teeth or similar. Any help or guidance? I have three different bikes that I want to check / modify the chain. Txs

Bill Kapaun

Wrap the loose chain around the biggest ring & cog (don't go through the RDER)
Round up 1"?
You want at least 1" but not more than 2" slack.

mack_turtle

there is no formula. I've not found anything more simple the method described on this Park Tool page. I know "engineers" think everything can be broken down to a theoretical equation, but mechanics don't waste their time with that. the only way to do it right is to do it on the bike. measure twice, cut once.

cxwrench

The safest way is small/small w/ just enough tension the chain doesn't rub on the upper pulley. If all parts of the drivetrain are in spec this will always work. Some of the new 1X drivetrains use different methods but this works on a double or triple every time. There are no formulas to determine chain length. As mack says in the post previous to mine "the only way to do it right is to do it on the bike". If you've watched 'several' videos you really should be able to figure this out, it's not rocket surgery.

MudPie

on the above Park Tool Page, #6 offers an equation for determining chain length. As an engineer, I also appreciate formulas.

Bill Kapaun

PARK shows an equation in the link you provided.
Personally, it's much easier to just wrap the chain around a ring & cog then take time to measure, write down numbers and do the math.
I wouldn't really trust a measurement from axle center to crank center to be THAT precise when juggling a snaky tape measure. I'm very far sighted (with a lazy eye to boot), old & a bit shaky. A monkey can wrap a chain.

mack_turtle

I find it funny that engineers ALWAYS start every conversation by telling everyone that they are an engineer.

Shame on me for never noticing that formula. When you work in a bike shop and replace chains ten times a day, you don't have time for that and stick to the practical method. I'd start with that, but check the chain on the bike to make sure it's correct. Real life > theory.

hokiefyd 

I do the small/small method as this gives a little extra chain length than the big/big method and I tend to prefer a little more chain slack/less chain tension in the system. I don't think there's a mechanical benefit to it and there are likely some drawbacks, but I think the drivetrain generally feels a little smoother to me this way.

Koyote

And I always wonder how many of those people actually have earned engineering degrees from accredited institutions and have then gained licensure.

https://www.allengineeringschools.co...g%20experience

delbiker1 

I have used the equation on Park Tool website and came up with the same length as doing the large/large. I no longer use the equation.

HillRider

Not that it matters but I have both a BS and MS in Chemical Engineering and a PE license in PA. So there!

Anyway, the big-big+1" method is safer if you have exceeded the rear derailleur published wrap capacity as small-small will give a too short chain. I have substituted a 26T or 24T granny chainring for the factory 30T on many triple cranks and the overall total teeth then exceed the rds's wrap capacity. Big-big+1'"assures the drivetrain is safe and small-small gives a slack chain which is unusable but not damaging.

HerrKaLeun

RTFM. Teh RD manufacturer has the proper procedure for THE DERAILLEUR in question. There are some differences between models (ie.. 2 or 4 added half links etc., full suspension correction etc.). RTFM

mack_turtle

watch out, we've got a badass over here!

DaveSSS 

I got my mechanical engineering degree from Kansas State University, in 1981, at age 28. After that, I spent 22 years working in a nuclear weapons manufacturing plant, with up to 6000 employees. A PE license was not a requirement to develop manufacturing processes for nuclear weapons components. I quit that job to follow my wife's more lucrative job, working for the US Treasury Department. Finding an equivalent position at age 50 proved impossible, so I retired, 18 years ago.

Badger6

If I am replacing a chain on an existing drivetrain, in which no further changes in gearing (meaning it will use the same number of teeth on big chain ring and the largest sprocket on the cassette), grip the chain in the middle, let it hang, and move the ends to equal length, then count pins and multiply by 2. If you use a quick link add 2, if you use a pressed pin, and you ended up with an odd number before multiplying, because the counted the side with one less pin, ad 1, and then multiply by 2.

But, you knew that...so, I assume you are putting a chain onto a new drivetrain for the first time....Google "chain length calculator." There are several calculators available for use, that ask for the sizes of the big ring and largest sprocket and your chain stay length. They've never failed me. I "build" a few bikes a year for myself and other people, I've had a 100% success rate using the available calculators. I am partial to this one: Chain Length Calculator. Good luck!

mack_turtle

true to form, an engineer rattles off his credentials, unsolicited, to anyone and everyone, for no reason at all. thank you for confirming the stereotype.

billnuke1 

I’ve got my “BT,DT...Been there, done that...”...

DaveSSS 

You're welcome, but you still can't be sure if any of that is true. Someone else seems to think that a PE license is needed to understand or advise on chain length. The park tool website has lots of info. No engineering degree needed to read it.

The big/big plus 1 inch method gives the shortest possible chain length, but the chain might need to be longer if a larger sprocket is used, sometime in the future. The little/little method gives the longest length that will work with any cassette within the RD's wrap capacity. I have a setup that slightly exceeds the RD wrap capacity, so the big/big method insures the shortest safe chain length. The chain is then slightly too long for the little/little combination that I never use.

HillRider

No, the original posting questioned whether those claiming to be engineers really are. We were just establishing our bonafides. As Walter Brennan (anyone else remember him?) used to say; "no brag, just fact."

And, no, you don't need a PE if you work for a corporation as your covered by your employer. You only really need it in private practice.

caloso

I've used both the big-big+1 and the small-small methods. In my experience, the big-big+1 method works well if you're currently measuring the biggest cassette you're likely to use. If you tend to switch cassettes or wheelsets for different routes, you may be safer with the small-small method.

Phil_gretz

Although I have an engineering degree, I never talk about it.

I do believe that the empirical method is best. For me, it's large-to-large plus 1 full link. But I prefer a shorter chain and snappier shifting.

Iride01 

Shimano give their suggested chain sizing for it's components in most every service instruction or dealers manual for every model line I've ever looked at. Sometimes their recommendation is dependent on whether it's a 1x, 2x or 3x crankset.

The main thing is, though the method used, you just know how to tell when it's too long or too short.

Too short and your pulley wheels might try to merge disastrously with your cassette cogs. Too long and your chain will be too slack when in the smalls. Either might give you less than par shifting. And probably a few other concerns.

Anyway, even for an engineer, I'd think you want to look at what the manufacturer recommends. And many do. SRAM and others don't make it as easy to find as Shimano does, IMO. But when I search hard on their site for it, I can usually come up with it.

Lacking that, then you can go by rule of thumb generalizations and common sense of what you see and experience after putting it on.

DaveSSS 

What the manufacturer lists for maximum sprocket size or wrap capacity has to include an extra 2-3T of wrap to handle the in between chain stay lengths that don't correlate to an even number of chain inches. That's where the rigorous formula at the park tool website can be helpful, but only if you have the geometry chart for your frame to get the most accurate chain stay length, measured from the center of the BB to the rear axle, along the centerline of the bike. That's hard to measure at home. I just make sure that the chain is long enough with the big/big method, then check for chain slack in the little/little.

SRAM claims that my standard Force AXS RD won't work with their new 10-36 cassette, but it does, with the aid of a slightly longer B screw. I have a crank with a 16T difference instead of SRAM's 13T and it still only comes up about 1T short on the wrap, at the small end. If I used a 48/32 crank instead of a 46/30 it would still work, with no slack at the little end.

Gasman1440

Thanks for all the great advice, Ill have to remember to call myself a toad or something in the next post! Didnt realize the impact of saying I’m an engineer would have. I didnt explain myself very well in that the other reason a formulae like many above mentioned is very helpful is I’m looking at changing a cassette and want to know if I need a new chain or if the existing, brand new chain, is long enough. I also have a 1X setup that had a 32t crank (9-46 cassette) that was changed to a 28t and a new chain. When I found the 28t to low my LBS installed the 32t and left the same chain. I’m having some minor issues and wonder if the chain is to tight which is another reason I was looking for a formulae. Txs

P.S. yes there is a formulae for virtually everything....so says the engineer

HillRider

One more time. The manufacturer's recommendations are based on your staying within their published limits for all parameters like max and min cog sizes, chainring sizes and total teeth. If you go outside their limits, you better be willing to adjust as needed.