Optimal Chainlength - It's A New Day

Back by popular demand, the general all-things Road forum!

Moderator: robbosmans

User avatar
wheelbuilder
Posts: 1193
Joined: Wed Feb 08, 2017 2:10 am

by wheelbuilder

Nice work Cal. I actually called Shimano last Friday about this exact thing. You are 100 percent correct. Chain should appear too long in small front/third from bottom rear. Take out some of the slop with B tension but keep chain wrapped. Just how it is.
Never cheer before you know who is winning

by Weenie


Visit starbike.com Online Retailer for HighEnd cycling components
Great Prices ✓    Broad Selection ✓    Worldwide Delivery ✓

www.starbike.com



TobinHatesYou
Posts: 12458
Joined: Mon Jul 24, 2017 12:02 pm

by TobinHatesYou

In summary:
If you size a chain on any drivetrain using the small-small routed (Campy) method, make sure your b-gap is exactly right. Otherwise just use the Shimano or SRAM method of big-big + 2 half links and it may actually help you notice if your b-gap needs adjustment.

For me, big-big + 2 links matches the simple equation to the nearest integer with every modern road derailleur geometry I've tried.

User avatar
Calnago
In Memoriam
Posts: 8612
Joined: Sun Nov 07, 2010 9:14 pm

by Calnago

@Wheelbuilder: Yup... sometimes the docs aren’t very clear. In particular the little diagram to tell where to cut the chain is more than a bit confusing. That’s why I created my own examples above as a little game... “where to cut the chain”.
Also, this is all related to what I started months ago in the thread about direct mount hangers. And just a day or two ago someone posted the statement to the effect of “... by all accounts the Direct Mount Hanger clears up the difficulty in rear wheel removal”. Aaach, i thought. No no no whether it’s a DM hanger or not, that is a separate issue entirely from the wheel removal issue. But it just made me more determined to finally create this thread which has been in the back of my mind for months now. Also, there was a thread about how noisy the new Shimano drive trains were with quite a few posters complaining about it. I asked if they could present some pics, but no one did. The reason I asked is because I thing the excessive noise they’re talking about could easily be caused by a chain that’s on the short side of optimal, which would be the case in Shimanos scenario that allows the addition of just one link. Like I tried to emphasize... always opt for the three in that case as one link extra is just not enough.

My next post on this will may be tomorrow morning and it will be just a continuation of that thread about DM hangers they I linked to earlier in this thread.
Colnago C64 - The Naked Build; Colnago C60 - PR99; Trek Koppenberg - Where Emonda and Domane Meet;
Unlinked Builds (searchable): Colnago C59 - 5 Years Later; Trek Emonda SL Campagnolo SR; Special Colnago EPQ

User avatar
Calnago
In Memoriam
Posts: 8612
Joined: Sun Nov 07, 2010 9:14 pm

by Calnago

@Tobin: that’s a very bad and incorrect summary.
Colnago C64 - The Naked Build; Colnago C60 - PR99; Trek Koppenberg - Where Emonda and Domane Meet;
Unlinked Builds (searchable): Colnago C59 - 5 Years Later; Trek Emonda SL Campagnolo SR; Special Colnago EPQ

TobinHatesYou
Posts: 12458
Joined: Mon Jul 24, 2017 12:02 pm

by TobinHatesYou

Lol there you go again. You have to find fault in whatever I say. Basically you've just said the equation-based sizing method is incorrect.

I used the Campy method right up to the point when I switch to WiFli. I then found it hard to eyeball the correct amount of "cage slack" with the longer cage. The first couple times I cut my chain, I left it 2 half-links too long because these longer cages tend to look fine even when they have virtually no spring-tension pulling them back. I have since reverted to using the big-big 2 half-links method, which is the same as the equation.

I just want to confirm: Are you saying the equation is wrong?
Last edited by TobinHatesYou on Thu Jun 28, 2018 12:59 am, edited 2 times in total.

User avatar
Calnago
In Memoriam
Posts: 8612
Joined: Sun Nov 07, 2010 9:14 pm

by Calnago

I don’t “have to find fault” in what you said. It’s just there, front and center. Please go away. Completely incorrect summary and I’m tired of correcting you. You’re a waste of my time.
Colnago C64 - The Naked Build; Colnago C60 - PR99; Trek Koppenberg - Where Emonda and Domane Meet;
Unlinked Builds (searchable): Colnago C59 - 5 Years Later; Trek Emonda SL Campagnolo SR; Special Colnago EPQ

User avatar
TonyM
Posts: 3376
Joined: Thu Jan 22, 2015 4:11 pm

by TonyM

Back in the days (80s/90s, 8 speed) Shimano was stating that the correct length was big chainring - smallest cog and the pulleys should be straight.

Maybe because it was the recommended « optimal » gear at that time?? Everybody had big gears...




TobinHatesYou
Posts: 12458
Joined: Mon Jul 24, 2017 12:02 pm

by TobinHatesYou

The simple equation is the mathematical equivalent of big-big sizing. It takes into account the amount of chain that goes around half the largest cog, half the the largest chainring, adds twice the distance of the chainstays and adds 1" (2 half links.) The only minor difference is the actual chain length between axle and spindle will be the hypotenuse of the right triangle where one side is the chainstay length and the other side is the difference in diameter between the cog and chainring. This can largely be handwaved away by rounding up the final value to the nearest integer.

There's no argument against this. It's primary education level math and whether you've incorrectly set up your rear derailleur is irrelevant. The output is the correct chain length.

In addition, I find it funny that Campy is using such a roundabout description for 12-speed. They're just saying a standard road bike with a 50t outer ring and a 32t largest cog requires 110 half links.

Let's do some math!

415mm = 16.3386in
2(16.3386)+(32/4)+(50/4)+1=54.1772
Round that value up to 55" = 110 half links!

Now you can say, but wait, Campy suggests the same 110 half links up to 53t chainrings. Well yes there is breathing room because we rounded up quite a bit in the 50t scenario and that number itself has some cushioning for the sake of safety. If we use a 53t chainring the math becomes:

16.3386*2+(32/4)+(53/4)+1=54.9272
Round that value up to 55" = 110 half links!

There's no magic here.

--

TL;DR If everyone just used the equation or its visual counterpart, life would be easier.
Last edited by TobinHatesYou on Thu Jun 28, 2018 2:18 am, edited 2 times in total.

kode54
Posts: 3750
Joined: Tue May 23, 2006 9:39 pm

by kode54

good to know!
Calnago wrote:
Wed Jun 27, 2018 11:50 pm
That loose chain that I pictured is in fact Di2, and it is on the 3rd cog from the smallest. I just loosened the B-screw to highlight the fact that depending on where the b-screw is set initially, it could be deceptive as to where exactly to cut the chain if you were trying to use the small/small method. That “loose”’ chain I depicted is actually the ideal length for that setup, it just needs the final b-screw adjustment.
- Factor Ostro VAM Disc
- Factor LS Disc
- Specialized Aethos Disc
- Sturdy Ti Allroad Disc
- Guru Praemio R Disc

syplam
Posts: 90
Joined: Thu Feb 02, 2017 7:30 am

by syplam

Thanks for the explanation Cal, so you have any advice for Red/Etap ? Should I use the Shimano method for Etap as well?

Sent from my SM-G9650 using Tapatalk


TobinHatesYou
Posts: 12458
Joined: Mon Jul 24, 2017 12:02 pm

by TobinHatesYou

syplam wrote:
Thu Jun 28, 2018 2:01 am
Thanks for the explanation Cal, so you have any advice for Red/Etap ? Should I use the Shimano method for Etap as well?

Use big-big + 2 half links or the "simple equation" for every modern drivetrain. Include the quicklink while measuring if using one.

Simple equation = 2(chainstay length in inches) + (largest cog tooth count)/4 + (largest chainring tooth count)/4 + 1 = chain length in inches. Round up to the nearest integer and multiply by 2 to get the appropriate number of half links.
Last edited by TobinHatesYou on Thu Jun 28, 2018 2:32 am, edited 2 times in total.

jlok
Posts: 2400
Joined: Tue Jun 30, 2015 3:30 am

by jlok

Thanks Tobin. I never thought about such simple equation. This make things such much easier.

And also thanks to Calnago. I thought the chain slack at 3rd small sprocket and small ring was caused by too long a chain. Now it's clearly the B-screwed up...
Rikulau V9 DB Custom < BMC TM02 < Litespeed T1sl Disc < Giant Propel Advanced SL Disc 1 < Propel Adv < TCR Adv SL Disc < KTM Revelator Sky < CAAD 12 Disc < Domane S Disc < Alize < CAAD 10

User avatar
Calnago
In Memoriam
Posts: 8612
Joined: Sun Nov 07, 2010 9:14 pm

by Calnago

@Tobin: Why I bother sometimes is beyond me... but here goes. I'm well aware of the math and the equation that you can use. It will get you a chain length that will fit big to big. But what do you think the +1 in the equation is for? It's to ensure the chain will indeed be long enough, including a standard derailleur cage. But after that, any rounding of the final answer is done to the nearest integer. Therefore, correct application of the equation would end up with 54 (108 links) as the answer, not 55 (110 links) as you conveniently bumped it up to. And what about a case where the largest cog is 29, not 32. And throw in 405mm chainstays. What's your equation going to give you then? Probably not 110 links. Point is, it's just not only about having the chain physically long enough anymore, that's easy... it's about getting the optimal chainlength to work perfectly with todays derailleurs and wider range cassettes, so that when all is said and done it can be fine tuned and adjusted to the cassette as closely as possible. Campy is saying that 110 links should be used for all combinations of their 12sp systems for chainstay lengths between 405-415mm. Doesn't get much easier than that. And Shimano has their system, quite different, and isn't so all encompassing. And as I pointed out, you could end up a little on the short side if you run into the scenario where the outerplates don't line up. And if it's too long, you could end up not being able to adjust the whole drive train optimally. There's bascially three scenarios with Shimano now... the case where the outer plates line up, in which case you add two links (use the same terminology as the manufacturers please, not "half links", as you can't add "half a link" to a chain). And the case where the outerplates don't line up, in which case Shimano gives you the option of adding 1 or 3 links. I'm saying don't ever use just the one link or you end up in the short chain scenario as I pictured above. So, it boils down to adding 2 or 3 links, depending on how the chain lines up when you wrap it around big/big.

The other thing is a matter of practicality. When working on a bike, most mechanics are going to install the crank, and other components and when ready, the chain. The most expedient and convenient method is going to be to just wrap the chain around and knowingly cut the chain where you know it needs to be cut, according to the guidelines above. As opposed to measuring the chainstays then running everything through an equation. Some bike mechanics don't like math and equations.

So, you can continue to run your equation method and hope it never lands you in the scenario of "it works, but it's kinda short", or you can be sure and just physically do it, mark it, cut it and join it. Plus, I still have my doubts as to just how many of the new Shimano systems, if any, you've actually installed from scratch and have experience with. Perhaps you've just not yet run into the scenario where things don't end up quite as you were hoping they would (the example I presented above in the pictures is exactly such a case). And I know you've said your drivetrain is quite noisy before. I said that's a shame, you should fix it. Perhaps you might want to start by questioning if you really do have an optimal chainlength on your bike, or just start over and get a less noisy drivetrain.

You've made your point, I rebutted it and explained why and how I think people can ensure they're getting the right length first time, every time. It's not the same today as it was even two years ago.

Oh, and for the people asking about SRAM... I have no idea if it's the same, probably, maybe, maybe not... but Tobin is clearly the SRAM expert. Perhaps he can start his own "how to" thread for SRAM. Gentlemen, start your calulators, and GTS!
Last edited by Calnago on Thu Jun 28, 2018 2:45 am, edited 2 times in total.
Colnago C64 - The Naked Build; Colnago C60 - PR99; Trek Koppenberg - Where Emonda and Domane Meet;
Unlinked Builds (searchable): Colnago C59 - 5 Years Later; Trek Emonda SL Campagnolo SR; Special Colnago EPQ

User avatar
Kayrehn
Posts: 1776
Joined: Thu Mar 24, 2011 6:06 pm

by Kayrehn

https://youtu.be/Vg2SoOOrpE8 this is good because they mentioned the different ways that you can go about it.

Sent from my SM-G930F using Tapatalk


by Weenie


Visit starbike.com Online Retailer for HighEnd cycling components
Great Prices ✓    Broad Selection ✓    Worldwide Delivery ✓

www.starbike.com



User avatar
Calnago
In Memoriam
Posts: 8612
Joined: Sun Nov 07, 2010 9:14 pm

by Calnago

Yes, the GCN video I've seen, but it doesn't say much about the 1 or 3 link option that Shimano recommends, and it's a little simplistic given what today's newest derailleurs have thrown at us. They talk about the small/small method as if it's interchangeble now, and back then it was. As I pointed out in this thread, using the small/small method with the newest Shimano stuff could leave you in a bit of frustration and buying a new chain. Plus, it was done before the new derailleurs were released... which is when things became just a bit more confusing. The Park website really has the best videos I've seen on this and they work. They also talk a bit more about mountain bike setups and pivoting suspension systems as well. It's very good. They also have the equation method stuffed in there as well.
Colnago C64 - The Naked Build; Colnago C60 - PR99; Trek Koppenberg - Where Emonda and Domane Meet;
Unlinked Builds (searchable): Colnago C59 - 5 Years Later; Trek Emonda SL Campagnolo SR; Special Colnago EPQ

Post Reply