Page 1 of 1

Chain Length & Rear Derailer Angle

Posted: Tue Oct 21, 2008 7:48 pm
by plasmapuff
Hey guys,

Apologies in advance for the flood of newbie technical questions but this was brought to my attention on the Bike gallery thread.

This is my new bike...
Image

Though apparently the chain is too short as the RD angle is pulled too far forward. I've double checked the bike since then and the RD appears only to be pulled this far forward on the larger chain ring, when on the big cogs Normally it does look instead like this on the small chain ring:

Image

This can
cause a greater wear on the jockey wheels, cassette and chain also greater stress on the mechanism itself as it is already nearly fully extended. "
It is a new 105 long cage derailer and funnily enough the LBS actually adjusted all my gears as I was having trouble with the FD...but they didn't say anything about it. Currently gears are running very smoothly, though I am concerned if there is going to be increased wear.

I sized the chain according to the Largest Cog and Largest Chainring Method on the Park Tools Website

So I guess the question is:
1) Is this a problem even if it happens only on some of the gears?

2) What can I do to fix it? Won't putting links back into the chain damage it by deforming the pins? (The LBS told me to use the chain breaker sparingly, thus the reason for the powerlink in the first place)

3) Why didn't the LBS pick this up when they were adjusting the gears? :twisted:

Thanks a heap in advance guys :D

Julian :)

Posted: Tue Oct 21, 2008 8:43 pm
by Kalgrm
G'day Julian,

If the gears are changing okay from and to all gears, don't worry about it. Yeah, it might be straight, but if there was a problem, you'd know about it by now. When the chain is too tight, you have extreme difficulty (maybe even impossibility) changing out of the gear you've shown in the main photo.

Having said that, you probably shouldn't ride in the gear you've shown very often. It's best to drop down to the smaller chain ring somewhere near the middle of the range of rear gears as you wipe speed off. In that way, you're never faced with having to ride "cross-chained", which can wear out chain, chain rings and rear gears due to increased lateral bending.

Cheers,
Graeme

Posted: Tue Oct 21, 2008 8:51 pm
by sittingbison
Hi Julian,
I did a bit of a response in the other thread. Does this rectify when you DON'T use the extreme gearing (either way)?

edit just had a thought, so feel free to think I'm a dill :roll: but is the crank "compact"? If so will the longer arm derailleur malfunction?

Posted: Tue Oct 21, 2008 9:25 pm
by twizzle
Jockey wheels should be vertical when the chain is on the big chain ring and smallest cog. If it vertical on the small chain ring then the chain is too short.

Edit : Shimano doc here. This is the 105 doc... but they are basically all the same.

Posted: Tue Oct 21, 2008 9:57 pm
by Kalgrm
Twizzle,

I reckon compact gearing and a long cage derailleur might change that equation. (Edit: ignore me. I can't find any evidence to back it up)

In MTB circles, the derailleur looks similar to how Julian has his RD set up when the chain is on the big chain ring and the largest rear cog. It's this combination which sets the chain length - any shorter and the chain can get stuck in that gear or break. Any longer and the RD can't take up all the slack in the smallest chain ring and the smallest rear cog (not that you should ever be in that combination ...)

Cheers,
Graeme

Posted: Tue Oct 21, 2008 10:10 pm
by twizzle
Shimano specs the chain length based on the derailleur, not the crankset. FWIW, I'm running a 105 triple rear on a 12-27 with 26-36-48 rings, and it's all within spec for shifting. IMO - that chain is far too short.

Posted: Tue Oct 21, 2008 10:17 pm
by plasmapuff
Thanks a heap once again guys. 8) Learning at a rapid rate of knots.
sittingbison wrote:This could be because the gearing is big chainring/big cluster cog, which is not the gear of choice. I usually use gearing such that the big half of the cluster is on the smaller chainring, and the smaller half of the cluster is on the big chainring ie 17-23/39 and 11-15/53 to avoid stretching the chain and derailleur.
Yep Sittingbison, Kalgrm and 32spokes have got it right. I had the bike cross chained when I took the shot. Gears shift flawlessly at the moment (as you would hope with a new drive chain).

I'll definitely keep this in mind in the future and change my habits. Thanks for letting me know about this early- I did a large part of the 100km ride on Sunday in this big-big combo, so hopefully haven't cause too much wear :D.

(AT) Sittingbison: Nah just a standard crank- 39/52 I believe. Cassette is the standard 12-25. Either I'm very strong (don't think this is the case) or I'm very light (55kg) as I found I was climbing hills very comfortably sitting down while doing ~90rpm on the small chain ring.

cheers,
Julian

Posted: Tue Oct 21, 2008 10:26 pm
by thirtytwospokes
twizzle wrote:Shimano specs the chain length based on the derailleur, not the crankset. FWIW, I'm running a 105 triple rear on a 12-27 with 26-36-48 rings, and it's all within spec for shifting. IMO - that chain is far too short.
Have you got a link to the Shimano chain length specs, or can you point me to where I might find them? I would've thought any chain length calculation/specification would have to take in to account the size of the largest chainring, largest cog, and length of the chainstays.

In any case, plasmapuff's picture of his bike appears to show the chain bending through each of the pulley wheels, which suggests the chain length is adequate.

Posted: Tue Oct 21, 2008 11:31 pm
by RobS
There's a few methods to determine chain length.

1) Wrap the chain around the largest front and largest rear sprockets. Find the shortest chain length that *will* (not nearly) allow them to connect. Add one full link.

The older Shimano docs implied this method but it wasn't clear.

2) Use a chain length calculator to calculate the chain length base on large front and large rear, round the value upwards. Add one full link. The calculation requires the chainstay length. This method is identical to (1) except it uses calculation. (but it doesn't allow for the very small increase in chain length due to cross-chaining)

In general you should use the complicated chain length formulas, some of the simplified formulas give slightly different (usually smaller) results.

3) Later Shimano docs. Place the chain on the largest front and smallest rear sprockets, routing the chain through the derailleur. The bolts for the derailleur wheels should be aligned vertically (ie. 90deg to the ground).

shimano_105

I checked a number of commercially assembled bikes using method (2) and they generally end up being with +/-1 full link from the calculated value (calculation includes the +1).

Some people say running one link shorter changes a little better. I'm trying this experiment on my MTB and it appears to change fine despite the fact the derailleur is buggered, it's possible the slight increase in tension helps holdback play in the derailleur.

Posted: Tue Oct 21, 2008 11:44 pm
by plasmapuff
1) Wrap the chain around the largest front and largest rear sprockets. Find the shortest chain length that *will* (not nearly) allow them to connect. Add one full link.
Thats what I did as per the Park Tools website...also did their calculation method as well to double check and they were both the same.

Posted: Tue Oct 21, 2008 11:56 pm
by RobS
> hats what I did as per the Park Tools website

Honestly I don't think you have anything to worry about. If you can get the gears in the large front/large back configuration it usuall means the chain isn't too short.

If you added a link the chances are it could go back the other way. One link is 1 inch so you can only control the angle in finite steps. It's only a guide.

Posted: Wed Oct 22, 2008 9:51 am
by twizzle
thirtytwospokes wrote:
twizzle wrote:Shimano specs the chain length based on the derailleur, not the crankset. FWIW, I'm running a 105 triple rear on a 12-27 with 26-36-48 rings, and it's all within spec for shifting. IMO - that chain is far too short.
Have you got a link to the Shimano chain length specs, or can you point me to where I might find them? I would've thought any chain length calculation/specification would have to take in to account the size of the largest chainring, largest cog, and length of the chainstays.

In any case, plasmapuff's picture of his bike appears to show the chain bending through each of the pulley wheels, which suggests the chain length is adequate.
Shimano Technical Documents. This has all of the usual crap you get when you buy some parts.