Gear set up problem

[Graham]

A friend of mine has just built up a new bike with what I believe to be a bit of a mismash of components and cannot get his gears set up properly. The components are as follows:

Shifters - Shimano dura ace 10 speed STI, front mech - Shimano dura ace, rear mech - Shimano XTR M971, cassette - SRAM 10 speed 11-32 , chain - KMC 10 speed and chainset - FSA compact 50/34.

He assures me that he has checked on a number of forums and with suppliers and apparently this lot is compatible. Can anyone confirm or deny this.

I have tried to set up the gears and whilst the front shifts perfectly I can only get accepable shifting on either the big ring or the small ring but not both. If I get acceptable shifting on the small ring then the chain is reluctant to come off the largest sprocket as if the cable were too tight. Slackening the cable by turning the barrel adjuster 2-3 clicks regains acceptable shifting on the big ring. Now if I shift to the small ring the chain tries to skip off the smaller sprockets to the one below indicating the cable is too slack. Reversing the clicks on the barrel adjuster gets acceptable shifting back on the small chain ring but it is now off again on the big ring as before.

So far I have checked the chain length and in the big/big combination directly i.e. not passing through the rear mech it is the classical overlap plus one link so that looks OK. By eye the hanger looks vertical and in line but I cannot be perfectly sure. There is no slack in the cable after the last change to the smallest sprocket and if I disconnect the cable from the mech and operate the STI whilst holding the cable there does not seem to be any appreciable resistance of the inner cable in the outer.

So Assuming this lot is actually compatible has anyone any ideas???

Thanks.

Graham.

[AMuzi]

Sounds like minor if frustrating setup issues.

Ensure rear changer is vertical, not bent or damaged.

Check your top pivot ("B") screw adjustment.
Ensure high and low limit screws are correct, especially low
gear with a firm thumb push on the changer body,
When you have all 3 set properly you should get good
shifting in and out of low gear.

Then connect the gear wire and set as usual for index response.

--
Andrew Muzi
<www.yellowjersey.org/>
Open every day since 1 April, 1971

[Sir Ridesalot]

Andrew, are the Shimano dura ace 10 speed STI brifters compatible with a Shimano XTR M971 rear mech?

Cheers

[AMuzi]

Yes. Santanas for example ship with XTR and STi Ten, 36t
cassette just as you describe.

Official statement:
http://techdocs.shimano.com/media/techdocs/content/cycle/SI/XTR/RD-M971/SI-5VZ0A-001-En_v1_m56577569830607913.pdf

[thirty-six]

type GS has a total capacity of 37 teeth, one can generally push the
large sprocket size but not the total capacity as this is set by the
cage length. If it's the SGS then you should be able to make it
work , but make a positive identification before further testing,
adjusting and bashing it about.

[Graham]

"AMuzi" <a...@yellowjersey.org> wrote in message news:k4a8hm$lib$1...@dont-email.me...

> Sounds like minor if frustrating setup issues.

> Ensure rear changer is vertical, not bent or damaged.

Thanks Andrew for the reply and your subsequent reply to another poster confirming the XTR should work with the STIs and providing a link to the Shimano tech doc.

After thoroughly going through the setup once again with no luck I finally removed the hanger from the drop out and discovered some paint on part of the mating face between the drop out and the hanger. Shoddy manufacure I guess. This is a relatively cheap carbon frame my friend imported from China. Having cleaned the surface and made sure it was as flat as possible I replaced the hanger after which I got better but not perfect shifting performance. Looking from the back of the bike with it upside down the jockey wheels look to be vertical but slightly toe out relative to the sprockets. I am not sure if that would make a difference but given it is not my bike and a carbon frame I am not up for trying to line things up perfectly. I will leave that to one of our LBSs if my friend cannot live with the shifting performance I have achieved - shifts into all gears but with slow upshifting on the big ring from the very lowest gears (tight cable?) and slightly noisy running in the higher gears (chain Just rubbing on the smaller cog - slack cable?) on the small ring. All intermediate combinations fine.

From the Shimano tech doc you referenced I noted that it recommended that this mech should not be used on a bike with internally routed cables. Unfortunately my friend's bike has internally routed cables. Disconnecting the cable from the mech and working the STI seems to indicate the cable is moving very freely but I accept this is effectively a no load test. When under tension from the mech spring things could be different. Could this be a cause of some of the residual gear set up problem?

Thanks again,

Graham.

[thirty-six]

On 1 Oct, 15:54, "Graham" <h2gt2g42-micenewgro...@yahoo.co.uk> wrote:
> "AMuzi" <a...@yellowjersey.org> wrote in messagenews:k4a8hm$lib$1...@dont-email.me...

Ah well, there's none so blind as those who don't want to see.

perhaps someone else might like to help to point out there is a GS
(long arm) and an SGS (longer than long) pulley-cage and only the sgs
cage is stated as suitable by the manufacturer for the gear range
Graham has quoted..

[Graham]

"thirty-six" <thirt...@live.co.uk> wrote in message news:30c5df42-c4b1-4012...@hj1g2000vbb.googlegroups.com...

On 1 Oct, 15:54, "Graham" <h2gt2g42-micenewgro...@yahoo.co.uk> wrote:
> "AMuzi" <a...@yellowjersey.org> wrote in messagenews:k4a8hm$lib$1...@dont-email.me...

Ah well, there's none so blind as those who don't want to see.

perhaps someone else might like to help to point out there is a GS
(long arm) and an SGS (longer than long) pulley-cage and only the sgs
cage is stated as suitable by the manufacturer for the gear range
Graham has quoted..

What I quoted was:

Shifters - Shimano dura ace 10 speed STI, front mech - Shimano dura ace, rear mech - Shimano XTR M971, cassette - SRAM 10 speed 11-32 , chain - KMC 10 speed and chainset - FSA compact 50/34.

What you advised was:

type GS has a total capacity of 37 teeth, one can generally push the
large sprocket size but not the total capacity as this is set by the
cage length. If it's the SGS then you should be able to make it
work , but make a positive identification before further testing,
adjusting and bashing it about.

So by my reckoning (50-34)+(32-11)= 37 total capacity. Just at the limit of the GS. So armed with your advice I did not think it necessary to worry about which model it was. Having now studied carefully the tech doc Andrew posted the link to it actually says 34 for the GS limit so I guess your 37 was a typo which I guess you did not "see".

My experience with road mechs indicates Shimano are very conservative over their capacities. I have the same chain ring and sprocket set up working smoothly with an Ultegra short cage with the B screw wound fully in. The cage on my friends XTR is way longer than the Ultegra and it was no where near the limits of travel on either big/big or small/small combinations.

Anyway just in case others advise that the capacity of Shimano mountain bike mechs is more critical than their road mechs and would have the effect on shifting performance I have described I have emailed my friend asking him to confirm which model he has. He has already taken his bike home and was happy with its shifting performance although after riding it for a while he might be back!

Cheers,

Graham.

[thirty-six]

On 1 Oct, 18:59, "Graham" <h2gt2g42-micenewgro...@yahoo.co.uk> wrote:
> "thirty-six" <thirty-...@live.co.uk> wrote in messagenews:30c5df42-c4b1-4012...@hj1g2000vbb.googlegroups.com...

>
> On 1 Oct, 15:54, "Graham" <h2gt2g42-micenewgro...@yahoo.co.uk> wrote:
>
> > "AMuzi" <a...@yellowjersey.org> wrote in messagenews:k4a8hm$lib$1...@dont-email.me...
>
> Ah well, there's none so blind as those who don't want to see.
>
> perhaps someone else might like to help to point out there is a GS
> (long arm) and an SGS (longer than long)  pulley-cage and only the sgs
> cage is stated as suitable by the manufacturer for the gear range
> Graham has quoted..
>
> What I quoted was:
>
>  Shifters - Shimano dura ace 10 speed STI, front mech - Shimano dura ace, rear mech - Shimano XTR M971, cassette - SRAM 10 speed 11-32 , chain - KMC 10 speed and chainset - FSA compact 50/34.
>
> What you advised was:
>
> type GS has a total capacity of 37 teeth, one can generally push the
> large sprocket size but not the total capacity as this is set by the
> cage length.  If it's the SGS then you should be able to make it
> work , but make a positive identification before further testing,
> adjusting and bashing it about.
>
> So by my reckoning (50-34)+(32-11)= 37 total capacity. Just at the limit of the GS. So armed with your advice I did not think it necessary to worry about which model it was. Having now studied carefully the tech doc Andrew posted the link to it actually says 34 for the GS limit so I guess your 37 was a typo which I guess you did not "see".

I made that mistake, an unfortunate rework on my intended post led to
some inappropriate editing with transposed value.

>
> My experience with road mechs indicates Shimano are very conservative over their capacities. I have the same chain ring and sprocket set up working smoothly with an Ultegra short cage with the B screw wound fully in. The cage on my friends XTR is way longer than the Ultegra and it was no where near the limits of travel on either big/big or small/small combinations.
>
> Anyway just in case others advise that the capacity of Shimano mountain bike mechs is more critical than their road mechs and would have the effect on shifting performance I have described I have emailed my friend asking him to confirm which model he has. He has already taken his bike home and was happy with its shifting performance although after riding it for a while he might be back!
>

One can push the largest sprocket size but AFAIK one cannot push the
total-capacity and any derailleur will offer improved performance if
this capacity is shied away from.

[Graham]

"thirty-six" <thirt...@live.co.uk> wrote in message news:bb0f957d-e31b-4a49...@g18g2000vbf.googlegroups.com...

As I have stated above the capacity point is not true from my own experience with Shimano road mechs. That aside my friend has just confirmed that his XTR is indeed the SGS long version which helps to explain why it had no difficulty in handling the big/big and small/small combinations. That being the case I think we can now rule out the cage length as a contributor to the shifting problem (except possibly too long and floppy!) as the spec for the SGS is:

Total capacity: 43 teeth, Largest sprocket: 34 teeth, Smallest sprocket: 11 teeth
Front chainwheel tooth difference: 22 teeth.

My friend's set up lies comfortably within that spec.

Graham.

[thirty-six]

On 1 Oct, 21:41, "Graham" <h2gt2g42-micenewgro...@yahoo.co.uk> wrote:
> "thirty-six" <thirty-...@live.co.uk> wrote in messagenews:bb0f957d-e31b-4a49...@g18g2000vbf.googlegroups.com...

The new chain will be stiff in its original grease and will need a
little light oil to free it up so that shifting is as expected from
cold.. A long arm sloppy cage will likely be even more sensitive to
this. Tell him to lightly oil the chain and go on a 10 mile ride
before bothering with the shifting and then see how it does. If the
chain is still stiff then a second oiling should free it up.

[John B.]

You can usually test cable binding on a stand by shifting rapidly from
one extreme, say low to low gear and then trying to move the derailer
further. If you can then check the cables.

--
Cheers,
John B.

[Graham]

"John B." <johnbs...@gmail.com> wrote in message news:tvck6896g72grhqvo...@4ax.com...
[Snip]

>> I will leave that to one of our LBSs if my friend cannot live with the shifting performance I have achieved - shifts into all gears but with slow upshifting on the big ring from the very lowest gears (tight cable?) and slightly noisy running in the higher gears (chain Just rubbing on the smaller cog - slack cable?) on the small ring. All intermediate combinations fine.

> You can usually test cable binding on a stand by shifting rapidly from
> one extreme, say low to low gear and then trying to move the derailer
> further. If you can then check the cables.

John thanks for your contribution. I have tried similar tests but they were inconclusive. I have never worked with one of these very long mountain bike cages before and to me they seem so floppy that its hard to decern whether there is a genuine very small movement in the parallelogram or the whole thing flexing. To get the effect I describe above with a 10 speed cassette we are only looking for tiny movements.

Given where I started:

"> I have tried to set up the gears and whilst the front shifts perfectly I can only get accepable shifting on either the big ring or the small ring but not both. If I get acceptable shifting on the small ring then the chain is reluctant to come off the largest sprocket as if the cable were too tight. Slackening the cable by turning the barrel adjuster 2-3 clicks regains acceptable shifting on the big ring. Now if I shift to the small ring the chain tries to skip off the smaller sprockets to the one below indicating the cable is too slack. Reversing the clicks on the barrel adjuster gets acceptable shifting back on the small chain ring but it is now off again on the big ring as before."

Then the problem seems to be related to forward rotation of the parallelogram and the backward rotation of the cage as it takes up the chain slack going from big/big to small/small. If those rotations are not in the exact plane of the sprockets I could see how the movement of the chain relative to the sprockets for each click of the chain could vary as a function of those rotations. Given that I achieved a partial solution by removing the hanger and then replacing it after cleaning the mating surface of the dropout I am still leaning towards a hanger alignment problem however slight.

I think other than curiosity and the dislike of been defeated by a bike technical problem I will send my friend to a LBS if he is not happy with the shifting. I could make a gauge/tool to check and correct it but I am unwilling to start levering a hanger in a carbon frame.

So thanks everyone for your contributions. I will report on any further developments.

Cheers,

Graham.

[Graham]

"thirty-six" <thirt...@live.co.uk> wrote in message news:6dec6ad7-5ed5-4a9b...@l18g2000vbv.googlegroups.com...

[Snip]

> The new chain will be stiff in its original grease and will need a
little light oil to free it up so that shifting is as expected from
cold.. A long arm sloppy cage will likely be even more sensitive to
this. Tell him to lightly oil the chain and go on a 10 mile ride
before bothering with the shifting and then see how it does. If the
chain is still stiff then a second oiling should free it up.

Thanks for that. I have never had such a problem with a road mech but these long mountain bike mechs are alien to me. However, as your potential solution is simple and something my friend can easily cope with on his own I will definitely pass it.

I would be great if it was the solution but if you read my response to John I am still suspecting a residual very slight misalignment of the hanger but have admitted defeat on this one given that I am not prepared to start getting physical with my friends carbon pride and joy.

Thanks again for your contributions.

Cheers,

Graham.

[John B.]

On Tue, 2 Oct 2012 09:17:50 +0100, "Graham"
<h2gt2g42-mi...@yahoo.co.uk> wrote:

>
>"John B." <johnbs...@gmail.com> wrote in message news:tvck6896g72grhqvo...@4ax.com...
>[Snip]
>
>>> I will leave that to one of our LBSs if my friend cannot live with the shifting performance I have achieved - shifts into all gears but with slow upshifting on the big ring from the very lowest gears (tight cable?) and slightly noisy running in the higher gears (chain Just rubbing on the smaller cog - slack cable?) on the small ring. All intermediate combinations fine.
>
>> You can usually test cable binding on a stand by shifting rapidly from
>> one extreme, say low to low gear and then trying to move the derailer
>> further. If you can then check the cables.
>
>John thanks for your contribution. I have tried similar tests but they were inconclusive. I have never worked with one of these very long mountain bike cages before and to me they seem so floppy that its hard to decern whether there is a genuine very small movement in the parallelogram or the whole thing flexing. To get the effect I describe above with a 10 speed cassette we are only looking for tiny movements.
>
>Given where I started:
>
>"> I have tried to set up the gears and whilst the front shifts perfectly I can only get accepable shifting on either the big ring or the small ring but not both. If I get acceptable shifting on the small ring then the chain is reluctant to come off the largest sprocket as if the cable were too tight. Slackening the cable by turning the barrel adjuster 2-3 clicks regains acceptable shifting on the big ring. Now if I shift to the small ring the chain tries to skip off the smaller sprockets to the one below indicating the cable is too slack. Reversing the clicks on the barrel adjuster gets acceptable shifting back on the small chain ring but it is now off again on the big ring as before."

I've always thought that the idler pulley (at the end of the arm) had
very little to do with shifting and was there mainly to take up chain
slack and that the top pulley was what actually shifted the chain.

( But who knows. After all there isn't any rope on a bike to run over
the pulleys and no sprockets for the chain to drive :-)

>
>Then the problem seems to be related to forward rotation of the parallelogram and the backward rotation of the cage as it takes up the chain slack going from big/big to small/small. If those rotations are not in the exact plane of the sprockets I could see how the movement of the chain relative to the sprockets for each click of the chain could vary as a function of those rotations. Given that I achieved a partial solution by removing the hanger and then replacing it after cleaning the mating surface of the dropout I am still leaning towards a hanger alignment problem however slight.
>

Sort of far out, but you did check the chain line? Not measured from
the seat tube but with a straight edge from the center cog on the
cassette to the center chain wheel on a triple crank or to the space
between the two on a double.

And even further out, is the rear shift cable run the right way at its
attachment to the derailer? I seem to remember that some Shimano
shifters could change their ratio slightly by rerouting the cable.
Muzi or Cholo can probably provide some specifics here.

>I think other than curiosity and the dislike of been defeated by a bike technical problem I will send my friend to a LBS if he is not happy with the shifting. I could make a gauge/tool to check and correct it but I am unwilling to start levering a hanger in a carbon frame.
>
>So thanks everyone for your contributions. I will report on any further developments.
>
>Cheers,
>
>Graham.

--
Cheers,
John B.

[thirty-six]

On 2 Oct, 09:25, "Graham" <h2gt2g42-micenewgro...@yahoo.co.uk> wrote:
> "thirty-six" <thirty-...@live.co.uk> wrote in messagenews:6dec6ad7-5ed5-4a9b...@l18g2000vbv.googlegroups.com...

Even once the chain has loosened up, the derailleur may still require
adjustment for best performance but there should be an obvious
improvement without tinkering.

[Lou Holtman]

Op zondag 30 september 2012 20:52:45 UTC+2 schreef Graham het volgende:

> A friend of mine has just built up a new bike with what I believe to be a bit of a mismash of components and cannot get his gears set up properly. The components are as follows: Shifters - Shimano dura ace 10 speed STI, front mech - Shimano dura ace, rear mech - Shimano XTR M971, cassette - SRAM 10 speed 11-32 , chain - KMC 10 speed and chainset - FSA compact 50/34. He assures me that he has checked on a number of forums and with suppliers and apparently this lot is compatible. Can anyone confirm or deny this. I have tried to set up the gears and whilst the front shifts perfectly I can only get accepable shifting on either the big ring or the small ring but not both. If I get acceptable shifting on the small ring then the chain is reluctant to come off the largest sprocket as if the cable were too tight. Slackening the cable by turning the barrel adjuster 2-3 clicks regains acceptable shifting on the big ring. Now if I shift to the small ring the chain tries to skip off the smaller sprockets to the one below indicating the cable is too slack. Reversing the clicks on the barrel adjuster gets acceptable shifting back on the small chain ring but it is now off again on the big ring as before. So far I have checked the chain length and in the big/big combination directly i.e. not passing through the rear mech it is the classical overlap plus one link so that looks OK. By eye the hanger looks vertical and in line but I cannot be perfectly sure. There is no slack in the cable after the last change to the smallest sprocket and if I disconnect the cable from the mech and operate the STI whilst holding the cable there does not seem to be any appreciable resistance of the inner cable in the outer. So Assuming this lot is actually compatible has anyone any ideas??? Thanks. Graham.

Are 10 speed DA shifters compatible with ATB 10 speed RD? From what I heard you have to match road shifters with road RD or ATB shifters with ATB RD, to shift 10 speed cassettes; road or ATB.

Lou

[Graham]

"Lou Holtman" <lou.h...@gmail.com> wrote in message news:1f8bb762-4100-4786...@googlegroups.com...

Op zondag 30 september 2012 20:52:45 UTC+2 schreef Graham het volgende:

[Snip]

> Are 10 speed DA shifters compatible with ATB 10 speed RD? From what I heard you have to match road shifters with road RD or ATB shifters with ATB RD, to shift 10 speed cassettes; road or ATB.

It turns out that the XTR M971 which is in fact a 9 speed mech is compatible with a DA 10 speed shifter. See Andrew Muzi's earlier reply to Sir Ridesalot yesterday. That is supported by my own research on the internet. As you surmised above the latest Shimano 10 speed ATB mechs are not compatible with their 10 speed road shifters.

Cheers,

Graham.

[Graham]

"John B." <johnbs...@gmail.com> wrote in message news:iigl68pas0ongi8q5...@4ax.com...

> On Tue, 2 Oct 2012 09:17:50 +0100, "Graham"
> <h2gt2g42-mi...@yahoo.co.uk> wrote:
>
>>
>>"John B." <johnbs...@gmail.com> wrote in message news:tvck6896g72grhqvo...@4ax.com...
>>[Snip]

> I've always thought that the idler pulley (at the end of the arm) had
> very little to do with shifting and was there mainly to take up chain
> slack and that the top pulley was what actually shifted the chain.

Agreed. But the tracking of that top pulley with repect to the sprockets is one thing I can see that would effect shifting with a mis-aligned hanger. Like you I am not sure what effect if any the bottom one would have in such circumstances. I am aware that I appear to be arguing in circles here. A sure sign of incomplete understanding!!!

> Sort of far out, but you did check the chain line?

Checked as you described and fine.

> And even further out, is the rear shift cable run the right way at its
> attachment to the derailer?

Yes. Checked against the tech doc Andrew Muzi provided a link to.

I have just got back from a quick 30 mile ride with my friend to road test his bike. We covered most conditions from steep climbing through rolling country to fast flat riding with a few sprints thrown in. Riding behind him and watching the changes I could not see anything untoward. The cage appeared vertical to the eye and seemed to be faithfully tracking the sprockets but I guess the tollerances on a 10 speed set up are so tight I should not expect to see other than significant misalignment.

The shifting remained the same throughout - a bit slow up shifting from the lowest gears (sprockets 1,2,3 - 1 being the largest) on the big ring and very slight overshifting in the highest gears (sprockets 7,8,9,10) on the small ring resulting in them chattering but not skipping. All other changes crisp followed by silent running. My friend pronounced he was OK with this and said I was being a bit of a perfectionist. Well as an enginner by training and still at heart I expect expensive high end kit to work perfectly and I get frustrated when I cannot set it up to do so.

The upshot of it all is I have suggested that he go to the place where he bought his XTR and ask them to check the alignment of his hanger and fix it if it is out. At least that will get that off my mind as more and more I am finding it difficult to see past that being the cause.

Thanks again for you continued interest in my problem.

Cheers,

Graham.

[Graham]

"thirty-six" <thirt...@live.co.uk> wrote in message news:df0c7630-8c7f-47bc...@r10g2000vby.googlegroups.com...

On 2 Oct, 09:25, "Graham" <h2gt2g42-micenewgro...@yahoo.co.uk> wrote:
> "thirty-six" <thirty-...@live.co.uk> wrote in messagenews:6dec6ad7-5ed5-4a9b...@l18g2000vbv.googlegroups.com...
>
> [Snip]

> Even once the chain has loosened up, the derailleur may still require
adjustment for best performance but there should be an obvious
improvement without tinkering.

See my response to John B. My friend oiled his chain before we set off on the test ride and there was no noticeable change in shifting performance after 30 miles of riding. We will monitor developments as we cover more miles over the next few days.

Thanks again,

Graham.

[Graham]

"John B." <johnbs...@gmail.com> wrote in message news:iigl68pas0ongi8q5...@4ax.com...

> On Tue, 2 Oct 2012 09:17:50 +0100, "Graham"
> <h2gt2g42-mi...@yahoo.co.uk> wrote:
>
>>
>>"John B." <johnbs...@gmail.com> wrote in message news:tvck6896g72grhqvo...@4ax.com...
>>[Snip]

> I've always thought that the idler pulley (at the end of the arm) had
> very little to do with shifting and was there mainly to take up chain
> slack and that the top pulley was what actually shifted the chain.

Agreed. But the tracking of that top pulley with repect to the sprockets is one thing I can see that would effect shifting with a mis-aligned hanger. Like you I am not sure what effect if any the bottom one would have in such circumstances. I am aware that I appear to be arguing in circles here. A sure sign of incomplete understanding!!!

> Sort of far out, but you did check the chain line?

Checked as you described and fine.

> And even further out, is the rear shift cable run the right way at its
> attachment to the derailer?

[thirty-six]

On 2 Oct, 15:38, "Graham" <h2gt2g42-micenewgro...@yahoo.co.uk> wrote:
> "John B." <johnbsloc...@gmail.com> wrote in messagenews:iigl68pas0ongi8q5...@4ax.com...

> > On Tue, 2 Oct 2012 09:17:50 +0100, "Graham"

> > <h2gt2g42-micenewgro...@yahoo.co.uk> wrote:
>
> >>"John B." <johnbsloc...@gmail.com> wrote in messagenews:tvck6896g72grhqvo...@4ax.com...

> >>[Snip]
> > I've always thought that the idler pulley (at the end of the arm) had
> > very little to do with shifting and was there mainly to take up chain
> > slack and that the top pulley was what actually shifted the chain.
>
> Agreed. But the tracking of that top pulley with repect to the sprockets is one thing I can see that would effect shifting with a mis-aligned hanger. Like you I am not sure what effect if any the bottom one would have in such circumstances. I am aware that I appear to be arguing in circles here. A sure sign of incomplete understanding!!!
>
> > Sort of far out, but you did check the chain line?
>
> Checked as you described and fine.
>
> > And even further out, is the rear shift cable run the right way at its
> > attachment to the derailer?
>
> Yes. Checked against the tech doc Andrew Muzi provided a link to.

Just a thought, all outer gear casing is long-lay casing trimmed
square and there is no movement at it's ends when the cable tension in
varied.

>
> I have just got back from a quick 30 mile ride with my friend to road test his bike. We covered most conditions from steep climbing through rolling country to fast flat riding with a few sprints thrown in. Riding behind him and watching the changes I could not see anything untoward. The cage appeared vertical to the eye

That's good enough.

> and seemed to be faithfully tracking the sprockets but I guess the tollerances on a 10 speed set up are so tight I should not expect to see other than significant misalignment.
>
> The shifting remained the same throughout - a bit slow up shifting from the lowest gears (sprockets 1,2,3 - 1 being the largest) on the big ring

So on a wide-spaced triple you are expecting perfect shifting in the
biggest cogs? I feel you are expecting too much from a derailleur
system. If you want perfect shifting use a half-step setup with a
Nouvo Record derailleur.

>and very slight overshifting in the highest gears (sprockets 7,8,9,10) on the small ring resulting in them chattering but not skipping.

Again I feel your expectations are perhaps too high for such a wide
chainring spacing which demands the use of an extra long pulley cage.

>All other changes crisp followed by silent running. My friend pronounced he was OK with this and said I was being a bit of a perfectionist. Well as an enginner by training and still at heart I expect expensive high end kit to work perfectly and I get frustrated when I cannot set it up to do so.

The rattling is a feature to remind the rider that such a combination
does not result in an efficient chainline. :-) It should be
documented as such.

>
> The upshot of it all is I have suggested that he go to the place where he bought his XTR and ask them to check the alignment of his hanger and fix it if it is out. At least that will get that off my mind as more and more I am finding it difficult to see past that being the cause.

If the jockey-wheel is cocked relative to the cogs then it may cause
some problem but it is the horizontal alignment which is relevant,
vertical alignment can be out by 15deg without affecting performance,
and I don't see that the closeness of cogs making any difference in
the outcome of such a V-misalignment.

As well as the hanger alignment, the frame-end alignment needs
checking and it could be that an error in frame-build is being
responsible and the hanger is being twisted with the frame-end when
the axle is tightened.

[AMuzi]

I linked the Shimano diagram which includes gear wire
attachment in my first reply.

[Graham]

"thirty-six" <thirt...@live.co.uk> wrote in message news:4fdf8197-38b2-488a...@p5g2000vbl.googlegroups.com...

On 2 Oct, 15:38, "Graham" <h2gt2g42-micenewgro...@yahoo.co.uk> wrote:
> "John B." <johnbsloc...@gmail.com> wrote in messagenews:iigl68pas0ongi8q5...@4ax.com...
> > On Tue, 2 Oct 2012 09:17:50 +0100, "Graham"
> > <h2gt2g42-micenewgro...@yahoo.co.uk> wrote:
>
> >>"John B." <johnbsloc...@gmail.com> wrote in messagenews:tvck6896g72grhqvo...@4ax.com...
> >>[Snip]

> Just a thought, all outer gear casing is long-lay casing trimmed
square and there is no movement at it's ends when the cable tension in
varied.

Yes as far as I know and they all have ferules on the ends. The only place there could be movement you describe is in the barrel adjuster. The XTR is designed to work with cable adjusters on the shifters and therefore does not have an adjuster on the mech. My friend has used an inline barrel adjuster, one would normally put in downtube cable fixtures, in the mech. He cannot use it higher up in the system because he has internal cable ways and it is not practical to fit the adjusters in the holes at the top of the downtubes.

> So on a wide-spaced triple you are expecting perfect shifting in the
biggest cogs? I feel you are expecting too much from a derailleur
system.

This is not a triple it is a 50/34 compact double chainset. Its the 16 teeth difference up front plus the 21 tooth difference at the back(11-32) that makes him think he needs this mech. As I said in an earlier reply "I have the same chain ring and sprocket set up working smoothly with an Ultegra short cage with the B screw wound fully in" Contrary to what the specs say I very much doubt that he needs to go to the extreme of a long cage mountain bike mech to get this setup to work on his bike.

> I feel you are expecting too much from a derailleur system.

Not based on my own experiences with road mechs - mostly Ultegras. I have never had any problems getting crisp shifting and smooth running even with an extreme road setup like the one we are discussing. Now you could be right once you move into the domain of these long cage MTB mechs, I don't know I have no experience with them. Apart from its length this mech is a 9 speed running against a 10 speed cassette - whilst it might nominally be compatible with the 10 speed DA shifters we could be running up against the higher tolerance limits required during 10 speed shifting compared with 9 speed.

>The rattling is a feature to remind the rider that such a combination
does not result in an efficient chainline. :-)

I would agree for the two extreme sprockets but not for 4 sprockets in on a 10 speed cassette. I often switch front rings around the "middle" of the cassette in anticipation of a rapid change in gradient. I would certainly expect any setup to do this 4 sprockets in from either end without rattling thereafter.

> As well as the hanger alignment, the frame-end alignment needs
checking and it could be that an error in frame-build is being
responsible and the hanger is being twisted with the frame-end when
the axle is tightened.

Good point I will add this to his list just in case the guys who sold him the mech forget and just check the hanger.

Thanks again for your continued interest.

Graham.

[Lou Holtman]

Op 2-10-2012 16:00, Graham schreef:

Compatible or kind of compatible. If your are a perfectionist it is not
compatible, if not it might 'work'. Not too fond of this mix and match
thing.

Lou

[Graham]

"Lou Holtman" <lou.h...@usenet.nl> wrote in message news:dff9e$506b272d$541d34fb$28...@cache90.multikabel.net...

Well being a sort of perfectionist I tend to agree with you. Please remember this is my friend's set up I am fighting with not mine. I would not and indeed did not chose to go this way to achieve the same gearing.

Graham.

[thirty-six]

On 2 Oct, 18:37, "Graham" <h2gt2g42-micenewgro...@yahoo.co.uk> wrote:
> "thirty-six" <thirty-...@live.co.uk> wrote in messagenews:4fdf8197-38b2-488a...@p5g2000vbl.googlegroups.com...

>
> On 2 Oct, 15:38, "Graham" <h2gt2g42-micenewgro...@yahoo.co.uk> wrote:
>
> > "John B." <johnbsloc...@gmail.com> wrote in messagenews:iigl68pas0ongi8q5...@4ax.com...
> > > On Tue, 2 Oct 2012 09:17:50 +0100, "Graham"
> > > <h2gt2g42-micenewgro...@yahoo.co.uk> wrote:
>
> > >>"John B." <johnbsloc...@gmail.com> wrote in messagenews:tvck6896g72grhqvo...@4ax.com...
> > >>[Snip]
> > Just a thought, all outer gear casing is long-lay casing trimmed
>
> square and there is no movement at it's ends when the cable tension in
> varied.
>
> Yes as far as I know and they all have ferules on the ends. The only place there could be movement you describe is in the barrel adjuster. The XTR is designed to work with cable adjusters on the shifters and therefore does not have an adjuster on the mech. My friend has used an inline barrel adjuster, one would normally put in downtube cable fixtures, in the mech. He cannot use it higher up in the system because he has internal cable ways and it is not practical to fit the adjusters in the holes at the top of the downtubes.
>
> > So on a wide-spaced triple you are expecting perfect shifting in the
>
> biggest cogs?  I feel you are expecting too much from a derailleur
> system.
>
> This is not a triple it is a 50/34 compact double chainset.

Sorry, that was my brain kicking in and trying to understand. i
should stop such inappropriate activity.

> Its the 16 teeth difference up front plus the 21 tooth difference at the back(11-32) that makes him think he needs this mech. As I said in an earlier reply "I have the same chain ring and sprocket set up working smoothly with an Ultegra short cage with the B screw wound fully in" Contrary to what the specs say I very much doubt that he needs to go to the extreme of a long cage mountain bike mech to get this setup to work on his bike.

Personally I agree that a short cage should be used unless there is a
particular problem that the user is experiencing.

>
> > I feel you are expecting too much from a derailleur system.
>
> Not based on my own experiences with road mechs - mostly Ultegras. I have never had any problems getting crisp shifting and smooth running even with an extreme road setup like the one we are discussing. Now you could be right once you move into the domain of these long cage MTB mechs, I don't know I have no experience with them.

I used one temporarily long time ago and was dissatisfied, but I was
comparing it to a crisp-shifting Dura-ace EX on a 14-24 5-speed with
the top pivot spring disabled and on a single chain-ring.

> Apart from its length this mech is a 9 speed running against a 10 speed cassette - whilst it might nominally be compatible with the 10 speed DA shifters we could be running up against the higher tolerance limits required during 10 speed shifting compared with 9 speed.

Well there is an option for attaching the control wire at a different
point under the cable clamp to effect a different pull-ratio. It
could help. You need to sight the jockey-pulley accurately with the
sprockets to see that it maintains alignment with the penultimate
sprocket in either direction.

> >The rattling is a feature to remind the rider that such a combination
>
> does not result in an efficient chainline.  :-)
>
> I would agree for the two extreme sprockets but not for 4 sprockets in on a 10 speed cassette. I often switch front rings around the "middle" of the cassette in anticipation of a rapid change in gradient. I would certainly expect any setup to do this 4 sprockets in from either end without rattling thereafter.

Yes, I hhad been thinking to understand. Very bad of me.

>
> > As well as the hanger alignment, the frame-end alignment needs
>
> checking and it could be that an error in frame-build is being
> responsible and the hanger is being twisted with the frame-end when
> the axle is tightened.
>
> Good point I will add this to his list just in case the guys who sold him the mech forget and just check the hanger.
>
> Thanks again for your continued interest.
>
> Graham.

A little more oil might still help but the grease should have
slackened. You could make sure that the chain is actually floppy in
your hand when the pulley tension is relieved.

[AMuzi]

My sentiments are the same as yours generally.

But in this case, it's all the same brand, same quality
level and Santana ships that exact combination on new
2012/2013 tandems. Santana is a product with which we have a
lot of experience and no gear train difficulties.

[Graham]

"Graham" <h2gt2g42-mi...@yahoo.co.uk> wrote in message news:e6WdnfpqFd1CtfbN...@bt.com...

"Lou Holtman" <lou.h...@usenet.nl> wrote in message news:dff9e$506b272d$541d34fb$28...@cache90.multikabel.net...
> Op 2-10-2012 16:00, Graham schreef:
>>
>> "Lou Holtman" <lou.h...@gmail.com> wrote in message news:1f8bb762-4100-4786...@googlegroups.com...
>> Op zondag 30 september 2012 20:52:45 UTC+2 schreef Graham het volgende:
>> [Snip]

> Compatible or kind of compatible. If your are a perfectionist it is not
> compatible, if not it might 'work'. Not too fond of this mix and match
> thing.
>
> Lou

My friend had no joy getting his hanger checked out by the place he bought his XTR from so I made up a tool to check and realign the hanger. It turned out to be 4mm out across the vertical measured at the position of the valve hole on the wheel rim and 3mm out across the horizontal. This hardly seemed significant but having discovered how soft the aluminium was that this hanger was made of I decided to bend it back into near perfect alignment. Unfortunately the XTR still exhibited the same shifting problems.

As it happens my friend had taken a short cage Dura Ace road mech off his bike to fit the XTR having been told he could not use it with an 11-32 cassette. After swapping the Dura Ace mech for the XTR I was able to dial in almost perfect shifting across all gears and by inverting the "b" screw with it fully extended managed to get acceptable running on the 32 sprocket when in the small chain ring.

So perhaps Lou is sort of right, not withstanding Andrew's experience, that for whatever reason the XTR is not a "workable" option in the context of my friends set up.

My only theory is that with the short chain stays on my friend's racing frame the chain line variation is more pronounced than it might be on the tandem Andrew mentioned. This coupled with the long flexible cage on the XTR might be putting sufficient side loads on the cage to move it off line just enough to prevent dialling in perfect shifting across the whole cassette in both rings.

Graham.

[Frank Krygowski]

On Oct 8, 11:59 am, "Graham" <h2gt2g42-micenewgro...@yahoo.co.uk>
wrote:
> "Graham" <h2gt2g42-micenewgro...@yahoo.co.uk> wrote in messagenews:e6WdnfpqFd1CtfbN...@bt.com...
> "Lou Holtman" <lou.holt...@usenet.nl> wrote in messagenews:dff9e$506b272d$541d34fb$28...@cache90.multikabel.net...

> > Op 2-10-2012 16:00, Graham schreef:
>

> >> "Lou Holtman" <lou.holt...@gmail.com> wrote in messagenews:1f8bb762-4100-4786...@googlegroups.com...

> >> Op zondag 30 september 2012 20:52:45 UTC+2 schreef Graham het volgende:
> >> [Snip]
> > Compatible or kind of compatible. If your are a perfectionist it is not
> > compatible, if not it might 'work'. Not too fond of this mix and match
> > thing.
>
> > Lou
>
> My friend had no joy getting his hanger checked out by the place he bought his XTR from so I made up a tool to check and realign the hanger. It turned out to be 4mm out across the vertical measured at the position of the valve hole on the wheel rim and 3mm out across the horizontal. This hardly seemed significant but having discovered how soft the aluminium was that this hanger was made of I decided to bend it back into near perfect alignment. Unfortunately the XTR still exhibited the same shifting problems.
>
> As it happens my friend had taken a short cage Dura Ace road mech off his bike to fit the XTR having been told he could not use it with an 11-32 cassette. After swapping the Dura Ace mech for the XTR I was able to dial in almost perfect shifting across all gears and by inverting the "b" screw with it fully extended managed to get acceptable running on the 32 sprocket when in the small chain ring.
>
> So perhaps Lou is sort of right, not withstanding Andrew's experience, that for whatever reason the XTR is not a "workable" option in the context of my friends set up.
>
> My only theory is that with the short chain stays on my friend's racing frame the chain line variation is more pronounced than it might be on the tandem Andrew mentioned. This coupled with the long flexible cage on the XTR might be putting sufficient side loads on the cage to move it off line just enough to prevent dialling in perfect shifting across the whole cassette in both rings.
>
> Graham.

Just out of curiosity, are all the pivots in that derailleur suitably
rigid? Or might some of them be worn and allowing things to be out of
alignment?

- Frank Krygowski

[thirty-six]

You mean like a Huret Svelto after 10 winters (at least)? :-)
Graham is most likely right, he's seen it. It's a question I asked
myself and then dismissed after seeing a Svelto and Nouvo Gran sport
and others with severely worn pivots working exceptionally well.
It's the expectation that a dual-pivot derailleur will work to the
same standards as a half-step setup with a single-pivot derailleur
which is the problem. I have what was billed as a RX100 (CBA
checking the number) short cage racing d on a med length frame-end
with standard racing 1" hanger drop, and it works fine for me up to 31
or 32 teeth rear sprocket, but I am only using a 5-sprocket freewheel
with 3 teeth between chainwheels I'm asking little of the mech
because of the small difference in chainwheel sizes and although not
snappy like a single pivot (which is my preferred derailleur, it works
well enough. If I was racing criteriums I'd still like the positive
clunk of the Nouvo Gran Sport.

Or is it the slant parallelogram which makes em sloppy? Isnot issit
cos Suntour were smooth not sloppy so it is the dual pivot that's the
general problem but is mostly overcome with reduced tooth height and
crackpot shaping. Oh well, that's commerce, you can take it or leave
it. I'll get my ancient stuff fixed up with new paralellogram pivots
and keep riding with a half-step. Whatever I use It's likely to be
the cheap end of the racing range, mind you, I do still have a
hankering for the old Mavic derailleur, What wassit, an 841 or summat?

[Graham]

"Frank Krygowski" <frkr...@gmail.com> wrote in message news:19cc8b84-8f66-4344...@e18g2000yqo.googlegroups.com...
On Oct 8, 11:59am, "Graham" <h2gt2g42-micenewgro...@yahoo.co.uk>

The mech is brand new. So I guess they are as rigid as they get.

Graham.

[Frank Krygowski]

On Oct 8, 7:08 pm, "Graham" <h2gt2g42-micenewgro...@yahoo.co.uk>
wrote:

>
>
> The mech is brand new. So I guess they are as rigid as they get.
>
> Graham.

OK. I was just wondering if it could be worn enough to contribute to
the problem. Apparently not.

- Frank Krygowski