Indexing adjustment mystery

Quoted message said:

Folks:

I've got a drivetrain problem that has me stumped.

I've got a Shimano 9-speed system (w/ triple chainwheels) that I am
unable to adjust so that it indexes well in the rear at all times. If I
adjust it so that it rear-indexes perfectly while the front is in the
big ring, the chain under load will invariably want to climb to the next
larger sprocket when the front is in the small ring. If I adjust the
indexing in the small ring, the chain under load will click and
occasionally auto-shift to the next smallest sprocket when the front is
in the big ring. The misadjustment is most apparent when pedaling hard
or sprinting.

I've checked the derailer hanger, and it appears straight and in line
with the dropout when viewed from the rear. The rear derailer, a
Shimano XTR M750 long-cage, has about 25k miles on it but does _not_
have an unusual amount of play in the pivots compared to my other bikes.
The rear derailer cable has only a few hundred miles on it. I found no
part of it fraying. The chain (HG72) and sprockets (11-34) have very
little wear that I can see.

Has anyone seen this problem before? What might the cause be?

This may sound obvious, but was anything changed around the time the
shifting problems started? Was this equipment in use on a different bike, or
did it work fine until one day it didn't?

For the range you're using, hopefully it's an SGS (and not GS, which has a
slightly-shorter cage) version of the rear derailleur. A derailleur with
too-short a cage might not like those big jumps and be a bit picky about
chain length, which, of course, varies greatly through its operation (with
such a wide range).

Make sure the "B-tension screw" is backed all the way out and see if the
problem is still there.

When you say that the chain and sprockets have little wear that you can
"see", have you actually measured the chain? By the time you can see wear on
the cogs they're toast, and a chain? I can't imagine how you could look at a
chain and tell it's worn, aside from lifting it off the chainrings to see if
it no longer matches the pitch.

The rear derailleur cable itself might be OK, but friction issues are
typically found in the housing or ferrules. Also, if this is anything but a
normal cable run (ie, tandem or recumbent), it's best to get rid of that
last little bit of friction by using either a teflon coated or at least a
stainless-steel (not galvanized) cable.

--Mike-- Chain Reaction Bicycles
www.ChainReactionBicycles.com

"Bill Bushnell" <[email hidden]> wrote in message
news:[email hidden]...

Quoted message said:

Folks:

I've got a drivetrain problem that has me stumped.

I've got a Shimano 9-speed system (w/ triple chainwheels) that I am
unable to adjust so that it indexes well in the rear at all times. If I
adjust it so that it rear-indexes perfectly while the front is in the
big ring, the chain under load will invariably want to climb to the next
larger sprocket when the front is in the small ring. If I adjust the
indexing in the small ring, the chain under load will click and
occasionally auto-shift to the next smallest sprocket when the front is
in the big ring. The misadjustment is most apparent when pedaling hard
or sprinting.

I've checked the derailer hanger, and it appears straight and in line
with the dropout when viewed from the rear. The rear derailer, a
Shimano XTR M750 long-cage, has about 25k miles on it but does _not_
have an unusual amount of play in the pivots compared to my other bikes.
The rear derailer cable has only a few hundred miles on it. I found no
part of it fraying. The chain (HG72) and sprockets (11-34) have very
little wear that I can see.

Has anyone seen this problem before? What might the cause be?

Thanks.

--
Bill Bushnell
http://pobox.com/~bushnell/

Bill Bushnell said:

Folks:

I've got a drivetrain problem that has me stumped.

I've got a Shimano 9-speed system (w/ triple chainwheels) that I am
unable to adjust so that it indexes well in the rear at all times. If I
adjust it so that it rear-indexes perfectly while the front is in the
big ring, the chain under load will invariably want to climb to the next
larger sprocket when the front is in the small ring. If I adjust the
indexing in the small ring, the chain under load will click and
occasionally auto-shift to the next smallest sprocket when the front is
in the big ring. The misadjustment is most apparent when pedaling hard
or sprinting.

I've checked the derailer hanger, and it appears straight and in line
with the dropout when viewed from the rear. The rear derailer, a
Shimano XTR M750 long-cage, has about 25k miles on it but does _not_
have an unusual amount of play in the pivots compared to my other bikes.
The rear derailer cable has only a few hundred miles on it. I found no
part of it fraying. The chain (HG72) and sprockets (11-34) have very
little wear that I can see.

Has anyone seen this problem before? What might the cause be?

Thanks.

--
Bill Bushnell
http://pobox.com/~bushnell/

Did it ever work properly? If not, perhaps the frame flexes enough to
pull the cables. It doesn't sound like a friction issue, but maybe the
cable housings are a bit sticky?

Joseph

Bill Bushnell said:

Folks:

I've got a drivetrain problem that has me stumped.

I've got a Shimano 9-speed system (w/ triple chainwheels) that I am
unable to adjust so that it indexes well in the rear at all times. If I
adjust it so that it rear-indexes perfectly while the front is in the
big ring, the chain under load will invariably want to climb to the next
larger sprocket when the front is in the small ring. If I adjust the
indexing in the small ring, the chain under load will click and
occasionally auto-shift to the next smallest sprocket when the front is
in the big ring. The misadjustment is most apparent when pedaling hard
or sprinting.

I've checked the derailer hanger, and it appears straight and in line
with the dropout when viewed from the rear. The rear derailer, a
Shimano XTR M750 long-cage, has about 25k miles on it but does _not_
have an unusual amount of play in the pivots compared to my other bikes.
The rear derailer cable has only a few hundred miles on it. I found no
part of it fraying. The chain (HG72) and sprockets (11-34) have very
little wear that I can see.

Has anyone seen this problem before? What might the cause be?

Thanks.

twisted hanger? evidently the upper pulley moves sideways when the cage
rotates.

--
---
Marten Gerritsen

INFOapestaartjeM-GINEERINGpuntNL
www.m-gineering.nl

On Mon, 23 Jan 2006 22:39:54 -0800, Bill Bushnell <[email hidden]>

Quoted message said:

Folks:

I've got a drivetrain problem that has me stumped.

I've got a Shimano 9-speed system (w/ triple chainwheels) that I am
unable to adjust so that it indexes well in the rear at all times. If I
adjust it so that it rear-indexes perfectly while the front is in the
big ring, the chain under load will invariably want to climb to the next
larger sprocket when the front is in the small ring. If I adjust the
indexing in the small ring, the chain under load will click and
occasionally auto-shift to the next smallest sprocket when the front is
in the big ring. The misadjustment is most apparent when pedaling hard
or sprinting.

I've checked the derailer hanger, and it appears straight and in line
with the dropout when viewed from the rear. The rear derailer, a
Shimano XTR M750 long-cage, has about 25k miles on it but does _not_
have an unusual amount of play in the pivots compared to my other bikes.
The rear derailer cable has only a few hundred miles on it. I found no
part of it fraying. The chain (HG72) and sprockets (11-34) have very
little wear that I can see.

Has anyone seen this problem before? What might the cause be?

In addition to the other observations posted...

If this is a new problem, recheck whatever was changed last; seemingly
trivial differences in cable routing can produce really peculiar
shifting wierdnesses, for example. Also, check to make sure that
dropouts are secure in the stays; one cracked joint could make many
odd things occur under high power input.

If this has been a problem from day one, then look at the length of
the chainstays. If the frame is built with very short chainstays (as
is more typically done with road bikes than mountain bikes), it may be
that the short run is making the angle just barely severe enough that
the chain is forcing the cage slightly to one side in the most
cross-chained combinations. If the tendency only appears when running
in the big/small and small/big ranges, I'd take that as a clue that
you can solve it by simply selecting sprocket pairs that avoid
significant chain angles. Your chain will also last longer if you do
this; the chain wears fastest when the chain's angle is greatest.
--
Typoes are a feature, not a bug.
Some gardening required to reply via email.
Words processed in a facility that contains nuts.

Bill Bushnell said:

Folks:

I've got a drivetrain problem that has me stumped.

Don't think it's been mentioned specifically in previous posts, but I
was just reading this (http://sheldonbrown.com/cables.html#autoshifting)
Sheldon Brown article in which he mentions friction in the cable routers
under the BB as possible causes.

L

Folks:

Thanks for your followups.

In article <[email hidden]>,

Mike Jacoubowsky said:

This may sound obvious, but was anything changed around the time the
shifting problems started? Was this equipment in use on a different bike, or
did it work fine until one day it didn't?

No. The indexing problem has gradually developed. When the bike and
components were new I had no such shifting problem.

Quoted message said:

For the range you're using, hopefully it's an SGS (and not GS, which has a
slightly-shorter cage) version of the rear derailleur. A derailleur with
too-short a cage might not like those big jumps and be a bit picky about
chain length, which, of course, varies greatly through its operation (with
such a wide range).

The derailer is a long-cage model, Shimano M953 (correction from my
original post).

Quoted message said:

Make sure the "B-tension screw" is backed all the way out and see if the
problem is still there.

I adjusted the B-tension screw almost all the way out. It has been like
this all the time. (See picture at the link below.)

Quoted message said:

When you say that the chain and sprockets have little wear that you can
"see", have you actually measured the chain? By the time you can see wear on
the cogs they're toast, and a chain? I can't imagine how you could look at a
chain and tell it's worn, aside from lifting it off the chainrings to see if
it no longer matches the pitch.

The chain has no measurable wear, measured in several places along its
length with an engineer's scale (20). I usually replace the chain when
I get a measurement that exceeds 1/20" in 12 inches (~0.5% elongation).
Some of the sprockets have a bit of wear, but I've seen this much wear
in the past and not had any shifting problems. (See picture.)

Quoted message said:

The rear derailleur cable itself might be OK, but friction issues are
typically found in the housing or ferrules. Also, if this is anything but a
normal cable run (ie, tandem or recumbent), it's best to get rid of that
last little bit of friction by using either a teflon coated or at least a
stainless-steel (not galvanized) cable.

The shift cable is a QBP stainless (3000mm), cut to fit. Yes, the bike
is a recumbent. I give the cable a light coating of Tri-Flow before I
install it. The housing is not continuous, but there are no sharp bends
in the cable or housing other than the rear loop. Usual cable failure
occurs in the shifter (Shimano Rapid-Fire). This cable is new and shows
no signs of fraying anywhere along its length.

In article <[email hidden]>,

Quoted message said:

Did it ever work properly? If not, perhaps the frame flexes enough to
pull the cables. It doesn't sound like a friction issue, but maybe the
cable housings are a bit sticky?

The indexing worked well until about six to nine months ago after which
it has gradually worsened. If I'm adjusted for good indexing in the big
chainwheel I need to adjust cable tension by turning the barrel adjuster
three to four quarter-turns tighter to get good indexing the the small
chainwheel. (Rear derailer is a reverse spring model.)

In article <[email hidden]>,

Linus said:

Don't think it's been mentioned specifically in previous posts, but I
was just reading this (http://sheldonbrown.com/cables.html#autoshifting)
Sheldon Brown article in which he mentions friction in the cable routers
under the BB as possible causes.

Interesting article. I had "autoshifting" on another bike with a flexy
frame. The solution in that case was to use continuous shifter housing.

My problem is that I can adjust indexing so that it is consistently good
when the chain is on one chainwheel, but I cannot adjust indexing to
give smooth shifts in the rear when using any chainwheel.

In article <[email hidden]>,

Werehatrack said:

Also, check to make sure that
dropouts are secure in the stays; one cracked joint could make many
odd things occur under high power input.

For better or worse, the dropouts are welded to the stays. I see no
cracks at these welds.

Quoted message said:

If the tendency only appears when running
in the big/small and small/big ranges, I'd take that as a clue that
you can solve it by simply selecting sprocket pairs that avoid
significant chain angles. Your chain will also last longer if you do
this; the chain wears fastest when the chain's angle is greatest.

The problem is more apparent in the "cross-chaining" gears, but the
chain angle is not severe in any case. The chain on this bike is 272
links, about 2.5 times the length of most road bike chains.

In article <[email hidden]>,

M-gineering said:

twisted hanger? evidently the upper pulley moves sideways when the cage
rotates.

Possibly.

I suspect I've got a slightly bent derailer hanger that I can't discern
upon visual inspection or a rear derailer that is bent or worn. The
variable seems to be the angle of the rear derailer cage and the
position of its upper pulley.

I took some pictures at the link below of the critical areas. Is the
hanger bent? Derailer worn? Maybe some more experienced eyes can take
a look-see and spot something out of whack.

Thanks.

http://tinyurl.com/7sfy8

--
Bill Bushnell
http://pobox.com/~bushnell/

Mike Jacoubowsky said:

Quoted message said:

I suspect I've got a slightly bent derailer hanger that I can't discern
upon visual inspection or a rear derailer that is bent or worn. The
variable seems to be the angle of the rear derailer cage and the
position of its upper pulley.

Picture 5/10 looks horrendous. Hopefully it's just the angle it was taken,
but it appears that your dropout bends in quite a bit from where it's
attached. Please check that photo closely and see if it's an accurate
representation or an illusion; it it's "real" then I'd say there's a
definite issue there. The derailleur hanger/rear dropout should fall in a
straight line from where its welded to the frame; it appears that the frame
spacing is too wide so the dropout is being forced inward. Besides causing
shifting problems, it's also an easy way to have a failure. Metals would
rather be bent once than live their life under pressure like that.

I *think* it's distorted by the camera lens. It looks like it was shot
close-up with a fairly short lens, which would cause pieces to appear
misaligned. Nevertheless, I'd have it verified with a good gauge.

Ti-Rush bikes should be well put together. Easy Racers charges enough
for them.

Quoted message said:

Quoted message said:

The indexing worked well until about six to nine months ago after which
it has gradually worsened. If I'm adjusted for good indexing in the big
chainwheel I need to adjust cable tension by turning the barrel adjuster
three to four quarter-turns tighter to get good indexing the the small
chainwheel. (Rear derailer is a reverse spring model.)

Could be new derailleur time. We see far more shifting issues as you
describe from moderately-worn reverse-acting derailleurs than standard. At
25k miles, I believe you'd find a definite improvement with a new one.

That's my guess, too. I had a long-cage Shimano that worked fine for
years, but acted up under tension. I'd suspect the extra weight of the
recumbent chain affects the shifting, too. Since you say you have other
bikes, I'd swap the derailleur with something newer and see if the
problem goes away. It won't cost you anything except a little labor.

Quoted message said:

Quoted message said:

I took some pictures at the link below of the critical areas. Is the
hanger bent? Derailer worn? Maybe some more experienced eyes can take
a look-see and spot something out of whack.

Thanks.

http://tinyurl.com/7sfy8

--
Bill Bushnell
http://pobox.com/~bushnell/

Jeff

Quoted message said:

I suspect I've got a slightly bent derailer hanger that I can't discern
upon visual inspection or a rear derailer that is bent or worn. The
variable seems to be the angle of the rear derailer cage and the
position of its upper pulley.

Picture 5/10 looks horrendous. Hopefully it's just the angle it was taken,
but it appears that your dropout bends in quite a bit from where it's
attached. Please check that photo closely and see if it's an accurate
representation or an illusion; it it's "real" then I'd say there's a
definite issue there. The derailleur hanger/rear dropout should fall in a
straight line from where its welded to the frame; it appears that the frame
spacing is too wide so the dropout is being forced inward. Besides causing
shifting problems, it's also an easy way to have a failure. Metals would
rather be bent once than live their life under pressure like that.

Quoted message said:

The indexing worked well until about six to nine months ago after which
it has gradually worsened. If I'm adjusted for good indexing in the big
chainwheel I need to adjust cable tension by turning the barrel adjuster
three to four quarter-turns tighter to get good indexing the the small
chainwheel. (Rear derailer is a reverse spring model.)

Could be new derailleur time. We see far more shifting issues as you
describe from moderately-worn reverse-acting derailleurs than standard. At
25k miles, I believe you'd find a definite improvement with a new one.

Still, I suspect that tweaked-looking dropout is the issue. But could just
be the way the photo's taken.

--Mike-- Chain Reaction Bicycles
www.ChainReactionBicycles.com

"Bill Bushnell" <[email hidden]> wrote in message
news:[email hidden]...

Quoted message said:

Folks:

Thanks for your followups.

In article <[email hidden]>,

Mike Jacoubowsky said:

This may sound obvious, but was anything changed around the time the
shifting problems started? Was this equipment in use on a different bike,
or
did it work fine until one day it didn't?

No. The indexing problem has gradually developed. When the bike and
components were new I had no such shifting problem.

Quoted message said:

For the range you're using, hopefully it's an SGS (and not GS, which has
a
slightly-shorter cage) version of the rear derailleur. A derailleur with
too-short a cage might not like those big jumps and be a bit picky about
chain length, which, of course, varies greatly through its operation
(with
such a wide range).

The derailer is a long-cage model, Shimano M953 (correction from my
original post).

Quoted message said:

Make sure the "B-tension screw" is backed all the way out and see if the
problem is still there.

I adjusted the B-tension screw almost all the way out. It has been like
this all the time. (See picture at the link below.)

Quoted message said:

When you say that the chain and sprockets have little wear that you can
"see", have you actually measured the chain? By the time you can see wear
on
the cogs they're toast, and a chain? I can't imagine how you could look
at a
chain and tell it's worn, aside from lifting it off the chainrings to see
if
it no longer matches the pitch.

The chain has no measurable wear, measured in several places along its
length with an engineer's scale (20). I usually replace the chain when
I get a measurement that exceeds 1/20" in 12 inches (~0.5% elongation).
Some of the sprockets have a bit of wear, but I've seen this much wear
in the past and not had any shifting problems. (See picture.)

Quoted message said:

The rear derailleur cable itself might be OK, but friction issues are
typically found in the housing or ferrules. Also, if this is anything but
a
normal cable run (ie, tandem or recumbent), it's best to get rid of that
last little bit of friction by using either a teflon coated or at least a
stainless-steel (not galvanized) cable.

The shift cable is a QBP stainless (3000mm), cut to fit. Yes, the bike
is a recumbent. I give the cable a light coating of Tri-Flow before I
install it. The housing is not continuous, but there are no sharp bends
in the cable or housing other than the rear loop. Usual cable failure
occurs in the shifter (Shimano Rapid-Fire). This cable is new and shows
no signs of fraying anywhere along its length.

In article <[email hidden]>,

Quoted message said:

Did it ever work properly? If not, perhaps the frame flexes enough to
pull the cables. It doesn't sound like a friction issue, but maybe the
cable housings are a bit sticky?

The indexing worked well until about six to nine months ago after which
it has gradually worsened. If I'm adjusted for good indexing in the big
chainwheel I need to adjust cable tension by turning the barrel adjuster
three to four quarter-turns tighter to get good indexing the the small
chainwheel. (Rear derailer is a reverse spring model.)

In article <[email hidden]>,

Linus said:

Don't think it's been mentioned specifically in previous posts, but I
was just reading this (http://sheldonbrown.com/cables.html#autoshifting)
Sheldon Brown article in which he mentions friction in the cable routers
under the BB as possible causes.

Interesting article. I had "autoshifting" on another bike with a flexy
frame. The solution in that case was to use continuous shifter housing.

My problem is that I can adjust indexing so that it is consistently good
when the chain is on one chainwheel, but I cannot adjust indexing to
give smooth shifts in the rear when using any chainwheel.

In article <[email hidden]>,

Werehatrack said:

Also, check to make sure that
dropouts are secure in the stays; one cracked joint could make many
odd things occur under high power input.

For better or worse, the dropouts are welded to the stays. I see no
cracks at these welds.

Quoted message said:

If the tendency only appears when running
in the big/small and small/big ranges, I'd take that as a clue that
you can solve it by simply selecting sprocket pairs that avoid
significant chain angles. Your chain will also last longer if you do
this; the chain wears fastest when the chain's angle is greatest.

The problem is more apparent in the "cross-chaining" gears, but the
chain angle is not severe in any case. The chain on this bike is 272
links, about 2.5 times the length of most road bike chains.

In article <[email hidden]>,

M-gineering said:

twisted hanger? evidently the upper pulley moves sideways when the cage
rotates.

Possibly.

I suspect I've got a slightly bent derailer hanger that I can't discern
upon visual inspection or a rear derailer that is bent or worn. The
variable seems to be the angle of the rear derailer cage and the
position of its upper pulley.

I took some pictures at the link below of the critical areas. Is the
hanger bent? Derailer worn? Maybe some more experienced eyes can take
a look-see and spot something out of whack.

Thanks.

http://tinyurl.com/7sfy8

--
Bill Bushnell
http://pobox.com/~bushnell/

Mike Jacoubowsky said:

Picture 5/10 looks horrendous. Hopefully it's just the angle it was taken,
but it appears that your dropout bends in quite a bit from where it's
attached. Please check that photo closely and see if it's an accurate
representation or an illusion; it it's "real" then I'd say there's a
definite issue there. The derailleur hanger/rear dropout should fall in a
straight line from where its welded to the frame; it appears that the frame
spacing is too wide so the dropout is being forced inward. Besides causing
shifting problems, it's also an easy way to have a failure. Metals would
rather be bent once than live their life under pressure like that.

Agree that in that picture and a few others, it looks like there are
dropout alignment problems going on. If that ends up being the case,
it's gonna bode ill because dropouts on aluminum frames shouldn't be
cold set. This bike really needs to have dropout alignment tools and a
derailer hanger alignment gauge put on it to figure out what's going on
exactly. The hanger looks fine in the 8th picture, but it would still
be good to get it checked with the tool if possible.

Hopefully it's just the derailer, which I also think would be good to
sub out to troubleshoot.

That the bike is having the problems you describe is really odd,
because normally LWB bents can't have this problem the way it occurs on
uprights, since the chain angle on the LWB can't get that extreme.

Bill Bushnell wrote:
..

Quoted message said:

I took some pictures at the link below of the critical areas. Is the
hanger bent? Derailer worn? Maybe some more experienced eyes can take
a look-see and spot something out of whack.

Check for a bent hanger yourself: put the rearwheel in backwards (cogs
on the left) and screw (with the skewer, remove springs) a same sized
frontwheel to the derailleurhanger. If something is bent it will be easy
to eyeball and rectify
--
---
Marten Gerritsen

INFOapestaartjeM-GINEERINGpuntNL
www.m-gineering.nl

On Mon, 23 Jan 2006 22:39:54 -0800, Bill Bushnell <[email hidden]>

Quoted message said:

I've got a Shimano 9-speed system (w/ triple chainwheels) that I am
unable to adjust so that it indexes well in the rear at all times. If I
adjust it so that it rear-indexes perfectly while the front is in the
big ring, the chain under load will invariably want to climb to the next
larger sprocket when the front is in the small ring. If I adjust the
indexing in the small ring, the chain under load will click and
occasionally auto-shift to the next smallest sprocket when the front is
in the big ring. The misadjustment is most apparent when pedaling hard
or sprinting.

Assuming that this was a working system before, I'd guess that the
rear derailleur cable is not routed properly at the at the attachment.
It may be on top of the screw instead of underneath.

It this not a working system before, I'd guess that the bottom bracket
is way wrong in size.

Bill Bushnell said:

Folks:

I've got a drivetrain problem that has me stumped.

I've got a Shimano 9-speed system (w/ triple chainwheels) that I am
unable to adjust so that it indexes well in the rear at all times. If I
adjust it so that it rear-indexes perfectly while the front is in the
BIG ring, the chain under load will invariably want to climb to the next
larger sprocket when the front is in the SMALL ring. If I adjust the
indexing in the small ring, the chain under load will click and
occasionally auto-shift to the next smallest sprocket when the front is
in the big ring. The misadjustment is most apparent when pedaling hard
or sprinting.

<deleted>

Quoted message said:

--
Bill Bushnell
http://pobox.com/~bushnell/

Have you tried adjusting the cable using the MIDDLE chain ring?

Thanks for all your replies.

In article <[email hidden]>,

Quoted post said:

Have you tried adjusting the cable using the MIDDLE chain ring?

Yes, that's how I normally adjust it, which is why I showed image 5/10.
If indexing is adjusted in the middle ring, it will work O.K. in the big
ring, and be out of adjustment in the small ring.

In article <[email hidden]>,

Paul Kopit said:

Assuming that this was a working system before, I'd guess that the
rear derailleur cable is not routed properly at the at the attachment.
It may be on top of the screw instead of underneath.

See images 2/10 and 4/10of the album (http://tinyurl.com/7sfy8). The
cable is anchored at the derailer according to Shimano's instructions.

In article <[email hidden]>,

JeffWills said:

Ti-Rush bikes should be well put together. Easy Racers charges enough
for them.

Polished aluminum. The weld beads are too coarse for Ti.

In article <[email hidden]>,

Mike Jacoubowsky said:

Picture 5/10 looks horrendous. Hopefully it's just the angle it was taken,
but it appears that your dropout bends in quite a bit from where it's
attached. Please check that photo closely and see if it's an accurate
representation or an illusion; it it's "real" then I'd say there's a
definite issue there. The derailleur hanger/rear dropout should fall in a
straight line from where its welded to the frame; it appears that the frame
spacing is too wide so the dropout is being forced inward. Besides causing
shifting problems, it's also an easy way to have a failure. Metals would
rather be bent once than live their life under pressure like that.

Yeah, the image does look bad. Fortunately the dropouts aren't so out
of alignment as the picture seems to show. That image was centered on
the top of the upper derailer pulley, so the dropout is seen at an
angle, the end of which is not in plane, so it looks really off. Also,
it was taken with wide angle zoom, so there is distortion off the center
of the image.

There is no bending of the frame required to close the QR. The dropouts
are spaced exactly at 135mm.

In article <[email hidden]>,

M-gineering said:

Check for a bent hanger yourself: put the rearwheel in backwards (cogs
on the left) and screw (with the skewer, remove springs) a same sized
frontwheel to the derailleurhanger. If something is bent it will be easy
to eyeball and rectify

I found this test so intriguing that I immediately got dressed, went out
to the garage, and tried it. I won't hazard posting any more distorted
pictures, though.

Using two same size wheels (ISO 559) installed as above, I discovered
that the rims varied in separation by a maximum of about 6.5mm. The
widest separation occurred at about 1:00, the narrowest at 7:00 (looking
at the bike from the drivetrain side, front of bike to the right).

If I did my trigonometry correctly this works out to about 2/3 degree
misalignment. So, there appears to be some error. Could this be due to
wear at the top of the right-side dropout slot as seen in image 9/10
(http://tinyurl.com/83tg3)? The direction of wear in the dropout slot
is consistent with the direction of observed misalignment. More
importantly, is this significant?

Thanks.

--
Bill Bushnell
http://pobox.com/~bushnell/

Bill: So what's the ending to this story?

--Mike Jacoubowsky
Chain Reaction Bicycles
www.ChainReaction.com
Redwood City & Los Altos, CA USA
"Bill Bushnell" <[email hidden]> wrote in message
news:[email hidden]...

Quoted message said:

Folks:

I've got a drivetrain problem that has me stumped.

I've got a Shimano 9-speed system (w/ triple chainwheels) that I am
unable to adjust so that it indexes well in the rear at all times. If I
adjust it so that it rear-indexes perfectly while the front is in the
big ring, the chain under load will invariably want to climb to the next
larger sprocket when the front is in the small ring. If I adjust the
indexing in the small ring, the chain under load will click and
occasionally auto-shift to the next smallest sprocket when the front is
in the big ring. The misadjustment is most apparent when pedaling hard
or sprinting.

I've checked the derailer hanger, and it appears straight and in line
with the dropout when viewed from the rear. The rear derailer, a
Shimano XTR M750 long-cage, has about 25k miles on it but does _not_
have an unusual amount of play in the pivots compared to my other bikes.
The rear derailer cable has only a few hundred miles on it. I found no
part of it fraying. The chain (HG72) and sprockets (11-34) have very
little wear that I can see.

Has anyone seen this problem before? What might the cause be?

Thanks.

--
Bill Bushnell
http://pobox.com/~bushnell/

Mike Jacoubowsky said:

Bill: So what's the ending to this story?

--Mike Jacoubowsky
Chain Reaction Bicycles
www.ChainReaction.com
Redwood City & Los Altos, CA USA
"Bill Bushnell" <[email hidden]> wrote in message
news:[email hidden]...

Quoted message said:

Folks:

I've got a drivetrain problem that has me stumped.

I've got a Shimano 9-speed system (w/ triple chainwheels) that I am
unable to adjust so that it indexes well in the rear at all times. If I
adjust it so that it rear-indexes perfectly while the front is in the
big ring, the chain under load will invariably want to climb to the next
larger sprocket when the front is in the small ring. If I adjust the
indexing in the small ring, the chain under load will click and
occasionally auto-shift to the next smallest sprocket when the front is
in the big ring. The misadjustment is most apparent when pedaling hard
or sprinting.

I've checked the derailer hanger, and it appears straight and in line
with the dropout when viewed from the rear. The rear derailer, a
Shimano XTR M750 long-cage, has about 25k miles on it but does _not_
have an unusual amount of play in the pivots compared to my other bikes.
The rear derailer cable has only a few hundred miles on it. I found no
part of it fraying. The chain (HG72) and sprockets (11-34) have very
little wear that I can see.

Has anyone seen this problem before? What might the cause be?

Thanks.

--
Bill Bushnell
http://pobox.com/~bushnell/

I'd suggest that the cable be given a good clean and maybe even
replaced. That's solved most indexing problems for me.

Marty

In article <[email hidden]>,

Mike Jacoubowsky said:

Bill: So what's the ending to this story?

I wasn't able to get out and test things until today. It seems that the
indexing is now consistently good in all chainrings.

The only thing that has changed is that I removed the rear derailer,
cleaned the hanger and interface, took the pictures that I posted
earlier, and then reinstalled the rear derailer. I noticed that near
the beginning of my ride at the first hill that required my using the
small chainring the indexing was off and required about six quarter
turns tighter on the barrel adjuster. After that I didn't need to fuss
with the cable tension.

The only thing that could have changed is the derailer mounting bolt
tightness. I suspect this was the cause. I don't recall ever loosening
or removing the derailer since it was installed 25k miles ago and never
thought to check the tightness of the bolt. Do rear derailer bolts work
themselves loose?

--
Bill Bushnell
http://pobox.com/~bushnell/