New derailleur shifting problem
#1
Thread Starter
Newbie

Joined: Jun 2019
Posts: 43
Likes: 3
New derailleur shifting problem
I replaced my old rear derailleur (Shimano Deore M591) with a brand new Shimano Deore M592 derailleur and installed a new derailleur cable (same housing). After the replacement, I could not seem to find the proper cable tension to shift smoothly both up and down the cassette, even after spending a lot of time trying to fine tune the adjustment. I could adjust it such that I could shift down perfectly, but then on the way up the chain would lag 1 sprocket behind where it should be, or vice-versa I could adjust it to shift up perfectly but would always lag behind 1 sprocket went shifting down. There never seemed to be a tension to induce the chain to shift properly going both directions and, in the middle, it would jump abruptly up or down between gears as a delayed effect from an earlier shift attempt.
The chain itself was essentially new when this replacement was made and, shortly thereafter, the cassette was also replaced with a new one.
Despite being similar to the old derailleur, the new M592 derailleur has a slightly different geometry with how it sits on the bike. One result is that the b-screw adjustment cannot bring the derailleur pulley as close the largest cassette sprocket as ordinarily suggested (i.e., turning the screw to its limit still leaves a relatively large gap), though this is often a fairly forgiving spec. Could this explain things? I am still using the chain length (number of links) that I was using with the old derailleur. Another idea is that maybe the chain needs to be slightly longer or shorter by a link or two? Though this, too, I would not normally expect to enable shifting perfectly in one direction but cause lags going back the other way up/down the cassette.
The chain itself was essentially new when this replacement was made and, shortly thereafter, the cassette was also replaced with a new one.
Despite being similar to the old derailleur, the new M592 derailleur has a slightly different geometry with how it sits on the bike. One result is that the b-screw adjustment cannot bring the derailleur pulley as close the largest cassette sprocket as ordinarily suggested (i.e., turning the screw to its limit still leaves a relatively large gap), though this is often a fairly forgiving spec. Could this explain things? I am still using the chain length (number of links) that I was using with the old derailleur. Another idea is that maybe the chain needs to be slightly longer or shorter by a link or two? Though this, too, I would not normally expect to enable shifting perfectly in one direction but cause lags going back the other way up/down the cassette.
#3
Senior Member

Joined: Jul 2005
Posts: 4,992
Likes: 710
From: Boulder County, CO
Bikes: '80 Masi Gran Criterium, '12 Trek Madone, early '60s Frejus track
For every rear derailleur I worked on, loosening the B-screw is how you get the jockey wheel closer to the cogs. Are you using the same cassette?
As for chain length, I've found that shorter usually gives crisper shifting because it reduces the distance between the jockey pulley and the cogs. Optimum chain length is different for 2x, 3x, and 1x. I can't suggest a change without knowing more, or seeing some photos. But the main rule is, you don't want the chain tight as a drum in big-big.
And using old housing and ferrules is always a bad idea.
As for chain length, I've found that shorter usually gives crisper shifting because it reduces the distance between the jockey pulley and the cogs. Optimum chain length is different for 2x, 3x, and 1x. I can't suggest a change without knowing more, or seeing some photos. But the main rule is, you don't want the chain tight as a drum in big-big.
And using old housing and ferrules is always a bad idea.
#4
Clark W. Griswold




Joined: Mar 2014
Posts: 18,402
Likes: 6,729
From: ,location, location
Bikes: Foundry Chilkoot Ti W/Ultegra Di2, Salsa Timberjack Ti, Cinelli Mash Work RandoCross Fun Time Machine, 1x9 XT Parts Hybrid, Co-Motion Cascadia, Specialized Langster, Phil Wood Apple VeloXS Frame (w/DA 7400), R+M Supercharger2 Rohloff, Habanero Ti 26
Always replace cables and housing when replacing derailleurs or shifters or brakes or anything like that. Unless it is brand new stuff just get new stuff it is pretty cheap even for the good stuff Jagwire Pro Shift is usually $32-40ish and the still fine sport stuff is $18-25ish and now they have 1x kits so if you have a 1x set up you can just buy for that though having extra supplies is never bad.
#5
Sock Puppet
Joined: Oct 2022
Posts: 1,701
Likes: 865
From: Planet Earth
Bikes: 2014 Cannondale Synapse Carbon, 2017 Jamis Renegade Exploit and too many others to mention.
Always replace cables and housing when replacing derailleurs or shifters or brakes or anything like that. Unless it is brand new stuff just get new stuff it is pretty cheap even for the good stuff Jagwire Pro Shift is usually $32-40ish and the still fine sport stuff is $18-25ish and now they have 1x kits so if you have a 1x set up you can just buy for that though having extra supplies is never bad.
This is also a possibility.
Last edited by Lombard; 12-31-22 at 09:00 AM.
#8
Thread Starter
Newbie

Joined: Jun 2019
Posts: 43
Likes: 3
Thanks for your suggestions.
Regarding the B-screw adjustment, what I meant was that I have loosened the screw (bringing the chain closer to the cassette) to a point where the screw is no longer pushing on anything and further loosening past this limit has no effect.
Yes, the cassette is the same model/type. My gearing is 3×9.
Although the old housing was reused, and thus suspect in the context of this problem, as near as I can tell it is undamaged and the cable slides effortlessly through it. The length of the housing was unchanged following the derailleur replacement. With the new derailleur, there is now noticeably less slack in the housing routing compared with using the old derailleur (i.e., there used to be a much larger 'loop' of cable/housing at the rear derailleur that is now mostly taken up).
I did check/realign the derailleur hanger (with the tool) in the course of the replacement and afterward (by eye) the derailleur appears to hang pretty straight.
A composite photo of the installed derailleur, for reference, can be seen here.
Regarding the B-screw adjustment, what I meant was that I have loosened the screw (bringing the chain closer to the cassette) to a point where the screw is no longer pushing on anything and further loosening past this limit has no effect.
Yes, the cassette is the same model/type. My gearing is 3×9.
Although the old housing was reused, and thus suspect in the context of this problem, as near as I can tell it is undamaged and the cable slides effortlessly through it. The length of the housing was unchanged following the derailleur replacement. With the new derailleur, there is now noticeably less slack in the housing routing compared with using the old derailleur (i.e., there used to be a much larger 'loop' of cable/housing at the rear derailleur that is now mostly taken up).
I did check/realign the derailleur hanger (with the tool) in the course of the replacement and afterward (by eye) the derailleur appears to hang pretty straight.
A composite photo of the installed derailleur, for reference, can be seen here.
#9
Senior Member


Joined: Feb 2017
Posts: 4,244
Likes: 908
From: Northern Shenandoah Valley
Bikes: More bikes than riders
Despite being similar to the old derailleur, the new M592 derailleur has a slightly different geometry with how it sits on the bike. One result is that the b-screw adjustment cannot bring the derailleur pulley as close the largest cassette sprocket as ordinarily suggested (i.e., turning the screw to its limit still leaves a relatively large gap), though this is often a fairly forgiving spec. Could this explain things? I am still using the chain length (number of links) that I was using with the old derailleur. Another idea is that maybe the chain needs to be slightly longer or shorter by a link or two? Though this, too, I would not normally expect to enable shifting perfectly in one direction but cause lags going back the other way up/down the cassette.
I've had a few of these types of derailers on several bikes and I just don't prefer them for this very reason. I find them very difficult to tune correctly because of that gap. Note that some of Shimano's newer derailers of this style (like the Deore M5100) have gone back to a more traditional cage design where the upper pulley rotates up as the cage rotates rearward. I think one of these would work better...but they're not available in 9-speed, unfortunately.
I bet you can put your old M591 back on the bike and get it shifting perfectly pretty easily. If so, I'd humbly recommend you buy a new M591 (or other derailer of similar design, like an RD-T4000) if you need to replace your current one. Another option might be switching to a Microshift Advent and accompanying shifter...these usually work very well.
#11
Thanks for your suggestions.
Regarding the B-screw adjustment, what I meant was that I have loosened the screw (bringing the chain closer to the cassette) to a point where the screw is no longer pushing on anything and further loosening past this limit has no effect.
Yes, the cassette is the same model/type. My gearing is 3×9.
Although the old housing was reused, and thus suspect in the context of this problem, as near as I can tell it is undamaged and the cable slides effortlessly through it. The length of the housing was unchanged following the derailleur replacement. With the new derailleur, there is now noticeably less slack in the housing routing compared with using the old derailleur (i.e., there used to be a much larger 'loop' of cable/housing at the rear derailleur that is now mostly taken up).
I did check/realign the derailleur hanger (with the tool) in the course of the replacement and afterward (by eye) the derailleur appears to hang pretty straight.
A composite photo of the installed derailleur, for reference, can be seen here.
Regarding the B-screw adjustment, what I meant was that I have loosened the screw (bringing the chain closer to the cassette) to a point where the screw is no longer pushing on anything and further loosening past this limit has no effect.
Yes, the cassette is the same model/type. My gearing is 3×9.
Although the old housing was reused, and thus suspect in the context of this problem, as near as I can tell it is undamaged and the cable slides effortlessly through it. The length of the housing was unchanged following the derailleur replacement. With the new derailleur, there is now noticeably less slack in the housing routing compared with using the old derailleur (i.e., there used to be a much larger 'loop' of cable/housing at the rear derailleur that is now mostly taken up).
I did check/realign the derailleur hanger (with the tool) in the course of the replacement and afterward (by eye) the derailleur appears to hang pretty straight.
A composite photo of the installed derailleur, for reference, can be seen here.
#12
Thread Starter
Newbie

Joined: Jun 2019
Posts: 43
Likes: 3
The M591 is of a traditional derailer design where the upper pulley rotates up as the cage rotates rearward...to keep the upper pulley tracking close to the sprockets as it shifts down the cassette. The M592 is a "Shadow" derailer and, as you're seeing, has a different geometry. It's designed for fairly large cassettes, so it may very well be that you can't get the upper pulley very close to the cassette, even on the largest sprocket. Making this situation worse is the geometry of the upper pulley -- its axis is on the rotation axis of the cage, so that upper wheel does NOT rotate up as the cage rotates rearward. So the gap you already have at the largest sprocket becomes even wider as you shift down the cassette, resulting in poor shifting behavior.
I've had a few of these types of derailers on several bikes and I just don't prefer them for this very reason. I find them very difficult to tune correctly because of that gap. Note that some of Shimano's newer derailers of this style (like the Deore M5100) have gone back to a more traditional cage design where the upper pulley rotates up as the cage rotates rearward. I think one of these would work better...but they're not available in 9-speed, unfortunately.
I bet you can put your old M591 back on the bike and get it shifting perfectly pretty easily. If so, I'd humbly recommend you buy a new M591 (or other derailer of similar design, like an RD-T4000) if you need to replace your current one. Another option might be switching to a Microshift Advent and accompanying shifter...these usually work very well.
I've had a few of these types of derailers on several bikes and I just don't prefer them for this very reason. I find them very difficult to tune correctly because of that gap. Note that some of Shimano's newer derailers of this style (like the Deore M5100) have gone back to a more traditional cage design where the upper pulley rotates up as the cage rotates rearward. I think one of these would work better...but they're not available in 9-speed, unfortunately.
I bet you can put your old M591 back on the bike and get it shifting perfectly pretty easily. If so, I'd humbly recommend you buy a new M591 (or other derailer of similar design, like an RD-T4000) if you need to replace your current one. Another option might be switching to a Microshift Advent and accompanying shifter...these usually work very well.
Thank you, hokiefyd, for the very helpful reply. My original intent was to do a like-for-like derailleur replacement, but inadvertently got the 592 instead of the 591. Since I thought that, in principle, both should work, I just went with it. And it nearly does work, it does shift the chain between sprockets, it just has this very annoying lag when reversing directions. I replaced the old, 591 derailleur since its jockey wheels were extremely worn and, more importantly, the derailleur frame seemed to be a bit bent/twisted from an accident, which is why I don't simply swap back to the old one. If I understand you correctly, the different geometry and movement style of the 592 derailleur results in a larger pulley wheel / cassette gap than typical (and larger than what I had before with the 591), which means there is a little extra slack in the chain at this critical point. So when the derailleur tries to pull the chain to change gears, it cannot get quite enough movement to make the change cleanly/completely. I probably compensate for this by adjusting (biasing) the cable tension so that at least one shifting direction works alright, but this results in the other direction being worse off and thus producing the shifting lag going to the other way.
Regarding Kontact and KCT1986's comment on the side that the fixing both that clamps the cable, could you elaborate on how that could possibly affect the shifting issue that I am describing? At the end of the day, despite the cable itself perhaps being a bit more deformed, that end of the cable is secure and transfers its tension to nearly the same location on the derailleur.
#13
Senior Member


Joined: Jan 2009
Posts: 4,423
Likes: 1,682
From: San Diego, CA
Regarding <span style="color:#222222;">Kontact and KCT1986's comment on the side that the fixing both that clamps the cable, could you elaborate on how that could possibly affect the shifting issue that I am describing? At the end of the day, despite the cable itself perhaps being a bit more deformed, that end of the cable is secure and transfers its tension to nearly the same location on the derailleur.</span>
Last edited by Crankycrank; 12-31-22 at 07:08 PM.
#14
SE Wis

Joined: Apr 2005
Posts: 11,556
Likes: 4,333
From: Milwaukee, WI
Bikes: '68 Raleigh Sprite, '02 Raleigh C500, '84 Raleigh Gran Prix, '91 Trek 400, 2013 Novara Randonee, 1990 Trek 970
Regarding Kontact and KCT1986's comment on the side that the fixing both that clamps the cable, could you elaborate on how that could possibly affect the shifting issue that I am describing? At the end of the day, despite the cable itself perhaps being a bit more deformed, that end of the cable is secure and transfers its tension to nearly the same location on the derailleur.
#15
Senior Member

Joined: Jan 2016
Posts: 1,524
Likes: 650
The point that the cable is clamped affects the distance and angle in relation to the pivot points on the derailleur. Although quite small visually, it has some effect on the actuation ratio of the derailleur. The resulting change could mean that the derailleur no longer has the 1.7:1 average ratio. It could now be + / - a few(?) % from the intended 1.7:1.
It would be interesting to see if the total movement of the derailleur, from low to high shift position, is substantially different depending on the attachment point.
It would be interesting to see if the total movement of the derailleur, from low to high shift position, is substantially different depending on the attachment point.
#16
Shimano had instructions for how to use 7410 Dura Ace 8 speed shifters with the 7700 derailleur by using the "B position" for the cable location. Despite 7410 using only 9 speed cable increments to shift 8, the B position (cable on the wrong side of the bolt) changed the standard pull of the 7700 to 7400 pull.
Yes, this is real. Why argue about it when it is the easiest thing in the world to try for yourself?
Yes, this is real. Why argue about it when it is the easiest thing in the world to try for yourself?
#17
Senior Member


Joined: Feb 2017
Posts: 4,244
Likes: 908
From: Northern Shenandoah Valley
Bikes: More bikes than riders
If I understand you correctly, the different geometry and movement style of the 592 derailleur results in a larger pulley wheel / cassette gap than typical (and larger than what I had before with the 591), which means there is a little extra slack in the chain at this critical point. So when the derailleur tries to pull the chain to change gears, it cannot get quite enough movement to make the change cleanly/completely. I probably compensate for this by adjusting (biasing) the cable tension so that at least one shifting direction works alright, but this results in the other direction being worse off and thus producing the shifting lag going to the other way.
I've read similar comments from others with their experience with these derailers, so I think this struggle is not specific to me or my bikes.
Having said that, it does appear that your cable routing is on the wrong side of the bolt. As others have noted, it can change the "pull ratio" of the derailer, so you get more or less derailer movement than intended, even if it's only slightly. Fixing that may or may not fix your shift quality, but you definitely want to start with as close as possible to how Shimano designed it.
#19
Senior Member

Joined: Sep 2006
Posts: 917
Likes: 250
From: Napa Valley, CA
Bikes: Wife says I have too many :-)
SamSam77,
Try putting the cable on the other side of the bolt as pointed out and let us know what happens.
Inquiring minds want to know.......................
Try putting the cable on the other side of the bolt as pointed out and let us know what happens.
Inquiring minds want to know.......................
#20
Thread Starter
Newbie

Joined: Jun 2019
Posts: 43
Likes: 3
Today I moved the derailleur cable to the other (proper) side of the fixing bolt and readjusted the tension. The issue is definitely not completely fixed but I think it might be slightly improved. The lag in shifting is most noticeable at the extrema of the cassette (i.e., cogs “1/2” and “8/9” in my case), but I think in the middle it might be better. If the cable being on the wrong side of the bolt was at least partly to blame for the issue, then I am very surprised and happy to be wrong about it. More testing to come in the coming week’s commute.
#21
Today I moved the derailleur cable to the other (proper) side of the fixing bolt and readjusted the tension. The issue is definitely not completely fixed but I think it might be slightly improved. The lag in shifting is most noticeable at the extrema of the cassette (i.e., cogs “1/2” and “8/9” in my case), but I think in the middle it might be better. If the cable being on the wrong side of the bolt was at least partly to blame for the issue, then I am very surprised and happy to be wrong about it. More testing to come in the coming week’s commute.
#22
Senior Member


Joined: Feb 2012
Posts: 19,373
Likes: 5,516
From: Rochester, NY
Bikes: Stewart S&S coupled sport tourer, Stewart Sunday light, Stewart Commuting, Stewart Touring, Co Motion Tandem, Stewart 3-Spd, Stewart Track, Fuji Finest, Mongoose Tomac ATB, GT Bravado ATB, JCP Folder, Stewart 650B ATB
Back to the cable casing- I have found that the last stay to der loop of casing to be very important to positive shifting. I cant count the number of times i have had sluggish shifting on a repair and have a lot of this clear up with a new inner cable and that last loop of casing. I believe that the cable wears a path on the casing liner and only when under some tension will this a lot of cable friction.
The next aspect that can hide from/fool some is the under BB cable guide. Millions of bikes have a slightly flexible but hard plastic version, various cable/shell specs vary but the material used is very common. Like the casing inner liner getting worn enough this guide can too. takes a lot more miles though. I've seen guides so worn that they rotate a bit when the rear der is shifted and the cable can't slide easily in the guide's grooves. Andy
The next aspect that can hide from/fool some is the under BB cable guide. Millions of bikes have a slightly flexible but hard plastic version, various cable/shell specs vary but the material used is very common. Like the casing inner liner getting worn enough this guide can too. takes a lot more miles though. I've seen guides so worn that they rotate a bit when the rear der is shifted and the cable can't slide easily in the guide's grooves. Andy
__________________
AndrewRStewart
AndrewRStewart
#23




