Front derailleur problems

[ricksavery]

I was just feeling at ease with the problem and you've introduced another variable - cage length
Got all Zen about it, then did a bit of lateral thinking (with my wallet out!)
I think the problem is that the LX front mech is more of a MTB mech. Checking the specs of Spa Cycles touring bikes - including Dawes Galaxy range, they all seem to use the Tiagra, with the same or very similar chainset to mine.
So I thought I would buy a Tiagra front mech and see - this will teach me to do better research next time.
Is this sensible does anyone think, or have i been a complete knob again?

 

[Moodyman]

Sounds plausible.

My advice would be to ring Spa. They're very good with their advice and their prices are v. good too.

 

[ricksavery]

Just to prove my point about research etc, I've just thought: I have flat bars so will the Tiagra be OK or should I get the 'flat bar' road front mech (FD-440 I think at Chain Reaction)? Is it essentially the same thing as the Tiagra ? I can hear mutterings and shaking of heads even from here !

 

[RecordAceFromNew]

Hi Rick if I were you I would check a couple of things before buying another front mech.

Firstly since mtb front shifter/mech are incompatible with road front mech/shifter respectively due to different cable pull, if you have mtb shifters now, replacing the lx front mech with a tiagra can give rise to a potential indexing challenge (in this case mech hitting limit screws before shifter clicks). Which shifter are you using?

While modern lx front mechs do specify a 12T minimum difference between the large and middle chain ring, in most cases while not ideal it will work with 10T difference by moving the mech up a little (difference of 38T and 36T middle ring radius is 4mm), as you mentioned you have done. This sometimes doesn't work because many things such as seat tube angle, chain stay length/angle, chain line etc. can all contribute to the problem/solution.

Since you said it is causing rubbing in low gears, to identify the solution, I wonder if you could describe where is it rubbing?

I suspect it is not rubbing at the tail (the horizontal pin joining the inner and outer plates at the lowest point of the mech) because generally that only get in the way when the total capacity of a front mech is exceeded (and LXs' capacity is generally 22T), and generally only when one is on low gear at the chainset and HIGH gear at the cassette, a gearing combination that is not exactly sensible or necessary.

More probably I suspect it is rubbing on the inner plate at low gear, if so I wonder if it is caused either by a) inner limit screw adjustment issue or b) a front mech rotational adjustment issue. The former is simple, but if the front mech's rotational (around the seat tube) adjustment is not optimum, the effective gap of the mech cage for the chain is reduced, and can cause rubbing.

Further there is a possibility that your lx front mech has a lower max outer ring limit than 48T. Do you have the model number? There are many different model variances just within Deore lx, both current and historic, and they don't have the same spec.

 

[ricksavery]

Thanks for the post. I'll try to be as accurate as I can in describing the problem.
First, the model of the LX front mech is T661, and according to the Shimano blurb, a 48T outer ring should be fine. there is a 12T min difference between middle and outer and a 22T overall difference.
I don't think it is a limit screw problem as the inner is at it's limit, so the rotation is the main problem.
When on the inner chainring and the largest sprocket, I rotate the mech until the inner plate jsut clears the chain. As i change up through the sprockets (still on the inner chainring), there comes a point when the chain rubs on the outer plate. I am not so worried about this as I reason I will have changed to the middle chainring by then.
Onto the middle chainring, and the mech rotated to suit inner chainring, the chain starts to rub at the second smallest sprocket.
If I then rotate the mech so that the outer plate just clears the chain now, the chain will rub in the inner chainring / large sprocket combination.
So I have the choice, pretty much all the gear options in the middle chainring and a couple on the inner, or the lower gears on the inner chainring and lose the smaller two sprockets on the middle - not great.
My assumption is that because the mech is mounted higher, this cause the chain to rub on the plates earlier (if only slightly) than would happen if mounted at proper height.
This is why I thought to change the mech to a Tiagra rather than change the chainrings - which seem a normal range on a tourer.
Apologies for the long reply, but i really wanted to describe it as accurately as possible.
Thanks for your help

 

[MacB]

Rick, just to add to my PM reply, I'd expect to be able to hit the full cassette from a middle chainring and all bar the opposing 2 cogs from the inner/outer chainrings. Some people avoid the cog at either end from a middle chainring but, considering how many people run a 1x9 setup, I don't think that's necessary.

 

[RecordAceFromNew]

I don't think it is a limit screw problem as the inner is at it's limit, so the rotation is the main problem.
When on the inner chainring and the largest sprocket, I rotate the mech until the inner plate jsut clears the chain. As i change up through the sprockets (still on the inner chainring), there comes a point when the chain rubs on the outer plate. I am not so worried about this as I reason I will have changed to the middle chainring by then.
Onto the middle chainring, and the mech rotated to suit inner chainring, the chain starts to rub at the second smallest sprocket.
If I then rotate the mech so that the outer plate just clears the chain now, the chain will rub in the inner chainring / large sprocket combination.
So I have the choice, pretty much all the gear options in the middle chainring and a couple on the inner, or the lower gears on the inner chainring and lose the smaller two sprockets on the middle - not great.

Two questions:


1) Is the inner position at its limit independent of the L limit screw because of cable tension (which can presumably be lowered if true) or is it because the front mech's moving parts hitting the frame? I am asking this because if properly adjusted at low-low a gap between the mech inner plate and chain of up to 4mm is still ok to avoid chain derailment. If it is not 4mm now, the extra gap from said 4mm will likely give you the room for rub-free low-low when rotation etc. are set to suit the high gear.

2) Is the front mech's tail pin detachable? The reason I am asking is because using spacer to widen the cage's tail (if undetachble by carefully bending and widening the cage there) of the front mech is indeed one of those things bike mechanics sometimes have to do to stop rubbing (if all else fails).

 

[ricksavery]

Pretty much what i wanted as a range of gears. Quite happy to lose the inner / smaller and outer / larger gear combinations, just don't think it's going to happen with the LX mech sadly

 

[Crepello]

Now that I know what shifters your using I can understand that you are having trouble tuning the mech without chain rub. I'd say that the 10T/12T chain ring difference is a bit of a red herring. It's the tuning without the chain rub you're having real trouble with. I don't have an answer in that many set up's such as Shimano that dictate an indexed front shift can be the very devil.

How old or how much use has the mech had? If it's a top swing design (clamp below cage), they are prone to wearing out rather quickly in my experience. The first thing you notice is increased chain rub, or not being able to adjust the mech without rubbing.

My own solution was to bend the mech cage at the appropriate point. This will work until the mech pivots get so worn that it starts rubbing again. This is not really a problem with non-indexed mechs. I've since manipulated a number of front mechs to achieve optimum shifting without chain rub, even a new Campag Record! I only do this when I'm satisfied that all other tweaks have been done.

 

[ricksavery]

To answer RecordAceFromNew first:
1 the inner position is at its limit independent of the L limit screw neither because of cable tension - slackened cable until there wasn't any to check - nor because of the mech hitting the frame. The mech just reaches the end of its range of movement.
2 I don't think the pin you mention is detachable. On some mech's I have seen there seems to be a screw fitting there, but mine seems to be riveted.
Thanks again

Crepello
I don't think the 10T/12T issue is a red herring particularly, as it is precisely the lack of a 12T difference that is the cause of having to mount the mech higher than normal - which I think is making the rubbing a problem. I also don't think that the indexing is making the problem worse - although I might be wrong.
The mech is only a couple of months old, and is a conventional design (ie not top-swing), so I am a bit loathe to start bending it about if it is the wrong mech for the job.
Sorry to sound so negative about your post :-) Really appreciate help and advice - which goes for all.

I know it's asking a lot of people to solve a problem / give advice about choices when you've only got my description to go on, and it's not physically in front of you
I bet if you were all standing in front of the bike there would be a collective intake of breath and 'of course... there's your problem..' :-)

 

[Crepello]

I can only add that to raise the mech the extra 2-3mm to accommodate for that 2 tooth difference is unlikely to cause any significant problems. As said previously in the thread, there can be other factors affecting the front mech behaviour, such as bottom bracket length. What are you running with your chainset? Is it a Stronglight Impact, in which case it's best run with a 107mm BB. By the way, I'm almost certain that Shimano road mechs are also designed to run optimally with a 12T difference between big and middle, but I've used one with a 48/38/28 chainset with no difficulty.

 

[RecordAceFromNew]

From what you said, perhaps the only other thing I would suggest is to check the chainline. Your front mech is designed for 50mm, if for whatever reason it is different then that could be another contributory factor to the problem, and may influence the solution.

I have already touched on potential compatibility issues between a new front mech and your existing (presumably Shimano mtb) shifter earlier. [url="http://bike.shimano.com/publish/content/global_cycle/en/us/index/tech_support/tech_tips.download.-Par50rparsys-0016-downloadFile.html/10)%20Drivetrain%20Compatability.pdf"]This[/url] may help inform compatibility between front mech and chainset (assuming your chainset is Shimano).

Good luck!

 

[ricksavery]

Crepello - in order to clear the middle ring the mech is mounted about 5mm higher than normal (ie ~8mm gap), which i think must be acontributing factor. The chainset is a Spa XD-2 with a Shimano 68-110mm BB. The LX isn't a road mech, so that (I suspect) is why there is a difficulty.

RecordAceFromNew - showing my ignorance but I have no idea what is meant by a chainline - in relation to this problem - but the Shimano techdocs say that my front mech is designed for a 45mm chainline not 50mm.

Thanks, as always

 

[RecordAceFromNew]

According to this tech doc, your Deore lx front mech (a T661 as you mentioned) assumes a chainline of 50mm. It is well known that 47.5mm and 50mm chainlines are standard configuration for the vast majority of Shimano's mtb setup, so no surprise there.


However, the 45mm chainline you have just mentioned is typical for road equipment. I haven't tried looking it up but I suspect it is your chainset that specifies/creates a chainline of 45mm, not your front mech as you have just said.


The difference between 50mm and 45mm, I suspect, is the key source of your problem (more so than the 10T vs 12T big/middle front ring difference). I suspect the chainset is so close to the frame, that your front mech bottoms out rather than being stopped by the L limit screw, essentially stopping the cage prematurely and leaving inadequate room for the low gears at the back when the mech is suitably set up for the high gears.


As you may recall, earlier I suggested L limit screw adjustment to allow a 4mm gap between inner plate and chain for Low-Low. Guessing again, is the current gap less than 4mm?


Since much of the above are guesses at the moment, it is best to check by measuring the current chainline at the front (which is easy for a triple, it is essentially bike centreline to middle ring), if unsure you can follow this.


To speed up solution identification, what shifters are you using?

 

[ricksavery]

RecodAceFromNew
Thanks for the reply, my apologies, my LX does have a chainline of 50mm, you're right, not sure where I got the 45mm from!
I'm sure it is well known about 47.5 and 50mm MTB chainlines, unfortunately not by me, wish I had, would have made my choice of mech easier.
Your post makes clear what the chainline is , thanks - I never understood that before, seems obvious when it's explained, feel a bit stupid that I didn't know this or ask what chainline meant - would have made it simpler - chainset and mech should have same chainline figure I assume?

Refgarding existing setup, what you say makes sense, the mech is bottoming out before hitting the limit screw, and yes, you're right again, the gap is less than 4mm - at the point where there is no rubbing in low-low, by rotation adjustment, the gap between chain and inner plate is probably 1 mm - very tight clearance!

The chainline at the front is 45mm (that's where I got it from!) - which, if I am following this right, means that the front mech will always be just that little too far away from the frame to get the setup perfect - is that right? The Tiagra techdocs quote a chainline of 45mm so presumably this will solve / help the problem.

The shifters are DeoreLX rapidfire pods.

Many, many thanks, I really appreciate you (and others) taking the time to help.

Even if I can't fix the problem without changing mechs, I'm fairly sure i understand why now.