Shifting problem, rear triangle misalignment and how to realign
 

I recently spread the rear drop outs from 126mm to 130mm to accomodate a 10 speed cassette. I've been riding on a trainer with it and am having difficultly shifting in the smallest cog (11 teeth).

I checked the rear triangle alignment with a Park F@G-2 and find that the left dropout is about 1.5 to 2mm further away from the centre line than the right dropout.

Would this amount of misalignment be the cause?

If so (and I gotta believe it would) how do I fix that?

..and why would you gotta believe it would be the cause? You changed two things - the rear triangle and the cassette/wheel. It is the latter that is the relevent variable.

The rear derailleur does not know where the triangle is in relation to the frame. The factors that affect whether it moves the chain to the small cog are the outer limit screw, cable tension and friction, spring tension and the distance of the cog from the dropout face. In cases of impact damage the derailleur hanger may come into play. You need to isolate the possible causes, which is best done by releasing the shift cable completely and shifting by hand to the 2nd cog and then letting go. Logically, if the derailleur now shifts to the small cog the problem is the cable is not releasing sufficiently. If it still fails then it's limit screws or a bent hanger, assuming the chain can correctly sit on the cog when placed there.

p.s. Though obviously off, correcting the alignment would only move your rear triangle over 1 mm at most.

I believed it might because the misalignment shifts the chainline to the left by 1mm.

Yes, I tried loosening the cable and adjusting the limit screw. Still had the problem. As for the hanger, I thought it made sense to first check the rear triangle alignment before messing around with the hanger. From your reply it sounds like I should just ignore what seems to be a minor misalignment and tackle the hanger as the culprit.

Yes it shifts the chainline - but the derailleur is attached to the dropout, so it moves with it, and the cassette is attached to the hub, which is also attached to the dropout. If the derailleur is shifting properly to the small cog you could detach the triangle completely from the frame and it would still shift properly. No, I did not say you should tackle the hanger. You should focus on what changed, which is primarily the cassette to derailleur relationship.

Also, you need to be more complete and accurate in your description of what you have done. Did you actually detach the cable, and does "adjusting" the limit screw mean you backed it off? Were you able to back it off enough so the derailleur pulleys were directly below the small cog? What happens when you try to shift to the cog? Do the pulleys sit below it (or even outboard) but the chain just makes noise, or does it not move to that position?

Finally, what else if anything beside the cassette did you change? Did you just add spacers to the hub or is it an entirely new wheel? Did you change the chain? Is the large cassette cog larger, requiring a longer chain? Did you adjust the b-tension screw?

We are not there to see it, so you need to make any text description specific enough so we almost feel we are there.

So aligning your frame will make the chainline in top gear worse, not better. Fix it anyway, check the hanger for alignment, cable friction, limit stop, usual suspects.

Please understand that the chain angle between front and rear when going to the small cog will be greater (worse chainline) if you move the triangle to the right, but again it will not affect the adjustment in the slightest.

Get the alignment fixed....check that rear derailleur hanger alignment as well against a properly dished wheel that is installed centered between the chainstays.

=8-)

When checking hanger alignment, strictly speaking the wheel doesn't need to be properly dished, only true.

But if it's on a bike with horizontal dropouts, the wheel pretty much does need to be properly dished in order to ensure the wheel is pointing directly ahead...

Not yet mentioned is the new cassette's high cog size change. As the cog looses teeth the gap from the guide pulley and high gear cog increases. More open chain is more chain links to flex before the chain climbs off the next small cog and onto the small one. This aspect changes with which front ring is in use, the rear cage's wind out changes the guide pulley's proximity to the cassette.

When the 10 speed wheel was installed were there any other chances? Like the cassette, chaindérailleurre, cable/casing, shifter? As kimmo said you need to establish the baselines properly. We don't know if that has been done well with whatever else might have been changed.

Could the alignment be part of the problem? Sure but as cny-bikeman mentioned chain lineline (rear triangle alignment) is not likely to be the big reason. Hanger alignment more so. But i suspect there's more going on that has not yet been said or corrected. If the frame was spread to fit a 10 speed wheel then likely the bike had a different spec before and the shifter, cable, casing, chain were all changed too. If so why not include this info in the OP?

The first thing i do in these cases detachetatch the cable completely and adjust the limits to shift fully across the cassette with my hand pushindérailleurillure. While this adjustment is not the final onexerciseersize does give you information aboudérailleurslure's movement and shifting action independent of the shifter/cable/casing.

The bike can only be seen and touched by the OP. It's up to him (her?) to figure this out or go to a LBS and ask them for help where another more experienced hand/eye can deal with this. Andy.

I'll join CNY bikeman in saying that a 1mm chainline error is totally meaningless in the scheme of things. If you think about it, the chainline varies by more than that simply because of slight spindle length, or crank position changes.

Odd are the hanger is off somewhat, though it's just as often a sluggish cable issue.

I'd start a diagnostic process by disconnecting th cable from the RD, and shift from high to 2nd high by pushing the lower body in. Then let go and it should snap back and shift cleanly. Adjust the limit until it does, and repeat that shift until it's consistent.

Connect the cable, adjust the trim and it should be just as crisp. If not, you know it's a cable related problem.

I really hate the spell check sometimes... Andy.

Not yet mentioned is the new cassette's high cog size change. As the cog looses teeth the gap from the guide pulley and high gear cog increases. More open chain is more chain links to flex before the chain climbs off the next small cog and onto the small one. This aspect changes with which front ring is in use, the rear cage's wind out changes the guide pulley's proximity to the cassette.

When the 10 speed wheel was installed were there any other chances? Like the cassette, chain, dérailleurre, cable/casing, shifter? As kimmo said you need to establish the baselines properly. We don't know if that has been done well with whatever else might have been changed.

Could the alignment be part of the problem? Sure but as cny-bikeman mentioned chain line (rear triangle alignment) is not likely to be the big reason. Hanger alignment more so. But i suspect there's more going on that has not yet been said or corrected. If the frame was spread to fit a 10 speed wheel then likely the bike had a different spec before and the shifter, cable, casing, chain were all changed too. If so why not include this info in the OP?

The first thing i do in these cases detach the cable completely and adjust the limits to shift fully across the cassette with my hand pushing the dérailleur. While this adjustment is not the final on the exercise does give you information about the dérailleur's movement and shifting action independent of the shifter/cable/casing.

The bike can only be seen and touched by the OP. It's up to him (her?) to figure this out or go to a LBS and ask them for help where another more experienced hand/eye can deal with this. Andy

When you align a derailleur hanger to a properly dished wheel that is centered between the chain stays - the hanger by default is also in parallel to the center line of a symmetrical frame - which is what 99.99% of us ride.

When you align the derailleur to a wheel that has not been properly dished - but is installed centered between the chain stays - all you have done is aligned the derailleur hanger to THAT PARTICULAR WHEEL.

...the hanger is still misaligned - i.e., it is not operating in parallel to the center line of the symmetrical frame. Swapping in different wheel later can cause problems.

If you have a hard time grasping this - read the freely available downloadable version of the Paterek manual....or visit a local master frame builder as I often do.

There's a reason they keep sets of properly dished wheels laying around as part of their frame building equipment.

=8-)

Your condescension is unwarranted.

When aligning a hanger to a poorly-dished yet true wheel in a frame with vertical dropouts, your reference is parallel to the centre plane, thanks very much.

Better think twice before you try slap me.

Glad to see you get it...

=8-)

Actually you want the hanger aligned to the cogs and not the wheel. Most use a wheel because most don't have a Shimano alignment tool and the wheel is attached to the cogs. The Shimano tool uses the hub's threading (as in the freewheel threads) to mount an arm that can be rotated around, taking the place of the rim. One shop that I worked for had a few of these tools, each on a hub with a different OLD.

But, again, most just use a well trued wheel. I do. At home this wheel is one of my building wheels and so is properly dished as well. BTW my rear triangle building wheels are laced on BMX front hubs. The axle diameter is 3/8", so to fit long enough axles for up to 145 OLD. And the dishing results in a symmetrical (no dish offset) wheel side to side allowing the sighting of the rear triangle alignment to be the same side to side. Andy.

The problem is Andrew, when you later throw in a properly dished wheel - there's a high possibility that the hanger is out of alignment again if the hanger was previously aligned to a wheel for which a dish check was not done.

I'll tell a story...about a year and a half ago...

A customer showed up - first timer. Wanted his wheel checked out AND also asked about noise and gearing problems. Now remember, right now I'm a wheel builder - that's my focus.

But I do have a stand and other tools...

Bike was a Specialized - one of those entry level commuter road bikes with mostly non-Shimano parts except for the brifters.

First thing I notice is:

Double washer setup on non-drive side - one flat and one serrated. Vertical dropouts - QR very tightly closed. Wheel appeared to be centered between the chain stays.

"Hmmm." my brain says, "Bet the wheel is off dish..."

"But it's vertical dropouts!" I say to myself.

"Yeah, the wheel is probably slipped too..." my brain says.

So on a hunch, I disengaged the QR and sure enough - despite the vertical dropouts, the wheel had been forced-angled in using the 1-2mm slack on both sides in order to force the wheel right in the chain stays in order for it to be centered. That's what the serrated washers in combo with the serrated QR nut on the driver had been used for - to lock the wheel in a non-settled position in order to center the rim between the stays.

So I had to re-dish and re-true the wheel. That part went fine.

Popped wheel back in and my brain immediately says, "Derailleur hanger was aligned to the old wheel!!!"

Sure enough - it was obvious as day and night. Banged in a little - but also angled slightly out - not in parallel with dished wheel, cogs or frame.

So I sent the customer off to my shop of choice - with a note to the folks specifying exactly what was needed - rear hanger alignment.

Next time I saw the customer, things were working much better - handling and shifting wise.

Moral of the story (long version):

Align the the rear hanger in parallel to a properly dished wheel AND the frame - regardless of whether the dropouts are horizontal or vertical.

Moral of the story (short version):

Anything is possible.

=8-)

The problem with MrRabbits case, is that even though the wheel wasn't dished properly and was somehow forced to an angle in the vertical dropouts, Aligning the hanger to that wheel, would have solved that bikes problems for that wheel, since alignment to wheel is what counts for RD performance.

You raise a niggling point, that though correct in theory, wouldn't be in practice. A wheel incorrectly dished bad enough, and canted in a frame to compensate (applies more to horizontal dropouts) will manifest other issues, such as sitting off to the side in the seat stays, or poor handling.

I do agree that having a reliable frame of reference is important, and the more likely issue is a wheel not pocketed fully in the dropouts -- which is common when wheels are installed with the bike in a repair stand -- can cause an error when used as the frame of reference. For 45+ years, I've made it a practice, and taught mechanics to do so, to ALWAYS mount wheels with the bike on the floor, letting gravity position the axle. This makes for a repeatable mount, uninfluenced by mechanic touch. Usually the wheel isn't actually installed on the ground, but the bike is taken from the stand, and both QRs opened and closed with the bike on the floor before brake and derailleur adjustments are made.

I often stop on the road to help disabled riders (mostly flats) and often find brakes rubbing when the wheel is replaced, because the wheel had originally been off, and the brake positioned accordingly.

But this is besides the point, because a wheel would need to be seriously off, for a dish issue to affect the hanger alignment enough to be an issue.

Mounting on the floor...yep what I've always done...so do most mechanics I know. Little down force on saddle and stem doesn't hurt either.

=8-)

mrrabbit- You are correct in that any wheel placed in the drop outs crooked will not agree with the hanger's alignment from a wheel that had been placed properly (given both wheels are true). I guess that the mistake i made in describing my view of hanger alignment was assuming that any one who knew about using a hanger aligning tool also knew enough to install their reference wheel fully. But i still maintain that a true wheel is the only quality that the wheel needs to be used as a aligning reference. The aligning tool does not care or know whether the wheel's rim is wider or narrower. And a well trued wheel, but not dished, is the same as far as the drive side's rim side wall placement WRT the hanger's outer face (where the tool touches the frame). So a wide rim trued to the NDS is potentially the same as a narrow rim trued on center. Will the aligning tool be able to differentiate between the two? I say no. If an off dish but true wheel is properly fitted into the drop outs the tool's extending part will just reach out more or less to touch the rim's side wall. If the rim is true the extending part will still touch the true rim as the tool rotates around the rim. That we need to go to this level of discussion surprises me.

The most common reason to have a rear wheel not properly seated into the drop outs fully IME is having the QR skewer spring back wards. No one has mentioned this yet. Andy.

If you are doing a 1/2 hour quickie 35.00-50.00 tune-up, that's what most mechanics do.

If you are doing a major tune-up or overhaul at 90.00 +, you do it right.

Leaving a wheel improperly dished but centered - and aligning the hanger to it introduces as FBinNY already mentioned poor tracking (handling) and also increases chainwheel/chain/cog wear and noise because they will no longer be operating concurrently parallel to the bicycle center line.

As I stated earlier, read the Paterek manual...it really drives home the concepts of symmetry, center line, parallel and tracking as part of the process of building a properly aligned symmetrical bicycle frame and assembled bicycle.

=8-)

Don't misquote me or take what I said out of context. I referenced a wheel dished off far enough to matter, as causing other issues greater than that of hanger alignment. I could (should) have added that rear wheels that far off are rare birds, and something to be dealt with if/when they arose.

Otherwise, my opinion is closer to that of A.S. You align the hanger to the wheel since the actual goal is to have the axis of rotation of the idler cage parallel to that of the cassette. Of course, if the owner changes things, it could introduce an error, but that cuts both ways, since the owner could at any time replace a correctly dished wheel with an incorrectly dished one and force it to center in the stays (not likely, but equally unlikely both before and after you see the bike).

As I said in the post you referenced it's a niggling point, but this being BF, I suspect it'll be debated ad nauseum until it descends into name calling and/or a moderator closes the thread.

I wouldn't be posting here at all, but felt I was misquoted, or taken out of context and wanted to clear the record.

BTW- I might add that the entire hanger alignment issue is overblown. Yes they need to be reasonable square, but this isn't rocket science, and no great amount of precision is needed. (the operative phrase here is reasonable aligned). Sram has for a number of years sold RD who's pantograph isn't a parallelogram and therefore the swivel axis of the cage assembly changes through the shift range. I'm not sure how many of the OCD hanger alignment people even noticed.

You were not misquoted:

"A wheel incorrectly dished bad enough, and canted in a frame to compensate (applies more to horizontal dropouts) will manifest other issues, such as sitting off to the side in the seat stays, or poor handling."

In that you recognize that installing and centering improperly dished wheels into an otherwise aligned frame can induce poor handling. And of course, the part where you say, "A wheel incorrectly dished bad enough" recognizes that the handling problems become readily recognizable the worse it gets.

I am merely recognizing what you have recognized - therefore it is by definition not a misquote - nor taken out of context.

When I take a polar position to someone else's argument - it's because I have a reason for it. Not just because I want to take the polar position.

=8-)

Have it your way. For the third time, It's a niggling point. Though I may not have said it clearly enough earlier here's what I meant. If we have the situation you posit, the resulting minor hanger misalignment will be the least of the bikes problems.

Like you I don't clarify my posts just for the heck of it, but to be sure that I'm not misunderstood.

You and A.S. can split this hair until the cows come home, my point is simply that it's a split hair and doesn't matter either way in terms of function.

BTW- all too often threads dissolve into hair splitting sessions about stuff that simply doesn't matter in terms of real world functions. I'll admit that I'm sometimes as guilty as anyone in this. But it doesn't change the fact that these non-issue debates are too similar to the nonsense we see on recbiketec a place I intentionally avoid because I don't own a sharp enough microtome.

Wow, I sure hope the OP is not any more confused than when he started after all that!

Same here. The answer to his question got a bit lost in the bickering.

Thread closed.