black02em2

I am currently overhauling and restoring an old Peugeot PB18 (Reynolds 501) and I am having issues getting the rear wheel to align with the seat tube. When the wheel is centered using the adjuster screws, it appears to tilt.

From looking at the frame from every angle, the driveside chainstay seems to sit lower than the non-driveside, causing the wheel to tilt as though they were twisted. There doesn't appear to be any scratches or bent tubes which eliminates a previous crash. Was it made this way?

What I've done so far:

• Wheel appears to spin true
• Tried two separate rear wheels in the dropouts (still leans to the left)
• Flipped the wheel with the cassette side on the left (still leans to the left)
• Used the string alignment technique (2-3mm off on non-drive side), but I don't think that affects it leaning
• Used a 2x4 to try and slowly bend the driveside chainstay upwards while lowering the non-driveside

So far I have no luck. Any suggestions?

Mr. 66

I think the one side could be a tad higher, I would measure the distance from the dropout to the seat pin, that will inform you of discrepancies. I would measure from bb to dropout also.

branko_76 

................................................................

Are the dropouts parallel to each other?...........Are they spaced properly for your rear axle so that you don't have to spread them or pull them in when installing the wheel?

SJX426 

@black02em2 - Nice set of pictures but.. none of them show the center of the frame being in the center of line of site (camera center). I suspect your dish is off, not that it is a contributor. @Mr.66 has good advice. Be sure to pick an easy reference point that is common to both sides, i.e put a fastener in the eyelet's and measure from the center of the fastener to the center of your seat clamp fastener for the set stay like distance and then to the center of the BB to the same for the chain stays.
One of the first impressions was that the rear was spread and one side was spread more than the other, but without pictures aligned with the center of the frame, it is hard to tell. You could do a string check too, for this alignment.

black02em2

That I am not currently sure as i don't have a dropout alignment tool. I have seen write-ups on making a DIY for now. Assuming the dropouts were slightly off, would that cause the wheel to lean vertically (my issue)? Or just horizontally?
Bare with me, as I'm new to this.

In terms of wheel dish, I fitted two separate wheels between the dropouts, then flipped each one and re-fitted them. The lean stayed on the same side. I assume that means the dish of the wheel is not the issue?

I've measured the rear and it's consistently between 125.9 and 126.1 based on my digital caliper. I've also done the string test around the headtube to the dropouts and the non-driveside is off by about 2.3mm. Would that be enough to cause the wheel to lean over?

In terms of taking centered pictures, and taking measurements from a reference point, i can definitely try. I've been doing a lot of reading but haven't been able to find anything relating to a vertically tilted wheel, mostly horizontally tilted wheels.

thumpism 

If you're sure the wheel is correctly dished, and if you're sure the rear triangle is centered (testing by string alignment or using a Park tool) and if you're sure the dropouts are aligned as demonstrated above and the wheel still tilts, you can always file the upper flat of the "long" seat stay's dropout. I've done that.

Aladin

Hard to really tell w pics.. guess is the one side has been lifted out to make a 126 spacing rather than the 120-122mm it might have been manufactured as.. (?).

The whole of the rear could be shifted that direction which throws centering off .. (?).

IF.. your sure the rear wheel is correctly dished.. your option is bending the rear back to alignment. Easy does it w care.. lay the bike on side
padded surface. Done this a few times.. few mm's is fine.

DOES.. the axle slide into the dropouts w ease? Have seen an issue where that cutout prevented a thicker axle going full into the cutouts.

Last week did a rear rim swap out.. guy fell on his son's bike. Might have been something similar to yours.. the previous owner didn't note off alignment.. as it likely didn't affect function to his useage.

SJX426 

Note: it doesn't' take much at the axle end to change the angle of the whole wheel. Those little 3mm alignment screws make a big difference with less than half a turn.

Aladin

Missed that one.. keen eye you have. And likely issue.

thumpism 

True, but those screws adjust the wheel in only one plane, that of the slot of the dropout. If one seat stay is longer than the other it will affect the position of the dropout on that side and, consequently, the position of that dropout's slot, making one slot lower relative to the other and tilting the wheel vertically. You can still center the front edge of the wheel between the chainstays and have the top edge of the wheel not be centered between the seat stays. Try it.

And let's not forget that the slots in "horizontal" dropouts are seldom actually horizontal to the floor or the horizon, or whatever. The slots of rear-facing track dropouts are more literally horizontal than the others.

repechage

remove the tire.
double check the dish of the wheel.
after that is confirmed, flip the wheel, reasonably easy as no driveline.
There may be something organically wrong with the frame.
a set of H tools, dropout alignment tools can confirm that the dropouts are parallel and that the adjuster screws are set properly.
With all that done, there is a chance that the dropouts are not set to the same drop on either side...
frametable time to double check before dire measures.

If the frame is "out" and poorly manufactured... it can be fixed but will be unsightly.

SJX426 

The screws were an example of how little it takes to skew the wheel. Since the seat stays are not vertical, it is still a risk that the wheel would be out of vertical if using only visual indication.
Theoretically, if the set stays are off in length, then the chain stays must be off too. There could be a position in the slots center lines where they intersect such that the axle is parallel to the BB. This is the problem the OP is facing, the axle is not parallel to the BB spindle centerline, which should be orthogonal to the plane of the major "triangle."

What might be interesting is to use a DO hanger alignment tool to see if the hanger is parallel to the plane of the wheel. Probably a useless exercise

repechage

frames do get built incorrectly. but strings and such won't really be definitive.
frame needs an alignment table.
bottom bracket fits over a "whipping" post. frame can be measured for twist, and errors in bottom bracket drop on each side.
really a lot of work to fabricate something that will provide anywhere equal results.
from your description, good chance it was a low QC frame.

dabac

Got hit by a car once. Impact on rear wheel/rear rack leg. Twisted the rear of the frame w/o visible damage to the frame. Resulted in a wheel with a sideways lean. Eventually I ground off a little on one side of the dropout slot and ran a weld down the other to maintain width.

Salamandrine 

Looks crashed, most likely. Could be congenital. A framebuilder could most likely straighten it. It's not that far out really. I've seen and bent back far worse.

Check dish and the other stuff mentioned first.

branko_76 

String is only good to get a general idea for rear triangle centering. You realy need something rigid to check alignment. Also, when cold setting the stays, you need to do it in a controlled manner. My solution was this...

branko_76 

After I got everything aligned, the wheel was still off-center. Doing some more exploring I found that one of the seat stays was longer than the other so I did as thumpism suggested, I filed the inner surface of the drive-side dropout so that the axel was perpendicular to the seat-tube. Since this frame had no integral derailleur hanger, I lowered the claw so that the axle did not flop around in the now oversized opening....

P!N20

Looks to me like the DS dropout is lower than the NDS. I find it hard to believe that would be caused by a crash without any other tell-tale signs. Just a Friday afternoon special?

Wharf Rat

I’ve got an 82 Miyata 712 with the exact same problem. I tried the string alignment check, less than 2mm off. Checked the dropout alignment, it was fine. Checked the wheels dish, also fine. I gave up. I ride it crooked. If I never noticed it being off center on the seat stay brake bridge I’d never have known. I guess I could measure seat & chainstays, but even if I found discrepancies I’ll never put the money into fixing it.

thumpism 

Good job.

I didn't know what tools you have or your capabilities, but once you eliminate the variables that you know can cause the problem but didn't you eventually run down the culprit. It had to be something like that.

Salamandrine 

IME it isn't uncommon to see secondary damage like that caused by a crash, typically if a car was involved. Bike shops used to do a lot of bending back crashed frames. A lot can be done with a park zombie tool and some basic measuring equipment, and experience.

I don't think the OP will find the skills there anymore though. Framebuilder would be the best bet.

Even though this is a Peugeot, I have a hard time believing it would have ever left the front door of a bike shop that crooked.

cudak888 

Everything you wanted to know about frame (and rear triangle) alignment, but were not afraid to ask:

https://www.parktool.com/blog/repair...rame-alignment

The tool isn't that expensive either, and you can probably build one from a few bits of wood and a scrap of steel.

-Kurt

79pmooney

+1 The problem here is that the OP knows, He now has to go to the ends of the world to resolve this, Had he not known, he could have simply ridden around the world quite blissfully.
One thing the OP can check - is it possible to put the wheel in and jockey it around the dropouts until it sits in line? I don't know if there is enough play in the axle/dropout fit to pull that off. If there is, the OP can simply find that position, tighten the QR and ride. (Not ideal. Having the axle snug up against the dropout and aligned means every time the wheel is put on, all he has to do is stand the bike up before tightening the QR and the wheel goes to aligned vertically automatically. But having the axle snugged against the dropout is not essential. A good tight QR is plenty strong enough to do all the work of keeping the hub right where the OP puts it. (A cheap modern asymmetrical steel Shimano QR with oiled or greased internals will do this as well as any QR out there.)

Dropout alignment tools here are not the solution. Yes, when all else is done, those tools should be used to align the dropouts but they are not going to solve the OPs issue and in fact using them before the issue is found means probably bending the dropouts twice. Those tools twist the dropout roughly about the slot. The do not change where the hub locknut sits much at all. So aligning the dropouts is doing very little to change how the wheel sits. If the wheel was sitting wrong, now the OP has dropouts aligned to a wrong position. The primary reason to align the dropout has nothing to do with frame geometry. It is about not bending the axle by twisting the locknuts.

If this were mine, I'd so a rough check with either dropout tools (which I do not own) or two squares or a square and a drafting triangle. If the dropouts are anywhere close to parallel. good enough. Now, find out why those dropouts aren't in the proper location relative to the BB, seat lug and centerline of the bike. Bend stays and or file the dropout until it locates the wheel in the proper orientation. Only then would I put the dropout tools and and bend (unless the dropouts were way off at the first check).

Ben

repechage

that is my hunch. Not an “expensive” frame...
have to think about the best fix.
if this is the case- Work
probably more work than reasonable save filing a curved notch with a 10 mm rat tail file to get the wheel to sit.
this would bug me till it was sorted.

Salamandrine 

For anyone that misses the joke, what i call the Park Zombie tool is in fact the FFS-2 Frame and Fork straightening tool. Works great for straightening frames and forks, and also intimidating teenage gang members. It was a very common and often used tool BITD.

That frame looks laterally bent aside from the dropouts not being level. Oftentimes when you bend stays back to where they were, they'll tend to fall back into place and realign vertically as well. At any rate, steel is malleable and you can force it back within reason, with a big enough lever.

The dropout alignment tool is the last step, after the frame is straight again.