You're jumping around a bit. By definition a dished wheel means different spacing on the right and left. And, so ordinarily also different spacers. Done to compensate for the hub design and the cassette or freewheel.

Just make sure the wheel is properly dished.

[MENTION=493423]Karri_R[/MENTION] apparently did the flip the wheel backward test, and the error was to the same side indicating it is probably not a wheel dishing problem.

[MENTION=493423]Karri_R[/MENTION], you won't be able to attach a photo to a message until you get 10 posts. But, if you try attaching one, it will go to your album where we can find it.

https://www.bikeforums.net/g/user/493423

Did you answer if you had horizontal dropouts or vertical dropouts?

Perhaps some close-up photos of the dropouts (right and left ones) without the wheel installed.

It may or may not be informative to viewers. Somewhere you missed something.

I went and did what mr. 66 said. Tubes are align to my eye, but I found out something else.

The chainstays are only tubes with lugged joints. All other tubes are welded. Crank-side lug is farther from the side of the BB. Maybe because of chainring clearance? Chainstays are not dimpled. Left-side lug starts from the edge of the BB. The difference is max. 1.5 mm.

I will try to align rear triangle couple of millimeter to left and share the outcome after that.

I know you're anxious to figure this out and fix it, but it might be best to post photos to your personal album (as suggested above) and get some feedback from the folks here before bending more metal.

If you post photos to your personal album, someone here will give a pic assist and post them to this thread on your behalf.

Most shops have a frame alignment gauge. They can check if for you. Or you can use the string test. Loop a string around the head tube at various heights and tie the ends to the rear dropouts. Measure the distance from the string to the seat tube on both sides.

so it sounds like the rear chain stays are asymmetrical from the factory.

That sounds like a bad idea. You are intentionally misaligning the frame now. A lot of classic frames cannot accommodate 28mm tires. That’s just the way there were designed. It makes no sense to misalign the frame to jam a 28mm tire in there.

Are you measuring from the respective outer edge of the bottom bracket to the center of the chainstay lug? Please clarify.

When I measure I use a 4' level and a thickness spacer fixed to the level. I place the level with thickness on the bb face with the spacer at the base of the seatube and then compare at the top of the seatube with spacer and level on the bb face.. I'll do the same with the he downtube. I'll then adjust as needed with bb isolated and secured.

.
...rarely, but still sometimes, a bicycle frame gets constructed with seat stays that are either uneven in length, or warped.
This results in a frame where one dropout is slightly higher than the other one. So when you insert the rear wheel, it tilts a little bit.

This can cause the problem you are observing. Have you actually measured both seat stays for length and straightness ?

It's not always noticeable to the naked eye, and it doesn't take much off level in the dropouts to cause the wheel to tilt. It's not something I would expect to see in Japanese built frame, but it does happen. Sometimes you can find one spot where the effect is minimized, by sliding your rear wheel back and forth in the horizontal dropouts. If your frame has vertical dropouts, and this is the problem, you're kind of stuck with it. It's not an easy problem to correct, if that's what it is.

Agree that unequal length stays, chain or seat, will cause a wheel to sit cocked off center WRT the unequal stays. I see many bikes with this to some small degree (and consider it to not be rare at all) and I also disagree that it's hard or difficult to correct for. Not a nice fix but usually able to be lessened or eliminated.

It is exactly this type of possible misalignment that I was alluding to when I said I would want to see the bike before making and claims as to what is going on. Andy

Photos added to my profile.

Pic Assist 1:

Pic Assist 2:

Pic Assist 3:




OK, I think these are all of them. Done before coffee, so I might have missed something.

Higher resolution images may be available in the original gallery, URL:

https://www.bikeforums.net/g/album/28904983

OP appears to be in Finland, so it may be a while before he can get there.

Per [MENTION=20650]T-Mar[/MENTION]'s Asian Serial Number Guide, not all Nishiki frames were made in Japan. Some of the (presumably) later ones were built in Taiwan by Dodsun, Fairly, and Giant. This transition appears to have occurred in the late 1980s - and I believe that's also about when TIG-welding started to come into its own re: frame production. The fact that the OP indicates (post 28) that all of his frame joints except the chainstays appear to be welded vice lugged and brazed argues for this being a strong possibility; ditto the vertical dropouts (see photos).

I'm wondering if perhaps the BB shell or (more likely) chainstays were replaced at some point. I'd expect those joints to be welded too if the other joints were welded. But I guess it's also possible that the OP's frame was from a transitional period where some joints were brazed and the rest were welded.

...yes, certainly the commonest workaround for this is filing the one of the dropouts to allow the rim to center in the frame. Still not ideal, but certainly cheap enough.
I've heard about people bending one or the other seat stays, to bring the dropouts more into alignment, but I've never tried that myself.

Maybe I will pop by during my summer holidays

I'll keep the beer cold and leave a light on. Andy

Yes, filing the slot is an easy method. I've filed the very end of axles a few times too, just like when the axle is bent one would turn the axle in the dropout till the bent axle cocked the wheel just right. I heard of someone who hacksawed the long stay and sleeved it shorter. But filing the slot is by far the most common way.

There's about a 3:1 ratio of stay displacement to amount filed off. To attain a 1mm centering of the wheel about .3mm needs removing (or about 0.012"). I have wood tube blocks to hold the frame in a vise close to the dropout to reduce the "file's song". A 10" rat tail (bastard or medium cut) file is about the perfect fit. One could Dremel the slot but with power comes bigger oops moments...

Besides the wheel needing to be dished and true, to use as a gage, it's axle needs to be straight. This is easily overlooked if it's a minor (remember that .3mm) amount. When placing the wheel in the dropouts I have the frame upside down so gravity will settle it. If you rotate the axle, and it's bent, you can watch the rim shift side to side as the bent axle pushes the hub shell around. Andy

I picked up a magnetic digital level a while back. I dont trust it absolutely, but its been useful for determining things like whether a head tube and seat are parallel enough to use as a reference to check the drop outs alignment.

Another check you could try is using your frame gauge to check the wheels alignment to the head and seat tube, in a similar manor to checking a derailleur hanger. If you trust the head and seat tube alignment, this would give better references than a tape measure between the rim and the stays and eyeballing it.

Hey, we've come back around to filing the dropout. Imagine that.

...I've done it before, but I was ashamed I couldn't come up with a more elegant solution.

Because it's an effective way to correct for some types of misalignment and has been proven over hundred+ years bikes have used dropouts.

I admit to being a bit careful to first suggest a method that is not easily reversable One could file the other dropout after finding that their assumption of needing to file the first one was wrong. This stuff sometimes reminds me of internet medical diagnosis. Without more data mistaken advice can cause more trouble then help. As the OP offered up more info the time for me to feel good about talking about a specific approach came to be.

I will give an example of why I can be hesitant to suggest filing drop outs. It was stated early on that for a 3mm rim shifting the dropout would be filed just under a .5mm. This is about half the amount that will do the job, thankfully the mistake was on the side of the "fence" that wasn't too far/much. But if the poster had said a different mistake and if the OP thought "hey the internet said to file off 1.X mm, I'll just go ahead" they wouldn't be happy to see the rim end up on the other side of the centerline.

But your post (#8) had what is now the likely technique of correction. Andy

Very cheeky, Andy. You know that was my post suggesting less the .5mm to move the rim 3mm.

The most you would need to do is a little over .5mm a side. I've done this a couple times and it never took many passes with the round file to get the wheel centered.