Novalite

A hub can be mounted away from the wheels center, and to avoid tyre also away, compensation is needed along the spokes connecting hub to rim. What are the rules in this, for ex a small out of center of the hub can be compensated along a spokes tension difference left/right. But there must be a limit to this, from which different spokes lengths have to been used. Is there a rule describing that point?
What max distance can be corrected along tension differences?
What are the consequences of tension differences in regards to wheel strength?

guy153

You can find the difference in spoke length by using a spoke calculator-- usually one of the inputs is the offset from the centre line to the flange.

A rear wheel for a road bike will often have different spoke lengths by about 1mm or so, often it's small enough that you will get away with using the same spokes left to right.

You would ideally like a good bracing angle for lateral stiffness. It's not ideal that the right side of a rear wheel has the spokes closer to vertical.

Assymmetry also gives you a difference in spoke tension from left to right and this can also be a problem if it pushes either side out of the acceptable range of tension. On a rear wheel for example it may be hard to get the left side tight enough for the spokes not to go loose without so much tension on the right that the rim fatigue cracks after a few thousand miles. A controversial subject but this can be mitigated by using a bit of thread lock or just linseed oil.
​

Andrew R Stewart 

When I initially read the OP I thought it was about wheels with the hub off center as in the wheel not being round when it spins

So much of the answers are shades of gray ones. The dished side of the wheel, as in the drive side for common bikes, starts to have too tight issues with much less then 1/2 the left side's flange to center. Or the left side starts to have nipple loosening issues from too little tension (Hence the rise of the thread locking nipples). I have no real numbers to quote and numbers alone serve as a poor substitute for experience and hands on learning. Andy

UniChris

Turning the nipples in and out changes the effective length of the spoke. Once that is opposed it also changes the tension.

Two things can resist the pull of one spoke: the geometric arch of the rim, and the tension of other spokes. So you can tension a wheel with the rim in different positions of dish.

Are you perhaps really asking how far you can move things by using different positions of the nipples on the same spokes, vs at what point you need to source spokes cut to different lengths?

That would essentially depend on the range of adjustment - in a double wall rim probably where the nipple bottoms out on the spokes on one end (possible extended with a washer) and where the spoke isn't threaded far enough into the nipple on the other, forcing the shoulder of the nipple to carry the spoke force in tension.

Precut spokes typically come in 2mm increments, so one simple answer would be "where the next size becomes a better choice"

Russ Roth

A hub can be mounted away from the wheels center, and to avoid tyre also away, compensation is needed along the spokes connecting hub to rim. What are the rules in this, for ex a small out of center of the hub can be compensated along a spokes tension difference left/right. But there must be a limit to this, from which different spokes lengths have to been used. Is there a rule describing that point?

Any decent hub should be engineered to make sure the spokes won't build up too far out of proper tension. There are no real rules, you build the hub so the rim is true to the hub. The only time I built with campagnolo hubs I found them to be really bad for the non-drive side spoke tension but the wheels still held up fine long term.

What max distance can be corrected along tension differences?

You really want to look at hub dimensions to answer this question. Most 130mm hubs the center of the hub to the drive flange is right about half the distance as the distance from the center to the left flange,much more than this can leave the spokes on the non-drive side rather loose. It is possible to deal with the tension difference by using different spokes, the last wheel I built I used 2.0/1.7 butted spokes on one side and 2.0/1.5 spokes on the non-drive. The narrower section spoke requires higher tension to pull against the thicker spoke. The result was them having nearly equal tension for a really stiff build. If I was really worried about tension I'd do a 1.8/1.6 butted spoke which would take a lot of tension to keep even a poorly designed hub properly built.

What are the consequences of tension differences in regards to wheel strength?
Varies but the rin can have side movement during sprints resulting in rubbing, early spoke wear and breakage, constant need to retension.

Dan Burkhart 

I thought the OP might have bee referring to hubs that are off center laterally such as Cannondale's Asymmetric Integration system. https://cannondale.zendesk.com/hc/en...c-Integration-
It puts the rim 6 mm off center from the middle of the OLD to even out the CTF dimension on both sides.

Novalite

I don't refer to specific hubs or so, just that I measured the hub as 5 mm shifted to the driveside.
The hub was referenced as "Surly hub Ultra singlespeed disc 135x10mm 36G".
A possible reason for the hubs offcenter may have been a chainline error reduction, since with a centered hub, the original chainline error would have been 10 instead of 5 mm. The chainline is now straight but I suspect that the rear tyre doesn't run in the fronts trail (tyres don't wear in their center but away from the driveside. Also, the bike is like hard to keep on a road line, especially without weight on its rear rack.
In case it is indeed the reason, I wondered why the wheel wasn't built like it should, being the reason for this topic.

Dan Burkhart 

Making the hub off center from the rim does nothing to alter the chain line. It's an asymmetrical rear triangle that does that. Shifting the position of the rim relative to the hub is then necessary to bring the rim back to the center line of the frame. This has the added benefit of equalizing the bracing angle of the spokes on both sides, making for a stronger, more stable wheel.
Check the link in my post above.

Novalite

The rear cog is mounted on the hub, so moving the hub towards the driveside brings the cog a same distance outwards.

Trakhak

Correctly built rear wheels have the rim centered between the hub locknuts, not between the hub flanges.

The right flange of the hub is closer to the centerline of the frame than the left flange is, because the right side of the hub has to accommodate the sprocket(s).

Since the right flange is offset closer to the centerline, the only way to center the rim on the frame's centerline with a hub with asymmetrical flanges is to use spokes on the right side of the hub that are shorter and at higher tension than the spokes on the left side.

The greater the difference in tension between the spokes on the left side of the wheel and the spokes on the right side, the more prone the wheel is to failure, at least theoretically. But modern rims and spokes and hubs are designed to work with extreme offsets, helped greatly by the wider rear dropout spacing in current use.

If the wheel and the frame are both correctly built but the chainline is off, that problem is addressed by selecting the correct bottom bracket spindle length.

Novalite

Due to lack of frame clearance for a sufficient sized chainring, the chainline had to be increased - both chainring and rear cog have to be moved outwards. Not just a longer bottom bracket spindle but also the cog with spacers between cog and its hub mount, being 6 bolts M5. If the hub wasn't moved out of center, it would have needed 10 mm spacers and M5 bolts 25 mm or so long, quite awkward since the cog then sits that distance away from its mount, resulting in a much increased force momentum on its mount.
See, that is a crucial difference between a brake disc mount and a cog that is mounted on the spindle itself.

Trakhak

Yes, of course. None of that contradicts anything in my most recent post. If the hub flanges are symmetrically spaced with respect to the frame centerline, the spokes can then be equal with respect to vertical angle and tension. The bottom bracket spindle must still be chosen appropriately for the correct chainline.

If that doesn't satisfy you, please state your question more clearly, if you can.

Novalite

Thanks for the information.
Maybe I need to tell the story from the begin, in telegram style:
1) Bike in production, got a mail about chainline problem, but a solution was found, I had to have a bit more patience.
2) Got mail bike ready to pickup. I noticed the bike behaved very unstable, it was hard to stay on a white line. But weird: after I installed a basket on top of the rear rack, with some luggage in it, the bike felt stable. Dealer said no idea how that came.
3) A year later, chain very weird: parts hanging tilted 45 degrees. Asked dealer several times for an explanation. Said he had no idea, said he had never seen such a torn chain.
4) Since there had to be a reason, I searched help on a forum. First answer given: wrong chainline, and instructions how to measure it. Did so, discovered a 5 mm error. Informed the dealer, answer was "he had followed my measurements and they were correct". Just like that. In fact, he finally admitted something he had tried to hide all the time
5) I found a solution as spaceNow I want to know what work this may involve. Shorter spokes on 1 side.rs, available from UK company Velosolo. These solved the problem.
6) Sometimes, after work on the bike, and riding it with no basket (see above), the bike all of sudden again behaved unstable. And that's where my topic here starts. I suspect that the rear wheel does not run in the trail of the front wheel. There is another element that indicates this: the rear tyre wears not in the center, but to the driveside. I have had two rear tyres, both wearing like that.
7) So, I suspect that the dealer didn't correct the wheel after reducing an initial 10 mm chainline error to a 5 mm, by moving the hub 5 mm to the driveside, and that he wanted to evade the work of having to redish the wheel.

My question is thus to get an idea what a wheel correction for an eventual confirmed case (rear tyre not in trail of front - still have to find some way to check that) involves. I have had the rear wheels rim broken over the entire circumference, crack crossing all spoke holdes. The rim was replaced under guarantee, but coincidently - they forgot to remount the spacers and the cog - I discovered that the dealer had sent the wheel back to the producer to replace the rim. This would have been an opportunity to correct the wheel (in case aboves suspection true). I asked the dealer several times about the wheels centering, so far he kept evading giving a clear yes/no on the question of the wheel was centered or not. So it looks like I'm gonna need absolute proof again, just like the chainlines error case (see above).

3alarmer 

.
...you can easily check the dish on any rear wheel with which I'm familiar with one of these. Don't buy it for just one use, but find someone with a similar tool for help. Your dealer probably has one similar. But they only work on frames with symmetrical rear triangles. I don't think they work on those purposely designed off center frames.

Bike Gremlin

For DIY: one could place the rim on two cups, then stack coins, and/or business cards beneath, until hub's flanges touch them. Flip the wheel over and check for any dishing errors. +- one business card thickness (under 1 mm) is considered acceptable - at least in my opinion.

dabac

Checking wheel alignment isn’t overly difficult, particularly if you can enlist a helper.
1) remove tires from rims.
2) turn bike upside down.
3) find a piece of string, longer than the bike
4) while keeping string taut between two people - one in front of the bike, one at the rear, bring string in towards the rims. If bike is OK, the string should make contact just about simultaneously with each rim at two places. You may need to adjust front wheel (turn) angle some.
if string contacts one rim at two spots while leaving an equal space at two spots to the other rim, then the frame/fork is off somehow.