Wheelbuilding tips - wheels going out of dish after build

mirv

I bought a Shimano R501 wheelset recently for its hubs, spokes and rims. The plan was to use those rims to replace the rims on a campy wheelset (the front rim wasn't machined right, so it pulsed when braking) which has novatec hubs, sapim laser spokes. The R501 hubs and spokes would then go to replace the hubs on a set of Shimano R500 wheels, and finally I would rebuild the original campy rear rim onto the R500 hub for use as a trainer or beater wheel.

Long story short, I took apart four, and built five wheels this weekend - R501 rims on novatec/2-1.5-2mm sapim laser spokes, R500 rims on new R501 hubs/2mm straight gauge spokes, and a beater wheel on R500 hubs/2mm straight gauge spokes. I'm using 12mm brass nipples for all the builds. All front wheels are 20H radial, and the rear wheels are all 24H 2x.

I'm able to get the wheels built fairly straight and round - probably less than +-0.5mm lateral true, and a bit less than +-1mm radially true. I used a little drop of fairly heavy oil on the spokes and nipples when building, and I tried my best to avoid spoke wind-up by turning 1/8 of a turn in the opposite direction after each adjustment. I'm using a turbo trainer for radial and brake blocks for lateral trueness, and flipping the wheel over to fine-tune dishing.

I did notice that my R501/novatec/sapim wheelset is going out of dish a few hours or overnight after building by maybe 2-4mm (at one point maybe 10mm but I decided to detension it entirely and start over), usually towards the drive side. The wheels stay laterally and radially true though despite the massive change so I don't think the nipples are unscrewing or the spokes are unwinding. I've washed the wheelset thoroughly to remove any leftover spoke lubricant in case that was the issue. The shimano wheelsets have done this to a lesser extent.

It feels that the spokes are stretching or 'straightening' after my build - perhaps from bending when lacing or crossing, but the effect is much more pronounced with the very thin 2-1.5-2mm sapim laser spokes than with the straight gauge 2mm ones.

Has anyone else experienced this sort of out-of-dish but otherwise perfectly true effect when building wheels, especially with very light spokes?

cdaly_

Last wheel I built, I did a spoke 'de-stressing' exercise where I squeezed pairs of spokes (wearing a pair of leather work gloves) at their mid-points to settle them in after the build was done.

I then put the wheel back on the building stand and re-checked/adjusted true and dish.

I used a home built trueing stand and gauges from the designs included in Roger Musson's Wheelbuilding book (only EUR10). Stand/gauges cost me about EUR5 for bolts and a couple of hours cutting up an old chipboard cupboard...

browsing

Try to "de-stress" the wheel as @cdaly_ suggests. it's very important. Also, I'm not sure how you tackled it but try to do little adjustments and then move onto the next worse part of the rim etc. Are you following a good guide? Try the one @cdaly_ suggests or you could try following the one on sheldonbrown.com. Sheldon recommends only lubricating the drive side as the tension is higher. If you're lubricating both sides perhaps that's the problem.

mirv

Done lots of googling recently, but I'm pretty sure that spoke/nipple lubrication is an ecumenical matter and the nipples are definitely not unscrewing themselves. If the spokes were to unscrew themselves from too much lubricant, then every single spoke one the NDS of the wheel (or every single spoke) would need to unscrew themselves uniformly in order to leave the wheel radially true and straight, but out of dish.

If anything, the sapim laser built wheelset have higher spoke tension than the 2mm straight gauge wheelset, but seem to be worse affected by the out-of-dish effects. I did bring the spokes up to tension in small steps rather than in one go, but I could try to be even more careful next time I sit down with the wheels as I try to bring them back in dish.

browsing

Love the Fr. Ted. reference! I'm no expert either as I've only built 1 wheel. However, I presume that lateral trueness is being affected too and not just a case of being out of dish? i.e. the 10mm buckle is typical of what's happening? Are you lubricating both sides then? For what it's worth, I used grease instead of an oil but can't imagine there's much difference.

mirv

I did say what happened in my original post, but perhaps you got lost in the large chunk of text!

The wheels were built with even tension, to < +-0.5mm of lateral true, and < 1mm of radial true, and dished correctly. After leaving them for a few hours or overnight they would go out of dish by between 2-4mm (and once 10mm, but I started over on that wheel). Despite being out-of-dish they would be both radially and laterally true or "straight" as I have said.

he rims are not "buckled" by any measure and you would not notice the error in dish unless you put them in a frame in order to sight them using a reference, unless you can eyeball 2-4mm dish from the spoke angles.

This is why I don't think lubricant or nipples unscrewing is related to the wheel going out of dish, unless they were unscrewing themselves in an unlikely uniform fashion. If anything, I used oil rather than grease so that the lubricant will wash or drain away in time after building.

Perhaps its the rim settling into the additional tension over time, or that the spokes are relieving their stress or straightening out from their slight curvature when lacing. Either way I'll dish them slowly and carefully and report back!

cdaly_

mirv wrote: »

I'm using a turbo trainer for radial and brake blocks for lateral trueness, and flipping the wheel over to fine-tune dishing.

Is there any chance the turbo trainer axle clamp is moving the wheel slightly when you flip for the dish check? You might be better off making the Musson dish gauge which checks against the axle face and rims rather than checking rim position against a surface as your turbo trainer method does.

The gauge is just a matter of a cardboard cut-out and a pencil...

You could imagine making something like:

and just use a pencil held against the axle end rather than a built in screw...

pprendeville

cdaly_ wrote: »

Last wheel I built, I did a spoke 'de-stressing' exercise where I squeezed pairs of spokes (wearing a pair of leather work gloves) at their mid-points to settle them in after the build was done.

I then put the wheel back on the building stand and re-checked/adjusted true and dish.

I used a home built trueing stand and gauges from the designs included in Roger Musson's Wheelbuilding book (only EUR10). Stand/gauges cost me about EUR5 for bolts and a couple of hours cutting up an old chipboard cupboard...

I second that book. Great advice in it and diagrams on how to build your own wheelbuilding tools.

browsing

Ah! I misread your post! I presumed it was fairly buckled and out of true. I thought one particular point on the rim (as opposed to a point in time) was out laterally by 10mm!!

mirv wrote: »

I did notice that my R501/novatec/sapim wheelset is going out of dish a few hours or overnight after building by maybe 2-4mm (at one point maybe 10mm but I decided to detension it entirely and start over), usually towards the drive side.

That is strange though. Perhaps there is some tension still built up somewhere. I.e. between nipples and rim or something that works its way out. You could perhaps try putting some lateral force on the hub to help relieve this. I came across this interestingly whilst watching a pro-lite "how its made" kinda video on YouTube. Fast forward to 3:30:
https://www.youtube.com/watch?v=H_q69F4HuOo

mirv

I think I found the culprit. It was the spokes bedding/straightening/stress relieving (take your pick) after truing. I approached it a bit more methodically last night and the thinner 2.0-1.5-2.0mm butted Sapim Lasers are far more prone to it than straight gauge 2mm spokes. The fact that I was building 20/24H wheels probably added to the difficulty of the build somewhat.

I ended up relieving the spokes by squeezing parallel pairs of spokes together quite hard, re-dishing, and repeating until relieving the spokes stopped causing any changes to the dish of the wheel. I also increased the spoke tension of my rear wheels so that the NDS spokes would relieve themselves a bit faster or to their permanent state quicker.

I've built Open Pro 32H rims on campy veloce hubs with 2.0-1.8-2.0mm spokes before and haven't had as much trouble with this before, but I think the higher spoke count and thicker spokes are a lot more forgiving on beginner builders!

mirv

browsing wrote: »

Ah! I misread your post! I presumed it was fairly buckled and out of true. I thought one particular point on the rim (as opposed to a point in time) was out laterally by 10mm!!

Yeah I see what you mean, but I would call that symptom 'out-of-true' at one spot, while being in dish for the majority of the wheel. In my head, lateral trueness is how straight the rim is laterally without giving any regard to dish.

mp31

This is great timing... I'm about to build by first ever set of wheels this w/e...in fact I'm sooo excited that I may just try lacing up one tonight.

I've Roger Musson's book as well and he doesn't use a spoke tension meter and instead suggests pinging spokes until they sound like the spokes on an existing wheel.

Does anyone here do that or do you use a spoke tensioning meter yoke.

mirv

Make sure the crossed spokes are crossing each other properly, i.e. elbows out spokes tuck underneath elbows in spokes.

doozerie

mp31 wrote: »

This is great timing... I'm about to build by first ever set of wheels this w/e...in fact I'm sooo excited that I may just try lacing up one tonight.

I've Roger Musson's book as well and he doesn't use a spoke tension meter and instead suggests pinging spokes until they sound like the spokes on an existing wheel.

Does anyone here do that or do you use a spoke tensioning meter yoke.

I use either, or both. I started out doing it by ear and found it a reliable method. Later on I bought a tensiometer (the Park Tool one) and tried using that exclusively for a while but I found it slower and more tedious.

I find the tensiometer useful if you want to compare two spokes with greater accuracy, or if you want to get a reference value from another built wheel, but mostly I use my ear still and then use the tensiometer to double-check at times.

cdaly_

mp31 wrote: »

I've Roger Musson's book as well and he doesn't use a spoke tension meter and instead suggests pinging spokes until they sound like the spokes on an existing wheel.

Does anyone here do that or do you use a spoke tensioning meter yoke.

I used pinging and all seems reasonably well...

mirv

I use pinging as well. When there's one tight spoke surrounded by two loose spokes on the same side, it's pretty obvious to average the tension of the three while maintaining radial and lateral true if possible. Although occasionally there's sometimes a couple of spokes that are always going to be tighter or loose than the rest, especially where the rim is pinned/welded or at the valve hole.