harshbarj

None. I use a hub till it breaks. Rims get tossed with the hub. I have a spare sturmey archer RXL-RD5 shell that I plan on building the next wheel on, then transfer the innards from the current wheel.

Trevtassie

"read Sheldon Browns Wheel Building Page" probably covers that, I figure if you are the kind of person who would do this on a store bought wheel, you're probably gonna check tensions anyway. And if it wasn't done from the factory, the same thing will happen with use, just slowly and unevenly.

cyccommute 

My point is that unless you have a fair bit of knowledge about wheel building, it's not a good idea to go doing something to the wheel after the wheel is built that will really mess up the wheel.

adipe

1. spokes break because they haven't been properly tensioned and therefore the nipples can unscrew whenenever spokes go slack from shocks;
2. even if nipples are glued to the spoke as to not unscrew the wheel will suffer - bearings at the hub too. the rim can twist if spokes go slack, the spokes can break from the small pings (not that you hear it) that occur whenever they go momentarily slack and then suddenly go back in tension at speed;
3. a wheel should not have spoke tension variance beyond a certain level because some spokes can go momentarily slack under load and therefore break from fatigue.
(later edit) a too large spoke tension variance affects the rim too because there is flexural stress on the rim which implies a portion of tensional stress and therefore fatigue. the rim is under compression (circumferential) but portions around nipples are under tension and it should be limited to a reasonable level.

so the solution is the following:
-140+ kgf DS spoke tension WITH tire pressure at normal psi (that would mean lets say... 155kgf DS with no tire on or tire pressure 0)
-spoke tension variance can be made as low as +/-5% especially with a new rim or if you work with patience and do your homework when repairing bent out of shape wheels even with visible flat spots; but i guess you could also settle to a maximum 15% total variance (+/-7.5%) if repairing bent rims;
-radial runout +/-0.1mm or even as low target as +/-0.05mm, that should be balanced with spoke tension variance;
-brass nipples because aluminum does not last; research galvanic corrosion and see that you cannot build properly with high friction between spokes and nipples;
-wax lubricate and maybe even mix some linseed oil for the nipples/spokes/eyelets; linseed oil has antibacterial properties and there are bacteria that degrade wax; linseed oil becomes sticky (polymerisation), a bit like glue over time so it should rather seal the rest of the mixture, not too much in the nipples/spokes threads;
-2mm DS spokes, butted spokes on the NDS - put labels on all of them as to observe the twist and ensure they don't remain under elastic twist;
-larger tires as to spread the load on a longer line on the rim;

always eyeleted rims because if you go old school and go high tension (170+kgf NDS, almost the same to a bit higher DS tension) and then lowering back to the target tension... non eyeleted rims are very fragile. remember that this should not be tried with aluminum nipples because of the higher friction (permamently twisting the spokes), even if you lubricate them, the lubrication will not stay put. ditch the aluminum nipples because of galvanic corrosion as well. this extreme tensioning ensures that the stress relieving is fixing the rim, beds in the nipples in the eyelets, the spoke elbows in the flanges... so that there are no residual stresses that remain.

and one important thing: stress relieve the wheel at every stage of raising the NDS tension after fixing the radial true at a moderate DS tension with rather slack NDS spokes; can't remember at which tension i did radial true but i guess about 70-75kgf would be alright. you might be lucky as to not need to dish later if you pick the average tension for radial true more exactly. ymmv. be sure to pay attention to spoke tension variance at that stage, you should try to somehow measure or at least estimate for each point on the rim delta elongation of the spokes and calculate how much the spokes are elongated from the tension variance as to balance these things out. if you want to do it you will find some solution and it can be done even without an expensive truing stand that is required for working with both accuracy and speed. if you build your wheels you can find other less expensive solutions.

adipe

...all this below is copy/pasted from a different post i cannot link now as i do not have yet 10+ posts. it is a guideline/method for weekend warriors that try building their own wheels without buying expensive tools but having time and some practical skills. for measuring radial runout for each spoke you could use an unexpensive laser line tool and maybe do it outside on a dark night for a 10 meter distance (say 5cm distance to the rim, 10 meters would make a 200x amplifying factor). the harder thing to do is to accurately measure the relative tension variance. you can disregard the error for not having an exact total tension for a spoke, all you really need is a low error regarding spoke tension differences which error would be low enough if all spokes are the same on the flange when using acoustics. there is a very very very small difference in spoke thickness variance at different tensions so all you have to do is discern between say 600 and 603 Hz using a tone generator and comparing it to the spoke vibrating between the nipple and cross. and if you have 2mm (that is constant because there is not butted portion) thickness for DS you can very accurately assess overall average final tension as well. here goes...
---

radial true should be sought before axial(lateral) true but you need to have the tension variance as to agree to the radial untrueness so to speak.
for example:
1.8mm spoke diameter (even though i'd rather recommend 2mm for DS);
290mm length;
179GPa modulus of elasticity in tension at about 70kgf average tension - when you should ensure radial true for a rear wheel or a disc front with the NDS spokes rather slack on the rear or non disc side for the front.
lets say one particular spoke is at 80kgf, therefore 10kgf (98N) higher than the average;

98N/2.544sqmm=38.5MPa higher stress than average
38.5MPa/179GPa*290mm=0.062mm - for 1.8mm thickness
0.05mm - for 2mm thickness

the rim would be ok to be farther out from the hub by this distance with respect to the average distance.
you should tolerate this runout for that portion of the rim with all these above parameters.

after you fix radial true at this lower tension (supposing you have a rather lateral true wheel) you raise the other spokes tension exclusively to a high level being careful to keep the lateral true. don't touch the DS nipples (for a rear wheel) after you went beyond fixing the radial true.

you might need to go back to fix radial true after the nipples have bedded in the eyelets and the spokes in the hub flanges in an other than uniform way.

you either fix the dish before the last stages of seeking a high tension (in order to be able to straighten the rim) and then backing to a lower level after all goes well or you fix the dish after all this.

it's useful to stress relieve at each stage of raising the tension - remember to don't touch the nipples of the spokes that are in high tension - DS rear, disc side front.

and the most important thing: don't use nipples other than brass and wax them. linseed oil would work but you would have a hard time later if you need to retrue. wax lubricants are best and its best to leave the wax solution be able to dry on the spokes thread and the nipples before assembly.

it's not very hard to achieve +/-0.05mm lateral and radial true with +/-5% spoke tension variance but you need to be careful to balance out the DS tension variance with the radial true from the get go.
find (google) a tone generator - i'm not allowed to insert links as i don't yet have 10 posts.

spokes vibrate together, when plucking one keep the other quiet to be able to discern which is higher and which is lower you sometimes need to pluch each one several times to discern between them.

find this equation:
T = 4μL²f²

tension as a function of linear density, length of vibrating wire, frequency.
google to check it out.

example:
T=4*600*600*0.15*0.15*3.14159*0.9*0.9*0.008*3.14159...
where
600 = frequency, Hz
0.15 = length between nipple and cross, m
0.9 = spoke radius (1.8mm/2), mm
0.008 = density for the stainless steels in respect to sqcm being 1000*sqmm
---
result: ~659.58 Newtons = ~67.3Kgf
for 603 Hz the tension (all these parameters above being kept) would be 67.6kgf

thats a very small difference (error) supposing you are careful and there aren't other noises/distractions and therefore are able to discern 603 vs 600 Hz.

radial trueness is fixed at a medium spoke tension for other reasons as well.

you don't need to worry about accuracy at the end tension values for spoke tension variance in regards to adjusting radial true considering spoke tension. radial true should not be normally be fixed at high tension for different reasons. if you need to go back to fix radial true you should lower the NDS tension first.

make sure you don't have a lateral untrue wheel when fixing the final radial true - with spoke tension variance being taken into account.

a new wheel could need two stages in fixing radial true because of bedding in - nipples/eyelets and flanges/spokes.

don't forget to stress relieve at every stage of raising tension NDS wise: raise tension, fix lateral, stress relieve.

put labels on DS spokes as to make sure they don't remain twisted. you might as well put numbers on them to keep track.

put labels on rim to be able to adjust dish in exactly n*90 degrees of spoke key turn. mark lines on the labeled rim as to be parallel with the spoke key and therefore end with exactly n*90 degrees for all nipples on the flange for this matter. you could use this trick for balancing radial true with spoke tension variance and therefore need to apply either another label on the rim or a different colour marked line.

you can neglect error due to spokes being thicker at the nipples; the error spread around all the spokes;
be careful not to mix different spokes as to be sure to have them at the same exact diameter;
you can neglect the very very very small variance of the spokes diameter at different tensions. if you doubt this you can research and even use online calculators. you only need to google "poisson's ratio" to find out the matter;

the only way you can be in considerable error is in regard to the estimated average tension because you won't be very accurately measure the distance (0.15m) for the vibrating chord etc. but for relative tension all this works very well.

you need not worry that you don't have an exact 140kgf average DS tension and it could be 136kgf etc.
you need to worry about uniformity in tension.
and you need to make sure stress corrossion does not happen. treat the surface on the inner rim around the eyelets with something to prevent corrosion.

i might have been slightly offtopic but all this for the sake of ensuring a long lasting wheel.
stress corrosion and fatigue being affected by the chemicals that need to be kept away from the vulnerable surface are not to be neglected things.

don't ask me exact figures for the fatigue being affected by corrosion, just make sure you deal with it. treat the surface of the rims with something no matter the spoke tension you choose to have as a target, 120kgf as written in the books or 150+ as what professionals use as a target BUT with very small tension variance achieved by proper stress relieving (done best in stages in raising the NDS after accurately balancing radial true with tension variance from the get go) and the lowering the tension after seeking a high tension and therefore cold working (strain hardening) the rim when straigtening small imperfections.

150kgf DS tension is needed to ensure the wheel is long lasting with a heavy load. no going untrue, no spoke breakage. if proper cold worked 1.5mm spokes do not break under heavy braking in DH use then surely you can't say thicker spokes need not be ensured a high tension so that they go at a very low amplitude in tension (delta, cyclic). the lower static tension the more amplitude in dynamic strain and the more spokes hit the flange after going completely slack under dynamic loads.

don't ask me exact figures for aluminum vs brass nipples. avoid aluminum nipples like plague for wheels that you build expecting them to last for decades. aluminum nipples are ok for competition, just to shed some grams.

if you don't believe me you can do your own research.

adipe

forgot to say that the thread pitch is usually 0.45mm for spokes that have normal nipples.
so for example 10 degrees (a full circle being 360 degrees) would make 0.0125mm distance. you surely can assess the angle when screwing/unscrewing the nipples after calculating the adjustments to be made to the spokes after measuring radial runout for each DS spoke and each tension difference (to the overall average calculated after) as well.

there are spokes that have a constant 1.8mm thickness and a different pitch so be sure to not blame me for those other very rare cases.
i only point out the method and principles, you can figure out all the other details.

if you are lucky as to have access to a truing stand with precision indicator gauges so that you need not find out other ways to measure in hard figures individual DS radial runout and a spoke tension meter then you can at least do the work in short time. a spoke tension meter is more useful than acoustics only if the measurement is at least as accurate because the main advantage of the mechanical (flexural strain) tension meter is a faster measurement.

i managed to finish it after taking the time and having no hope (also being on a low budget) of finding a LBS to do it this way and not trick me, taking my money for a low quality repair.

all this because i had to repair a rear wheel on a bike i bought and now the NDS goes only 18% slacker in respect to the load on the wheel (measured it at different loads). that would make about as large a load (dynamic load, not static) as 400kgf until the straight to the ground NDS spoke goes slack. so i expect to never have to repair it again. i could very well curse if someone steals the whole bike though... IF. not again!