himespau 

I'm going to be building my first wheelset and have the hubs in hand (waiting on delivery of cone wrenches to service them) and the rims are in the mail. Now I want to order some spokes and nipples so I can get rolling.

So far, I've used 3-4 different spoke length calculators and they've all given my slightly different estimates for how long of spokes I should order.

I've used:
https://www.prowheelbuilder.com/spoke/index.php
https://www.wheelpro.co.uk/spokecalc/
https://www.bikeschool.com/tools/spoke-length-calculator
https://lenni.info/edd/

For example, for the front wheel I get values from 267.2-265.8 (first two calculators, respectively) for the spoke length depending on the calculator I use. For the rear, I'm seeing 266.3L and 264.5R; 264.9L and 263.1R; and 265.4L and 263.6R.

My hubs are older shimano 105 hubs hb-1055 and fh-1056 and my rims are mavic xm 317. I've found good info from a shimano site on the specifications of my hubs, but the mavic site only gives spoke support diameter not ERD (and I don't know how to convert them), so I've just used the ERD from the two calculators that have the hub in their database.

Given that these will be the first wheels I'll have built I don't have spare spokes/nipples lying around to do the whole ERD measurement trick myself by measuring the distance between two of them, and I know my tape measure doesn't give me the level of accuracy needed to measure the whole distance, so I'm sort of at the mercy of the calculators.

So is there one that I should trust more than the others?

I will be getting my spokes from that nice guy on ebay who seems cheap enough and cuts them to length for you (still debating between straight 14g and butted, but will probably go butted as the cost difference isn't that huge).

SumoMuffin

I'm not trying to hijack the thread, but I'm actually about to undertake the same project for the first time. I just wanted to ask an additional question.

How much leeway is there usually for spoke length? and is it better to be over or under the recommended length?

himespau 

That was my other question. Do you round up or down? Or to the nearest number either way?

For my numbers there's basically a 1.5 mm difference between the high and the low and I don't know how much room there is for error in that.

seeker333

If the inputs are the same, the results should be the same; the math is simple. I use this calculator - it has a good hub and rim database, although the data is old:

https://sheldonbrown.com/rinard/spocalc.htm

Pay attention to the dimensions used in these calculation - they may not be accurate. Shimano hub dimensions don't change often and are usually well known. Other hubs, not so much.

Same goes for rims. Sometimes the ERD varies for a given rim model by more than 2-3mm due to manufacturing variance or design changes. Your best bet is to measure it yourself - measure twice, at 0-180 and 90-270 deg, take average. I always measure rims myself since the one time I got burned from relying on bad ERD info and ended up wasting money on wrong size, unusable spokes. This slows your project down since you have to have hub and rim in hand before you can order spokes. If you want to keep wheel costs low, you need to anticipate and build spares well before you actually need them. This allows you to take your time and buy parts when available at lowest price.

https://sheldonbrown.com/rinard/spoca...asureHubAndRim

Spokes are often sold or stocked at more than 1 mm increments. If you have to choose, go 1-2mm shorter rather than longer. Spokes usually slowly elongate and detension with use. They're just thin gauge wire subject to stretch under load. If they're too long you could run out of thread to tighten - so shorter is better.

Ideally you want the spoke threads centered in the nipple with 100% engagement. Spokes are usually threaded longer than the nipples, so this is not hard to achieve if you are careful in measuring component dimension and selecting proper spoke length.

Finally, do yourself a favor and use brass nipples. Al nipples are much more likely to break, and they are hard to remove when broken

SumoMuffin

One more question from me: How much leeway is there with spoke diameter? Say you have 2.6mm diameter holes on the hub flange, how low can you go with the spoke diameter?

Drew Eckhardt 

It depends whether you're using brass or alloy nipples.

DT spokes have 9mm of thread, and DT 12mm nipples (the standard length) have 8 including the slot so a spoke will bottom once its end protrudes 1mm pat the end of the nipple (2mm beyond the end of the slot).

With alloy nipples ending up 1mm below the nipple slot after the spoke has stretched you're eventually going to have problems with broken nipples, so you given the choice of ending there or at the end of the nipple (due to spokes sold in 2mm increments) you round up aiming for the end of the nipple although this gives you just +/-1 mm of tolerance.

With brass nipples you don't have to worry about breakage, although running out of threads is equally bad so you aim for the slot and round down thus allowing for spokes 2mm too long (or rims 4mm too small in ERD) and ones that are even shorter.

Spokes don't elongate since they aren't loaded anywhere near their elastic limit and the tension drops when you roll along the ground or over obstacles.

They stay exactly the same length and maintain their tension until the rim requires replacement because you bent it or wore out the brake tracks.

You might have problems with the nipples unscrewing due to insufficient tension or failure to counter wind-up, but those are build problems not "stretching."

You might be able to bend outbound spokes around the hub flanges in use so they form a straighter and shorter line with the nipple sockets in the rim although that's also a build problem (you neglected to correct spoke lines) and not stretching either.

ultraman6970

I built my own wheels and many times what I do is use the DT, the sapim and one from another guy, if there is 2 mms difference I just take the average or just go with the smaller one, all depends too of what the LBS has available too. +- 1.5 or 2 mm wont screw up the built just in case, the only thing I would be worried about is if the spoke don't go inside of the hole at least 2 or 3 mms because if its at the edge of the rim pretty much the nipple is the one that will be working too much and at some point it will snap, actually sooner than later.

krome

I use the spokecalc that is on the Sheldon Brown website. I think that whatever you decide on, stick with one. After a few wheels under your belt, you can learn the calc's "bias". I typically round down to the nearest 2mm if that is what is available. If 1mm increments are available, use that. There are two common lengths of nipples, that can give you some extra wiggle room.

I uploaded the spokecalc spreadsheet to my google docs. I add my own data as I need it. So far I haven't had a problem with spoke lengths.

FBinNY 

The basic math on all calculators is the same, but there are subtle variations when correction factors are added.

The biggest adjustment is accounting for the height of the nipple head. Some assume you're entering the diameter at the top of the nipple and bring the spoke back to that same height, others assume you're entering the rim diameter at the nipple seats and add 2-3mm for the height of the nipple to the calculated length.

Also, the spokes are usually laced over/under and do not take the shortest route to the rim. Some calculators add a bit for that extra distance, some don't.

Those are two reasons that entering the identical data in various spoke calculators will yield different answers, within a range of 2-3mm. I always use the same calculator, because once I know how it's results compare the real world results and know exactly how much to adjust the given length.

Drew Eckhardt 

It depends. If you aim for the end of the nipple 1.5mm will be enough to bottom the spoke. If you aim for the nipple slot -1.5mm will break alloy nipples.

ultraman6970

I agree, that's why i mentioned that if the spoke is at the edge of the rim then the nipple will snap sooner than later.

davidad

I use spoke calc from SB's site and the ERD on the DT site is 604.

mrrabbit 

https://www.mrrabbit.net/wheelsbyflemingapplications.php

=8-)

himespau 

Hmmm, the guy I've been planning on going with cuts/threads them himself, so I guess I'll have to ask him what increments he cuts them at. If he goes with 1mm, I think I'm going to just go with the median value and round down as the numbers are all so close. I guess I'm sort of surprised that the drive side spokes are only going to be 1 mm shorter than the non drive side. I would have thought there'd be more than that with dishing and all.

mrrabbit 

Pablo uses a Phil Wood spoke machine...he should be able to cut to .5mm increments if you really want.

=8-)

3alarmer 

himespau 

On a related question. I have some cheapo avenir (I think) spoke wrenches that grab on 3 sides. I was reading that for building a wheel you really want some of the better wrenches that grab all four sides so you don't strip things. If I buy normal 14g spokes with the standard brass nipples, is there one size of wrench I'm going to need? I'm sort of confused by the sizing of the wrenches/colors of the handles as they don't just say 14g or whatever on the ones that I've seen (or maybe they do and I'm just blind).

Bezalel

For typical builds with brass nipples a standard spoke wrench is enough. For aluminum alloy nipples or low spoke count wheels that require a higher tension a 4 sided wrench helps to avoid rounding the nipples.

rydaddy

Sorry but I gotta call you out on this. Spokes elongate linearly with tension until reaching the elastic limit (ie yield). Simple stuff.

I tend to round down my spoke lengths no more than 1 mm for this reason.

As for spoke length calculators, I use spokecalc and the DT swiss one, and sometimes edd. I always use 2 of the 3 as a check.

mrrabbit 

Are you observing actual spoke stretch...

or are you observing:

1. Nipple gouge into the rim each side of approx. .5mm for single wall non-eyeletted rims
2. Inward eyelet deflection of about .25mm under tension each side.
3. Elbow lean of approx. .5mm for standard spokes in thin flanges.
4. Failure to factor a deduction for larger diameter holes when calculating lengths.

=8-)

Before you can claim stretch - you gotta eliminate other stuff first.

=8-)

rydaddy

No.
No.
No.
No.

Simple mechanics of materials is all I'm saying. Whether or not it matters is up to the wheelbuilder.

=8-)

mrrabbit 

All the knowledge of materials mechanics in the world still dictates that you eliminate other things as part of making your case. It's no secret that nipples eat on average a 1/2 mm into the rim material of single wall rims - creating the illusion that the spokes have stretched because the spoke ends end up higher in the nipple than the builder was aiming for.

Bottom line is - you need to equivalent of a lab test environment to show that for a typical bicycle wheel - 15g and 14g spokes stretch a considerable amount (enough to drive calculation corrections) absent all other possibilities.

=8-)

rydaddy

I completely understand your point. More than you probably credit me. However, I never said spoke stretch was considerable. Just that it exists. Contrary to Drew's post.

himespau 

Well, the guy says he stocks butted spokes in 2mm increments (though, if I want, he could cut some down to 1 mm differences, but it's not usually necessary to be that precise), so I'm probably going to go with:
36 266 mm spokes (32 spoke front wheel, the 3 lengths calculated were 267.2, 266.2, 265.8 and this is the first wheel I'll build so I'll want the 4 spare spokes)
19 264 mm spokes (32 spoke wheel, non drive side, the three lengths calculated were 266.3, 264.9, and 265.4, this will give me 3 spares)
20 262 mm spokes (32 spoke wheel, drive side, the three lengths calculated were 264.5, 263.1, and 263.6, this will give me 4 spares as these are the spokes most likely to break, right?)

Does that make sense? Basically, I'm taking the average of the 3 calculated lengths and then rounding down to the nearest even number.

I can get 75 double butted 14-15-14g steel spokes for roughly $40 shipped, so that's why I'm distributing them the way I am. Is that a sensible arrangement for my spares? 4 for the first wheel I'm building because it'll be my first and I might screw it up, then 4 for the drive side and 3 for the non-drive side rear as the drive side rear is where they're most likely to break while riding?

Drew Eckhardt 

Right, they elongate when you initially tension them - up to a full millimeter for 1.5mm butted spokes like DT Revolutions front and rear drive side.

They will not stretch with use as a previous poster stated.