ppg677

I bought some Light Bicycle AR56 carbon rims. ERD "not including nipple head dimension" is stated as 527mm.

Now I have attempted to measure myself using the technique of using 2 spokes threaded into a nipple (where I thread the spoke to just below the screwdriver slots). And I've done this a couple times now and got 533mm and 535mm. Clearly some measurement error as I'm not getting perfectly repeatable measurements. But in any case, I am definitely seeing higher than what Light Bicycle states "not including nipple head dimension".

I guess the question is whether online spoke calculators assume the nipple head dimension is included?

I did ask Light Bicycle sales rep for suggested spoke lengths with the hub specs I gave them. And I got results that more-or-less match other online spoke calculators assuming an ERD of 527mm.

Advice?

Andrew R Stewart 

I don't know this rim specifically, and haven't laced up more than a few pairs of carbon rims a few years ago, but each wheel I have built seemed to have a lot of spoke "bedding in". Meaning it took a lot of nipple turns after the slack was taken up to when full tension was attained. It felt like more than common the spoke stretch amount, as though the nipples were having to bed into their rim hole seats deeper than would be the case for an Al rim.

So I might follow the rim brand's advice in this case. I hope you'll be using Brass nipples... Andy

79pmooney

Maybe some have higher expectations than I have. I never go by published ERDs. The rim in hand is what matters. I also take two measurements perpendicular to each other and average them. Rims, for the most part, aren't perfectly round.

Now, all of my experience is with aluminum rims. The carbon fiber ones may be far more uniform. I use to build fiberglass boats using molds. That experience does not enhance my faith that the carbon fiber stuff from similar molds is always going to be right on perfect. And molds wear out and get replaced. New mold, different ERD, same old publicized ERDs? Totally believable in my mind.

Edit: I've never had to deal with more than trivial bedding of nipples in the rim. Andy's post and your measurements make it sound like there might be about 5mm of it here. I'd listen to the manufacturer as this is something they would know well.

Russ Roth

I'm not sure that many calculators take into account the spoke head nor do I tend to think they need to. When I built up some mavic carbon rims I used mavic's recommended erd and the spokes came out just over 2mm too long and unusable for my purposes. Found out mavic adds about 4 or 5mm for the spoke head. When I measure using spokes, assumeably similar to how you do it, I measure to the flattest section on the bottom of the spoke nipple. Between compression of the rim from being pulled in and stretching in the spokes, the end of the spoke nipple is usually close to the bottom of the screwdriver slot in the nipple when built which holds up fine. A couple things I may be doing different; I bottom out the spoke nipple on the spoke so I know exactly where the nipple was set when I lay them out to measure, I also make three marks on the two spokes I line up so when I pull them out to lay together I can be more certain of their exact alignment before measuring, and I use three sets of opposite holes and average them. Personally, I would lean towards using LB's measurement.
Not sure why someone said only use brass, mavic supplied nipple washers but they build with aluminum nipples, lightbicycle has no problems building with aluminum and even said they don't bother with washers, and the pub rims when I bought them also said that as long as the nipples are anodized there isn't an issue with aluminum and they don't bother with washers either. Contact LB to verify, they're always helpful, but I wouldn't worry about nipple material.

Bill Kapaun

I'd take a tape measure and check the extreme outer diameter in a couple different spots.
It seems conceivable to me that it could be a bit out of round.

FBinNY 

Most (don't bet on all) spoke calculators yield a spoke length where the spoke ends at the ERD, or top of the nipple head, or 1mm shy of that.

As for measuring ERD, I prefer to measure down from the rim's outside edge, then subtract 2x ( radius v. diameter). You can measure with or without the nipple, just be consistent.

Kontact 

The problem is that ERD has two different definitions in common use - and they are 4mm different. One to the base of the spoke bed, the other to the assumed base on the nipple head.

ppg677

which do I want to use for spoke length calculations?

Kontact 

The best you can do is read the documentation of the spoke calculator.

However, these articles make it sound like the error is very small if you are building a cross spoked wheel rather than radial. So maybe it doesn't matter? But I think it is safe to use an ERD that includes the heads of the nipples because spokes that are 2mm too long are a better choice than spokes that are 2mm too short.

https://spokecalc.io/wrong-spoke-length.php
https://spokecalc.io/measuring-rim-erd.php

FBinNY 

As I said in post 6. Most calculators have the spoke end at the ERD entered, or 1mm shy. So, you'd want to include the nipple's head.

However, rather than roll the dice, try various online calculators. If you find some with differences of 3mm or so, that would be the nipple, with the shorter result assuming you included it already, and the longer one adding it for you.

FBinNY 

You're better than that. I gather it was an example of typing faster than you were thinking.

I don't know or care what the articles say, the difference is significant and always 3-4mm, regardless of pattern, since that's the thickness of the nipple.

Like you, I prefer to err long on double wall rims. But that requires nipples with threads short enough to all overrun of 2-3mm. Otherwise, running out of thread when nearly finished is beyond frustrating.

ppg677

I've always used the ERD quoted by the manufacturer which I don't think includes the nipple's head. The only difference here is that I plan on using expensive Sapim CX-Sprint spokes rather than cheapo DT Competition (where I wasn't too concerned about nailing down the right length because I wasn't out that much).

Either that or my technique to measure is just not very good...because in reality...spokes under high tension are going to have zero slack whatsoever and maybe even stretch a bit. And the spokes I used to measure ERD were fairly straight but not super straight (since they were used).

FBinNY 

If you've already built a wheel using a calculator and still have it, you have all the info you need.

Start by confirming the ERD, by measuring down to the top of the nipple, and subtract double that from the measured rim OD. The spec will either match that closely or be about 8mm less if it excludes the nipple.

Double check by running the same data set, and (hopefully) you know the spoke you actually used.

Now, assuming you were consistent, note where the spoke ends in the nipple. Look at where the spoke ends and decide if it's just right, or if it's short or long by Xmm.

This is now your correction factor which you'll add or subtract whenever you use the same calculator with comparable data, ie. ERD with or without nipple.


BTW - because I've lost track I went back and read your OP. Your measurement of 533-535 is perfectly consistent with thd rim spec of 527, with 7mm being twice the nipple head engagement. Your 2mm variance is normal because a use inbuilt rims typically a bit out of round.

Lastly, if you're concerned that the thinner spokes will "stretch" more, don't be as the difference is tiny.

So, immunize yourself against the "too much info" disease and build your wheels before you talk yourself out of it.

Kontact 

My bad - I misunderstood the example given. It certainly will change your spoke length be 2mm.

I do stick by my assertion that using an ERD 4mm too small is unlikely to tank your wheel build because 2mm of extra spoke is unlikely to bottom out the nipple. But I think we agree on that.

FBinNY 

I happen to disagree. Regardless of pattern, spokes reach the rim at close to radial. ERD differences change spoke length by almost exactly half the difference (ERD is a diameter while spoke length is a radius).

The opposite is true for hub spoke hole diameter, where the effect varies from 0 for full tangent, ie. 4x36h to half the change for radial. I've long suspected that this is one reason that 4x36 was such a popular pattern for decades. I, for one, loved needing only 305mm spokes to build a large variety of wheels. Aaah the good old days......


BTW my "your better" was simply an acknowledgement of your knowledge on the subject, whether we agree or not.

Kontact 

I was editing when you were posting. See above. I was wrong.

FBinNY 

No problem. As I said, I chalked it up to fast typing.

Mtracer

First off, I've actually used Light Bicycle rims and found them to measure very close to spec. I don't understand exactly how you measured this. And maybe your technique is a given and needs no further explanation for those in the know. But what I've done is made a set of measuring spokes. It makes no difference how deep the spoke is threaded into the nipple. I just threaded them in a nipple and applied thin CA to glue the nipple to the spoke permanently. Then I measured from the bottom of the slot in the nipple head 300 mm and cut the spokes off. This length of 300 mm just needs to be a known value and longer than half the ERD.

Note: Because the nipple seat is tapered as well as the holes in the rim, the bottom of the slot seems to line up pretty well with the rim bed. So, this is one reason this measurement should be pretty close to what Light Bicycle says.

Then I thread them through the rim, opposite each other of course. With the rim laid flat, I used whatever spacer I need to hold the spokes up off the work surface and inline. Knowing the exact length of the spokes, I measure how much they overlap. So, if the spokes were 300 mm long and overlapped 70 mm, the ERD would be 2x300 - 70 = 530 mm.

Ultimately I've used the DT Swiss calculator results, but as others suggested, I've used others as a reality check. DT Swiss provides full details on exactly how the measurements are made. In fact, here's the DT Swiss drawing of how they define ERD for their calculation. It is showing measurement to the bottom of the slot in the nipple. So, if you cut your measuring spokes to length referenced to this slot (as I did), you're going to get exactly the measurement needed for the DT Swiss calculator.

To minimize error, I also use DT Swiss spokes and nipples, and hubs. FYI, DT Swiss nipples get a little weird in that I think if the nipple is 2 mm longer (like 14 vs 12 mm), the threaded part is only 1 mm longer and the un-threaded part is of course also 1 mm long. At least it's something like that. Of course, their calculator factors this in as well as I think spoke stretch for the smaller bladed spokes. Point is, there are these details to double check if working with other spoke and nipple manufacturers.

Kontact 

So DT goes against other industry sources and does not measure ERD to the base of the nipple. In fact, it looks like they split the difference.

FBinNY 

Not sure we can criticize DT here. The rim is countersunk, so the nipple pockets so low that the slot lines up with the edge if the hole.

As a result, we can't know what DT is thinking.

The entire ERD debate is an example of why I have the Admiral Hopper quote under my signature here. We simply can't know what rim ERD specs really mean.

While many in the industry argue that ERD is properly measured to the top of nipple, I feel that's inherently flawed. IMO, specs (all specs) must conform to the four corners rule, so a rim ERD must be based on the rim alone. Otherwise, it includes an unreliable variable of the ASSUMED nipple height. The exception would be if the rim demanded a specific nipple, and that were explicity stated in the spec.

So, my advice is to trust nobody and measure for yourself.

Kontact 

The biggest problem with ERD isn't measuring, it is knowing what the calculator thinks you should have measured.

Regardless of the nipple design, a spoke that penetrates the base of the nipple by 2mm is going to prevent nipple failure just as well as any other nipple. So I can see the argument for just saying "ERD is the shortest distance across from spoke end to spoke end". But it isn't an argument, because different groups are defining it differently.

DT is not measuring to the rim diameter or the end of the spoke. They are measuring to the slot, which is about 1mm short of the spoke end. But then their calculator is going to reflect that type of measure.


Does anyone know how the old Wheelsmith yellow sticks measure out? Are they 350mm long? Or 348mm to reflect the 2mm of nipple head thickness? That's the closest to an industry standard, because nearly every shop wheelbuilder has used them to get ERD at some point.

FBinNY 

Yes, agree 100%.

This is the other side of the wooden nickel.

The published spec is dubious, then we have to guess what the publisher of the calculator guessed the spec writer had in mind. This is like the fictional device where the hero chases the villian onto a train and has to decide whether he jumped back off or not.

For this reason I've long advocated messuring rims yourself, then always using the same calculator.

Once you've built the first wheel, compare the results to your expectation to find a correction factor to apply from then on.

For example, if you expected the spoke to end 1mm shy of the top of the nipple, and it ended 2mm beyond the top, you now KNOW to subtract 3mm from all future calculations consistent to that first one.

ppg677

I did try the Sapim calculator (https://www.sapim.be/spoke-calculator) which uses a DIFFERENT method of measuring the inside diameter, and then entering the thickness of the rim (at the nipple bed). The best I can measure my rim with a tape measure gives me 520mm. I think the thickness of the carbon at the nipple bed is around 3.5mm or 4mm (I can get a precise caliper measurement). If it is 3.5mm, then this matches the 527mm ERD of the manufacturer at least for one interpretation of ERD!

In any case the difference is about 1mm for their calculated spoke length. So I'm either going with 260mm/261mm spokes (for the front wheel) or 261mm/262mm spokes. I suspect both will work and that I should flip a coin...or just go with the former since that is what Light Bicycle suggested to me.

FBinNY 

You should be OK.

People get overwrought over spoke length. One of the hidden benefits of double wall rims is the wide spoke length tolerance.

Assuming a nipple with 7-8mm thread length, spokes can end anywhere between 2mm shy of the top of nipple to 2mm+ above, giving you a 4mm margin of error.

Andrew R Stewart 

To further muddy the ERD waters is that some Al rims have nipple hole reinforcements (eyelets) which add about 1mm of radius above the rim's spoke bed. Some companies (like one that starts with a M and ends with a C) don't include this added nipple seating height to their published ERD stats.

This is another reason why measuring the actual rim's ERD is a better way if possible. I use the BBP/DT spoke sizing chart for this. It has a linear scale and two specific nipples that are placed in opposing rim holes. One lines up one nipple end with the ) line on the scale and the opposite nipple end lines up with the ERD readout on the scale. Very quick and easy to do a few measurements around the rim to average them. Andy.