KendallF

OK, I'm thinking about moving one of my Powertap hubs to a different set of wheels. The catch: it's a 32 spoke hub and the rear wheel I'd like to put it in is a 24 hole rim. My thoughts are to lace it 16 spokes 2 cross on the drive side, and 8 radial on the NDS.

I am fairly certain that PT doesn't recommend radial NDS lacing, but I'm wondering if anybody has tried it with success? Could I do a 1x or 2x pattern on the NDS? In any case, I know I'd need some whacky spoke lengths, and I'm going to have to google that article on non-integer cross counts for calculating spoke lengths again..

CrossCountry2k7

From the manual:

For your safety, the non-drive side of the PowerTap hub must be built with at least a 2x lacing
pattern. Because of the patented PowerTap design, torque is transferred through the hub to the
non-drive side. Failure to adhere to this precaution will void the warranty.

So if you don't value the warranty or your safety, go for it.

rdtompki

It's a training setup - why not buy a proper rim?

KendallF

Pretend you don't know better and answer the questions above, if you have any relevant data...

AEO

you need to use either...
3x = 2.38x
2x = 1.63x (doesn't really work on NDS because the spoke holes will overlap)
1x = 0.88x

this is 2.38x on both sides.


powertap having a larger flange diameter, will obviously put the exit angles a bit steeper than what I have here, depending on your rim.

nitropowered

You need 2x on the NDS.

Rims are cheap, get a proper rim

KendallF

Hey, I knew I'd get a relevant response eventually.. S/N ratio is currently 4:1, about what I expected. Thanks! I guess I don't get the comment re: 2x on NDS..can you expound or refer me to more info? I'm googling now for the article I remember seeing..from Damon Rinard?

Edit: reading it now... https://www.sheldonbrown.com/rinard/36-24.htm

AEO

what happens when you use 2x on NDS (1.67x when adjusted for 32-24) is that...
if you can see from my hub, the spokes go leading-trailing-skip-skip-repeat with a 2.38x and 1.67x mixed. I think I did leading 1.67x and trailing 2.38x because I ran out of the correct length spokes for 2.38x all the way around.

when you use 1.67x what happens is that the spokes will now share the same holes for both leading and trailing. so you get leading/trailing-skip-repeat which just doesn't work.

straight up 2.38x or 0.88x will give you trailing-skip-leading-skip-repeat
but as you can see, on my shimano hub there, is that 2.38x gives a spoke exit angle very close to 90 degrees. Ideal angle is 90 degrees, but there are no benefits of having more than 90 degrees. Now the problem with using a taller rim or hub with taller flanges is that the angle becomes greater and you have to reduce the cross number so your spokes don't become too long for the wheel and overshoot 90 degrees.
As a bonus, By only using 8 out of 16 holes, you don't have a problem with spokes overlapping the elbows of other spokes, even at these ideal 90 degree exit angles.

I did make a decimal calc off of the rinard page.

waterrockets 

I've been riding a 32h PT for 2+ years now and am really considering this triplet lacing. I would not recommend radial NDS since the torque arrives at the left flange before the right flange. You don't want the hub shell doing all that work w/out the help of the left flange.

I'm thinking about taking the risk. It's just a cool lacing pattern.

Note that I race my training wheels... or train on my racing wheels... you know.

operator

Hilarious.

You're going to risk wheelbuild labour, and spoke cost and hub failure and warranty for a $30 rim. Relevant info indeed.

KendallF

So I think I can do the spoke calcs for 12 each side, and that will work, I think.. but are you saying you'd do the 16 DS/8 NDS? If so, that adds one more complication with the irregular hole interval. I think I am getting this though..the NDS calcs will be easy as the spokes are regularly spaced on the rim, but the DS calcs need to be "faked" somehow.

AEO

you can do 16 and 8 or 12 and 12 it works either way, but the key part to triplet lacing is the 2:1.

In a regular wheel the NDS spokes are dished to a point where they only have about 60% of the tension in the DS. what a triplet lacing does is allow the NDS spokes to come into around 85~90% of the tension in the DS by using half the spokes. In wheel building, even spoke tension is very important, as a wheel with spokes tensioned evenly will last longer than unevenly tensioned spokes.

on a well built rear wheel with cassette, the spokes that fail first are the NDS side from metal fatigue at the elbows. The metal fatigues because the spokes are allowed to go slack. higher tensions will prevent this mode of failure, but we all know this is impossible in a dished wheel because the rim will pull to the NDS too much.

KendallF

Drew it out tonight so I could visualize it. NDS 1x using 8 holes is simple; calculate spoke lengths per 16 spoke rim and hub, 1x. DS 3x using 16 holes gives leading and trailing spokes the same length, 2.38 effective cross. The NDS 1x pattern using 8 holes gives close to tangent angles leaving the hub, which is what we need to take up the PT hub torque on that side.

Here are my calculations using Damon Rinard's spocalc, and the decimal cross count from DR/AEO's spreadsheet.

Rim: Gigantex 50mm clincher, hidden nipples, ERD 563. Hub, PT-SL, WL 31.8, WR 18, flange diameter 70.

NDS: 16 h rim 1x, 258.5 length
DS: 24 h rim 2.38x, 263.2 length

Can anybody confirm this? Thanks!

Edit: Note what the spoke lengths imply: the exit angle leaving the hub is higher than a traditional 32 spoke 2x pattern on both sides (obvious from the 2.38 cross number on the DS, not so obvious from the "1x" on the NDS). As AEO mentions above, the spoke tension should be much closer left/right also. I searched for a spoke calculator that would give exit angles and spoke tensions, but didn't find one. If I build it, I'll take pictures and measure tension.

AEO

563 ERD and 70mm flange diameter?

you can see my 2.38x spokes going out at near tangent on a 45mm flange and 594 ERD rim.
I'd be curious to see if 2.38x will cause the DS spokes to go beyond 90 degree exit angles.

waterrockets 

KendallF, looks like a good plan, thanks for posting the research.

sstorkel

It may be an accurate plan, but I'd hardly call it good. A good plan would be to spend $40 and buy a 32-hole rim...

KendallF

I don't really have time to inventory my wheels and my plans for the naysayers. Suffice it to say that I already own a set of Kinlin XR300s with a 32 hole PT in the rear; I don't need another. You might also want to read up on the advantages of the triplet lacing pattern.

waterrockets 

+1 this isn't about saving money. This is some experimentation on what may be a really really sweet build when it's done.

I agree a 32h rim is the reliable way to go.

GGDaddy

Okay, very old thread I'm bumping, but the subject is exactly on target.

So I have a Power Tab G3 hub power meter, and the LBS wheel builder guy apparently didn't read the manual, because he built it with radial spokes on the NDS.

And dumb me, I'm new to this stuff, so I didn't know it's an issue, and put 140 miles on the thing in the week since picking it up configured this way.

According to the manual, I've apparently voided the warranty. I'm going to ask him to thread it right today. But my questions--why is the warranty voided? Have I actually broken the thing? Panicked here... thanks for any thoughts you can share.

GGDaddy

Okay, just following up my own post--no need to respond. I've taken it back to the guy, and he's going to see if Quark will rebuild it and keep it under warranty. If not, he'll eat the cost of a new hub and build it into my wheel for me.

Pretty decent of him.