[lazysod]

I got a 32-hole Mavix CXP33 rim and an old Ultegra rear hub quite cheap. I'd like to try building a wheel with a mixed lacing pattern, like one of these on Robert Torre's site:
32-Spoke Hybrid Half Crow's Foot
Jeremy Parfitt's 3-cross / Radial Mix
or maybe I'll chicken out and use the more conventional
2-cross NDS / 3-cross DS combination

I find Jeremy Parfitt's 3-cross / radial mix particularly interesting, and consider using it DS, with 2x lacing NDS. Compared to Torre's hybrid half crow's foot, the DS crossing spokes are 3x instead of 2x, giving somewhat better drive torque transfer. But the radial spokes aren't as evenly spaced as in the hybrid crow's foot - could that be a problem? Otherwise the two look quite similar.

Some wheelbuilders prefer 2.0/1.5/2.0mm DT Revolution spokes NDS, and 2.0/1.8/2.0 DT Competitions DS. Is that still OK with patterns like 2-cross NDS / 3-cross DS, which intend to balance the lateral strengths better?

Opinions? I'm looking for arguments that would enlighten me. One-liners like "Use 3x both sides" or "I build with 14 gauge spokes everywhere", without saying why, will be ignored, so please don't waste your time on that.

Edit: Fixed description of Jeremy Parfitt's lacing.

 

[Creakyknees]

Er... beyond just being different, can you find any real (vs "internet theory") performance or durability difference?

For example you mention better drive torque transfer... come on, you really think a wheel flexes that much that it matters, in that plane?

OTOH, if it's something cool to do, have fun with it.

 

[Mike T.]

The stuff you talk about does better in latte shop discussions than it does in actual riding on the open road. 3x wheels with equal gauge spokes don't have built-in issues that are magically cured by [email protected] spoking. So ignore me if you wish or think I'm wasting your time as I don't have (nor do I care to dream up) "arguments that would enlighten" but when you've been into wheelbuilding as long as I have (5th decade) you'll know that I was right.

 

[Lectron]

Yes, Mike T is right. Forget about the hogus bogus. 3X all the way if
28+ spokes.

I whrote this some time ago 'bout lacing patterns
-just as a short summary.....

 

[bopApocalypse]

I find the half crow's foot / radial mix particularly interesting, and consider using it DS, with 2x lacing NDS. Compared to Torre's hybrid half crow's foot, the DS crossing spokes are 3x instead of 2x, giving somewhat better drive torque transfer. But the radial spokes aren't as evenly spaced as in the hybrid crow's foot - could that be a problem? Otherwise the two look quite similar.

* I imagine the torque transfer difference between 8 spokes 3x and 2x is minimal.
* Doesn't seem like it would be an issue. Tension should be pretty even between the crossed and radial spokes, so I wouldn't expect any issues in terms of broken spokes or uneven loading of the rim.

Assuming all else (especially bracing angle) stays the same, changing from a simple, even pattern (1x,2x,3x,4x: leading,trailing,leading,trailing...) to a more variable pattern (32h crow's foot: radial, leading,radial,trailing) seems like it should be weaker both laterally and radially. Changing the bracing angle (heads-in crow's foot) will gain back some (but how much?) of that lateral strength. If I was dead-set on doing a mixed pattern, I'd probably try something like a 1x heads-in on the drive side, and 2x on the NDS.

 

[eddie m]

The number of crosses doesn't matter much. Radial spoking is a lot stiffer laterally, but not strong enough torsionally for a rear wheel. With the same number of spokes, more crosses will be slightly stronger torsionally, and slightly more flexible laterally, but not enough that you will notice a difference. I actually had a real analysis of that from some engineering school, with real data and everything, but I haven't been able to find it recently.
The drive side needs twice or more the tension of the non-drive side, so it's reasonable to use lighter spokes on the non-drive side, but I wouldn't use anything less than 2.0/1.7/2.0, just because it's too hard for me to control wind-up with lighter spokes.
Spoking pattern has almost no effect on the tension balance between the 2 sides. The site that recommends crows foot lacing is completely wrong on that. If I had better graphics capability I would draw the correct free body diagram, but I don't so I won't.
The only other thing is you should be careful if you use fewer crosses on the drive side. That requires that you transfer torque across to hub to left side. That's fine if you have a hub that is designed for that, but I wouldn't do it with a conventional Campy or Shimano hub.

em

 

[lazysod]

Thanks, bopApocalypse and eddie m!

Mr. Torre writes about the 2x/3x lacing: "... but the difference in the tensions is not as great as they would be if the wheel was laced conventionally". Shouldn't it be the opposite way? The "torsional" force component on the spoke should be less for 2x compared to 3x (and zero for radial), while the "axial" components (parallel to the hub axle) must be equal and the radial components almost equal, thus the tension will be less for 2x and even less for radial.

Since the angle for 2x and 3x isn't that different, I agree with eddie m that "spoking pattern has almost no effect on the tension balance between the 2 sides".

So I'm discarding the 2x/3x combination, as the logic is flawed. If tension balance was the only concern, fewer crosses or radial on the DS could be better, but I'm not sure how an Ultegra hub from the mid 90's will handle that.

 

[sevencycle]

Cross 3 drive/ cross 2 non-drive. Good setup that seems to equal force at hub.No weird lacing. Dont think too hard just build a wheel and ride it. You will luv the wheel that YOU built. Personal satisfaction

 

[rruff]

So I'm discarding the 2x/3x combination, as the logic is flawed. If tension balance was the only concern, fewer crosses or radial on the DS could be better, but I'm not sure how an Ultegra hub from the mid 90's will handle that.

It would twist in the middle, and since the lightly tensioned NDS spokes are taking all the torque, the pushing spokes will go slack pretty easily.

Sorry to spoil your fun, but it would be a good idea to lace the used hub the same way it was originally.

 

[Lectron]

.....What do one win with going radial DS. The spokes gets nearly 20mm shorter and the angle from the rim toward the flange increases with 6.9%.

The elasticity/spoke elongation is reduced by the same factor (6.9%)*. This is compared with 28h 3X.

2X vs 3X will give a difference in length of 10mm.(ERD=600 Flange=46 FD=19)........

Going fewer crosses DS will even tension out a bit (negotiable?),
but maybe even more important:

-The NDS spokes gets longer, and will stretch/warp more making
the wheel stronger (read able to carry more load).

Going Revolution spokes (14/17) NDS and Competition DS (14/15)
will give you the same benefit you're looking for when going RAD DS.
The elongation of the rev is quite a bit more than the competition
as it's not only thinner, but the swagged part is also longer

As long as your NDS spokes don't go slack, your fine. If it's strong
enough, it's (in most cases) stiff enough

So on an Ultegra hub, my recommendation would be 3X 3X
For a mid size flage DS, maybe down to 2X DS.
High flange like powertap can do with 1X (head in)

 

[Dave_Stohler]

On heavily dished rear wheels, the "leading" spokes on the non-drive side can actually become completely slack when you crank really hard. As soon as you stop cranking, they snap back into tension, and this will eventually lead to fatigue, either in the spoke heads, or on the hub flange. As a result, it's often considered wise to do a semi-radial pattern, which means that all the non drive spokes will be equally tensioned while sprinting.

Problem is, this usually means 2 different lengths of spokes. As a way of trying to use the same size spoke on a dished wheel, the 2x/3x pattern is often used. It will still suffer from the slack spoke problem as noted above, but to a lesser extent.

 

[eddie m]

Judging from the various wheels that are common now, there seems to be no consensus on how best to build a rear wheel. For every possible solution to the problem of insufficient tension on the non-drive side there is another theory that the exact opposite solution is better. The only consensus seems to be that the non-drive side spokes will suffer fatigue failure if they are too heavy. You can use a very light butted spokes on the non-drive side, but they may not be strong enough or resist twisting enough to use on the drive side. Just about everything else anyone “knows” about wheels is based only on anecdotal experience, not any real data or verifiable analysis of the actual structure. Even the idea that heavy gauge spokes routinely go slack is questionable. If it is possible to fully unload a heavy gauge spoke, why is it not possible to fully unload a lighter spoke? Judging by the large variety of different spoking patterns that all seem to work OK, it seems to me that the basic structure of a tension spoke wheel must be strong enough that you can build a pretty reliable no matter what crazy spoke pattern you use.

em

 

[rruff]

I've played around with some FE models of wheels, but the unknown is knowing exactly what forces the rider is putting on the wheels. Some designs are better for certain kinds of forces and some are better for others. There are several ways of lacing wheels that work quite well... and a few that are just stupid. The interesting part is when you are trying to optimize the strength/weight/aero relationship... then you need to be pretty precise. But at any rate you won't be losing much at all by lacing your wheels in a conventional way. There is usually a good reason why things become "common"...

 

[Dave_Stohler]

Judging from the various wheels that are common now, there seems to be no consensus on how best to build a rear wheel. ..........

em

True. It depends on what kind of a rider you are. Sprinter? Go semi-radial. Recreational? Go 2x/3x.

 

[wankski]

well i can kinda see the OP's point... it may not amount to much, but whats the point doing a custom when its a ubiquitous 3x... boooring... for a campag hub, what do ppl think about 1x heads in DS and 3x nds?? all revo 32h ...kinda try to fix up tension imbalance on an OP rim....

 

[Lectron]

Kida already aswerd, and not much to gain.....

Improves lateral and vertical stiffness a bit, but a decrease
in torsional stiffness....Also a tougher on the DS flange.
The angle of the spokes are not much changed, but they get
quite a bite shorter and are not bend that much in the cross.

Saves in a few grams in weight, not as durable, looks cool

 

[eddie m]

well i can kinda see the OP's point... it may not amount to much, but whats the point doing a custom when its a ubiquitous 3x... boooring... for a campag hub, what do ppl think about 1x heads in DS and 3x nds?? all revo 32h ...kinda try to fix up tension imbalance on an OP rim....

That build has 3 problems and no particular advantage over a “boooring” 3X wheel. First, DT Revolutions are not strong enough for the drive side. Revolutions have a yield strength of about 400 lbs and possibly much less, and a breaking strength of about 500 lbs. The drive side (particularly on a Campy hub, which is the most dished hub available) needs as more than 200 lbs tension, and maybe more than 300 lbs. That’s too much for that spoke. Also, it is extremely hard to avoid windup of such a heavily swaged spoke at such high tensions. The last problems that using 1X on the drive side does nothing to balance tension between the 2 sides, but transfers some of the drive torque across the hub, which has a narrow, light gauge spool.
If you find reliability ”boooring” go with your hyper light weird build.

em

 

[eddie m]

True. It depends on what kind of a rider you are. Sprinter? Go semi-radial. Recreational? Go 2x/3x.

You missed my point. If everyone agreed that semi-radial was better for sprinters, that would be a consensus on the best way to build wheels for particular applications. But there is no agreement on that, there is not even agreement on which side should be radial in a half radial wheel.

Most of the new weird spoke patterns are based on marketing, not actual evidence. One of the few real improvements since in the last 40 years is the widespread understanding that butted spokes have fewer fatigue failures than straight gauge spokes. But even that is misunderstood. There is really no point of swaging a spoke so much that the narrow section is smaller than the root of the treaded section. The bigger threads don't add any usable strength, but they increase the difficulty of twisting during during tensioning. Yet those spokes are becoming pretty common now.

em

 

[lazysod]

Thanks for the feedback, everyone!

Regarding rruff's comment about lacing the used hub the same way it was originally: I was aware that the notches from the old spokes can stress the elbows of the new spokes if the new pattern is different. I hope this problem can be solved either by having the head of the new spoke on the side with the notch, with the elbow on the "clean" side, or by lacing the spoke in the opposite direction, so the elbow doesn't touch any of the sharp corners next to the notch (provided I can get the pulling spokes on the outside where spokes cross). In some cases I may have to file down the sharp corners, e.g. for some of the radial spokes in Jeremy Parfitt's 3x/radial mix.

Another issue with this hub is that the left flange obviously has been laced twice, so each hole on this flange has notches on both sides. (See 3rd image.)

After re-reading Sheldon Brown's webpage on wheel building, I also noticed that the notches from the old spokes could act as stress risers and weaken the flange, increasing the risk of tearing the rim apart with radial lacing. Therefore I'm now somewhat more sceptical to using any of the patterns with partial radial lacing on the drive side of this hub. Radial on all of the DS was never an option with this hub. As you can see from the first picture, the hub is quite narrow, only 16mm at the label.

But as tension on the NDS is much lower, the risk of tearing apart the left flange should be very low. Thus radial NDS, with 3x DS, as Dave Stohler hinted at, may be an interesting option. Using different spoke lengths is not an issue for me, as I'll use a supplier that sells DT spokes individually, at an affordable price. So if the preferred pattern would require 8 different spoke lengths (0x, 1x, 2x and 3x on both sides), I'd still use it!

Lectron, regarding thinner (e.g. DT Revolution) spokes NDS: I understand that thinner spokes, for a certain slack (in mm), will see less change of tension. But since they stretch more for the same force, won't that cause a wheel that flexes more when applying force on the NDS spokes, e.g. when tilting the bike in a left curve, maybe making the rim touch the brake pad? Or will the extra "pre-stretch" they get during the build to some extent prevent that? (Non-linear relation between stretch and force?)

Wanski, you're right. The point of doing this build is trying a wheel that's a bit different from those tried and tested 3x/3x ones I've used so far. Let's call this my "experimental" wheel. But I don't see any point doing an experiment that I expect to fail anyway, like radial both sides on a rear wheel.

Some images of the FH-6402 hub:

 

[Lectron]

The thingy 'bout going lighter spokes NDS has to do with preventing
them from going slack under heavy vertical (and torsional) load.

The tension of the spoke has got very little to do with the stiffness of
the wheel until it gets over tensioned (plastic) or going slack.
The relation between force and elongation is (ideal) linear in the
∆F they're designed to work. Hooke's Law works also here, and
k can easily be determined by type of steel and cross section area.

I'll leave that to the ones interested as I'm on my girlfriends laptop
(vacation) with no scientific data

So. For a light rider one could use same thickness on NDS as DS,
but the wheel would in-fact withstand less load before loosing it's
qualities than the one having thinner NDS. It would be (not mentionable)
stiffer though.

When it comes to using rev DS, I really have no problem with that.
They are just as stiff as a sapim CX-ray, in fact a tad stiffer.
I do prefer using some tougher stuff, but one of my CX
wheels have rev all over. No probs. bringing them up to 1200N DS.
Takes a little bit more skill to work with...And of course it doesn't get
as stiff as with a comp.

Anyways. All this talk about extreme lateral stiffness from manufactures
is very often to hide the fact that it's not a particular strong set of wheels.

A strong build wheel is stiff enough. Not necessarily the other way around