Paired spokes, and the dreaded "get away with"

Paired spokes, and the dreaded "get away with"

The questions keep rolling in, which is fantastic, and the questions not only give good topic ideas on their own, they also kick that lonely brain cell into hyperdrive and dredge out more good stuff to talk about. We recently got two questions, one from Anthony in Washington State and another from Donny in Singapore, and though they're very different questions they wind up relating back to the same start point. Anthony has had good experiences with both Rolf paired spoke wheels and "normal" wheels (ours), and wants us to weigh in on the validity of the paired spoke concept. Donny is going to try his first wheel build, with Kinlin XR31s, Bitex or Novatec hubs, Pillar bladed spokes, and brass nipples. He wants to know if he's chosen good components (there are obviously ones I prefer, but yes these are just fine) and, more importantly, since he's quite light, can he get away with fewer spokes, or if that will compromise the build's stiffness too much. We'll excuse Donny since he certainly didn't know my enmity for that phrase of "get away with," and off we go.

The concept of paired spokes is straightforward and I'd say valid. You don't sell however many million wheels Rolf has sold and continues to sell if paired spokes doesn't work as a concept.

As you can see from these 5 star professional illustrations, the paired spoke design aims to balance the pull of opposing spokes, one against the other, and avoid the zig-zag stress pattern that the conventionally spoked rim shows. 

Rolf tell us that this allows for fewer spokes, and higher spoke tensions, and lighter overall weight than conventional wheels. There are far better and more valid ways to illustrate a wheel's strength and resilience than throwing it off a building, for what it's worth. 

When you look at Rolf's wheels, they indeed generally have fewer spokes than most wheels. Their weights are on the light side but not holy cow, and we don't get a spec on spoke tension but we also know that spokes, once tensioned, gain nothing from more tension. No strength, no stiffness, no nothing. They just need enough tension to keep from going slack and that's that. 

What about aerodynamics? Rolf claims their wheels are pretty fast, and you can read their rundown on that for yourself. They test with a 20mm clincher, and against a Mavic Ksyrium SL, which we know has similar aerodynamic benefits to your average cereal box. 

I think that Rolf wheels are generally ok. The whole reason that we sell wheels built with Astral rims, which are Rolf rims sold to the wheel builder market with conventional spoke drillings, is that I was in a shop in San Diego that had a bunch of Rolf wheels that looked like they had quite nice rims and I struck up a conversation with them just as the Astral project was starting. I've never seen a set of Mavic Ksyrium SLs and thought "oh man, we're missing out if we can't build with those rims!" And on their higher end wheels, Rolf's hubs are made by White Industries, and you don't need to be told how I feel about White Industries

The issue with balancing the stresses as paired spokes do is that this leads to stress concentrations. It's not uncommon to hear of spoke pull throughs with paired spoke wheels, and then stuff like this can happen (not to be snuff porn about it, but I came across this photo recently entirely unrelated to this post).

Not a WI-built hub

When you concentrate stresses, stresses sometimes bite back. And yes, we've seen hub shell failures on other "normal" brands, too. Also, I have no idea what's going on with the bigger non-drive flange. That's just weird and backward to me.

So overall, it's none surprising to me that Rolf wheel owners would have a generally good experience, but I certainly don't think that their engineering is a dead lock world beater either. 

So how does this relate to Donny and his wheel build? Well, this is Donny's first wheel build, and if he's anything like me, his first wheel build won't be as good as his 100th, and his 100th won't be as good as his 1000th, and he'll still continue to improve after that. So some room for error on the first build is good. And he's using good components, but not great components. 

The fewer spokes you use, the more important each one is to the structure, which lowers room for error. Also, each spoke is a point of control. Committed readers will be familiar with my use of that phrase. "Points of control" is sort of how I articulate "avoiding the zig-zag deal" from the illustration at the top. And, just plainly, it's going to take some skill to get the wheel durably straight and round for a first time builder using that few spokes. I sure as heck wouldn't recommend fewer points of control and more importance of each point.

Kinlin rims are good, not great. They don't have the extrusion quality that HED does, and their joins not infrequently require some artful manipulation to get right in the build. A good builder can sort these out and repeatably build good wheels with them, but they will challenge a novice builder. Bitex and Novatec hubs don't typically have superb geometry, so making sure all the non-drive spokes have enough to keep from going slack will require excellent tension equalization on all of them. The Kinlin rim's extrusion quality is going to add a bit of challenge there, but Donny's inexperience will add much more challenge. We sure don't need to make this harder.

And then finally we get to the standard phrasing of "how few spokes can I get away with?" When I tell you that this question is always phrased this way, it is ALWAYS phrased this way. Why? Is it some kind of guilty conscience that the zeitgeist imparts on people trying to build with few spokes? Do they know they're tempting physics and fate? Tough saying, but it's gotta be something. There are a few reasons to try and use as few spokes as possible. The big one is just plain artistry, and I can get that. It's not my gig, but I get it. Another is losing every gram you can, and though some of this falls under art, we know that weight doesn't matter nearly as much within these levels as people perceive that it does, and there's a point of reversing returns - making stuff lighter makes it worse. And then there's making things as aerodynamic as possible. 

I didn't address Donny's build stiffness concerns in this one, and that will come at another time. This one is long enough as is.

A brief post-script. We enjoy writing these, and though we're not crazy about having the time to do it, we're happy to do what little part we can to offer this distraction to help increase the likelihood that you'll be able to do the right thing and stay away from other people. And the emails and comments have been awesome. So thank you all for that. We of course would love to increase the "doing well" factor of "doing well by doing good" when the world returns to a semblance of normalcy, and one of the big ways we do that is through reach. So if you enjoy what we're doing, it's helpful to us if you share and let other people know about it. Thanks. 

Back to blog


Binx – I might have, will have to look for it. It’s very difficult to do, because if you change the hub the recommendation might change, change the spokes and same, change the rim and same. Altamont Lites would have a completely different matrix than Altamonts. And then intended use, tire selection, what are you trying to optimize for, blah blah blah. Not easily cataloged.


Dave, you’ve likely already done this in another post, but do you have a “Get-Away-With Table” that sets out should-do’s and can-get-away-withs regarding recommended spoke counts for different weights, ride surfaces, uses, disc/rim, etc.?


Thanks Dave :-)
Yeah, I don’t ride very agressively, so since I allready have one 32*3 set (reussed Chris King Classic hubs from a custom 9 speed set I’ve had converted to 11 speed via the Lasco Concept Machined Shimano Ultegra Casstte – by the way it’s a magnificent solution for not “retiring”…;-) those hubs that are still working for me since 2004/2005, and Boyd Altamont lite rims) so I think I will go 28/24 or 24/20 when I’ll go ahead with my next project.
Many thanks,
Stay well and safe !

Raul Gelbermann

Hi Raul – Glad we have a relevant post for you!

It’s a good question of “what would be best” in terms of lacing. There’s no one answer. You can go 32/32, for sure, and there is nothing wrong with that. I think you will be well past the point of diminishing returns in terms of adding durability to the build or really increasing any positive attributes. 24/28 would be a good, conservative build that should be tolerant of a ton of rough handling. If you are looking to maximize low weight and aerodynamics and raciness, then at your weight, unless you ride VERY aggressively, 20/24 will behave well and I would expect it to last a very long time.

Most of our customers, most of the time, would be best served by a 24/28 or 20/24 build of these components in this situation.


Thanks for the great knowledge !
I personaly have “retired” rolf vector pro wheelset (still intact but needs an extensive overhaull) in favor of 1 custom build 32*3, a ksyrium sl exalith set and a dt swiss pr 1400 oxic set, so this article is hitting a few points interes for me :-) apart from doing both good and well in this times !!
The thought of building another custom wheelset with Chris King r45 hubs – spoke count options 32*3 f&r, or 28 r/24 f, or 24r/20f with boyd altamont lite rims is preocupying me for some time – what wold be best for body weight 68 kg / 71-72 kg with riding equipment on.
Waiting for future same kind of knowledge !
Again thanks !

Raul Gelbermann

Leave a comment

Please note, comments need to be approved before they are published.