Thicker drive side spokes for rear wheel build?

Tech Tech Help
1

Hi everyone, I’m currently building a new set of wheels and have stumbled upon an interesting approach to (mostly) rear wheel spoke choice.

On https://www.sheldonbrown.com/spoke-pitch.html , John Allen writes about a trick for ending up with a stronger wheel which I’d call pitch matching. In short, the idea is that by roughly matching the tone at which the drive side and non-drive side spokes ring, you end up with a stronger wheel. This is achieved by using thicker drive side spokes. From what I understand the basic idea is that due to wheel/hub (and cassette) geometry, the non-drive side spokes see less tension. By making these less-tensioned spokes thinner, you should end up with the same spoke stretch as on the drive side therefore getting a stronger and/or more durable wheel. In the above article, he describes matching 2mm thick spokes on the drive side with 1.6mm on the non-drive side. I’ve tried to reason my way through this argument, and as a first step tried to figure out where he comes up with these values, 2mm and 1.6mm.

With the equation for musical pitch of any string: F = 1/2L * (sqrt(T/m)) (described here: Why correct musical pitch depends on spoke length and not on thickness ) where F is the resulting frequency, L the length of the spoke, T the tension and m the mass per unit of length, you can plug in the example above and get the following:

Since our goal is that drive side and non-drive side spokes ring at the same pitch we can set Fdriveside = F_nondriveside. With the length spokes being (amost) equal, we can cancel those terms and remove the square root which leaves us with:

T_ds/m_ds = T_nds/m_nds

The mass per unit length is proportional to the square of the diameter of the spokes, which gives us, for the above example, a tension ratio of T_nds/T_ds = m_nds/m_ds = r_nds^2/r_ds^2 = 1.6^2/2^2 = 0.64 or in other words, the target spoke tension of the non-drive side is 64% of the drive side tension. In this case both spokes should produce the same pitch.

I have found further evidence of this practice on the sapim website, specifically their CX-Sprint spokes ( Spokes - SAPIM ) where they state: Many professional cyclists prefer using the CX-Sprint on the drive side and the CX-Ray on the non-drive side for balanced stiffness, enhancing overall wheel performance.

With the above formula that gives us the tension ratio as a function of the spoke weight we can check what that would give us for the CX-Ray and CX-Sprint combination. Sapim specifies 279g for the CX-Ray and 334g for the CX-Sprint (both for 64 spokes at 260mm not that it matters).

This gives us 279/334 = 0.84 or 84% tension ratio. Since from what I know tension ratios of rear wheels with non-offset rims are typically closer to the mid 60% range this would not give us an even pitch but at least bring the two sides closer.

Now finally to my question:

Given two otherwise identical rear wheels, one has the same spokes on both sides, the other one has thicker drive side spokes (possibly pitch matched), is there any difference in performance/durability/etc…?

In my mind I’m a bit torn between:

  • The approach is totally bogus
  • The argument is sound, but the actual effect on wheel performance/longevity/… whatever is negligible
  • This is a good way of improving a wheel

An argument in favor of the first two options would be that Jobst Brandt (iirc) does not talk about it in The bicycle wheel. At the same time the method is mentioned on sheldonbrown and the website of Sapim which makes me pause for sure.
Does anyone have any experience or opinions on that? Would be super curious to hear.
Thanks :slight_smile:

Edit: Added a concrete question to my ramblings

1 Like
2

Plucking a spoke and listening to its vibrational frequency is a manner of quantifying the spoke’s tension. It seems like a better way to do this is with a spoke tensiometer (using 3 point bending). Maybe there’s some advantage to the acoustic method, but not sure what that might be.

1 Like
3

Yes, sure, I’ve been using a combination of plucking spokes and using a tensionmeter for my wheel builds so far. The advantage of the plucking is that it’s a lot faster than having to position a tensionmeter over and over again. This means that you can go through a wheel fast and find spokes that are way off quite quickly. For the fine tuning and the absolute tension I then like to use the tensionmeter to make sure I’m putting the correct amount of tension.
My question was more, perhaps I didn’t phrase it correctly, is there an advantage of using thicker drive side spokes compared to using the same thickness spoke on both sides? The pitch argument then just quantifies how much thicker you should ideally go.

2 Likes
4

Since on regular wheels the drive side spokes are “more upright”, you get a more even (for both sides of the wheel) lateral stiffness by using a slightly thicker (stiffer) spoke on the drive side.
I have done this on some mtb wheelbuilds for myself, but stopped doing it for years now. Now I am probably an insensitive guy (or a lightweight that rides too slow and easy trails) , since I don’t find it matters anything for me. Berd spokes are fine too for my mtb wheels, so there is that.
Spoke preload L/R ratio (kgf per spoke during build) is a given once the wheel geometry (hub/spoke number and length/rims/offset etc) has been chosen, as to arrive at a centered wheel.

1 Like
5

Starting with a clean slate, this is a great argument for designing wheels around 2:1 spoke lacing.

Using conventional hubs and rims and lacing patterns, there are a couple issues with what you’re describing. Firstly, most 2.0 thickness spokes are straight gauge. They are stiffer but heavier and weaker. Their extra stiffness places greater strain on the rim hole. Secondly, it can help even out left/right rim flex behavior, but it makes vertical rim flex behavior much worse. As I don’t really notice uneven lateral rim flex even on a mountain bike, I’d optimize for the smoother ride, lower rim strain, and greater fatigue life of thin spokes all around.

2 Likes
6

Ah yes, true, great point, I had not realized that but of course that way you’d get even spoke tension but offsetting the geometric difference between drive side and non-drive side with the number of spokes.

Ah yeah, fair, any potential benefit would probably neglected if you end up going for straight gauge spokes and as you say you’d probably end up with a worse wheel in the end.

The spokes that I’m considering are CX Ray and CX Sprint. They’re bladed and the center sections are 0.9mm and 1.2mm in thickness. Both are 2.2mm wide. So they should be nice and springy and have high fatigue life.

I guess in the end there’s no way around a trip to the hardware store and firing up my 3d printer to build a little wheel stiffness testing rig. Build the wheel twice, once with all the same spokes, once with thicker drive side spokes and measure the lateral deflection for both :grinning_face:

7

I know nothing about wheel building but, are these to be disc brake wheels?

If so, then surely the conventional wisdom of rim brake wheel building has been thrown out as the forces applied to the NDS have fundamentally changed since your reference texts were published?

I’ll follow along here as this is an interesting topic. Hats off to wheel builders!

8

That assumption isn’t valid if the cross sectional area changes along the length, as happens with butted spokes. To get the actual ratios you need to consider the strain of each section of the two types of spokes. The thinner section strains more than the ends (this being the point of butted spokes).

If you do the calculation the decrease in stiffness is slightly greater than the decrease in weight. I get a ratio of 0.79, not 0.84.

I can run you through the calculation if you are interested.

1 Like
9

The wheels that I’m intending to build are disc brake indeed.

From what I understand the core principles stay the same with some exceptions. On rim brake front wheels you could get away with radial lacing. Radial spokes, in an idealized way, cannot transmit any torque. For a rim brake front wheel this is not necessary since it’s not the driven wheel and the braking forces go directly from the top of the rim to the bottom where the wheel contacts the ground.

For rim brake rear wheels the drive side transmits the pedaling torque so that one always had to be cross laced, i.e. the spokes to not connect hub and rim in the direct way, radially, but go out at some angle. The more tangential, the better the spokes are at transmitting torque. For the non-drive side, on a rime brake rear wheel you could still get away with radial spoking. There are a few wheels that feature this combined pattern. Cross laced on the drive side, radially laced on the non-drive side.

For disc brakes, now, the change in lacing a wheel is that the side where the disc is mounted needs to be cross laced as well, radial won’t work. The braking torque needs to be transmitted from the hub through the spokes into the rim. This means that for a rear wheel the drive side and the non drive side, i.e. the brake side both need be cross-laced. The drive side transmits the pedaling torque, the brake side the brake torque. For a disc brake front wheel you could again technically get away with only cross-lacing the brake side but I haven’t seen a wheel where the side opposite of the brake disc is laced radially.

That’s how I understand the difference between rim brake and disc brake wheels. So in short, there is a difference but with regards to wheels it mostly affects the spoke pattern and the forces that the hub needs to be able to withstand. In addition, the brake side of the fork and the brake side on the rear triangle probably need to be reinforced as well since the braking force is not applied off-center.

1 Like
10

Uh, thanks a lot, very interesting. I think I can intuitively understand what you mean. The tension causes the spokes to elongate and that elongation means that the cross section has to go down as the material has to come from somewhere. This reduced cross section changes the weight per unit length and should thus lead to a higher pitch. Is that the correct argument?

I’ll try and put my physics degree to good use and see if I can figure this out on my own.

I’ll keep you posted though if I can get to the same 0.79 as you did.

11

1/k.tot = 1/k.1 + 1/k.i + 1/k.n with k = EA/L of each spoke section along its length.
In the end it does not matter all that much. And 3d printing (parts of) a testing jig for wheel stiffness sounds like a challenge, given the relatively small deflections involved - fully assembled wheels are remarkably stiff structures given the lightweight individual components. Perhaps better to buy some Bosch alu profiles and connections? But a nice experiment for the sake of it anyway!

1 Like
12

I was going to use some thick plywood for the main parts and 3d print connector pieces and other little bits that I’d need here and there but I’ve always wanted to play around with alu profiles, nice idea.

And yes, I’m just trying to find an excuse to perform a nice experiment for the hell of it :smiley:

1 Like