Building an Asymmetric Rear Wheel: Different Spoke Gauges and Counts, Left vs Right

Two levers a workshop uses to fight a rear wheel's slack non-drive side — with the recipes, the costs, and links to the original posts.

1. The problem: a dished rear wheel is slack on the left

A rear wheel is dished — the cassette forces the drive-side flange inward, so the drive (right) side runs at high tension and the non-drive (left) side runs slack. A slack non-drive side is the root of most rear-wheel complaints: it flexes under load, the spokes go loose, the wheel rubs the brake when you sprint. Almost everything distinctive about this workshop's rear-wheel building is aimed at one thing: keeping the non-drive side from going slack. There are two levers for that — gauge and count.

2. Lever 1 — different spoke gauge, left vs right

The rule is consistent: a thicker, heavier spoke on the drive side, a thinner, lighter spoke on the non-drive side. The drive side carries higher tension and does the driving, so it gets the stronger spoke; the slack non-drive side gets the lighter one, which is also where you can save weight without hurting the wheel.

The workshop's house recipes, expressed as relative spoke cross-section (drive : non-drive, drive = 100):

The default — double-butted on the drive side and a thinner bladed spoke on the non-drive side — is the one most riders are best served by.

Sources: Nomu Lab #554 — On different-gauge builds, #786 — When the drive side is the thinner spoke

3. Lever 2 — different spoke count, left vs right

The second lever is to put more spokes on the drive side and fewer on the non-drive side. The well-known extreme is 2:1 (drive side twice the non-drive count) — on a 24-hole wheel, 16 drive spokes and 8 non-drive. This freedom is exactly what most straight-pull hubs lack, and a big reason the workshop prefers J-bend for a performance rear — see straight-pull vs J-bend.

What it buys you:

• The left/right tension comes much closer together — the headline benefit.
• More drive-side spokes means more spokes pulling in the drive direction (the trailing spokes that actually transmit pedalling torque), which can improve drive efficiency.

What it costs you:

• The non-drive side now carries that whole side's load on very few spokes — higher load per spoke, and a real risk of spoke breakage, especially at the elbow on J-bend spokes.
• Higher load per nipple on the rim (the workshop calls this RK — the force trying to pull the nipple through the rim).
• Uneven hole spacing. Asymmetric counts force spots where two same-side spokes come consecutively, which makes precise radial (vertical) trueing essentially impossible. This is why factory asymmetric-count wheels run stiff, heavier rims — and why they tolerate far more vertical runout than a normal hand-built wheel.

Source: Nomu Lab #939 — On 2:1 and other asymmetric-count wheels

4. Does your count ratio even work?

Not every left:right ratio is buildable. Before committing, draw the wheel's flat layout (lay the rim holes and hub holes out on a line) and check whether the spokes land without a phase clash. Two rules from the workshop:

• Use the real counts, not the reduced ratio. Write 16:12, not 4:3 — reducing the fraction can cut through the actual repeating unit and mislead you.
• The hub and rim hole counts have to agree. 2:1 (16:8) works cleanly on a 32-hole hub with a 24-hole rim. A 16:12 (28-hole) build only works if you can drill the hub at custom phases; on an ordinary evenly-drilled hub it can't be done without a phase clash.

If you've invented a clever new left:right ratio, draw its flat layout before ordering parts. The workshop's view is blunt: there is probably no practical ratio that hasn't already been tried — the layout will quickly show you why.

Source: Nomu Lab #1294 — How to judge whether an asymmetric-count build is possible

5. The workshop's stance

Putting it together: gauge asymmetry is the practical default — a half-Competition rear gives most of the benefit with none of the downsides. Aggressive count asymmetry (hand-built 2:1) is treated with caution and not offered as a product, because the non-drive J-bend spokes carry too much load and the vertical-true compromise is too large. The lesson that runs through all of it: a good rear wheel is mostly about spoke selection and lacing; tricks like tying are only the finishing touch. On that "lacing" point, see also why a hand build should cross the non-drive side rather than lace it radial.

Sources: Nomu Lab #939, #1294