Spoke Tension, Left vs Right: Why a Rear Wheel Is Uneven and What Actually Fixes It

The bracing-angle physics behind the imbalance, the fixes that work and the one that doesn't, and why radial lacing really does build a stiffer wheel — with links to the original posts.

Every rear road wheel is built with unequal spoke tension, left versus right. This isn't a building mistake — it's geometry, and you can't make it go away, only manage it. Understanding exactly why is what separates the builders who manage it well from those who chase their tail. This guide lays out the physics one workshop has written about at length, with links to the source. Throughout, "right" means the drive (freehub) side and "left" the non-drive side.

1. Why the two sides can't be equal

A rear hub has to make room for the cassette, so its drive-side flange is pulled inward toward the centre — the dish (Japanese ochoko, "sake cup"; English "dish"). That makes the right spokes stand up at a steeper, more vertical angle and the left spokes splay out flatter.

The closer a spoke is to vertical, the less of its tension is "lost" to the angle — so the steeper drive side ends up at higher tension, and the flatter non-drive side at lower tension. The gap is large: if a rim's tension limit is 110 kgf and you build the drive side to the limit, the non-drive side might land anywhere from 80 kgf down to 60 kgf, depending on hub dimensions and lacing.

And a slack non-drive side is not cosmetic. When you sprint, the left side goes loose and the wheel flexes enough to rub the brake. Hub geometry matters enormously here — a difference of just a few millimetres in flange width visibly shifts the left/right tension balance.

Source: Nomu Lab #46 — On left/right spoke-tension difference (part 1)

2. The one fix you must never use

There is an obvious-looking way to make the tensions equal: narrow the left flange so the two sides are symmetric again. Don't. Narrowing the left flange equalises tension at the cost of a drastic loss of lateral stiffness — you've fixed the numbers by making the whole wheel weaker, "levelling down to the weak side."

Source: Nomu Lab #46

3. The fixes that work: hi-lo flanges and offset rims

If you can't move the flange in, the legitimate move is to attack the angle instead — bring the left and right spoke incidence angles closer together without touching flange width. Two tools do this:

• Hi-lo flanges — make the drive-side flange taller. A larger drive-side flange diameter steepens... actually evens out the incidence angles between the two sides. Campagnolo was doing this decades ago.
• Offset rims — drill the rim's spoke holes shifted toward the left. That lays the right spoke down a touch and stands the left spoke up a touch, pulling the two incidence angles closer. Often combined with hi-lo flanges; rarely seen on very deep rims, where the left nipple angle gets too harsh.

Evening out the angle doesn't fully equalise tension — the two sides' spokes are still different lengths — but it helps a lot. A worked example from the workshop: refitting a 126 mm hub into a 130 mm frame by adding a 4 mm spacer on the left, then re-centring the rim by half that (2 mm), measurably narrowed the left/right tension gap — and it's felt on the road, not just on paper. The other lever, changing the left/right spoke count, gets its own guide.

Source: Nomu Lab #46

4. The radial line: why cross count costs tension

Now view the wheel from the side. Extend a spoke in a straight line and, for a radial-laced wheel, that line passes through the hub centre — call it the radial line (a line taken from the circle's radius). A spoke pulls the rim most directly when it lies on the radial line, i.e. radial lacing.

As you move to 2-cross, 4-cross, 6-cross, 8-cross, each spoke deviates further from the radial line, and more of its pull is "lost" to that deviation. So a more heavily crossed spoke has to be tensioned harder to do the same job. This is the side-on counterpart to the bracing-angle loss from section 1.

Source: Nomu Lab #76 — On left/right spoke-tension difference (part 2)

5. ST and RK: why radial lacing is vertically stiffer

The workshop uses two terms. ST is spoke tension — how tight the spoke is, and what a tension meter measures (or rather infers — see how a tension meter works). RK is the force the nipple exerts trying to pull through the rim. RK is what actually drives vertical stiffness, and it's what destroys rims — but you can't measure RK in a workshop, so rim makers publish an ST limit as a proxy.

Here's the consequence that surprises people. For the same RK, a radial spoke needs lower ST than a crossed one (the crossed spoke wastes some pull deviating from the radial line). Flip it around: at the same ST limit, radial lacing produces the highest RK. So:

Source: Nomu Lab #76

6. What breaks when you over-tension

Push tension high enough and one of four parts — rim, nipple, spoke, hub — gives. Ranked from most to least common in practice:

This is the whole reason rim makers specify an ST ceiling: to keep RK below the level that splits the rim at a spoke hole. Stay under it, keep the two sides as balanced as geometry allows, and a rear wheel will hold.

Source: Nomu Lab #76