CR-V With an Off-Center Wheel and a VSA Light

What went wrong

• Re-indexing the steering wheel on the splines only rotates the rim relative to the column. The rack is still off-center, so the steering angle sensor still reports a turned wheel, the turn-signal cancel cam is out of position, and the clock spring is now wound unevenly - that is how clock springs get destroyed.
• A 25° offset after tie rod replacement almost always means toe was set with the wheel not held centered, or one tie rod was adjusted much more than the other.

Correct method

1. With the front wheels straight ahead, count turns lock-to-lock and find the true center of the rack; mark the wheel.
2. Set total toe with the steering wheel held centered by a fixture (or a helper) - not by eye after the fact.
3. If the wheel ends up off-center, correct it at the tie rods: shorten one and lengthen the other by the same amount so total toe does not change while the wheel rotates to center.
4. Only after the rack and wheel are centered do you verify the clock spring is centered per service info; if it was installed off-center, re-center it (or replace it if it has been wound tight).

Do not correct wheel position by pulling and re-indexing the wheel - that hides an off-center rack and mis-times every steering-angle-dependent system.

Why it set

• VSA/ESC compares steering angle sensor input against yaw rate and wheel speeds. With the wheel 25° off center, the SAS reports a constant turn while the car goes straight; the module flags an implausible signal and illuminates the lamp. The re-indexed wheel made the electrical zero wrong and the rack was off center, so the error persisted.
• A clock spring that has been wound unevenly can also produce an SAS signal fault or an open in the SAS circuit.

What must happen first

1. Fix the mechanical center: rack centered, wheel centered at the tie rods, toe reset, and confirm on an alignment printout with the wheel held straight.
2. Inspect the clock spring and SAS wiring - no stretched harness, connector seated, no stored SAS circuit codes.
3. Then perform the steering angle sensor zero-point calibration/relearn with a scan tool, per the manufacturer's procedure (typically wheels straight, on level ground, engine running, scan tool-initiated, then a short drive).
4. Clear codes and road test; confirm no VSA/ABS lamps and that the SAS reads near 0° going straight.

Calibrating before the mechanical center is correct just stores a new wrong zero - the light comes back.

SRS and column (critical because the wheel was removed)

• Clock spring centered and correctly indexed; verify no binding or scraping through full lock both directions.
• Horn, cruise control, and all steering-wheel buttons function.
• No SRS or SAS codes stored or pending; airbag warning lamp proves out and goes off.
• Battery was disconnected during airbag service and the system was not probed with a meter - confirm no one did.

Steering and alignment

• Both tie rod ends tight, jam nuts torqued, boots not twisted, cotter pins/prevailing torque nuts correct.
• Total toe and thrust angle re-verified with the wheel held centered; printout attached.
• No interference or tire rub at full lock; steering returns to center evenly left and right.

Systems verification

• SAS zero-point calibration completed and confirmed (reads ~0° straight ahead).
• Road test at highway speed: no pull, wheel centered, VSA/ABS lamps off after a full drive cycle.
• Written note to the customer explaining what was corrected and why re-indexing the wheel was the wrong fix - this protects you if the other shop's work is questioned later.