Brake judder comeback six months after pads and rotors
A 2019 SUV, 78,000 km, returns six months after we fitted front pads and rotors, with steering-wheel shake when braking from 90–110 km/h. Measurements: rotor thickness 27.6 mm both sides (new 28.0, minimum 26.0); disc thickness variation 0.045 mm left, 0.018 mm right; lateral runout with the rotors on the vehicle 0.11 mm left, 0.04 mm right; hub flange runout 0.06 mm left, 0.02 mm right (limit 0.03 mm); wheel nut torque found at 150, 155, 160, 160 and 165 N·m against a 140 N·m specification, with the wheels last off at a tyre shop 2,000 km ago. Pads have 7 mm remaining with even wear, and the caliper guide pins move freely. The customer wants new rotors under warranty and has already said he will post a review.
1. Rank the causes by what the measurements prove and decide whether the rotors are defective.
Read the worked answer
- Hub flange runout, 0.06 mm left against a 0.03 mm limit. Root cause. Dirt, corrosion or a burr under the rotor is enough, and the error is transferred and magnified into the rotor. This originates in our last brake job.
- Wheel nut torque of 150–165 N·m against 140 N·m, applied by a tyre shop with an impact gun. Uneven over-torque distorts the hat and the rotor and promotes disc thickness variation. It happened 2,000 km ago and matches the return of the judder.
- Disc thickness variation 0.045 mm left. A consequence, not a cause. At 27.6 mm the rotor is 1.6 mm above minimum with most of its life left.
Verdict: the rotors are not defective and were almost certainly sound when fitted. The left runout of 0.11 mm splits neatly into about 0.06 mm of hub and 0.05 mm of rotor. Correct the hub and most of the judder disappears with the same rotors. Fitting new rotors onto the existing hub error would simply reproduce the complaint.
2. Set out the repair sequence, including the measurements that must be taken before the vehicle is released.
Read the worked answer
- Strip both front hubs, clean the flange face and the rotor-to-hub interface to bare metal with a Roloc disc, and remove any burr around the stud holes.
- Measure hub runout with a dial gauge on a torqued-down rotor; it must be ≤ 0.03 mm. If it is not, the hub itself is bent and must be replaced - no amount of rotor work will fix it.
- Index the rotor: mark its position against a stud, measure, rotate 180°, measure again, and keep the best position. This alone often removes 0.05 mm.
- Torque the wheel to 140 N·m with a calibrated wrench, star pattern, two stages, wheel supported - never an impact gun.
- Confirm caliper guide pins and piston knock-back; a sticking pin keeps a pad in contact and builds deposits even when the runout is correct.
- If runout remains above 0.05 mm, machine on the car or replace the rotor - but only after the hub face is proven good.
Bed in with 8–10 moderate stops from 60 km/h, not one hard panic stop, and record final runout and torque on the job card.
3. How do you handle the warranty claim, the complaint risk and the workshop process that let this happen?
Read the worked answer
Be straight before he posts the review. Explain that the judder traces to hub preparation and to wheel-nut torque applied elsewhere, that the rotors are within specification, and that we are correcting the hub faces at our cost.
Offer: free hub clean and runout correction, plus a goodwill contribution towards machining or replacement if the runout cannot be brought into specification. Do not simply fit new rotors under warranty - with the hub still at 0.06 mm the judder returns in a few thousand kilometres and the second complaint is unrecoverable.
Record: before and after measurements, torque wrench setting, the bedding-in procedure given, and a copy for the customer. This is also the evidence that protects the workshop if the complaint escalates.
Process change: hub clean and runout check on every brake job, a calibrated torque wrench on every wheel that leaves the building, and a re-check whenever a vehicle has had wheels off elsewhere. The failure here was not the parts; it was two assumptions - that the hub was true and that the wheel would be refitted properly.