Road-Speed Whine and a Clunk on a High-Mileage Minibus Driveline
A rear-wheel-drive 15-seat minibus with a five-speed manual gearbox has covered 240,000 km. The driver reports a whine that rises with road speed, audible in every gear and also with the gearbox in neutral while the vehicle is coasting, but silent when the vehicle is stationary. There is a single metallic clunk when the clutch is released in first gear, and a vibration at 70–80 km/h. On inspection: propshaft universal-joint play about 0.5 mm; differential backlash 0.45 mm against 0.13–0.18 mm; crown-wheel runout 0.12 mm against a 0.10 mm limit; the pinion flange nut is loose with the stake broken away. Only 1.2 litres of oil drain from a differential rated at 2.0 litres, and the oil is grey with a metallic sheen; the magnetic plug holds a fine paste but no chips. Gearbox oil is clean and at the correct level, and the whine is present in fourth (direct) gear as well. The tyres are new and have been balanced.
1. Decide whether the whine originates in the gearbox or the differential, and set out the evidence for your conclusion.
Read the worked answer
It is the differential.
- Road-speed dependence, not gear dependence. A gearbox fault normally changes pitch or disappears when a different gear pair takes the load, because a different set of teeth is in mesh. A noise that tracks road speed in every gear is generated downstream of the gearbox.
- Neutral coasting. With the box in neutral and the vehicle rolling, no gear pair is transmitting torque; the mainshaft gears are simply spinning on their bearings. Noise that persists there is coming from the propshaft, differential or wheel bearings, not from a loaded gear mesh.
- Silent when stationary. A worn input-shaft or countershaft bearing is characteristically noisy at idle in neutral with the clutch engaged. That test was negative, which argues against the gearbox.
- Direct physical evidence in the axle: 0.8 litre of oil missing out of 2.0 litres, grey oil with metallic sheen (fine wear particles held in suspension), backlash 0.45 mm against 0.13–0.18 mm, crown-wheel runout over limit, and a loose, un-staked pinion nut.
The loose pinion nut is the likely root cause: preload is lost, the pinion walks axially and radially, mesh position drifts, the oil overheats and the level was never topped up. The 70–80 km/h vibration is explained by the 0.5 mm universal-joint play together with the same worn pinion bearings; with new, balanced tyres eliminated, the propshaft and pinion are the remaining rotating suspects. Before stripping, confirm with a stethoscope or vibration pickup on the axle nose during a road test and check wheel bearings for play and heat.
2. Explain the clunk on taking up drive, and why simply tightening the pinion flange nut to the specified torque is the wrong repair.
Read the worked answer
The clunk is lash being taken up through a chain of clearances. At 0.45 mm, backlash is nearly three times the upper limit, so the crown wheel can rotate a visible amount before it touches the pinion. Add the 0.5 mm universal-joint play and the axial float of a pinion whose bearing preload has collapsed, and every torque reversal - clutch release in first gear - produces a metallic impact. Worn spider and side gears in the differential would add a second knock when turning.
Why not just tighten the nut:
- On this type of axle the pinion bearing preload is set by crushing a collapsible spacer. Once the nut has run loose, the spacer and bearings have been hammered and the preload is lost; re-torquing the nut either leaves the preload wrong or over-crushes the spacer and overloads the bearings, which will fail within a few thousand kilometres.
- The nut is a single-use staked or crush nut; re-using it means it will not hold torque.
- Pinion depth and backlash are set by shims and by crown-wheel position, not by nut torque. Tightening the nut changes preload only and leaves the 0.45 mm backlash untouched.
- The bearings have already run with low oil and metal in suspension; the surfaces are damaged, so re-tightening merely hides the symptom.
Correct route: strip the axle, replace all bearings and the collapsible spacer and seal, fit a new nut, set pinion preload with a torque gauge, then set backlash and check the tooth contact pattern before assembly is closed.
3. At 240,000 km with this evidence, would you rebuild using the existing crown wheel and pinion, or fit a new matched set? Describe the setting-up and running-in checks that decide whether the repair is a success.
Read the worked answer
Fit a new matched crown wheel and pinion. The case for replacement: 0.12 mm runout (over the 0.10 mm limit), backlash nearly three times specification, oil starvation and grey oil all indicate that the teeth have been running out of position and overheating. Worn teeth cannot be made quiet by re-setting backlash - a new contact pattern on a worn, slightly barrelled tooth flank concentrates load on the tooth tip and the whine returns within a few thousand kilometres. The set is also a matched pair, so replacing one half alone is not an option. Re-use the differential case and side gears only if they pass inspection and the side-gear thrust washers are renewed.
Setting-up checks:
- Pinion bearing preload with a torque gauge and a new collapsible spacer.
- Backlash 0.13–0.18 mm measured at three or four points around the crown wheel, with the variation between points noted.
- Tooth contact pattern with marking compound on both drive and coast sides - centred, not at the toe or heel, not at the tip or root.
- Crown-wheel runout and bolt torque, and total turning torque of the assembled axle.
- Correct oil grade and quantity to the filler level (about 2.0 litres), new seal, breather clear.
Running-in and hand-over: first 500–1,000 km at moderate load, no towing, then drain the oil and inspect the magnetic plug; re-check for leaks and noise at that point. Also replace or recondition the propshaft universal joints, since 0.5 mm play will keep producing a clunk and vibration even with a perfect axle.