The physics of judder in ATC transmissions

Why Clutch-Based All-Wheel Drive Judders

Why a driveline with an electronically controlled multi-plate clutch shudders under acceleration, what role shaft compliance plays in the process, and why a transfer case fluid change makes the symptom disappear

An expert feature by AWD.TECH — a practitioner's perspective from the all-wheel-drive field, grounded in published research on driveline torsional vibration

You press the accelerator, the car surges forward — and suddenly a rapid series of small jolts runs through the body, as if someone were drumming on the driveline from below. Lift off the pedal — it vanishes. Press again — it returns. Owners of BMWs with xDrive know this well, and they are not alone: the same symptom appears on any vehicle whose all-wheel-drive system is built around an electronically controlled multi-plate clutch in the transfer case.

The first suspect that comes to mind is the gearbox. But the gearbox is usually innocent. All-wheel-drive drivetrains are our field at AWD.TECH, and this symptom is one we encounter constantly — over time we arrived at an explanation that differs from the usual "the clutch is just worn out." Curiously, the literature offers no ready-made answer for transfer cases specifically: for decades, serious research into this class of vibration was carried out on manual-transmission clutches and automatic-transmission wet clutches, while clutch-based transfer cases remained hidden in manufacturers' internal reports. Yet the physics is one and the same — and once you carry the accumulated scientific results over to the transfer case, the picture snaps into focus. To do that, you have to look at the driveline from an unfamiliar angle: not as an assembly of rigid steel parts, but as a spring.

The Clutch That Always Slips

Let us start with the hardware. Transfer cases such as BMW's ATC series have no center differential — in its place sits a pack of friction discs running in an oil bath, compressed by a servo actuator. Hundreds of times per second, the control electronics adjust the clamping force, metering anywhere from zero to half of the engine torque — and momentarily more — to the front axle at the command of the stability control system.

Look at the schematic: the rear axle is coupled to the engine directly and rigidly. Every bit of torque headed for the front axle, however, passes through the friction of sliding discs and a chain — the single "soft," modulated link in the entire driveline. And here is the key detail that almost every description overlooks: this clutch is practically never fully locked. It operates in a regime of controlled slip — the discs are always sliding slightly against one another, and it is precisely the amount of slip that sets the transmitted torque. As long as the frictional properties of the disc–fluid pair are healthy, everything runs smoothly. But this elegant design has an Achilles heel, and to see it, we need to understand what the torque path from the clutch to the wheel is actually made of.

The Driveline Is a Spring

A propshaft looks like an unbendable bar of steel. In reality, it twists under load — as do the half-shafts, the splined joints, the drive chain inside the transfer case, and even the tire sidewalls. Every element is compliant, and they are all connected in series: the whole chain from the clutch to the tire contact patch behaves as a genuine torsion spring. Under acceleration, as torque rises, this spring winds up through measurable angles and stores elastic strain energy — exactly like a wound watch spring.

This is not a loose metaphor. Engineers who study driveline vibration describe the powertrain in precisely these terms: the classic model by Crowther and Zhang is a torsional system of four inertias — engine with flywheel, clutch, gearbox, and driveline — connected by elastic shafts. Like any pendulum, such a system has a natural frequency; for automotive drivelines it falls in the range of a few to a few tens of hertz. Remember that number — we will come back to it.

Stick and Slip

Now let us put the clutch and the spring together and watch the mechanism in full.

While the clutch discs are held by static friction, engine torque winds up the shaft spring. The torque in the spring climbs and climbs — until at some point it exceeds what static friction can hold. The discs break loose into sliding. Sliding friction is weaker than static friction, so the transmitted torque drops abruptly, and the spring, suddenly freed, fires off part of its stored energy — spinning up the wheels and shafts. The slip speed across the discs collapses, they grip once more, and the whole sequence begins again: stick, wind-up, breakaway, release. Cycle after cycle, at that very natural frequency of a few to a few tens of hertz.

Here is what this drama looks like if you record the torque at each axle over time. On the left, a healthy clutch: both curves are calm. On the right, the moment frictional stability is lost: the front axle starts sawing, and — note this — the rear axle answers with oscillations in antiphase. This is the desynchronization of the axles: the energy fired off by the shaft spring at every breakaway does not vanish — it is slammed across to the rigidly coupled rear axle, and the whole car shudders.

The phenomenon is called stick-slip. The same physical mechanism makes door hinges squeak, a bow sing on a violin string, and chalk screech on a blackboard. In a driveline it manifests as the rapid-fire jerking that engineers call judder.

At this point an attentive reader will spot a catch. A spring cannot sustain oscillation on its own — any pendulum with friction eventually comes to rest. So why doesn't the juddering die out instead of repeating mile after mile? Something must be pumping energy into every cycle. And that something is the most interesting part of the whole story.

The Negative Damper

In a healthy clutch with fresh fluid, the coefficient of friction rises slightly with slip speed. This property is not a gift of nature but the work of the additive package: the friction modifiers in the transmission fluid are formulated specifically so that the friction–velocity curve has a positive slope. The physical meaning is simple: the faster the discs try to slide, the harder friction reins them in. Any random disturbance damps itself out — the system is stable.

Over time the picture changes. The fluid oxidizes and degrades under heat and shear, the additives are consumed and deposit onto the discs, and wear debris from the friction linings contaminates the fluid. The slope of the friction curve flips: static friction climbs above dynamic friction, and friction begins to fall as slip speed rises.

This has been confirmed experimentally with exhaustive directness: researchers measured friction–velocity curves with a scanning force microscope on real wet-clutch friction materials and showed that fluid degraded in service produces a flat or negative slope on any friction material, new or worn — whereas fresh fluid always produces a positive one.

A negative slope of the friction curve is mathematically equivalent to negative damping. Instead of quenching oscillations, friction feeds them: on every cycle, the system siphons a portion of energy from the engine's power flow and invests it in the vibration. A spring that behaved quietly for years turns into a self-excited oscillator. Note that nothing changed mechanically in the shafts — only the sign of one curve's slope changed. That is enough.

From there a vicious circle takes over. A worn clutch slips more and runs hotter; overheating accelerates additive degradation; degradation amplifies the vibration. Add the purely mechanical contribution — growing clearances in the disc pack and wear in the clamping servo mechanism, which make the electronics miss their torque targets — and the symptom begins to progress.

Why Under Acceleration, Specifically

Now the answer to the original question assembles itself. Under hard acceleration, the clutch transmits maximum torque while slipping — the best possible conditions for breakaway. The rising load keeps winding the shaft spring tighter, increasing the energy dumped at every breakaway — which is why the jolts under acceleration are the harshest. And at the same time the slipping pack heats up intensely, finishing off the fluid's frictional properties.

The same conditions — high torque plus forced slip — arise in long corners under power and when parking with the steering at full lock, where the difference in path length between the front and rear axles drives parasitic power circulation through the locked driveline. The three regimes in which owners notice the symptom are exactly the three regimes in which the physics predicts it.

For completeness: stick-slip is the principal culprit, but not the only one. Researchers also identify a negative friction gradient without full sticking (the clutch slides continuously while the torque oscillates), geometric defects — disc warping and runout — and the more exotic parametric excitation. Experiments show, however, that the largest vibration amplitudes coincide precisely with the onset of stick-slip: breakaway from a stuck state, with the wound-up shafts firing off their energy, hits hardest.

What to Do About It

The physics yields remedies — and our experience bears them out.

If the jerking has only just appeared, a transfer case fluid change (even of fluid nominally "filled for life") followed by clutch adaptation with diagnostic equipment will very likely cure it. Fresh additives restore the positive slope of the friction curve — the negative damper disappears, and the very same compliant driveline stops oscillating. A beautiful example of how a liter of the right fluid flips a sign in the stability equation. From what we see at AWD.TECH, this is exactly how a substantial share of judder cases resolve — when the problem is caught at an early stage.

If, however, the discs are glazed and the clamping mechanism is worn, a fluid change buys only a reprieve — what follows is a rebuild of the clutch pack or replacement of the unit. And procrastinating is unwise: wear debris circulates in the fluid and methodically destroys the bearings and the chain inside the transfer case. Time and again we come across units where an overdue fluid change turned reasonably priced maintenance into expensive repair — the condition of the discs and chain leaves little doubt how long the vibration had been building.

In Closing

The juddering of an all-wheel-drive vehicle with a clutch-based transfer case is not a mysterious electronics glitch, nor the car's "character." It is friction-induced self-excited vibration: the shaft spring stores energy under load, a worn clutch with tired fluid pumps it up through a negative friction gradient, and the stick–wind-up–breakaway cycle reaches the body as a series of jolts.

Let us be clear: we discovered none of the individual links in this chain — the torsional models, rig experiments, and tribological measurements listed below were performed by researchers long before us, on other assemblies. AWD.TECH's contribution lies elsewhere: we connected those results with hands-on experience of clutch-based transfer cases and showed that classical judder theory explains a symptom that all-wheel-drive owners encounter en masse yet, until now, had no coherent public explanation for. In our view, this is a case where a bridge between academic tribology and field experience was long overdue.

So the next time you feel that familiar shiver under acceleration, you will know: it is the song of a spring that nobody remembered to soothe with fresh fluid.

References

1. Crowther A. R., Zhang N. Analysis and simulation of clutch engagement judder and stick-slip in automotive powertrain systems. Proceedings of the Institution of Mechanical Engineers, Part D: Journal of Automobile Engineering. 2004. Vol. 218, No. 12.
2. Modelling, Analysis and Simulation of Clutch Engagement Judder and Stick-Slip. SAE Technical Paper 2016-01-2355. 2016.
3. Clutch Shudder Correlated to ATF Degradation through Local Friction vs. Velocity Measurements by a Scanning Force Microscope. Tribology Transactions. 1996. Vol. 39, No. 3.
4. Parametric excitation as a cause of clutch judder: theoretical study and experimental validation. Mechanical Systems and Signal Processing. 2023. Vol. 193, 110256.
5. Effects of vehicle driveline parameters and clutch judder on gearbox vibrations. Proceedings of the Institution of Mechanical Engineers, Part K: Journal of Multi-body Dynamics. 2021.
6. Study on the effect of glycerol monoisostearate friction modifier on anti-shudder performance of ATF. Tribology International. 2022.
7. Modeling and analysis of friction clutch at a driveline for suppressing car starting judder. Journal of Sound and Vibration. 2018.
8. A novel monitoring method of wet friction clutches based on the post-lockup torsional vibration signal. Mechanical Systems and Signal Processing. 2013.