Every compressor and turbine wheel has a speed limit. Every one is different, based on wheel diameter, trim size, blade thickness and design, cutback angle, operating temperature and material. Every wheel is "burst tested" which means that we spin them up to well above their normal operating range until something lets go (usually part of a blade or hub). Needless to say, we don't publish the speed limits - the list is pretty big and there is no way on most vehicles to monitor turbo speed.

However, when we match a turbo to an engine, maximum speed at sea level and max speed at altitude (when the air is thinner) are considered and then approved by the engine builder.

So what can go wrong? Quite a few things, depending on type of turbo and installation.

On a wastegated or VNT turbo, leaking charge pipes/hoses a blocked airfilter or restricted inlet pipework (softened pipes sometimes " suck in" and reduce their capacity) mean that the volume of air passing into the compressor is reduced. This reduces the pressure at the inlet (which can cause compressor oil leakage at low engine speed) and because the outlet pressure is directly affected by the inlet, this reduces the pressure at the compressor outlet. On a wastegated unit, this means that there is not enough pressure getting into the actuator, so the wastegated is further towards the closed position than it should be. This causes more exhaust gas to pass through the turbine wheel/housing and speeds up the turbine/compressor. More speed = more pressure out of the compressor. The upshot of all this is that the wheels are having to spin faster than normal to compensate for the blocked airfilter. This can also happen in a similar way with a VNT, but it is the engine's ECU which senses that there is insufficient pressure at the compressor outlet and closes the vanes up to increase speed and compensate.
Of course, exactly the same thing happens on a turbo which has been re-calibrated to run above it's normal pressure or when a higher rated actuator is used. Higher pressure out of the same turbo = higher speed, pure and simple!

Re-chipped engines. I cannot generalise on these - they're not all bad, but basically the only way to get more power out of an engine is to burn more fuel. To burn more fuel, in most cases you need more air. More air = more boost pressure = more speed. I say in most cases, because on some diesel engines there is already an excess of air at high engine speeds, so not every engine will need to have the turbo speed tweaked to get a little more power, but for some re-chips - they may really overspeed the turbos! Nobody seems to publish any info on this.

When a turbo "fails" due to overspeed (I contend that it is not a failure - it has become damaged!) there are a few typical signs. First to show is usually the compressor wheel, which exhibits an "orange peel" type of finish on the back disc of the wheel. This is because as it overspeeds, the wheel grows in size. This causes minute cracks between the grain boundaries of the material. In mild cases, the material returns back to it's original condition (like elastic), but in most overspeed cases, the cracks continue to grow and spread until they become quite visible to the naked eye. Eventually part of a blade or section of the hub may "let go" in a typical fatigue failure, but the orange peel is the giveaway to the cause.

Similarly, when a turbine wheel grows due to overspeed, this is often characterised by the back disc taking on a dished or concave shape. This is because the centre hub is very solid and doesn't expand as much as the blades, this effectively bends the blades backwards towards the compressor and eventually, the wheel assumes this shape as it passes beyond the point where it will "spring back" to shape. Another typical symptom is the loss of part of the exducer portion of a blade or blades. To be sure of the cause, we clean the wheel in an ultra-sonic tank and view the fracture surface under high magnification. From this we can usually see quite clearly where the initiation of the fatigue crack is. If it is on the high pressure side of the blade (the inside of the "bucket" shape of the blade, then is is almost always overspeeding or cyclic overspeeding as we usually refer to it - it is the expanding and contracting during speed cycles that makes the crack grow.

Of course usually all of this is masked by loads of other damage, losing a blade causes extremely high out of balance forces which bend the shaft and sometimes snap it between the journals, hammered bearings and centre housing, rubbed turbine inducer blades and this often leads to losing part of the inducer blades (looks similar to foreign object damage) due to the high frequency vibration caused by the rub. The same often happens at the compressor side and of course all of this is usually accompanied by clouds of blue or white smoke as the oils spews past the damaged seal area at both turbine and compressor end. All in all, the turbo is a right mess and all caused by overspeeding!

Garrett Troubleshooting Guide
Check it OR regret it PDF
Check it image
Check it image
ISO 9001 bsi image
Mitsubishi image Garrett image Borg Warner image