Thoughts on turbocharger behaviour (1)

Supercharging, turbocharging, charge cooling
Guy Croft
Site Admin
Posts: 5003
Joined: Sun Jun 18, 2006 9:31 am
Location: Lincoln, UK

Thoughts on turbocharger behaviour (1)

Post by Guy Croft » Tue Jun 20, 2006 4:36 pm

Ignore the gases for a moment and imagine the turbo as standalone unit - a mini engine (where the shaft power comes from the turbine) driving a centrifugal compressor.

The compressor's operating behaviour, given enough shaft power to drive it, depends on keeping the impeller (inside it's surrounding architecture of volute insert, diffuser, air filter) away from two zones:
surge - too much pressure, insufficient mass flow - leads to sudden and violent reverse flow & broken vanes, choking - too much mass flow, insufficient pressure - leads to overspeed

Within those borders the mass flow and boost pressure the compressor can develop varies according to the impeller efficiency and the shaft speed. There are defined zones of good efficiency in which a well-matched compressor will perform well, giving useful power and economy and outside of them regions that yield poor - or even catastrophic results. The boost pressure on a turbo test can be readily controlled/altered, but on real engine it will vary hugely according to the general flow of air and gas across the cylinder and the effectiveness with which a particular engine burns fuel and converts it to energy to drive the piston and the turbine.

It is well known of course, that the power to drive the compressor and make it perform useful work on the intake air (ie: convert velocity into pressure) comes from the turbine, in turn from the exhaust gas energy (enthalpy - a combination of heat, pressure and mainly velocity). The turbine is constrained by similar map characteristics to the impeller and depending on those characteristics, it is usually possible to generate more turbine power without changing the wheel itself by altering the ex cam, gasflowing the head, reducing downstream back-pressure or improving the exhaust manifold - using pulse tuned headers eg. But there is a limit at which the turbine will need to be changed for one with more torque. It must go outside its map parameters. And, you will be familiar I am sure of the problems relating to lag from turbine inertia where a large wheel is used because the original one did not have enough power to drive the bigger compressor.

Bigger turbos produce more power for same boost and inlet temp, this is because - if we're talking about a higher capacity impeller - if it's coupled to a turbine powerful enough to drive it - and it is operating stably within its efficiency regime - it can develop higher mass flow at the same manifold pressure and be well clear of the critical surge and choking regimes.

Does this really happen, well on properly modified and tuned engines yes, but not really on standard trim production ones. The typical 'exception' is where the owner decides to run higher boost (and the engine is capable of giving it), but he takes a huge chance on wrecking the whole engine due to mechanical, thermal or other stress overload to the reciprocating or turbo parts way beyond design limits. There is always some latent power capacity to enable this, how much; well your guess is as good as mine.

No, production cars are so well matched by real experts these days, using meshing solution software like Wave, that the overall power gains (averaged across the rpm load spectrum) from altering the turbo at either or both ends are largely imaginary. After all, going back to my 1st point about considering the turbo as 'mini engine', where is the extra turbine power coming from? Bigger impeller? Well, is the engine capable of generating more exhaust gas energy as it stands? Well, safely? Without detonating or melting something, probably not. Usually what you get when it's just a turbo change is a wallop of power at the top and less down below, very dangerous for road use.

I think it's a very bad idea to swap turbos without the advice of a matching engineer. Very little can be assumed from A/R ratio and wheel diameters. Added to that the nozzle If fitted) and diffuser are so critical to get the best out of either end.

Really substantiable, safe results only come from dyno test coupled with accurate measurements of in/ex mass flow, pressures and shaft speed. One cannot assume that a larger impeller and/or a bigger turbine will safely develop more power, as it is easy to put either end into an unstable area. A turbocharger can withstand mild surge and choking but usually the first warning is the last and it takes out the whole engine due to part ingestion, and it is so easy to run in those zones and never know. Similarly bad news, a bigger turbine can overspeed a small impeller, and a bigger impeller can take the turbine out of the loop too. Overspeed can cause stress damage, broken blades ... Getting either wrong can generate terrible (undriveable) lag problems too.

The whole trouble with petrol turbo engines is that we're asking a machine with a very confined operating range to do the impossible. They are fabulous for diesels with a 3000 rpm band, but on petrol engines with a 7000rpm + band, it is inevitably a case of adding it on at the top and taking it away at the bottom, and there we're stuck with the joint problem that we have a low CR at the very time (off boost due to low ex gas energy) we need a high one.

Do all compressors generate a different map (pressure ratio vs. corrected mass flow) according to the turbine they're coupled to? No, the map is the unique aerodynamic signature of the impeller wheel for a given diffuser and compressor intake duct. The turbine merely determines the attainable shaft speed under load and of course to an extent the inertia. So 'the zone of the map in which a given compressor unit will operate will vary according to the turbine they are coupled to'.

There can be a difference of many percentage points in efficiency between two impellers of the same diameter, according to the blade and hub aerodynamic design and layout. The performance is also hugely affected by running clearances between the impeller tips and the housing. I've seen impellers that looked identical to the human eye but had 4% efficiency difference just due to different backsweep angle and splitter vane profile. When they are being mapped up the turbine is retained and only the impeller changed back to back. When they are being matched, the entire combo - impeller, intake duct, diffuser, turbine housing, turbine entry duct, nozzle and turbine itself has to be selected in order to determine an optimum unit setup.

Hybrid was just a sexy name coined to describe aftermarket built-to-order units. Most of it is based more on experience than science. Get it right and the car goes well. Is it optimised? ONLY those guys in the aftermarket who do on-engine mapping with turbo speed probe can do this effectively.


Who is online

Users browsing this forum: No registered users and 1 guest