Wednesday, Nov 02, 2005 at 17:04
Hi Richard and John,
Indeed an interesting discussion. My apologies to Coyote - we seem to have thoroughly hi-jacked your original thread about turbo timers - but it's all good stuff, no?
Let me say again, I make no claims to be an expert on engine and turbo-charger design - my experience is in instrumentation and control systems. But I did study (and pass! - strangely enough) units in thermodynamics & thermo-fluid mechanics, albeit many moons ago now. So, while I think the following is fairly accurate, I'll happily stand corrected if some (or all!) of this is wrong:
As I understand it, there are two basic types of turbine, impulse and reaction. The difference is easier to understand if you think of hydro-electric (
water) turbines.
Impulse turbines, such as Pelton
Wheels, rely on relatively low flow rates but high pressure
water (say, several hundreds of metres of 'head' pressure) to create a jet of very high velocity
water (ie. lots of kinetic energy) which impacts upon the turbine blades (more like little buckets really) and the kinetic energy is converted into mechanical energy in the turbine shaft. The 'exhaust'
water has no pressure and low velocity.
Reaction turbines, such as Francis turbines used in many low-speed vertical shaft hydro generators, use large flow rates and relatively low pressure (head). They rely on the 'reaction' of
water flow over the turbine blades, in the same way an aeroplane wing creates lift.
With gas turbines it gets a bit messier, as gases expand and contract and change in temperature, while
water (in general) does not. Power generation steam turbines use, AFAIK, a combination of impulse and reaction turbine blade designs. IIRC, the first few blade stages are primarily impulse blades to take advantage of the velocity (kinetic energy) of the steam emerging from the throttle jets, while the majority of the stages are reaction blades, using the flow of the expanding (and cooling!) steam to convert thermal energy into mechanical energy.
I believe turbo-charger exhaust gas turbines are reaction turbines. Therefore, as you both say, it is gas FLOW over the turbines blades that produces the mechanical energy to drive the compressor side. And that the gas will only flow if there is a pressure difference between the inlet and outlet sides of the turbine to create the flow. BUT, it's a basic law of physics that if you lower the pressure of any gas, it gets cooler [and vice-versa, if you increase it's pressure it gets hotter] (- after all, this is the fundamental principle that makes diesel engines work in the first place).
So the gas coming out of a turbine must be lower in pressure AND temperature than what went in, otherwise you've not extracted any energy from it and the turbine can't develop any power.
As an aside to back this up, AFAIK, all power generation steam turbines of any real size (say, >60MW) use superheaters and RE-HEAT. The steam comes from the boiler's superheater stage, where it's deliberately heated to temperatures
well above the boiling point at whatever is the operating pressure, and is expanded across the first turbine stage (HP - high pressure). Then the steam flows back to the boiler and is re-heated to a higher temperature again before being expanded through the MP and LP turbine stages. The steam pressure doesn't increase during re-heat (it can't, otherwise the flow would start to go backwards!) - only the temperature.
At the exhaust of the LP stage, the pressure is kept at nearly full vacuum (about -100kPag) and the temperature is below 100C. The object is to have the steam going into each turbine stage at the highest practical pressure AND TEMPERATURE and have it coming out at the lowest pressure AND TEMPERATURE (but still in the gaseous phase). This gives the maximum possible power output and efficiency.
Back to automotive turbos: Richard, I don't think I'm doubting the abilities of the design engineers by questioning the placement of EGT sensors. It is an undeniable and easily measured fact that there is a large temperature drop across a turbo-charger's exhaust gas turbine at high loads. Engine designers are quite able to quantify the difference for any particular 'standard' engine design and specify an appropriate 'downstream' limit in their operating instructions.
Many of my customers are quite happy with their downstream measurement. But I maintain it is inherently better to measure the upstream temperature directly, rather than infer when it might be too high based on the downstream temperature and an assumed maximum difference. More so for any 'non-standard' engine ('chipped', non-standard pump settings, aftermarket turbo, etc.), as more unknown variables come into the mix.
I agree that "Heat has everything to do with it" is over simplistic. And that warm up and cool down are both important. And that, in general, the manufacturers guidelines should be followed. [But then, few of us do this religiously. Anyone with any non-standard part or accessory on their vehicle is not following the guideline I've seen in every owner's handbook I've ever seen: "Fit and use only genuine xxxxxx parts and accessories"...]
John, not sure exactly what your point is about intercooler pressures? While I've not measured it myself, I be surprised if a clean, original 4WD vehicle intercooler had a drop of 3 psi as standard. However, if you're concerned about too much pressure drop across it (and assuming it's not clogged or excessively fouled internally) a simple way to overcome the pressure drop in non-electronically controlled engines, is to move the pressure sensing point from the compressor outlet port (where is usually is as standard) to a tapping off the actual inlet manifold. Then the wastegate will be operating to maintain constant manifold pressure rather than constant compressor outlet pressure.
The small hose from the compressor outlet to the wastegate will usually have a tee-off going to the injection pump's pressure compensator. Make sure this is moved to the new tapping point too and remember to plug the compressor port.
Eagerly await the next instalment(s). Maybe we should start a new thread - Turbo Talk????
FollowupID:
391137