Turbo speed sensor ever used ?

[Bernd Felsche]

Espen Hilde writes:

>> mmm - Interesting, I'd assumed air flow measurement and MAP would be
>> sufficient to determine volume of air entering the chamber. Are you
>> suggesting AFM doesn't give an accurate enough figure due to backpressure
>> at the turbo and that MAP in no way assists this - in the circumstance
>> where we have an AFM and MAP ?

>The AFM dont know the backpressure and therefore the ecu cant predict 
>the right advance....we are wanting to be right on the edge of detonating
>here....at a 700hp 2l turbo engine 0,5 degree wrong advance at max torque
>is 40 hp lost.

The state of the turbo is defined by the operation of the engine.
In theory, the ECU should be able to work it out given sufficient
knowledge and the appropriate algorithm.

If the *backpressure* is so critical, then you should arrange for
phasing the exhaust pressure pulses before the turbo so that the
exhaust still "scavenges" the combustion chamber as the valve
closes during overlap.

A reduction in overlap may also be worth considering. You don't
really want internal exhaust gas recirculation if your sole quest if
for maximum power.

>AFM and a MAP sensor on the exhaust could do the trick if the AFM is 
>fast enough.

Those sensors would not be fast enough and sufficiently resistant to
the 500+C temperatures in the exhaust. If you're getting 700bhp from
2 litres, you're spinning at close to 10,000 rpm if not more. A
pressure sensor needs to pick up the peaks and throughs at the
exhaust valve to adapt the airflow.

Similarly, to accurately measure the airflow, you need a fast
pressure measurement at the inlet as well. There are pulsations in
the airflow even with (turbo-)supercharging. Transducers will
typically not give you enough information at high speed.

Pressure sensors take about a millisecond to respond to a pressure
change. So at 10,000 rpm, you get all of 6 samples per revolution -
not very accurate in identifying the pulsations. (vis Motorola
MPX5700 or MPX4250A) Those sensor need to be mounted remotely from
the engine - else they become soup - so you need to have a pressure
line to the ECU that introduces minimal delay. An oil-filled tube
would improve the response time - you still need to allow for the
speed of the wave propagation to determine precise phasing.

>How the algorithm or map for backpressure vs .boost should work beets
>me....Ignition retard under accleration when backpressure is higher than 
>steady state...
>Its complex......
>Turbo speed VS. engine rpm wot table?
>A backpressure table of %influens on volumertric filling at wot?
>Use a valve on the exhaust pipe for adjusting backpressure under dyno
>runand fill the tables....a hard day work? 
>Different engines responds different to backpressure at different rpms.

Because of the dynamics of the flow as noted above. Having no
overlap would go a long way to solving the problem.
As you don't have a long exhaust tract, you lose out on taking
advantage of an "under-pressure" to help evacuate burnt gases from
the combustion chamber as you have a turbocharger sitting in the
way, a short distance downstream from the valve.

	If you had freedom in designing your cylinder head and
	exhaust valves, I'd suggest that you have one exhaust valve
	from each cylinder directed to the turbocharger, opening
	first and a second (or third) valve opening later to a
	turbo-bypass with the first closing slightly to prevent
	return flow.  That permits a degree of overlap on the second
	exhaust with the possibility of useful scavenging.

	With say three camshafts, a variable timing of the second
	exhaust camshaft and hence the valve timing will also
	alter the boost pressure - the duration of the second
	exhaust opening need only be (guessing here!) about 100
	degrees. The maximum duration would be determined by the
	boost requirements and the limit of overlap.

	Also, the greater the advance on the second exhaust, the
	greater the scavenging, but the lower the boost. Applying
	traditional exhaust tuning techniques (resonance, etc) to
	the bypass exhaust would mean more effective scavenging
	during overlap as the bypass exhaust is only carrying a
	portion of the full exhaust flow - between a quarter to a
	third (again, guessing).

>> I've alwayus wondered what the relationshiup is between boost and less
>> ignition advance - is it simply chamber filling, propensity to ping, or
>> some other presumably non-linear relationship ?

>Exhaust backpressure is driving the exhaust temp to the roof..more residual
>hotexhaustgasses .more prone to detonation. 

The higher temperature will also contribute to detonation because
the exhaust side stays hot.  That may be good for increasing boost,
but not so good if you have to lose power by retarding the ignition.

If the static backpressure is too high and phasing the exhausts
doesn't contribute to better throughflow, then it may be necessary
to split the inlet and exhaust systems to equally-phased cylinder
groups, necessitating multiple turbo-chargers.

-- 
Real Name: Bernd Felsche
    Email: nospam.bernie at perth.DIALix.com.au
	http://www.perth.dialix.com.au/~bernie - Private HP
----------------------------------------------------------------------------
To unsubscribe from diy_efi, send "unsubscribe diy_efi" (without the quotes)
in the body of a message (not the subject) to majordomo at lists.diy-efi.org