dprantl

Hmm, many factory turbo systems choose to have the MAF before the turbo. I haven't noticed any problems running 26psi in my '91 SAAB 9000T, and the car is about to cross 200k miles. It is true that you must have a much longer air path sealed from leaks, but I'm not sure the lag you talk about is significant.

Dan
'91 928GT S/C 475hp/460lb.ft

tveltman

Maybe not. I seem to remember reading something about throttle plates and MAF sensors, maybe in Corky Bell's book. I can't recall ATM, and I could definitely just be talking out of my *** anyhow. I am of the opinion that MAP/MAT is the way to go regardless. MAF is too finicky and has leak problems. This is one reason why I am going electromotive in my rebuild.

PorKen

For N/A cars, a LH2.3 dual rev limit should be easy to program using the idle-speed-drop input, and then you don't have issues with afterburning.


If I ever get around to making a mid-turbo, I will try the anti-lag on the EZF. The WOT input (and codespace) could be rewired for the clutch switch.


BTW: to access the terminals on the clutch switch you have to remove it. There are two small phillips head screws which hold it. Note that there are two sets of holes for the microswitch. (It's easy to try and install the switch incorrectly the first time you do it, but only the upper set should be threaded.)

The clutch switch is a normally open type, but as installed, it is normally closed. IE when you push in the clutch, the input to the EZ/LH will be OFF.

John Speake

Niklas is visiting me later this week and we will be discussing these topics and others...

RKD in OKC

On the 944 Turbo the MAF is pre-Turbo. It goes Air Filter - MAF - Turbo - Intercooler - Throttle body. The BOV dumps from between the intercooler and throttle body to between the turbo and the MAF so air stays metered. Boost pressure to operate the wastegate is taken from between the intercooler and throttle body. Boost measurement for control is taken from the intake. Pressure to operate the BOV is taken from the throttle body.

John Speake

I can't remember if John's TT has twin inlet tracts, but it would be possible to have a MAF in each leg and use a combiner box.

We are wortking on a MAF-less system as well...

dprantl

A MAF-less system would be awesome if it could be tuned to be smooth under all operating conditions.

Dan
'91 928GT S/C 475hp/460lb.ft

John Speake

The system we are working on is for the race car in my avatar, this will run with Jenvey 50mm ITBs and work on Throttle pot for load with airtemp and MAP sensor absolute airmass correction.

A development of that could be a MAF-less system to go with the stock intake, but not sure whether there is a viable market for that product ?

ptuomov

Here's what I believe:

There's precisely one reason to place the MAF upstream of the turbo. That's to accommodate crank-case breather system that is fed to the intake. Unless the air-oil separator is perfect, in a system where the MAF is downstream of the turbo, the blow-by oil may interfere with the MAF wires. More blowby oil, worse the problem.

If you either have perfect air-oil separation (like for example a well designed dry sump system) or if you don't hook up the crankcase ventilation system to the intake (either vent it to atmosphere or to the exhaust), then MAF as close to the throttle plate as possible is generally the best thing to do.

If you are making a single rear-turbo kit, this will definitely be useful. Three other things to look into. First, you might want to buy some used variable-turbine-geometry diesel turbines. Those can't hold up in gasoline engines because fo the heat, but the rear turbo's downside of cooler exhaust gas can be used as an advantage by using a diesel VGT. Second, you might want to ceramically coat the entire pre-turbo exhaust. Third, you may also want to route your bypass valve to the exhaust instead of to intake tract. This, combined with disabling the DFCO should keep the turbo spooled nicely -- especially if you can run the engine at 4500 rpm and 40 degrees retarded with your EZK anti-lag.


John's system has dual intake tracts. If the MAF were to be moved to upstream of the turbo, that would mean using two MAF's anyway, because one would need double the housing area for the low-pressure side. The voltage averaging circuit is trivially easy to make and digital dual-to-single MAF emulator only marginally harder. (I've done both, which tells you that it's REALLY easy.)

The only reason to do the pre-turbo MAF version would be to create a California-legal version of the kit with all the testing procedures and plastic bags of exhaust gas. I doubt that this is commercially feasible and/or that John Kuhn has any interested in dealing with that hole insane asylum... ;-)

The same applies to MAP conversion with the stock intake manifold. MAF system is superior in every other way except that the hot wire is sensitive to oil from the crankcase breather, which is a problem for boosted cars that need to feed in the evacuated gas upstream of the turbo.

AO

Not exactly... I have dual intakes - one MAF. (Eewww... that reminds me of a video that circulated on the internet a few years back - GROSS)

With the STer when you set your tune you're only measuring a % of the are coming in and then plugging in a fuel value for it. As long as that doesn't change over time, it's fine. So if a single MAF setup measures 20% of the air, I'm only measuring 10%. Still works. It also avoids having to use a special-built SMAF - not saying that's bad, but it's easier to source a standard MAF.

123quattro

Yes, that's how it works.

Mine is set up as a secondary rev limiter. The primary rev limiter is set at 8200 rpm with a fuel cutoff. The secondary is set at 6000 rpm with spark retard. It's only active when it sees another input. I'm using the clutch switch. When that switch is triggered, it defaults to the 2nd rev limiter. The limiter has a lot of spark retard authority. Mine is setup to pull about 65 degrees of timing. Fueling is not changed. So spark goes from ~30 BTDC to 35 ATDC. It throws a TON of energy into the exhaust. I also richen the mixture from about 12:1 while under boost to about 10:1 during anti-lag to try and lower the exhaust temp a little.

This works for both flat shifting the car at wot, and also launching the car at full boost. You just have to be careful not to use it too often or too much or you'll tear up the turbos and exhaust manifolds.

ptuomov

Have you experimented with routing the bypass valves from the intake side into the exhaust manifold? With that 10:1 AFR, that fresh air is going to UNLEASH THE FURY in the turbine! ;-)

That will work, if your system is completely symmetric. Since my trubos are clocked differently, it's not guaranteed to work for me. Using one MAF in a dual system also removes some of the incredible robustness of the MAF system to almost any changes or faults in the physical engine.

If you ever get the urge to add another MAF to make your system symmetric, the MAF signal averaging circuit components are less than $20 I recall.

tveltman

If the MAF output is a voltage signal, then the averaging circuit is indeed trivial, but should cost you something like 20 cents...It's two resistors and a common ground...

ptuomov

You probably want some isolation with opamps, connectors, and a box?

tveltman

Bah, who needs that ****?