shiv@vishnu

Hi guys,
Quick question: What is the upper voltage limit on the factroy MAF sensor on a 996tt. Is it 5v? Or is it 12v? IIRC, the modded 993tt i've tested went over 5v under heavy load. I have not been able to see anything over 5v with our car, suggesting that max value is indeed 5v. But our car is running stock turbos so I can't be sure. Further confusing matters is that there is both a 5v and a 12v wire going into the MAF sensor. Perhaps the 12v is for the MAF and the 5v for the IAT? Just guessing here....

Thanks,
shiv

cpu77

I can't believe Shiv from vishnu tuning is actually posting here! You made a real nice turbo kit for the GC8 RS. You definately turn out a quality product! Nice to see ya here. Sorry I don't know the answer to that question I am sure Kevin from UMW will chime in

BTW are you tuning 996TT now?

JasonAndreas

The maximum output is 5V.

The supply voltage is 12VDC and the reference voltage is 5VDC. The IAT is a negative temperature coefficient sensor so you would need to measure resistance.

IrocMan

Hi,
In addition to what JasonAndreas says, if you set a custom EFI (and if you ask for voltage, I guess you do) you should think about having another temp sensor (like the basic Bosch in copper one - don't have the ref here) that has a more linear response than the factory NCT. I'v had issues with that.

Cheers

shiv@vishnu

Jason- Thanks for the info. That make a lot more sense. So those 500+hp setups basically max the MAF at 5v? Or do they swap out a larger diameter MAF housing or install a new MAF?

IrocMan- You hit it right on the head We're currently implementing our PROcede computer on our car. Been using it in the new twin turbo BMW 335 for some time now. And so far, things are looking equally nice with the 996tt. The MAF limitation took us by surprise though.

cpu77- Thanks for the kind words. And yes, we will have a tuning option soon for the 996tt/997tt

Cheers.,
shiv
www.vishnutuning.com

roadsterdoc

Most just max out the stock MAF (which is weak and commonly blows). Some are using a larger diameter MAF housing, and there are two different setups using a different MAF sourced from another vehicle. I know that one in use is made by Hitachi, but on at least one car it creates a lot of CELs. A new MAF solution is ideal.

shiv@vishnu

Thanks for the info. Very helpful. What about a T-MAP sensor conversion? Seems like it would be a good fit in a car that constantly runs closed loop wideband control. It should be quite easy to program a calibration table that spits out a pseudo-MAF signal based upon MAP and charge temp. Failing MAFs isn't too surprising given that even a stock car is bumping into its ceiling. It just strikes me as odd that mos tof those high boost setups offer the ECU no reliable load reference above 500hp. It's like running a 2 bar maps sensor in a car that runs 20psi of boost.

-shiv

ebaker

I believe the 996 TT switches to open loop under most full load situations.

Jean

The ME7.8 has the ability to do pressure sensing over throttle position, but only uses it when the MAF is broken. And when that occurs, everything else is in limp home mode, the ABS/PSM lights are on.


The stock MAF makes it's peak voltage at about 51lbs/min, with these engines having an efficiency of about 9.5hp/lb/min of air, you will be seeing around 490HP or so. You need to put a bigger MAF housing.

The problem with stretching the MAF capability is that the calibration will need to be boost specific or only tuned at the particular range where ou want to run the car.

Some tuners do use the O2 sensors as their primary feedback sensor because they are beyond the limits of the MAF however it also has its complications on high HP engines since the tubing will be wider than stock and the sensor will not read very accurately.

All the Info comes from a very knowledgeable tuner who has been doing this succesfully for a while now.

shiv@vishnu

I know for a fact that my 2001 996t stays closed loop all the time. In fact, comparing the 2001 ecu pin out to that of a 2002 shows extra inputs for wideband o2 sensors. Looks like 2002 and up may be more convention in this respect.

shiv

shiv@vishnu

Excellent info and 100% inline with what I've been measuring. I've noticed that there is no high voltage fault trigger with the MAF. That is, the ecu is completely happy being fed a 5v signal. Overboost protection seems to be triggered by the MAP sensor which can easily be soft clamped. And it's possible to bias the widebands so that it adds in big short/long term fuel trims under boost. I'd love to know, for sure, the ECU changes made in 2002.

-shiv

sfl993t4

Very interesting info. There must be another factor than just the MAP as adding a resistor to the MAF signal line appears to work. If the suggested MAP based overboost protection is functioning, it would need to only work when the MAF is at max voltage, otherwise reducing the MAF signal would have no effect.
Of course I could be wrong.
Steve

IrocMan

I definitely agree. This was used a lot in Formula 1 and Supertourism in France when I still was an intern, a long time ago.
A while ago also -but mostly because of money- my first turbo engine only had a WB O2 sensor signal mapped to a [0-5V] as a compensation of the basic TPS/RPM map. It was cheap and worked pretty well, later we added exhaust temperature.

Mr C

shiv, I think you should actually look into real tuning options for Bosch Motronic ECUs, piggyback isn't a very nice solution, especially for fuel/timing.

just my 0.02.

roadsterdoc

Thread revived. Shiv certainly has experience with piggybacks, stand-alones and flashed stock ecus. Anyone remember Frankenstein from the early 90s?