a MAF too small

[Jason Haines]

At 11:57 AM 7/24/99 -0500, you wrote:
>>
>>Regarding the minimum flow rates, you do need to make sure you are not at
>>the bottom of the scale at warmed up idle but we have used this technique
>>(the two MAF method) as have other companies when nothing else would work
>
>Nothing else meaning didn't try MAP?.

For a boosted V8 application using a GM OE computer system, switching to
MAP isn't that easy. On NA cars, many of our customers do switch the
1985-1989 vehicles to the MAP based 1990-1991 systems. On the newer
vehicles (1994 and newer), switching to MAP is not really an option and the
MAF works so well for why would you want to.

>>(I have even seen a GM development vehicle with two MAFs but only one
>>connected but I was told they had also doubled the MAF g/s to frequency
>>table in the calibration - in the example described with the 1986 to 1989
>>that is not possible the table is already at the maximum of 255 g/s and
>>with the voltage type MAF the range is fixed 0 to 5 volts unlike the
>>frequency type that often have much more range than the vehicle ever uses).
>
>????, not in the case of the GNs..

OK. Granted, some of the very high horsepower GNs are probably past the
roughly 420 to 500 g/s that is available in the tables of the newer
vehicles, but for many vehicles 420 to 500 g/s (roughly 550 to 675 HP) is
plenty of range.

>>
>>On modified engines you are usually idling at a higher g/s than stock
>>(especially if you have a larger engine and a higher than stock idle speed)
>>so you can usually half the scale at still not have gone below the bottom

>>of the table. One thing to do on voltage based systems would be to check
>>your warmed up idle voltage from the MAF - stock is 0.5 to 0.7 volts. If
>>you are higher than this on a modified engine then you have some room left
>>in the table. Keep in mind a lot of the calibrations are designed for very
>>low idle speeds (525 rpm in some cases).
>
>But, double?, I just haven't seen it.  Mybe if idling at 1K, but who wants 
>that?.

You don't need double. If you look at the 7165 vette hac someone posted and
look at the data I posted before you will see that the stock 1985-1989
vehicles idle at roughly .5 to .7 Vdc or 6 to 8 gps and the table goes down
to 3.2 gps plus it can average between the 0 value and the 3.2 gps value. A
383 or 420 CID engine with a larger cam (more overlap) and a 700 or 800 rpm
idle (or a supercharged car with the supercharger load) will usually idle
at 10 to 14 gps depending on the engine/cam/etc. So half of that is 5 to 7
gps which still falls into the table values. I am not saying this is the
best way or the preferred method but it can be made to work if you are
trying to work around ECM limitations.

>>Finding a larger (higher flow range - not restriction) 0 to 5 Vdc Bosch
>>type meter has not been easy either. It seems that in those cases where
>>higher ranges were needed, some of the OEMs also went to two MAFs (Nissan
>>Skyline, BMW M5, Mercedes V12 etc...).
>>
>>So who feels like making an analog averaging circuit for two 0 to 5 Vdc
>>signals? You could leave the burn off circuit attached to just one MAF or
>>make it operate both MAFs if you wanted.  8-)  (What would the time delay
>>in such a circuit be?)
>
>I still say a $50 MAP is the better answer to the $300 Hi Flow MAFs, just 
>wish we had more versions of code to run 2-3 bar maps.
>For the $300 MAF, might as well run a Haltech, and then be able to have full 
>time Baro calibration map

A GM based 2 or 3 bar MAP system for a V8 would be nice. The aftermarket
EFI systems just don't seem to have the driveability features usually found
in the production base systems (idle spark, spark scatter, torque converter
lockup, etc.).


Jason




---------------------------------------------------------------------
Jason R. Haines				jhaines at lingenfelter.com
Lingenfelter Performance Engineering	219-724-2552 / FAX 219-724-8761
1557 Winchester Road, Decatur, IN 46733 USA
http://www.lingenfelter.com		catalog, shop talk and more
sales at lingenfelter.com			sales and tech questions
---------------------------------------------------------------------