a "cost-effective" AFM->MAF conversion (long)

[Rob Clark <rclark@xxxxxxxxxx>]

Background:
- -----------

So I was reading the "Tired of Living In Vane?" (Nov 98 _Roundel_) 
about the AFM to MAF conversion.  I figured it should be a relatively 
cost effective (ie cheap) upgrade... I replaced the MAF in my Ford 
Ranger shortly before selling and buy my 325is, and the part was 
about $40, if I remember correctly.  So, then I check 
http://www.splitsec.com to learn about the conversion kit... they
want $850 for the damn thing!



So I was thinking:
- ------------------

The job of the conditioner that converts the MAF signal to a AFM-like 
signal seems pretty easy.  Remembering the ideal gas law from freshman
chemistry:

   PV = nRT

Where P = pressure, V = volume, n = # of moles, T = temp, R is a
constant.

The MAF measure the mass of the air (per unit time), while the AFM
measures volume of air (per unit time).  Since the mass of air is
proportional to n we get:

   V = (mass) * (some_constant) * T / P

So far, it sounds like what the "conditioner" does is pretty simple.
The _Roundel_ article also mentions that the "conditioner" also 
dampens the response of the system, to avoid stability problems.
This is also pretty easy.  Just throw a cap in parallel to the
MAF, right?  (Ok, I admit it... I am a CS major, not an EE.)

A circuit that does this sounds pretty simple, but I think it is
going to involve op-amps... this pretty much pushes the envelope
as far as my knowledge of electronics goes.



So then I got a better idea:
- ----------------------------

Why go to all the effort of making your nice modern MAF look like a
mechanical AFM?  Why not reprogram the fuel curve in the EPROM to
use the MAF signal instead?

I don't really know what the fuel signal is a function of... I would
guess that it would be RPM, O2 signal, AFM signal, temp, and barometer.
The fuel curve would at least need to be a function of AFM signal
temp, and barometer for this to work.

Another idea would be get rid of the barometer and temp signals, and
trick the computer into thinking that there is constant temp and
pressure.  This won't work if the ignition is a function of temp 
and/or pressure.

Any thoughts?



Conclusion:
- -----------

This seems like a relatively easy DIY job, if I can figure out the
electronics, and line up all the required parts.  (If you have
expertise in either field, feel free to help!  :)

Once the plans are drawn, and any necessary circuits are assembled,
this should be a *real* easy installation... no more than a short
afternoon.  (A 3 newcastle brown ale job... sorry, I'm not much of a
Rolling Rock fan... does this mean I am forever disqualified from
bubbaship?  :)

And this would probably yield a bigger HP increase than an electric
supercharger!  (Sorry, couldn't resist)

BTW, If I do go ahead with this, I will make any plans I draw up 
freely available.


- ---
Rob Clark  (Rob_Clark@domain.elided)
San Diego
BMWCCA 156543
87 325is

------------------------------

End of bmw-digest V9 #468
*************************