[Diy_efi] pic based afm -> maf conversion

[Tom Visel]

I may be not smelling the dead horse, but in my experience - moving from 
Phoenix, AZ at 1200 feet to Flagstaff (7100) and then Prescott Valley 
(5200 feet,) the fuel injection systems which don't have a baro sensor 
(some earlier Nissan/Datsun, Toyota) do tend to run richer at the higher 
altitudes.  Yes, this has been seen in real life.  It's enough to foul 
out plugs after a couple of weeks of winter driving.  Even after the 
corroded coolant temp sensors and their connectors have been replaced 
and the idle mixture adjusted, the cars without a baro sensor just don't 
get the part/full throttle mixture close enough to correct to keep from 
fouling plugs unless hotter plugs are installed.  Yeah, I know, it's a 
kludge, but people would rather not come back with fouled plugs in the 
name of "doing it right."  Customers gotta drive...

Same thing happens when a baro-equipped (and dependent) vehicle - like a 
late-80s Ford - loses its baro signal and is too far from nominal 
conditions.  IDK if this E30 has a baro or cares much about it; it was 
simply something to consider.  Of course, one could put on the 
rose-colored (coloured on that side, right?) glasses and always drive at 
sea level.

It might be worthwhile to substitute an audio-trim 50K potentiometer 
with a 330-or-so ohm resistor in line for the intake air temp sensor to 
do some testing.  Since the air temp is a thing that can vary wildly, 
the ECM shouldn't throw a fit, unless it thinks the sensor is open or 
shorted.  When you adjust the pot, you can watch the mixture and timing, 
and see if tweaking this input is a worthwhile tool.  Even if it's not, 
logging the old vs. new intake air temp sensor curves would be 
worthwhile.  Maybe no translation needs to be done, maybe more work for 
the PIC. 

TomV


Daniel R. Nicoson wrote:

> Comments in line below:
>  
>
> Dan Nicoson
>  
>
>     -----Original Message-----
>     *From:* diy_efi-bounces at diy-efi.org
>     [mailto:diy_efi-bounces at diy-efi.org]*On Behalf Of *Tom Visel
>     *Sent:* Wednesday, January 18, 2006 6:46 PM
>     *To:* diy_efi at diy-efi.org
>     *Subject:* Re: [Diy_efi] pic based afm -> maf conversion
>
>     "very similar info to what a  AFM would" ... but not the same.  If
>     "close enough" tuning were OK, the AFM would still be there, right?
>
>     The output curves in the test rig described will have a certain
>     relationship for one density of air - barometric pressure, temp,
>     humidity - and another relationship altogether when the density
>     changes.  The same number of CFM that max out the AFM's output
>     could be so dense that the MAF is maxed out, too, or (at altitude,
>     on a summer day) they could make for a much smaller (raw) reading
>     from that same MAF.  That's the primary weakness in vane air flow
>     meters - they measure volume, not mass.  Most L-Jet based systems
>     have some way to determine barometric pressure and intake air temp
>     - this last usually in the AFM, by the way - so they can derive
>     the mass of the air.  Since the sensor will be giving a signal
>     that relates to the air mass, a way must be found to remove the
>     barometric pressure measurement (the only remaining measurement
>     external to the now-absent AFM) from the computer's calculations. 
>      
>     I haven't researched the vane meters, never needed to.  I think my
>     inclination was that the dynamic flow of air moving that meter
>     would include the effects of density (since dynamic
>     pressure=1/2*roe*v^2).  I guess it doesn't really show up in real
>     life, Tom, you sound like you know this one cold.  Point conceeded.
>      
>
>     A couple of thoughts:
>
>     If you remove the baro sensor (by making its value constant,) the
>     computer will not - if it were going to - advance or retard the
>     timing in response to weather conditions and/or altitude.  This
>     could be no big deal - the computer might not do any futzing with
>     the timing anyway based on baro readings.  It could be a problem
>     manifested as lack of torque (high altitude) or pinging (low
>     altitude.)  Probably nothing, but the only way out of it would be
>     to remove the baro component from your MAF signal before supplying
>     it to the ECM.  Any takers on this last one?
>      
>     I would just leave the stock baro and temp sensors in a similar
>     position.
>
>     If the ECM is counting on air door overswing to provide
>     acceleration enrichment, the engine may end up with a stumble on
>     sudden throttle opening, perhaps necessitating the watching of the
>     MAF or TPS for rapid increases.
>
>     In most of the systems I've dealt with, air temp has more
>     influence on spark control than fuel.  Does your old AFM have the
>     ACT/IAT sensor built into it?  IDK about your Bosch sensor, but
>     there are MAFs out there that have ACT sensors as part of the
>     package.  If it keeps your ECM from getting stupid and going into
>     damage control mode, it would be worth the effort to include an
>     air charge temp sensor.
>
>     TomV
>
>     ------------------------------------------------------------------------
>
>     _______________________________________________
>     Diy_efi mailing list
>     Diy_efi at diy-efi.org
>     http://lists.diy-efi.org/mailman/listinfo/diy_efi
>      
>
>------------------------------------------------------------------------
>
>_______________________________________________
>Diy_efi mailing list
>Diy_efi at diy-efi.org
>http://lists.diy-efi.org/mailman/listinfo/diy_efi
>  
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.diy-efi.org/pipermail/diy_efi/attachments/20060119/dfe18c01/attachment.html