[Diy_efi] Bosch L-Jet AFM question...

[Chris Conlon]

At 07:16 PM 4/1/03 -0800, Chris Ice wrote:

>Am attempting to replace the AFM with a MAF sensor (using a signal 
>conditioner)...and am not sure how/where to send the voltage.
>
>Pin 6 - as far as I can tell is ecu ground
>Pin 7 - afm signal voltage to ecu (at idle, 3.2v across 6&7)
>Pin 9 - 12v supply (w/o car running)
>Pin 8 - what the frick is this?  7.6v?
>(36/39 = fuel pump, 27&6 is air temp...those are easy)
>
>If I simply send signal down pin 7 (leave 8 and 9 disconnected), the ecu 
>is clearly unhappy with this and just dumps gas pretty much w/o regard 
>to the signal voltage on 7.
>
>I've read in various places that the afm voltage is corrected for dips 
>in battery voltage.  If I put a 200 ohm resistor between pins 8 & 9, I 
>can get the car to run, but it's now way lean.

Pin 8 is almost certainly "Vc". Older Bosch AFMs, which are powered
by Vbatt rather than regulated +5, often output a lower voltage back
to the ECU as a full scale reference value. From what I've seen this
tends to be a fixed, regulated voltage, which is low enough that it
should be able to be maintained even when battery votage drops pretty
low. Rather than providing an output range from 0v to Vbatt, which
will vary, the AFM provides an output voltage that ranges from 0
to Vc, which does not vary.

If you mentioned whether your AFM was a rising-voltage or falling-
voltage model, I missed it. However, most of the Vc-style AFMs I
have seen have a rising voltage with rising airflow. In this case
leaving the Vc pin open or (essentially) shorting it to Vbatt will
cause just what you've described. In one case the ECU thinks the
full scale output range is 0v to 0v, and so even 1v translates
to a huge airflow. In the other case, pulling Vc up to +12v would
make any airflow signal seem less by comparison to that higher-
than-usual voltage than to the usual voltage (7-9 volts).

My suggestion is to carefully measure Vc with Vbatt at slightly
different levels. If it is always constant, get an adjustable
regulator to match that voltage exactly. (And if not, pick a
likely value from the observed range and go with that.)
Whatever values your converter box outputs are *relative* to the
Vc voltage, so if Vc is going to change at all, the signal you're
sending down pin 7 has to change in ratio to the Vc value.

   Chris C.


_______________________________________________
Diy_efi mailing list
Diy_efi at diy-efi.org
http://www.diy-efi.org/mailman/listinfo/diy_efi