[Author Prev][Author Next][Thread Prev][Thread Next][Author Index][Thread Index]

Re: CIS-E fuel modification @ WOT

At 05:55 PM 3/12/98 -0600, you wrote:
>Q-list resident-theorist Tony Lum proposes:
>My main concern is how the ECU reacts to the sudden change in resistance
>(i.e. voltage level?) it sees at the temp sensor.  Will it instantaneously
>jump into is cold-run enrichment mode, cranking the pressure actuator
>current from ~10mA to ~40mA (give or take, depending on the value of 
>your FS resistor)?  Perhaps the CIS-E III systems will suspect something
>is wrong and flag a fault code?

Sorry don't have to worry about CIS-E III on JT engine :)

>Also, you might have to be careful with tripping the WOT switch at
>low rpms ( < 3000?), since the mixture may be too rich for the car and
>cause the engine to bog.

You could re-calibrate the position of the switch to close later. Or how
about a 2 stage switch with different resistors. Or digitally programmed
resistors using PIC... nahh was supposed to be easy.

>I still like the idea of changing the 3mA value in the ECU to something
>higher, perhaps around 10mA.  If this value is stored in a PROM, perhaps
>it can be found and changed.  It is certainly a much more elegant

Certainly would.  TAP lists that +10 HP chip in their ads, but when I asked
him about it, he said there was none.

>For those with a "Bosch Fuel Injection & Engine Managment" book, there
>is an interesting picture on page 28 of the "Continuous Injection - Theory"
>section.  It should that, assuming the engine is running at 10mA of
>pressure actuator current at 14.7:1 air/fuel, 12.6:1 is somewhere near
>20mA.  This is only an increase of ~10mA.  Can you achieve that by
>using the fake-temp-sensor method?

I think Bentley troubleshooting procedure calls for 15K shunt inserted into
ecu plug and you were supposed to see 40 to 80 mA (sorry, values off the
top of my head) to the diff press reg.

>Another interesting idea that occured to me is that it is better to
>trigger the WOT switch later in the RPM range rather than sooner, which
>seems to be supported by my driving experiences.  If you trigger the
>WOT switch at ~3000rpm and hold the throttle down, the engine goes
>open loop right away and enriches the mixture, giving you a bit of
>extra power.  However, as the rpms increase, the ECU (I am guessing)
>holds the pressure actuator current static, so that the engine starts
>to approach stoichiometric, then surpasses it and begins running lean.
>Now, if you run just shy of the WOT switch until, say, 5000 rpms, the
>ECU will maintain the air/fuel ratio at stoichiometric.  Then, when
>you close the WOT switch at 5000rpms, you get that extra bit of
>enrichment, giving you a tad bit extra power, which is better than
>the lean condition theorized about in the prior paragraph.
>'85 Coupe GT
>Eric J. Fluhr                                Email:  ejfluhr@austin.ibm.com
>630FP Logic/Circuit Design                   Phone:  (512) 838-7589
>IBM Server Group                             Austin, TX
Let's see... 1. Get datalogging voltmeter attached to OXS sensor.  2. Put
laptop in car with observer to record lambda readings.  3. Make multiple
high speed runs to validate data.  4. Connect black box and calibrate. 5.
Make more multiple high speed runs to read change. 6. Recalibrate
resistor...wait a minute-I'm supposed to rebuild my calipers, change my
rotors & pads this weekend-oh nuts!

One these days I swear to stop fixing Audis and start driving 'em ;^)



Tony Lum                                        1987 5000CS Turbo Quattro
Berkeley, California, U.S.A.                	1985 4000CS Quattro
Audi Owner/Driver/Mechanic by Necessity ;^)	1980 5000S  Sedan