1997 F-Body ECM

[kur4o]

Does anyone know the correct relationship of crankshaft position sensor to Opti's low resolution? I believe there to be four lobes on the CPK reluctance wheel, which will result in one pulse for every low res pulse.

Thinking I need to add that function to my opti simulator.

NOt really needed. The signal is used only for misfire detection and obd2 emission stuff.

With the bench setup you will make some very good progress, A good time to test the onbboard flashing. I am sure you will want to change some of the stock bins to explore more of the hidden stuff.

The limp home mode is much more used on older pcms. Maybe older system can give you better understanding of the chip evolving.

[Tom H]

NOt really needed. The signal is used only for misfire detection and obd2 emission stuff.

With the bench setup you will make some very good progress, A good time to test the onbboard flashing. I am sure you will want to change some of the stock bins to explore more of the hidden stuff.

The limp home mode is much more used on older pcms. Maybe older system can give you better understanding of the chip evolving.

If I don't drive the crank CKP, will the PCM enter limp?? Or a better Q... what causes the PCM to limp. If there is no map, no maf I don't think the PCM would know how to fuel: would expect no limp home. (?)

[Tom H]

With the bench setup you will make some very good progress, A good time to test the onbboard flashing. I am sure you will want to change some of the stock bins to explore more of the hidden stuff.

Been a while since I was messing with flashing. Perhaps I should have got rid of VATS with a flash... PITA

[kur4o]

The true limp will be triggered by catastrophic internal hardware failure within the pcm, cpu is fried or something,

Than some fixed spark advance will be used with some very basic fueling based on rpm only.

[Tom H]

Data in this post is for the wrong part... Formula and graph are for a 3bar map... Sorry for the confusion
I will re-work based on Kur4o's data in the next post

On the test bench, I started to look at a few of the PCM inputs. Fuel for open loop looks heavily at the MAP, so I started there.

On the web I found documents that detail the MAP sensor. The basic formula for MAP voltage is Vout = Vsource ((0.00318 * P ) - 0.00353)
Unloaded of any other peripherals, my Vsource is almost exactly 5.0V
This ignores temperature effects, and part allowable error. If anyone needs the data, I have it.

Here is the nominal graph I will use in test

MAP_Graph.jpg

I wonder what a normal MAP sensor value is at idle (KPA or volts) ?

-Tom

[kur4o]

On healthy engibe 30-35kp map is normal. It will be 1.0-1.3volts measured as voltage.

Map is used for spark caclulation, and for fueling if maf fails.

the pcm uses this formula for map streaming (0.369004*x)+10.3542
x=map voltage in hex

[Tom H]

the pcm uses this formula for map streaming (0.369004*x)+10.3542
x=map voltage in hex

Thanks! Perhaps this mistake was found earlier also, I have now re-named all the files for 3 Bar MAPs to stop me from using bad data yet again.


Slightly different topic... Once the PCM is powered, at some point the ignition is turned on. At that point the PCM reads the MAP to establish a baseline barometric pressure. I probably need to simulate this for a properly working test bench. Your thoughts?

-Tom

[Tom H]

Hi Kur40

I am thinking "x=map voltage in hex " is the a/d numeric output value. That would give the Map a range from 0 - FF of 10.3 through 104.5. sound right?

[In-Tech]

Thanks! Perhaps this mistake was found earlier also, I have now re-named all the files for 3 Bar MAPs to stop me from using bad data yet again.


Slightly different topic... Once the PCM is powered, at some point the ignition is turned on. At that point the PCM reads the MAP to establish a baseline barometric pressure. I probably need to simulate this for a properly working test bench. Your thoughts?
-Tom

Hiya Tom and sorry to butt in,
I really enjoy watching your incredible tracing ability. Just one thing I would like to point out is, unless you are trying to break into the "boot" sequence, always power the ecu THEN ign on after boot. All of the math equations internally are based on time. Gm spent quite a "bit of time" figuring that out Anyhoosit, pretty much does a baro check on key on THEN calculates map from there.

[Tom H]

Hope this is more like the expected Xfer function...

1BarMap.jpg

[Tom H]

Butt_ins R Good.

At the moment, I am just trying to simulate the start/run sequence & not break anything. We will get to breaking stuff later...

My test bench need to be able to simulate the startup sequence. Concern is that if the Opti signals are doing their thing, that might mess up the baro measure. It takes sooooo long looking through the code to figure out some of these details...

I probably need to build some sort of test board to bias the MAP input at baro with key off, THEN when key on, delay, start Opti Simulator, bias map to throttle closed idle (use the number Kur4o gave me)

(??)

-Tom

[Tom H]

Butt_ins R Good.

At the moment, I am just trying to simulate the start/run sequence & not break anything. We will get to breaking stuff later...

My test bench need to be able to simulate the startup sequence. Concern is that if the Opti signals are doing their thing, that might mess up the baro measure. It takes sooooo long looking through the code to figure out some of these details...

I probably need to build some sort of test board to bias the MAP input at baro with key off, THEN when key on, delay, start Opti Simulator, bias map to throttle closed idle (use the number Kur4o gave me)

(??)

-Tom

Also.... getting tired of loading the optisim everytime I want to try something. I need to rejig the code to run in internal EEPROM and set the board up to run my code on power up. Also need an inhibit input to stop the OptSim when key is off/ in a delay before starting. I guess I can't complain, nothing else to do when Covid is roaming round

[In-Tech]

Hiya again,
I built a poormans logger back in the day that logged bootup to ram and wrote to eeprom, it wasn't on this stuff but still might work with a bit of changes. That was back in the 90's and was for dynamic code stuff so it will take me a while to find if you are interested, otherwise it can remain in the dust bin :)

[Tom H]

Hiya again,
I built a poormans logger back in the day that logged bootup to ram and wrote to eeprom, it wasn't on this stuff but still might work with a bit of changes. That was back in the 90's and was for dynamic code stuff so it will take me a while to find if you are interested, otherwise it can remain in the dust bin :)

Thank you for your offer! At the moment I have SO many half done projects in this PCM. I hesitate to take on more. At one point I even started to write a 'hc11 simulator that would permit me to single step through code without thinking about the state of flags and so on. Rather than having you do the dust bin dive right now, perhaps I could request later if it becomes a factor.

As of today, I now have a faux MAP sensor. It is just a few resistors and lets me switch between a canned barometric pressure and an Idle pressure as was suggested a few days ago. I have LEDs connected in place of injectors (with bias resistors to take care of OBDII type faults). I can see the pulse width increase as pressure goes down. LEDs are OK for a rough monitoring of activity, just hope all the flashing doesn't trigger an epilepsic event <grin>

I have also modified the Optisim so that I can turn on/off the high/low res pulses. will now put in place RPM change from keyboard and that bit is done.

-Tom

[kur4o]

I can`t wait you finish the opti simulator and the test bench.

It will be real fun discovering new features and doing some more patch work.