Resetting the '97-'01 PCM

Standard wisdom: disconnecting the battery overnight on a '97-'01 XJ resets the PCM.

Here's what I'm running into: either I don't have or can't find anything from Chrysler that definitively supports this assertion.

What sparked all of this: a few months back I did the 0280155917 injector upgrade, replacing spark plugs and the throttle body gasket at the same time. Took the XJ out for a short (10-15 mile) test run to let everything bed in. Came home, unplugged the battery overnight, and the PCM presumably reset itself.

Over the next couple of weeks, I saw an improvement of around 0.5mpg which has held steady since then. Definitely better, but not quite what I was expecting. Some poking around over the next few weeks turned up a flaky oxygen sensor; new ones plus a MAP sensor (being replaced due to age, not overt failure) are on their way. Once they're in, I'd like to clear the PCM and have it be learning with the full gamut of new equipment in there.

So, my question: is there any definitive reference that states that a PCM reset can be achieved by disconnecting the battery? I'm not arguing that this isn't necessarily the correct method, just that I'd be interested to know where the idea comes from. Also, if it's possible to check with an OBD-II scantool for a reset in the PCM's stored history anywhere I'd very much like to know if that's possible.

This very simple procedure will Erase the “Adaptive Memory” stored inside the PCM and allow a new “Adaptive Memory” to be developed.

After performing this procedure the PCM will re-learn and store into Adaptive Memory your engines performance characteristics.

Please perform these steps exactly as they are written, in the order they are written. This will cause the adaptive memory in the PCM to be erased and cause the PCM to go into Fast Learn Mode adaptive mode.

Disconnect both battery terminals and touch them together for 30 seconds. (This is to discharge the PCM capacitors, which maintain the Adaptive Memory.). Alternatively just ground out the positive cable.
Reconnect the Battery Cables
Turn Ignition Switch to the “On” position but DO NOT start the engine
Turn Headlight “On”
Turn Headlights “Off”
Turn Ignition Key “Off”


The PCM Adaptive memory has now been flashed, or erased from the PCM.

When you start the engine it will be running off a set of pre-programmed tables that come with the PCM from the factory.

When you get the engine up to operating temperature the PCM will start to collect data for the “Adaptive Memory”.

The PCM will collect data for Adaptive Memory for the first 50 Warm-up Cycles.

like said above discharging the capacitors is the key. just disconnecting the battery is not going to get the job done. mike

x99j said:

like said above discharging the capacitors is the key. just disconnecting the battery is not going to get the job done. mike

Click to expand...

You are misunderstanding something, because discharging the capacitors is what is done by disconnecting and shorting the battery terminals.

Doing this will clear the memory, but as I understand it, the ECU will not realize it's been cleared and won't learn, it will just use the cleared values. Switching the lights on and off (the steps after shorting the battery cables out) tells the ECU that it has been cleared and should start learning again.

kastein said:

You are misunderstanding something, because discharging the capacitors is what is done by disconnecting and shorting the battery terminals.

Doing this will clear the memory, but as I understand it, the ECU will not realize it's been cleared and won't learn, it will just use the cleared values. Switching the lights on and off (the steps after shorting the battery cables out) tells the ECU that it has been cleared and should start learning again.

Click to expand...

This is correct

dukie564 said:

This very simple procedure will Erase the “Adaptive Memory” stored inside the PCM and allow a new “Adaptive Memory” to be developed.

After performing this procedure the PCM will re-learn and store into Adaptive Memory your engines performance characteristics.

Please perform these steps exactly as they are written, in the order they are written. This will cause the adaptive memory in the PCM to be erased and cause the PCM to go into Fast Learn Mode adaptive mode.

Disconnect both battery terminals and touch them together for 30 seconds. (This is to discharge the PCM capacitors, which maintain the Adaptive Memory.). Alternatively just ground out the positive cable.
Reconnect the Battery Cables
Turn Ignition Switch to the “On” position but DO NOT start the engine
Turn Headlight “On”
Turn Headlights “Off”
Turn Ignition Key “Off”


The PCM Adaptive memory has now been flashed, or erased from the PCM.

When you start the engine it will be running off a set of pre-programmed tables that come with the PCM from the factory.

When you get the engine up to operating temperature the PCM will start to collect data for the “Adaptive Memory”.

The PCM will collect data for Adaptive Memory for the first 50 Warm-up Cycles.

Click to expand...

Cool, thanks. It was the discharge part of all this that I had missed.

Out of curiosity: the above looks like it was copied from something Chrysler-related, judging from the language used. What was the source for it? If it was the FSM (which, unfortunately, I won't have in front of me until tomorrow), just letting me know the section and page would be a huge help; even a TSB number would be great.

kastein said:

Doing this will clear the memory, but as I understand it, the ECU will not realize it's been cleared and won't learn, it will just use the cleared values. Switching the lights on and off (the steps after shorting the battery cables out) tells the ECU that it has been cleared and should start learning again.

Click to expand...

And this is definitely a case of needing it to relearn - I found out a few days ago that I've been running a CA-specific oxygen sensor upstream on a Federal-spec XJ for quite some time. Although I doubt that made much if any difference (and probably less so now that it's shot), I'd like to eliminate as many points of possible contention as I can.

I'm surprised fed vs CA sensors are different at all. From what I see in the catalog they also split euro II vs euro III - perhaps it's a difference of where different fuels are in use? My initial guess was that the only difference was wire length but clearly that is wrong.

i don't believe it's in any FSM. But i did verify it a while back with a Chrysler engineer over the phone.

kastein said:

I'm surprised fed vs CA sensors are different at all.

Click to expand...

There's some mix & match going on there, definitely. I can't remember all of the specifics, but I ended up having to call Bosch last week to get a straight answer on what goes where depending on the emissions setup in use. My recollection is that the CA-specific sensors were intended to be used in only two locations: between the pre-cat and first cat, and between the first and second cat. Makes sense, given that only the CARB XJs would have had that system.

The fun part is that even with the two middle sensors being CA-specific, they're listed in some parts books as being either CA- or Federal-spec - hence the call to Bosch to confirm, since I was getting conflicting info all over the place. In my case, I had ended up with one of the two 'sandwiched' sensors in the upstream position. My fault for not checking part numbers correctly when I put it in a few years ago, but there's also a tendency in CA (where I lived at the time) for parts stores to hand over the CA-spec sensor out of force of habit even if you specifically stated that you had a Federal-spec vehicle.

From what I see in the catalog they also split euro II vs euro III - perhaps it's a difference of where different fuels are in use? My initial guess was that the only difference was wire length but clearly that is wrong.

Click to expand...

That one (IIRC) more likely comes down to emissions standards tightening from Euro II to Euro III. By the time the '97 XJ came out, virtually all EU markets had been transitioning to unleaded fuel for 4 or 5 years at that point if they hadn't gone fully-unleaded already, so the fuel grades were pretty much the same across the board.

dukie564 said:

i don't believe it's in any FSM. But i did verify it a while back with a Chrysler engineer over the phone.

Click to expand...

Hm, fair enough. I'll put it on the 'to track down' list - not saying you're wrong by any means, but I really like being able to refer to as-official-as-possible sources for stuff like this.

Wow, this just got brought up in a thread on facebook. Been a long time and I forgot I posted in this thread. Feel I should clear the air on this subject - I looked at the wiring diagrams for 97-01 and I can't see any way the ECU would sense the headlight status, so turning the headlights on and off probably does nothing to clear the ECU. Unless someone can prove to me that it does...

kastein said:

Wow, this just got brought up in a thread on facebook. Been a long time and I forgot I posted in this thread. Feel I should clear the air on this subject - I looked at the wiring diagrams for 97-01 and I can't see any way the ECU would sense the headlight status, so turning the headlights on and off probably does nothing to clear the ECU. Unless someone can prove to me that it does...

Click to expand...

I thought the same thing when I read this thread.

I believe the misunderstanding was to cycle the headlights while the battery was disconnected as the method to discharge the caps that maintain the adaptive memory, not cycle them after the battery was reconnected. I never heard to do this with the battery connected until I read it in this thread.

That does make sense. Not sure it would work, I would have to look at the ECU I have torn apart at home to see if there is a reverse-bias protection diode in the ECU or not and which side of the voltage regulator(s) the larger filter capacitors are on.

I've always saw no reason for the headlight steps on this method. There is no BCM and no way for the PCM to know anything about the condition of the headlights.