ECMs, their data collection and storage function, and infrequent glitches

[BonzaiDuck]

This inquiry is about my 1995 Isuzu Trooper LS. I have a history of posts in the last few years about my Trooper projects, which gave me great satisfaction and continue to give it whenever I drive the vehicle. The 1995 is a 21st Century Android Trooper now.

My most recent extended activity in renovating the vehicle ranged over 2020 and this year. This was an effort in the aftermath of the Little Repair Shop of Horrors and its Mechanic From Hell, encountered in my market search for a new repair shop between 2018 and end 2019.

In 2020, I cleaned the contacts for various items, but particularly the Mass Air Flow Sensor on the common-chamber or intake manifold. So after giving that sensor a cleaning, I took the car out for a short errand, and found at some point coming home that it the car would suddenly fail to hold the idle at a stoplight and it would stall. This behavior lasted for about a minute or so, until later the vehicle was behaving properly with a rock-solid idle speed at warm temperatures of 750+ and cold ore pre-warmup of 1,000 rpm.

Without anything further to do affecting electrical contacts or vacuum hoses, this never happened again -- until the other day.

A couple events or factors changed over a day's time just before I had the stalling problem again. First, the ambient temperature here in So-Cal during the same afternoon hour had dropped from weeks-long value of 100+F to something just over 80F. There had also been a generous amount of rainfall. Second, I had been treating the crankcase with a treatment of Marvel Mystery Oil and Synthetic oil, which had done a great job to eliminate the noise of a sticky valve or valve-lifter. This had evidenced significant improvement over six months with two oil and MMO changes, but there was a quantum leap recently. That is, sometimes at certain temperatures one could rev the engine and there would be a faint sound like a cylinder not firing when it was precisely supposed to do so. This has suddenly disappeared as an intermittent engine behavior.

Somewhere I had read -- and at the time I'd cleaned the MAF sensor -- that the ECM collected data from all the sensors and stored them in some sort of scratch-pad RAM or PROM. I think the sources said that the ECM could be reset by disconnecting the battery and then turning the ignition key to full "ON" and holding it for 30 seconds.

So I am only suspecting that the recent stalling episode was again a case where the ECM had "bad data" so that it misbehaved in how it preserved the proper idle speed.

Can anyone offer any further experience or confirmation about this? The car now runs like a peach, holds its idle, runs like a top.

For that, I and my newly-converted believer and Solid-Gold Repair-Shop service advisor highly recommend Marvel Mystery Oil for removing varnish, gum, sludge -- noise and related shortcomings.

 

[mindless1]

I don't know about your Trooper in particular, but usually with vehicles of that era, you merely disconnect the battery for a few seconds and the learned parameters are lost from the volatile memory, no need to wait a long time or turn the key or anything.

I follow what you stated about it happening right after cleaning the MAF sensor, but this more recent event, how do you account for "bad data" when nothing was disturbed again, that would change air/fuel ratio?

If it is setting no trouble codes, you are left waiting until it happens again and making a diagnosis then. Not knowing the full vehicle history, there are still things you can do or check now such as fuel pressure, put some fuel system cleaner in the next tank of gas, smoke test the intake for leaks, do a cylinder leak down test, check (and change as needed) spark plugs and wires, and more.

Your issues are probably going to be handled better in a Trooper owners' forum, though if you don't have a copy yet, here is the Workshop Manual for it. You might want to focus on section 6E2:

1995 Isuzu Trooper Factory Service Manual.pdf

drive.google.com

You need an editor to trim your posts down to an internet-human-digestable amount of text.

 

[deadlyapp]

Quite common for cars of that era to have poorly manufactured ECMs as well, often represented by bulging capacitors. I had a 91 mitsubishi that I had to replace/rebuild the ecm several times, one of which it actually caught fire.

 

[BonzaiDuck]

I don't know about your Trooper in particular, but usually with vehicles of that era, you merely disconnect the battery for a few seconds and the learned parameters are lost from the volatile memory, no need to wait a long time or turn the key or anything.

I follow what you stated about it happening right after cleaning the MAF sensor, but this more recent event, how do you account for "bad data" when nothing was disturbed again, that would change air/fuel ratio?

If it is setting no trouble codes, you are left waiting until it happens again and making a diagnosis then. Not knowing the full vehicle history, there are still things you can do or check now such as fuel pressure, put some fuel system cleaner in the next tank of gas, smoke test the intake for leaks, do a cylinder leak down test, check (and change as needed) spark plugs and wires, and more.

Your issues are probably going to be handled better in a Trooper owners' forum, though if you don't have a copy yet, here is the Workshop Manual for it. You might want to focus on section 6E2:

1995 Isuzu Trooper Factory Service Manual.pdf

drive.google.com

You need an editor to trim your posts down to an internet-human-digestable amount of text.

[also for deadlyapp . . ] I ran it by my repair shop this morning, to see if I could get an opinion about my "theory". I actually think I had read somewhere a couple years back referring to bad or inaccurate data stored by the ECM. I'm going to continue looking for more indications at forums and other online resources.

Usually, the service advisor would ask "Tom" the Isuzu dealership veteran to consult with me. Sometimes I'd slip them a $20; today I brought donuts. Tom was busy, and another mechanic came in. He suggested I should bring the car in after he found his adapter for an OBD-I vehicle to use an OBD-II scanner. I said maybe I'd call back in a couple weeks.

But I'm also thinking about what my cousin told me: something so intermittent not likely to be diagnosed for a definite cause if I take the car in. In the meantime, it's running perfectly. warmed up, the idle is rock-steady when I slow down and stop at a stop-light, or 750rpm. When cold, also as it should be -- it starts at 1,000 and slowly drops to 750 as it warms up. It's been at least a couple weeks now since "The Event", and through several errand trips here and there every other day it seems totally reliable and behaving as I just described.

I may wait to let them throw the scanner on the Trooper until after New Years, if I don't experience any further trouble. So, for the service of doing it and the possibility that something needs further attention, it will better fit into next year's repair and maintenance budget.

At the moment, I'm thinking what mechanics always tell me: "If it ain't broke, don't fix it." It doesn't SEEM broke. It seems to run fine. I'm just going to watch it and deal with it if a problem arises, I suppose. It may be related to the aging AC compressor, but I haven't noticed the AC exhibiting any noises or odd behavior. And as I'd said, when I had the stalling problem a couple weeks back, it continued to stall for another minute or two with the AC turned off, so . . . .

MINDLESS1 also has some thoughts, and therefore reasons to actually run the car into the shop sooner than next year. I'll probably go down to Autozone or O'Reilly's and find some fuel system additive. And I'll think about this some more for a couple weeks, since I told Jeff and the mechanic I'd get back to them in that amount of time.

UPDATE: I did some more poking around, locating a website for an "ECM Repair" outfit. It gave some symptoms for ECM failure/deterioration, but pointed to some other causes like Throttle position sensor or Mass Air Flow Sensor. I sent an e-mail to the service advisor. Will likely (cross fingers and hope!) to take it in after beginning November and have them investigate.

I love the vehicle, haven't planned on replacing it. I've got a Nissan hardbody pickup for a backup ride. If there's anything wrong or beginning to go wrong, I'm just going to have it fixed. Amen. Period.

MORE: (Sorry, Mindless. Yeah, I got this problem with my fingers, brain and keyboard . . . )

Somebody asked "what had changed" to possibly throw the ECM data storage out of whack, or cause the stall at idle. I've been using Marvel Mystery Oil to clean the engine. I've watched and listened over six or more months how that stuff works. Most recently, there was a quantum improvement -- disappearance of valve-lifter or related noise -- that happened suddenly. The engine is running better than it was, but it was ALWAYS running good. That . . . the change in ambient temperature with the weather . . . those are the things I'd noticed before the stall at idle occurred.

I took it out tonight and ran it up and down the hill (and up and down and up and down again again). Rock steady idle. No problems at all. Nothing guaranteed with that, but -- no problems at all for now.

 

[deadlyapp]

Just a note, if you're not actively throwing a CEL, the scanner won't tell you anything on an OBD I vehicle. If you are presently having the issue, it may help diagnose (eg you're getting incorrect reading for the TPS, IAC, etc.)

I used to have terrible idle issues on my old car where going to neutral or clutch in while still rolling would cause the RPM to come up to about 2000 rpm and then once stopped it would come back to normal idle. Sometimes it would attempt to stall at a stop but not neutral and rolling, so you'd have to go clutch in, feather the accelerator, and also try to brake (often I used the handbrake). God I don't miss that vehicle.

Edit: whatever you decide to do, I would caution against throwing new $100 sensors over and over at it to try to make it "perfect" despite your obvious OCD. It's a 30 year old car with near first generation electronic sensors and they are going to start failing at a rapid rate my guess over the next few years.

 

[BonzaiDuck]

Just a note, if you're not actively throwing a CEL, the scanner won't tell you anything on an OBD I vehicle. If you are presently having the issue, it may help diagnose (eg you're getting incorrect reading for the TPS, IAC, etc.)

I used to have terrible idle issues on my old car where going to neutral or clutch in while still rolling would cause the RPM to come up to about 2000 rpm and then once stopped it would come back to normal idle. Sometimes it would attempt to stall at a stop but not neutral and rolling, so you'd have to go clutch in, feather the accelerator, and also try to brake (often I used the handbrake). God I don't miss that vehicle.

Edit: whatever you decide to do, I would caution against throwing new $100 sensors over and over at it to try to make it "perfect" despite your obvious OCD. It's a 30 year old car with near first generation electronic sensors and they are going to start failing at a rapid rate my guess over the next few years.

This doesn't seem at all like what you experienced in its frequency. While I don't keep a notebook to record every little event with the car, I keep an Excel spreadsheet history of repairs, mileage, expense and the dates. In my mind, I'm pretty sure the last time it happened prior to the September event was in early 2020 when I cleaned the contacts for the MAF sensor and took the Trooper out for a short drive.

In the matter of sensors, I had acquired good aftermarket replacements for the Heated Oxygen and MAF sensors, but never installed them. The former is screwed into the catalytic converter, so we'll cross that bridge when I get to it.

If something happens, I'll run the Trooper by my repair shop. If anybody can fix something properly, they can. They inspire confidence. However, your remark about the limitations of OBDI make me wonder if it's something I even want to do in November -- the current tentative plan.

There is one other minor quirk that "visits" my driving experience here and there -- per a particular idiot light. The Trooper has two fuel gauges or indicators: the traditional dial "E-to-F"; and an idiot light (a little gas-station pump) which is supposed to light up when the tank is severely low. Owner's Manual states that the owner should not worry if the light goes on from time to time. In my case, it goes on for a trip-duration just after filling the tank, then it will go off, then maybe come back on at a stoplight -- but goes on less frequently when the tank is maybe half full.

But there's also a history that MAY be behind these things going back to 2019, when I was so unlucky to do business with The Little Repair Shop of Horrors and its Mechanic From Hell. They had recommended -- pressured me! -- to replace the valve cover gaskets. I succumbed. To replace those gaskets, the mechanic has to pull the common-chamber (intake manifold). In so doing, the MFH broke the nipples connecting the interior radiator core to the cooling system. And to fix that, he had to pull the dashboard. He was worse than careless. I still have thoughts of murdering the f***er. Of course, you can extrapolate from this: the wiring harness is behind the dashboard; the ECM and the Anti-Theft box is behind the dashboard.

But nothing ever really changes. I went maybe 8 months most recently without that idiot light turning on, after I'd first noticed it following the horror shop and the MFH, when it was intermittent through last year. The experience with the MFH was 7,000 miles prior to the current odometer reading, or June 2019.

You ARE CORRECT, though, about old vehicles. You can keep track of every subsystem and its repair and replacement, but there are so many other things that can eventually go bad and which you wouldn't anticipate.

The biggest dilemma I face here is two-fold. I absolutely love the Trooper (especially after upgrading the audio, blue-tooth, Wi-Fi, night-vision-rear-view), and like anyone else at this time, I'm not eager to shell out for a new vehicle. I'd want an EV; I could settle for a hybrid; I'm not eager to buy another used vehicle. This one -- purchased used -- has seen me through exactly 20 years. The spreadsheet on "average maintenance/repair per year or per month" is very respectable; the insurance savings were stellar.

People lose patience and get their panties in a twist if they have an 8-year-old vehicle and have to shell out $600 to fix an electric window or door lock. So they pitch the old vehicle and either buy a new one or another pre-owned. I didn't let myself do that. I figure $1,000 per annum is respectable and reasonable for upkeep, and there are "lucky" years. Maybe I just don't fully appreciate what might be around the corner.

It's nice to have a spare ride! That's all I can say at this point!

 

[deadlyapp]

I'd ignore the fuel light. In a vehicle the age of yours, just use the trip odometer to determine when to fill up. I'd personally pull the damn bulb out of the gage cluster which is probably ten minutes of work

 

[BonzaiDuck]

I'd ignore the fuel light. In a vehicle the age of yours, just use the trip odometer to determine when to fill up. I'd personally pull the damn bulb out of the gage cluster which is probably ten minutes of work

Back again to comment and update.

The "fuel empty" idiot light is now misbehaving less.

I followed the advice of someone who posted here and put some Lucas "one-time" fuel-system cleaner in the gas-tank.

For the phantom "stalling event", this seems to be much like the previous occurrence 2 or 3 years past. I've taken the Trooper out almost every day or every other day (here and there) since it happened in September. It hasn't repeated itself since then.

Instead, the idle speed at various temperatures is proper and rock-steady, whether I flip on the AC or flip it off.

None of the pro mechanics with whom I've conferred will say anything about my theory concerning bad data in the ECM. They don't dismiss the theory, but they don't mention it or even speculate. Otherwise, people suggest sensors, particularly the Idle Speed or (Idle-something) sensor.

Sometime soon, I will disconnect the battery and twist the ignition key to reset the ECM and memory.

I love this old Trooper so much, that driving it exclusively is preventing me for running my Nissan truck out on the road once in a while. The idea being -- cars need to be driven with some regularity. I'll have to make an appointment with my truck and stick to it . . .