'95 LeSabre Dies Without Warning

[timrice]

I have a '95 with 113K. A week ago, I was cruising down the street and the engine died with absolutely no warning. No CE light or code was set. I coasted to the shoulder, stopped, and the car started right back up, and I drove home.

It did this again once each of the next two days, and both times, it started up fine immediately and ran OK.

One of the times, it finally set a P0341 code--camshaft sensor.

I started noticing that you could go just about five miles after the engine was first started from cold (in the morning, or leaving work) before it would happen.

Then yesterday, it *didn't* start right back up like usual. It would crank, but not fire. I could hear the fuel pump energize with the ignition in the ON position. Battery connections looked clean and were snug. I left it on the side of the road for an hour while I went to get a camshaft sensor. When I got back, I tried starting it, and it did. (engine cooled down?) I got about another 5 miles, and dead it went. I put the camshaft sensor in then--took about 10 min.--and it started right up. I was pretty happy, thinking I'd fixed it.

Wrong. Next morning on the way to work--about 5 miles out, dead--wouldn't start. An hour later or so, it started up and ran another 3-5 miles over the the mechanic I took it to out of frustration.

He put it on his Sun diagnostic machine all afternoon and it didn't die once.

I'm wondering if I don't need to be looking for an intermittent open circuit in the wiring between the cam sensor and the ECM somewhere--maybe that's why I get it driving down the road (a major source of vibration in my county) and my tech doesn't with it idling in his bay.

I would also *like* to think that the 341 isn't a "false code" due to the engine dying for some other reason.

[timrice]

PCM? What is that exactly?
(You don't mean ECM - Engine Control Module - do you?)

[timrice]

Well, it turned out to be the ignition control module under the three coil packs on the engine. The part (AC Delco #1977A) was about $140, and with 2 hours labor the bill ran just short of $300. (ouch) I'm a bit hacked that the 341 code was false and I sunk another $40 into a cam sensor I didn't need--chalk that up to experience, and probably worth it. But I guess the way I have to look at it is as if I were just making a car payment this month where I usually don't have to.

I got to looking on eBay to see what the going rates are for these modules. They had a few listed, and some of the bids were starting down around $40 for new ones! My local big-name auto parts store wants $200. So I'm thinking for that $40 price I may pick one up off eBay or at the salvage yard and throw it at my '90 LeSabre--it was doing a similar thing, dying with no warning. (Although I still wonder if there's some overheating thing going on there...)

Found out too that Snap-On and maybe other companies make "load testers" for these modules, and I'm thinking about looking into what those cost. The ones I've seen so far look like they'll do about any brand of car, and if it were less than $100, it could pay for itself the first time you used it.

[timrice]

No, I spoke too soon. This problem was not fixed by putting on a new ignition control module. A couple weeks after the ICM was replaced, it started doing exactly the same thing. I took it back to the same shop, and this time they put a camshaft sensor on it (yes, after I had written down for them that I'd already done that to no avail). When I found this out, I said I didn't think the car was fixed, and we took it on a quick test drive, and my statement was proven when it acted up again. (Also found out that a bad cam sensor won't kill a running engine--it'll make it hard to start, but it won't kill it. So why they put a cam sensor on it, I don't know. I sure wouldn't have put mine on if I'd known that fact.) At any rate, they had the car in their shop for two weeks monitoring, probing, checking, etc., and got it to fail once, but didn't capture the data for some reason. They had a portable monitor on the crank sensor outputs, and I drove it for them and it didn't act up. So I took the car back home because it seemed like we weren't making any diagnostic progress, and on the very first run to work today, bam. Engine died about 5-6 miles from the house. Tried to start it right away, and it wouldn't. I waited a minute or two, and tried again. It fired right up, and I drove it the remaining 10 miles to work. Crapped out just like that again on the way home tonight. I'm not 100% certain, but tonight when I turned the ignition back on to restart it, I don't believe I heard the fuel pump whirr. After 30 more seconds, I tried again, and it did whirr, and it started. So I'm wondering now if I have a fuel system problem, and I want to replace the fuel pump relay. Trouble is, I haven't found it yet. It's not one of the six relays in the "Maxifuse / Relay Center" on the firewall. By the way, crappy job of labeling the relay functions there, GM!! What's more fun than pulling relays out one by one and checking to see which function doesn't work? But I guess we can't afford a 5-cent label on a $20,000 car, huh?

I'm thinking it's buried in a smaller relay center under the right-side sound insulator--anyone know for sure?

I don't think it's a bad fuel pump so far because I had chronic fuel pump problems on my '89 LeSabre, and when it overheated, it always needed way more than a minute or two to come back and work (like 1-2 hours or more) and finally, it just quit coming back altogether. This one doesn't act that way. After all the hide-and-seek hassle this problem has been, I'd gladly pony up the dough right now for a new OEM fuel pump if that would fix it!! I guess I could rule out the fuel pump itself by taking a jumper wire with me, and the next time it dies and won't start back up, jump that green-tabbed fuel pump test connector to the battery and see if I can start it then. If not, fuel pump. If so, it's something causing the power to the fuel pump to be cut. I guess that is unless the relay is included in the test circuit too. (??)

In the "diagnose fuel system" section, my Haynes manual says that there's a possibility the oil pressure switch could be bad. But it doesn't give any info about where that switch is or how it's wired into the fuel pump / pump relay circuit. Anybody know? So after the fuel pump relay gets ruled out, I'm going to focus on that switch next. I would note that I'm not getting any oil pressure lights on the dash at any time.

This ever happen to anyone? Any help wouuld be appreciated.

[rustbucket]

That green tabbed connector goes directly to the fuel pump (bypasses the relay and oilswitch). If you put a 12v test lamp between it and ground, it lights when the fuel pump is getting voltage, thereby checking the relay (and oilswitch if present). I am guessing crank sensor. Its kind of hard to test when it is intermittent. My microfiche for 87 says put a timing light on a sparkplug and crank it. If the crank sensor is working, the timing light flashes, if not it could be the crank sensor (among other things). Maybe you can figure out how to drive it around with the timing light attached till it dies on you and then quickly see if the timing light flashes when you are cranking it.

[CustomChevyGuy]

that sounds totally familiar. ive got a 94 and about 120k it started acting up like that, and they've had to replace a number of my computer components. have them check your Ignition Moduel, crank shaft sensor, and maybe even the Main Computer. i took it to Firestone, and they were able to help me out quite a bit

[tdenkler]

just fixed my 96 lesabre. same prob. there is an 8 way connector in front pass side rail underneath carpet. If car ever in a flood or spill water in pass front carpet, connector corrodes. controls fuel pump,pw, and power seat. clean, apply grease, wrap with tape.
I spent 1500 on fuel pumps,relays,etc, all for naught.
Goodyear and Firestone don't know squat!

[timrice]

Again, I've already had the ignition control module replaced. (Not that it means that it's guaranteed to be good--but as soon as it acted up again, I took it right back to the shop that installed it, and after another two weeks, they still didn't have it figured out.) It's funny...everybody I talk to about it says "it's going to be something simple." That may be true, but it's apparently not *that* obvious. I have yet to meet the person that can tame this car. Wonder what's in the Yellow Pages under "Exorcists"...

Interesting suggestion you made there, tdenkler. I don't believe this car's ever been in a flood, and I don't think I've ever had the carpet get wet except for a few hours about 10 months ago when I shampooed it. But that might be worth looking at.

By "rail" are you talking about that plastic trim piece that covers the edge of the carpet alongside the door jamb where you step in and out of the car?

By the way, here's the latest in my almost 2-month saga:

Put a fuel filter on it yesterday, and drove it 18 mi. w/o idling first, and it didn't die at all. Got my hopes up just a little, but had them dashed again this morning. 8 miles after leaving the house, bam. Took about 3 minutes to get it started again. Crank, wait. Crank wait. Repeat. Sounded like the fuel pump was priming each time, too.

I noticed that the vacuum hose going to the fuel pressure regulator is pretty rough looking--has some cracks. So I'm going to replace that next.

I don't see how any of this explains why it won't die if you just idle it for 10-15 min. before you decide to drive it any distance. If you don't idle first, it's certain to die (or has every time thus far).

I saw in the Park Avenue forum where somebody was grousing about GM's cheaper rubber on their vac hoses, and another person took issue with that criticism, saying that even at 4 cents' savings per car, that it adds up to real money when you build a lot of cars. May be, but even if they built a million units on one particular model in one year, 4 cents would be a savings to GM of $40,000. Wow. That's a piddly drop in the bucket to GM, and any warranty claims arising from that cost-cutting move will eat into that. 10 times that would be a piddly drop in the bucket, too. Cutting corners on design and materials is something I can be more forgiving on when I buy a Yugo or the low-end GM model or whatever. When you drop $20K+ on a nicer car, though, those kinds of cost cutting strategies seem a lot less acceptable. I work for a major OEM (not cars), and I see this cost-cutting (we call it "value improvement") stuff every day, and when the customer inevitably gets left in the lurch, my phone rings. We have a very expensive machine down because we "saved" 10 cents by going to a cheaper electrical terminal design. In an industry where downtime can be very costly. Not smart. Just one warrany claim blows away all the yearly savings the bean counters claimed we'd reap, plus more. It's like OEMs today have completely forgotten that reliability, not just cost, is a selling point. Reliability is a force for developing longstanding customer loyalty. Companies today are in it for the quick kill--how do we make the shareholders happy today. What about tomorrow's shareholders, people? And tomorrow's employees?

[Now stepping off soapbox...sorry. I will try to remember this is not the op-ed section.]

[Jed Rule]

Tim, I agree with you completely.
What is GM thinking when they charge $300 for a plastic upper intake manifold that cannot withstand the heat of the EGR. Didn't any of those engineers ever burn their fingers on a EGR valve? Have they even opened a hood? Then when your warantee is up they come up with a $$$$ replacement of both the upper and lower manifolds. It takes two guys in Canada to design an effective repair.
Back to your problem:
When the ign is turned on the PCM energizes the fuel pump relay which powers the circuit to the fuel pump. It will remain on as long as the engine is cranking or running and the PCM is receiving fuel control pulses from the ignition control module. The fuel pump relay is #N in the relay center to the right of the glove box. You can apply 12v to the test connector to by-pass the relay and PCM control circuit. Did you try to spray some carb cleaner in the throttle when it won't restart? That would tell you if it's a fuel delivery problem. You will need to test the fuel pressure too. Also check for gas in the vacume line to the fuel pressure regulator.
If you remove the fuel pump relay while the engine is running it should continue to run. If it dies you have a bad oil pressure switch.
I don't understand why it doesn't set the check engine lite. You could have a bad PCM or poor connections to the PCM Check the PCM grounds on the back of the engine next to the oil pressure switch (under the #2 spark plug). Good luck.

[timrice]

Latest update: New vacuum hose did nothing. Not real surprising, but one more thing ruled out. It died once on the way to work this morning; I wasn't hearing the fuel pump prime every time when I turned the key to the on position. So I thought I might be on to something. Got it started and went on to work. Then tonight, I didn't want it dying on the way home, so I idled it in the lot for 20 min. I came back out, and for the first time in two months, it finally had died idling!! So that's new info. Couldn't start it back up, and again, the fuel pump didn't always make a noise when I turned the key to the "on" position. Got my fuel pressure gauge out, but I never saw the pressure below 40 psi at any time. So that's making me think now that maybe the fuel side is OK after all, even though I don't understand why sometimes the pump makes noise and other times it doesn't. At that point, it still wouldn't start, so I poured 12 oz. of cold water on the crank sensor (that's a trick I've seen other people write about here in the forum, so I figured "why not?"). That didn't have any immediate effect. After a few more minutes and a few more crank attempts, it started back up, and I got about 4 miles with it before it died again. Another five minute delay and I got it to go and made it back to the house. Is it possible the ignition switch itself is bad and causing the engine to die? I'd think that if the ignition switch were bad, you wouldn't even be able to get it to crank. And it's always at least cranked after it died.

I already bought a fuel pump relay, so I'm going to try to put that in tonight and see what happens tomorrow, even though after today's events I'm again a bit skeptical that I'm dealing with a fuel problem.

[Lesabrelover]

Replace the ECM. My 94 had a similar problem, and I went through the same thing you did. Fuel pump, filter, relay, plugs, wires, ign module. Replaced the ECM and all was fine. You can pick up a reman ECM for about $150 Hope this helps

[rustbucket]

If you put a 12 v lamp (or voltmeter) between the green tabbed fuel pump test connector and ground, it lights when the fuelpump is receiving a signal to run. If the bulb lights when it is supposed to, that eliminates the fuel circuit. (ie the fuel relay is receiving an "on" signal from the ECM, and the relay is working and the wiring to the pump is OK)

[timrice]

Thanks for the tips, LeSabrelover and rustbucket.

The plan today was to carry my buddy's timing light along for the ride to work, and when the car crapped out, to use it to decide whether there's any spark there when it cranks but doesn't start.
The idea being to confirm whether the problem is on the fuel or spark side.

Well, so much for plans. It rained all day today, and the car didn't foul up to or from work, nor during an hour of idling once I got to work. It's like the car knows it's being watched. Who knows, maybe the humidity plays a role in the problem somehow.

Hopefully it will fail tomorrow and I will have more to report. I'm starting to like this new ECM idea more and more, I have to admit.

[timrice]

It failed twice today. Both times I got the timing light on it almost immediately after the car stopped rolling, and both times it started immediately when I cranked it. So didn't learn much there. I may have to rig the timing light up so it's attached all the time and the gremlin won't have a chance to escape while I jump out, pop the hood and clip the light on.

Priced ECMs today. Salvage yard has 'em for $75, but no warranty. The two name-brand parts outlets here in my little burg both sell 'em for $110 after core. One offers "lifetime warranty," and the other 90 days. It's getting awfully tempting to go get one and try it, as much as I hate throwing parts at cars. I will check the (+) and ground wiring from the battery back as I saw suggested on another thread in this forum.

I think it is past time to get the real factory service manuals, too, so I can figure out what wires go where.

[pcmos]

Just one quick thought, if the car is really quitting almost instantly and not chugging or hesitating otherwise, it probably isn't a mechanical failure meaning an air/fuel delivery problem. How did you find out that you set a cam sensor code that first time? Do you have your own scan tool? The reason I ask is because the CE lamp doesn't come on for every failure, and you may have stored codes in the computer that you don't know about right now. If I had to put all of my money on something, I would definately bet on the PCM ground connections. Oh by the way, GM refers to what is normally called the ECM as the PCM or (Powertrain Control Module), they do this in all of their shop manuals. But since the ECM/PCM works by switching grounds, its absolutely critical that the ground connection is stable and solid, try filing the connections down and replacing any bolts that are holding the ring connectors in place. You can easily trace PCM wires using their color coded insulation, and of course solid black is always a ground, so trace those back first and check the connections. As im sure you know even if the connector is firmly held in place against a metal part for a ground, this doesn't mean that corrosion between the connector and the metal part isn't destroying the flow of electricity. Use a good metal file to clean all of your PCM connectors. Now on my older 1992 LeSabre there was a ground that ran directly off of the big ground terminal on the battery, it was a little say 16 or 14 gage wire that split off right at the battery terminal and went straight to the PCM. But GM put a connector in that line about 12 inches back from the battery so that the battery cable could be easily replaced without having to splice that line. That connector rotted out on my car and caused almost exactly teh same symptoms you are describing, until finally one day I started getting all kinds of crazy lights on teh dash board including ABS and Security indicators. Basically the PCM wasn't getting enough clean power to operate.