97 Metro LSI dies when it gets hot

I have about 140,000 miles on my Metro. It is 4 cylinder, auto with AC.
Every year, before summer, she starts having issues after engine warms. The problem is, it would stop running and wouldn't start agian till several hours later. I have a good battery and starter and she turns over every time.
Key on injection pressure is 27 Psi. Idle runs at 21 Psi. Compression is 172 on all 4 cylinders. I have new plugs and rottor. Air filter is new. Timing is at 5 BTDC.
I took in the injectror for testing to a local shop and they fried the selinoid so now she has a new injector...
Tried inspecting the timing belt, but can't take the cover off due to the fat crank pully. I guess I will have to cut a slit in the bottom of the timing belt cover to remove it.
I did get a code PO 134 at idle one time.
She starts right up and purrs like a kitten for about five minutes and then gets sluggish and dies. I can keep it going by giving it gas, but the engine doesn't run right.
So I have fuel and sprk with no trouble codes... where do I go from here?
Help.:banghead:

I would look really hard for any vacuum leaks from bad hoses.

Sounds like a bad sensor to me. Failing when it gets hot, or a temperature sensor that's always reporting cold. When it's actually cold, it works fine. Intake air temp sensor might do this, when the engine warms up enough to go into closed loop operation, the computer sees 20 degree air coming in and loads up with fuel, too rich to run with the actual 80 degree air that you're getting. Also could be the ignition module, a lot of solid state components work fine up until a specific temerature, then they quit. Electronics shops sell a spray can that freezes components. When it's failing, you spray things one at a time, if it starts working, it's the last thing you sprayed.

I would look really hard for any vacuum leaks from bad hoses.

OK, thanks I didn't think of that. I wonder if it has a vacuum advance. I will look into it.

[quote=Woodie83]Sounds like a bad sensor to me. Failing when it gets hot, or a temperature sensor that's always reporting cold. When it's actually cold, it works fine. Intake air temp sensor might do this, when the engine warms up enough to go into closed loop operation,

I had a problem with sensor in the air filter housing before. That was four years ago and if memory serves me right, that thing is actually a temperature sensor in there. That one gave me some sort of trouble code so I chased it down and it was a broken wire at the sensor. So that was an easy fix. This time no trouble codes, ok except with the PO 134 that didn't show up a second time.

I will go look for a can of cold stuff tomorrow. Back in the day we used to use a similar procedure to locate bad ICs in electronics...

Thank you all for the suggestions. I will keep plugging away...

Woodie has coherant input a bad ECT/IAT is definitely something to look at. If you have a scan tool you can monitor the temperature as read by the PCM, and determine if it is ball park or not.

On the other hand PO134 is for the HO2S11. This is important, because the
O2 sensor is only used in cloesd loop (after the car heats up, not when cold), and if it is reporting erroneous oxygen content, the car can have serious driveability problems.

You can monitor input voltage from the oxygen a/f sensor with a scan tool too. Or if you have a DVOM you can backprobe the blue o2 sensor wire. Either way, what you are looking for is a regular fluctuation between .2 and .8 volts after the engine is hot. If it goes negative, or operates above .9v the sensor is bad. If it oscillates slowly, sticking around a similar voltage too long, the sensor may be bad.

Look at your long term fuel trim values (LTFT), if they are in the 10 or above range + or -, and you can rule out vacuum leaks, leaky injector, etc... you most likely have a bad o2 sensor.

-MechanicMatt

Hello, I hope you haven't cut that timing cover yet. I would avoid doing that. It doesn't seem to be your problem anyway. Is this a recent problem or has it gradually gotten worse? There are many possibilities. If it's recent you can consider bad gasoline (which could have gave you that brief code). What type of spark plugs are you using as you may want to check the gap if it's not a pre-set plug. You may also want to consider checking your catalytic converter. That could also possibly explain the code you described. I believe the code you described means one of the o2 sensors is not reporting data.

[quote=91Caprice9c1]Woodie has coherant input a bad ECT/IAT is definitely something to look at. If you have a scan tool you can monitor the temperature as read by the PCM, and determine if it is ball park or not.

Thanks Matt,
I don't know what ECT/IAT is? I will look that up.
I have an OBDII reader so I can only read codes and clear them. I am pretty handy with DVOM so once I locate the sensor, checking singnals should be no problem.

Thanks for the heads up on the PO134. I couldn't find anything useful in my manuals on that code.

Hello, I hope you haven't cut that timing cover yet. I would avoid doing that. It doesn't seem to be your problem anyway. Is this a recent problem or has it gradually gotten worse? There are many possibilities. If it's recent you can consider bad gasoline (which could have gave you that brief code). What type of spark plugs are you using as you may want to check the gap if it's not a pre-set plug. You may also want to consider checking your catalytic converter. That could also possibly explain the code you described. I believe the code you described means one of the o2 sensors is not reporting data.

Hello Doug,
Didn't have to cut the timing cover. All I had to do was dorp the engine about 1/4" and that gave me the needed clearance.
The timing belt looks brand new. Not bad for 140,000 miles... Did I say I love these cars?
I know about the O2 sensor on the exhaust pipe. Is that the one you are talking about or are there other ones I should be looking for?

This problem happens just about May or June every year. Starts gradually and gets worse. Now it is the worse it has ever been.

I use Bush Plantinums and set the gap at 1.0 mm. My new spark plugs have a fine carbon dust on them now. I have never seen that before. The engine has only been run for about 20 minutes...I think it is running way too rich. I wonder if the carbon built up and possible running rich relates to O2 sensors.

I only got the PO 134 once. So could have been the fuel. I drained the tank and put some Chevron gas in there, just in case...

What should I be checking on the catalatic convertor?

Thanks,
Tony

Well, to check a cat, you either gotta remove it and see if it's plugged, or simply examine the exhaust flow from the rear of the car with an assistant revving the engine. The flow should be enough to push a relaxed hand back (don't burn yourself) and not restricted.

These cars also have a "buzziness" sound to them even with a new exhaust system under high RPM. Should sound similar to a pissed off hornet in the muffler.

If it's not the cat, it seems like an electrical component is failing under high temps. Maybe the ignition coil, or ECM.

You can monitor input voltage from the oxygen a/f sensor with a scan tool too. Or if you have a DVOM you can backprobe the blue o2 sensor wire. Either way, what you are looking for is a regular fluctuation between .2 and .8 volts after the engine is hot. If it goes negative, or operates above .9v the sensor is bad. If it oscillates slowly, sticking around a similar voltage too long, the sensor may be bad.

Look at your long term fuel trim values (LTFT), if they are in the 10 or above range + or -, and you can rule out vacuum leaks, leaky injector, etc... you most likely have a bad o2 senso


Hello Matt,
Here's a progress report.
I tapped into the upstream O2 sensor's blue wire. Reference to ground the voltages are .915 for idle engine cold. .870 high rev. engine cold, .945 idle engine hot and .918 high rev. engine hot.
If I understand you correctly, the upstream O2 sensor may be bad. Would it help to remove and clean the sensor?
Also checked the exhaust pressure at idle and high rev. Very strong and pushes back my hand. There's also a little of the carbon soot in the pipe. The catalatic convertor has no dents so I am assuming it is ok.
Not too many problems on a test run last night. A couple of times coming to a fast stop at red lights the idle would drop real low and the engine would die. It started right up each time.
Taking her for a torture test today. It is supposed to be in the high 90s here.
Haven't found any vacuum leaks so far.
Going to get a can of freez air to check the temperature sensor in the air filter housing. If I take the sensor out, what resistance should I be reading for cold and hot? I am thinking of comparing the reading with the tip of the sensor in hot and cold water to see if it's working.
Thank you all for all your help...
Tony

Excellent work, Tony.

It sounds like your o2 sensor may be suspect.

I want you to do two more things:

With the engine hot (operating temp; w/ temp gage centered) backprobe the o2 sensor's blue wire and ground your meter as you did before. Create a large vacuum leak by removing the brake booster hose from the brake booster. Your idle will climb considerably, but not dangerously, and you should observe a quick drop in reference voltage from the sensor. It should drop immediately below .45v at least.

Then bench test the sensor with a Propane torch. Clamp the sensor in a vice, then point the torch at the sensor's business end - get it red hot. Probe the blue wire and ground your meter to the sensor body. With the torch on the sensor, voltage should spike to .9 volts, then drop to .1v immediately after you pull away the torch. Then spike back to .9 when you reapply the torch. If the sensor responds quickly to the presence of the flame with the range I described, there is probably something giving you a rich condition, if the sensor does not respond properly, replace it.

-MechanicMatt

Excellent work, Tony.

It sounds like your o2 sensor may be suspect.


-MechanicMatt

Hello Matt,
Thanks for all the good heads up. I am continuing work on the Geo. I think it was John Lennon who said: “Life is what happens to you while you are busy making other plans.” So here’s a progress report.

I replaced the timing belt and took the car for a torture test. It was 98 degrees out at 2 PM. I made a 40 mile round trip with ac running with two stops one short and one long. The Geo ran perfectly at all speeds without any problems.

In the mean time, I replaced three of the tires. I also replaced the ATF gasket and filter. I also installed a computer and it is showing 41 mpg on level road with ac running and going about 60 mph. Not bad at all…

So have I fixed it or is it the good Chevron gas? :confused: I don’t know.

I the mean time, I scrubbed the engine and under-carriage clean. There is nasty oil leak from the oil pan that could be coming from crank shaft bearing. There’s a good deal of red silicone gasket stuff around there so the previous owners must have had a leak problem before and that’s how they dealt with it.

I am going to drop the pan and put a new gasket on to see what happens.

Once I have that taken care of, it will be time to check the O2 sensor per your instructions.

Thank you again, I will keep plugging away.

Till next time, Tony

No gasket on the oil pan, please. They seal better without it. Also if the gasket is thick, it'll keep the crankshaft sensor from working right.

[quote=Woodie8
3]No gasket on the oil pan

Hi Woodie,
Got the pan off. I see what you mean about the sensor. Ok, no gasket. I am using chop sticks to scrape off the old Red silicone RTV. Is there an easier way, or maybe some solvent, I could use to get it all cleaned off?
Thanks,
Tony

sailorton08, buy a gasket remover solvent from your auto parts store. This works well to help remove the RTV. After the solvent dries you should be able to remove the RTV with a rag. Some stubborn spots might call for the use of a wooden pop-cicle stick or something like that that won't scratch the oil pan flange surface.
When re-installing the oil pan, make sure to use a bead of high-temp RTV silicone all the way around (except for bolt holes.)

[quote=sas95]sailorton08, buy a gasket remover solvent from your auto parts store. This works well to help remove the RTV.

Thanks, I will see if I can find the stuff locally. I used a lot of elbow grease and tooth brushed clean. I didn't get the leak stopped 100% so I might be dropping the pan again during the next oil change...
Tony

With the engine hot (operating temp; w/ temp gage centered) backprobe the o2 sensor's blue wire and ground your meter as you did before. Create a large vacuum leak by removing the brake booster hose from the brake booster. Your idle will climb considerably, but not dangerously, and you should observe a quick drop in reference voltage from the sensor. It should drop immediately below .45v at least.


Hello Matt,

I did the first part of the o2 test last night. I started with engine temperature at 170f, and ended at 198f. My voltages to blue reference to ground are as follow:
RPM V Vacuum
900 0.1-0.8 on
1400 0.6-0.8 Off

As you can see, with the vacuum hose off, my idle went up, but only a little. The voltages were fluctuating and never stablized. With the hose off, I got a higher low voltage reading...

The last time I checked the upstream O2 sensor the voltages were a pretty solid 0.91 volts and dropped a fraction with reving the engine high.
Does this mean the the upstream O2 sensor is going bad?:sadwavey:

Thanks a million for your time and expertise.
Tony