I own a 2009 Pontiac G5. The car has been idling rough lately and has stalled a few times. The engine light came on recently so I did a scan. I received the following 5 DTCs: P0172, P013A,B,C and F, all indicating Heated Oxygen Sensor 2 slow to respond to lean to rich or rich to lean. I did a real time response for the two oxygen sensors and found sensor 1 had millivolt readings of 900 or so while the engine was idling properly. That reading dropped to the 20 - 30 range when idling became rough. Sensor number 2 had 800 mv when idling properly and then changed to "delayed" when idling became rough. Sensor 1 mv readings ALWAYS changed first on rough idle, i.e. sensor 2 response was well after that of sensor 1. If sensor 1 recovered soon enough after dropping from 900 the engine would continue to idle properly and sensor 2 would not change.

I don't know what all this indicates so I was thinking of changing sensor 1 to see what happens because it is the easiest to replace. I have to raise the vehicle to get at sensor 2 so I am not keen on that. I will be turning 82 soon and my old bones and eyes aren't what they used to be. I have done all car repairs in my driveway for my whole life and am too stubborn to change. LOL

Can anyone help me with some advice?

 

Sounds like you're on the right track. Do you already have a spare? Or just swap them and see if the issue moves with the swap???

Best, Nick

 

Your second O2 is a catalyst monitor, it doesn't affect how the engine runs. You need to observe the O2 in graph form to see if the cross counts are in the correct range. It's very difficult to just observe the voltage values changing. Does your scanner have a graphing function?

 

Just ran some graphs. Sensor 1 graph at idle varied between 500 and 800 or so mv. At 2,000 rpm variation was between 60 and 800 mv - higher value most of the time. Sensor 2 ran between 500 to 800 and occasionally dipped to 30 when idling. At 2,000 rpm it varied between 400 and 800 and occasionally dipped to 130 or so.
I ran a second graph a few minutes later and sensor 1 did not change much on idle however at 2,000 rpm it became a beautiful sawtooth pattern varying between 60 and 800. Sensor 2 did not change much for either idle speed or 2,000 rpm.

During all the graphing there was no rough idle and the engine had been running for about 20 minutes.

Do you know what this means as far as the sensors functioning properly?

 

First of all, sensor 2 data will only serve to confuse you. Focus on sensor 1. As I stated, sensor 2 is a downstream catalyst monitor. No matter what it reads, it cannot change engine operating conditions, only command a check engine light if operating outside of its parameters. There's likely another issue going on causing your rough idle, but the sensor being slow to respond tells me it's probably in need of replacement. You should fix that regardless if you want to eliminate your CEL, but you'll have to run more diagnostics in order to determine what is causing your rough idle. Is this condition repeatable or just intermittent? If it happens every time under the same conditions, that makes it easier to diagnose. I'd get a fuel pressure gauge ( you can do a free "rent a tool" at most major parts stores ) and watch your fuel pressure while the engine is idling rough. If it's fluctuating badly, your fuel pump may be failing. If the pressure stays constant, you may have a fuel injector sticking either closed or open. You could also have a failing or dirty idle air control motor. Does your scan tool have bidirectional control? I'd be pulling up the IAC and watching to see if it's opening to where the ECM is commanding it to, and also using the scan tool to command it open and shut to see if it's moving in a linear manner through the correct range. Unfortunately, it's difficult to diagnose things over the internet, I can only give suggestions as to what I'd be looking at if I were trying to fix it. Hopefully I've been able to help a bit.

 

Thanks for the quick reply.

Rough idle does not always seem to happen. Sometimes, rarely, the engine will stall. Most times the rough idle will correct itself and the engine will idle smoothly. I will try some of your great suggestions.
As I've said I'm an old geezer and not too familiar with all these electronic gizmos. However I am an electrical engineer so I am well versed with electrical concepts and have learned much from this forum because of all the help from people like you.
The vehicle only has 70,000 Km (42,000 or so miles) and it is mostly driven around town. I have never inspected or changed the spark plugs in 15 years because it has been running beautifully until now. In the old days (50 or so years ago) if I had a problem with a rough running engine I would take it out on the highway for a good run to clean out any deposits that may have built up and that usually did the trick until I changed the plugs. Do you think a misfiring spark plug could cause this problem? If so, I will take the time to pull and inspect them before I do anything else.

 

Yes, the engine could probably use a tune up. I had similar problems on my VW Golf and plugs/wires fixed it. Can't hurt to try, spark plugs are cheap. Does your car have plug wires or is it coil on plug? If it has wires, they could very well be deteriorated from age, even if they look okay. Replace them. My old school method of determining if wires are good is starting the car in pitch darkness with the hood up, and if they've deteriorated, you'll actually be able to see the electricity running through them. But you can also check them with an DVOM. Although I've had wires be junk that tested fine on a DVOM.

 

LS1Formulation

Replaced the plugs and upstream O2 sensor today and nothing changed. When I removed the old O2 sensor it was quite black from soot so I thought I had the problem locked but that wasn't the case. The problem seems to be getting worse. Whenever the car has warmed up it starts it's rough idle and eventually will stall. I will try your fuel pressure and IAC suggestions next.

Somewhere I read that the problem could be caused by a fuel injector leak or a vacuum leak. What are your thoughts on those?

Thanks in advance for any help.

 

I received your PM. Yes, a leaking injector could be the problem, and a vacuum leak could also be an issue. I bought a quick and easy injector tester off of Amazon that helped me diagnose a sticking injector, and it can also could be used to identify a leaking one, although that may be more difficult because you'd have a constant pressure drop. Here's the tester I bought: The fuel pressure test would be the first thing I would do. Hook it up, start your car, record the value running, then shut it off and record that value as well. Leave the gauge hooked up and come back after 15 minutes and check the pressure. It should maintain within about 5 PSI of what it was when you shut it off. If it doesn't, you could have a bad check valve in your fuel pump, or a leaking injector while sitting static. If it maintains pressure, check for vacuum leaks using a can of throttle body cleaner, spray around all the intake seals and piping, and see if the idle changes. Do it on a cold start and when it warms up, as things can expand and a leak that wasn't there cold can appear when hot. If that doesn't give you a good result, get that tester and follow the instructions, and see if any of your injectors are sticking while running. Hope this helps.

 

Thank you for your reply. I will be testing the fuel pressure and injectors tomorrow. I will also check for vacuum leaks as per your suggestion.

I was thinking about also trying a quick check of the ecu to see if it is reacting properly. I have not been able to find a test to determine if the ecu is OK regarding the control of lean or rich mixtures. Maybe the ecu is not correcting the fuel trim properly. So I would like to run something by you before I give it a try.

Since a low sensor 1 voltage output (a few millivolts) indicates a lean mixture and a high voltage output (900 or so millivolts) indicates a rich mixture, I was thinking I could ground the signal lead of the sensor to input zero voltage into the ecu. If I monitored the ecu fuel trim at the same this should cause the fuel trim to rise. And also If I applied just under 1 volt on the signal lead this would indicate a rich mixture and therefore the ecu would try to correct by decreasing the fuel trim. Would this not show me if the ecu is responding properly? The other thing is I would not want to damage the ECU by doing this. Do you think it is safe to do this?

 

I've never heard of trying it, and I'd be of the opinion that attempting to apply any voltage to your ECM could very well fry it. Years ago, when I used to change chips in an ECM, I'd wear a ground strap on my arm because even a tiny static shock could fry the whole thing. It is extremely unlikely that the controls in your ECM aren't functioning. One thing you may want to try is, once you are fully warmed up and the problem is occurring, unplug your O2s and see if it runs better. I'm not 100% certain on your car, but it should make your ECM go into open loop mode. If the O2 is malfunctioning, the problem will go away when it's unplugged. If it's still there, you can rule out the O2 as the cause.

 

As stated before I have already replaced the upstream O2 sensor and nothing changed.

Late last night I received the injector pulse tester you suggested. I am totally unfamiliar with testing injectors. I read the instructions but would like to ask a couple questions not answered in the instructions. What pulse mode do you suggest I start with?. Also is it necessary to pay attention to the polarity of the terminals when hooking up those leads to the injectors?

Thanks again for your assistance.

 

I'm suggesting disconnecting for a reason. If you force the ECM into open loop and the car runs better, it can help with diagnosis. As you'd stated, it runs fine cold, and then when it warms up, it begins to run like crap. When it's cold, it's in open loop. However, you want to see it run in open loop when it's warmed up. Disconnecting the O2 sensor should do that. Another thing to look for on your scanner, check to see if your coolant temp sensor is reading correctly. If it's reading incorrectly, it could cause the engine to run rich ( i.e. reading a cold temp when it should be warm ).

As for the tester, it's pretty easy. I just used the shortest pulse count - You don't want to flood the cylinder. Polarity doesn't matter. I started my engine after every injector to clear out the fuel each time, but that probably wasn't necessary. You're looking for consistent fuel pressure drop, so if you're at 60 PSI for example, and you do a short pulse and the pressure drops by 5 PSI, every injector should, in theory, drop pressure by approximately 5 PSI. If you find one or more that aren't dropping as much, that will indicate a sticking or clogged injector. Also, if you test on the shortest pulse count and the fuel pressure drop isn't easy to see ( only dropping 1-2 PSI, for example ), go up to the next higher pulse count. Starting the engine each time also primes the system for you to test the next injector while clearing the extra fuel from the tested cylinder.

 

Finally completed some additional testing. I think the fuel pressure looks OK. I ran the tests twice and they were quite consistent. Pressure went up to 60 psi on start up. Dropped to 50 when engine was switched off. Spec for pressure is 50 to 60. After 15 minutes pressure had dropped to 45 psi.
Fuel injector tests were quite difficult to perform, the connectors were hard to get at, I had to remove intake air assemblies so I was not able to restart the engine after pulsing each injector. I did some jiggery pokery to attach to the injector connectors - I used some of the wife's small sewing needles to insert into the connectors and then attached the pulse unit's leads with alligator clips. Anyway the pressure drop for all four injectors with the one pulse was 12 psi so I think that indicates the injectors are in good condition.
The engine coolant temperature sensor looks OK too. Before I started the engine I disconnected the oxygen sensor as you suggested and the scanner indicated open loop status. The engine temp started off at 16 degrees C. After idling for 6 or 7 minutes or so it was up to 78 degrees. The idling turned rough and the engine stalled. Next I reconnected the O2 sensor and restarted the engine. It ran quite smoothly for a few minutes and then started to falter. I stepped on the gas pedal and raised the rpm to 2,000 for a few minutes. The sensor output looked good varying between 60 and 850 mv and had a sawtooth shape - short term fuel trim varied between -2 and +1, injector pwm was 1.9 ms. Coolant temperature reached 82 degrees. I reduced the rpm to idle and everything was OK for 5 or so minutes so I shut the engine off and erased the trouble codes hoping they would not come back because things were running properly when I shut the engine off. However when I restarted the engine it only idled for about 30 seconds and then the idling turned quite rough and it finally stalled.
From what I could find the engine does not have an idle air control motor nor does it have a pvc valve which are other components you suggested could be a problem.
I still haven't done a vacuum leak test - just ran out of time. Will try to get it done tomorrow.

What are now your thoughts. You've been a great help up to this point and I thank you very much.

 

The engine must be drive by wire if it doesn't have an idle air control. I wonder if there's something sticking in the throttle body or if one of the TPS sensors or wiring is faulty. It's hard to diagnose problems over computer, I'm sorry I couldn't be more help. You may want to try wiggling wire connections to see if you have a break in the wiring. Also, probably less likely, but you may have a dirty MAF sensor, which would misreport air mass going into the engine. Not quite sure what steps to take next, I've been out of the daily driver car repair game for many years.

 

You have been very helpful in educating me. I knew nothing of fuel injection engines when I started diagnosing this problem but with your advice I have been learning an awful lot. Thanks for your patience. I am an old carburetor guy that has done all my own vehicle repairs for the past 50 years in my driveway in both summer heat and winter cold.

I was also thinking that the throttle butterfly valve may be sticking so I took the plunge this morning and removed the throttle body. There is no wire control and the body seems to have some sort of motor attached right near the electrical connector. I have cleaned the valve and body with WD40. I noticed that after removing the throttle body off the intake manifold that there appeared to be some sort of liquid down at the bottom of the manifold. The manifold is quite deep so I used a flashlight and sure enough there were very little puddles of liquid pooling in the lower parts of the manifold. I used a nice clean rag with a long screwdriver to soak up some of that liquid and it appears to be gasoline, Do you have any thoughts as to why gas would be in the intake manifold? When I first noticed the liquid I thought it might have been accumulated condensation. Maybe this is the reason the engine is running rich and idling rough when warmed up.

 

When I say "drive by wire", it means there's no accelerator cable, just an electric motor, which is what you're describing. You definitely shouldn't have fuel puddling in your intake manifold, I looked at a photo of the style of engine in your car and it's a fully dry manifold. So, what I would do next is look at any and all connections to the intake ( vacuum lines, etc. ) and see if any of those have evidence of fuel coming through them. I don't know much about these engines, so I don't know how they're configured, but it doesn't appear to have a fuel rail mounted fuel pressure regulator, which if faulty could cause the issue. The only thing I do see is what appears to be an EVAP purge solenoid mounted adjacent to the throttle body, and that should only be purging vapor, not liquid fuel. How do you fuel up your vehicle? Do you let the pump shut off by itself and return it to the holder? Or do you squeeze the handle a few more times after it clicks off? Older cars can tolerate the latter, but newer cars cannot. There's always a possibility of your purge canister being filled with liquid fuel, which would allow fuel to enter your manifold. A good test would be to disconnect the purge valve electrical connection and start and warm up the car. It will throw a code or two, but if that IS where the fuel is coming from, disconnecting it will prevent the engine from going into that "rough idle" condition once it warms up. Try that and let me know your results.

 

Eureka!!! I believe we've found the reason for the rough idle. But how and what to repair is beyond me. Let me explain after I inform you of my agreement with what you said about the accelerator pedal. I finally found a section in the manual describing the pedal operation. The manual is so confusing because it is hard to find the proper section which describes the functioning of components I am not familiar with. The manual states that the ECM determines the driver's intent based on input from the accelerator pedal position sensors. I finally was able to find info on the pedal sensors and throttle actuator control. So now I have a much better understanding thanks to you. Not bad for a carburetor, distributor guy. I am learning an awful lot thanks to you and I really appreciate it. My wife told me only yesterday that maybe I should just take it to a repair shop and let them find the source of the problem. But as she's told me over the years when I try to fix something, "You sure are a bulldog"- because I just won't give up until I figure it out and fix it. That still applies to me today even though I will be turning 82 in a few short months .


So getting back to it. I cleaned the throttle body which appeared to be quite clean to me. I moved the butterfly valve back and forth numerous times and it wasn't sticking as far as I could tell. I reassembled everything and started her up but after 5 minutes or so it started it's rough idle routine again and then stalled, so cleaning the throttle body didn't seem to do a thing.


Next I tried what you suggested, disconnecting the EVAP purge valve electrical connector. I mistook what you were saying and thought the valve was near the intake manifold. I paid no attention to the evap hose that attached to the intake manifold when I was cleaning the throttle body. I could see no way of getting down there after assembling the intake air ducts to the throttle body so I did what I thought would be the next best thing and that was to disconnect the plastic pipe from it's upper connection, which I did not realize at the time was the evap purge valve. Anyway I started the car up and heard a sucking noise. The car was idling just great even though it was warmed up. This had only happened once in the multiple times I've been trying to diagnose the problem so I turned it off. Started it right up again and again it idled beautifully. I did this off and on thing multiple times and each and every time the idle was good. The O2 sensor was varying in a nice saw tooth pattern between 40 mv to 800 or so. The short term trim was also very good varying between +2 and -3. Prior values were 800 to 900 mv for the O2 sensor and -22 to -10 for short term trim.

After completing all this idling testing I was shocked when I got out of the car. There was fluid running down the driveway behind the car. I never had this happen before so I checked out the source of the fluid. It did not smell like gasoline and appeared to be just water and it left a black streak which looked like soot as it ran down the driveway. It was coming out of a hole in the muffler which I assumed was a drain hole for condensation drainage.

So as I mentioned I think we have found the source of the problem but I don't know what to do to correct it. I know you will have some good advice.

 

So, what you did by disconnecting the hose is introduce another source of air into the engine, unmetered air. The fact that it began to run fine leads me to believe that you may have a malfunctioning MAF ( Mass Air Flow ) sensor. The condensation you witnessed is a byproduct of proper combustion, totally normal. First thing to try is, with that same hose disconnected and the car warm, start it up and replicate the "good running" scenario. When it is running well, place your thumb over the end of the hose. Does it make the car start sputtering and running rough? If so, disconnect the electrical connection to your MAF sensor ( it will be in the intake ducting ) and see if the car runs better with the hose connected/plugged. If it does, your MAF sensor is probably dirty and / or reading incorrectly. They're cheap ( under $50 ), so I'd just replace it. If the car continues running well with the hose plugged with your thumb, I'd still be leaning towards the EVAP purge valve letting liquid gasoline into the intake. The fuel puddling is the strange part here. A bad MAF shouldn't cause that. Try what I said and see what happens.

 

Being the cheapie that I am I have ALWAYS topped up my tank and filled gas cans when prices dropped. Are you saying this is a bad thing for the evap system? Could I have damaged something by doing this?


I was really relieved when you said you thought the liquid dripping out of the muffler was just normal condensate. As I have said I had never seen it that bad before. I imagined all kinds of bad things until I received your response. Thank you once again.


I just finished trying what you suggested. When I plugged the hose as best I could the engine really started to idle roughly - I did not wait to see if it stalled but I am sure it would have. As soon as I disconnected the MAF sensor electrical connection the engine died. I reconnected the MAF and started her up again with the hose still disconnected. Again as soon as I disconnected the MAF it died. It appears to me as though the MAF sensor is OK. Do you agree? I replaced my original MAF sensor after 14 years just last May so I didn't think it was that but I tried what you said just in case. I think I will purchase a new EVAP purge valve and see what happens. It's only $26 on Amazon. Do you agree?