Engine Idles Unevenly and Almost Stalls When Cold

[Schurkey]

Then explain these....and the fuel pressure regulator is on the rear of the injector assembly.

Well...crap. Mia Culpa. I totally forgot about that, on the big-blocks.

How can I accomplish that without hunting down expensive and hard to find dedicated equipment?

Define "expensive". $500 for a scan tool and accessory bundle is cheap compared to anger-management therapy, endlessly replacing parts that don't solve the problem, or selling the truck at a loss because you can't figure it out.

Hard to find? Heaps of used, professional-grade equipment on eBay, which is where I've purchased my scan tools.

As far as a PC-based solution goes, I have a variety of laptops with serial, parallel and USB that can run anything from Windows 9x to 10, so I can use whatever software and hardware, new or old, is available, so long as it is relatively cheap.

Others on this forum have used the software + cable or dongle solution. I have not. What I know is that folks with "cheap" scan-tool "apps" come back on the forum complaining that their new tool doesn't do this or that, or they have to pay extra for functionality that should have been included with the tool to begin with.

 

[grampadirt]

Well...crap. Mia Culpa. I totally forgot about that, on the big-blocks.


No prob,it's all good

 

[cc333]

Define "expensive". $500 for a scan tool and accessory bundle is cheap compared to anger-management therapy, endlessly replacing parts that don't solve the problem, or selling the truck at a loss because you can't figure it out.

Well, less than $500.

I can do maybe $200, but less is better.

c

 

[jdjonesdr]

Well, less than $500.

I can do maybe $200, but less is better.

c

I have a Bosch 1300 scanner I got on Amazon for a little more than 200 bucks. It's the only Bosch scanner that will scan OBDI systems.
I love the thing. I scan my truck every time I do any sort of work on it, just for schits and giggles.

 

[jdjonesdr]

I have a Bosch 1300 scanner I got on Amazon for a little more than 200 bucks. It's the only Bosch scanner that will scan OBDI systems.
I love the thing. I scan my truck every time I do any sort of work on it, just for schits and giggles.

I just checked EBAY and there's one there for $225. That's a good deal.

 

[Caman96]

I just checked EBAY and there's one there for $225. That's a good deal.

Aside from reading/clearing codes, what functions do you get?

 

[Sabinoerc]

If the light is "on" more than it is "off", the mixture is rich. If the light is "off" more than it's "on", the mixture is lean.

Interesting, I didn’t know that. I’ve used the field service mode to see if it was in open or closed loop before I got TunerPro but didn’t realized the dwell of the flash meant anything.

Well, I ran the truck, locked into closed loop, until it warmed up a bit, and it was idling like normal, unlike before, when it goes into open loop.

As was mentioned somewhere above, the light flashing fast or slow only tells you whether its in open loop or closed loop - you aren’t forcing a mode of operation.

The intention/goal is to see if the bad behavior is occurring when in open loop only or also when it gets into closed loop. I’m assuming when you say “locked into closed loop” you were thinking the act of shoring the pins put it into closed loop and you didn’t see the problem after it warmed up. Did the problem happen at the initial cold start?

Now that the confusion is cleared up, sorry for not being clearer earlier - the suggestion would be to start the engine when cold, put it into field service mode shorting the pins, and watch the light rate of blinking. It will start out in open loop. I’m assuming the problem will be doing its thing as it typically does. Watch the light and see if the rate of blinking changes to indicate the engine has gone into closed loop. See if the problem goes away before it gets to closed loop, at the time it goes into closed loop, or persists after getting into closed loop. There’s a possibility it never gets into closed loop also even when warm.

This could give some clues to what the problem could be and perhaps rule out the O2 sensor as cause of hunting if the problem is only occurring in open loop. As mentioned data logging is much more pinpointed at the actual problem but if like me when first starting working on these, I didn’t have that capability and was trying to get it running that day. So better to have a tool/TunerPro but if you don’t have that in the house at the moment, doing the above is something you can do now and may help, may not - good luck.

 

[Road Trip]

Always good to have a known good O2 sensor.

if you want some confidence on whether the O2 is a contributor check whether problem only occurs in open loop goes away in closed loop. If problem is occurring in open loop only it’s not likely caused by O2 sensor.
The O2 is only used for afm after getting into closed loop.

Of course, you could have a bad O2 and never being able to get into closed loop. Old engines often have more than one problem which are not noticed until chasing a more obvious problem. I just mean that a bad O2 is not going to cause a idle hunting up and down if never getting into closed loop.

Adding to the “it’s worth investing in data logging capability”. You can check if the O2 is working by watching the output as the PCM changes the afm target. The PCM moves the afm up and down and expects to see a response in the O2. If it doesn’t reach expected values, too slow to respond, & some other stuff it will drop out of closed loop. I had a “lazy“ O2 sensor once which was dropping me out of closed loop during hot idle and causing me to failing emissions. It would get into closed loop on cold start fine, stay in closed loop driving but drop out on hot idle which is part of the emission test.

That said, getting going on data logging takes some time and effort. Worth it if you are planning to keep the truck for a long time. Probably too much time when in middle of problem now but something to think about for the future.

for now, I’d narrow done whether the problem is in open or closed loop or both before messing with O2. They aren’t that expensive but sometimes hard to remove. Easier to use a paper clip to decide if it’s a candidate first.

EDIT: Reply collision alert with @Sabinoerc above. But it sounds like we're on the
same page, so I'll leave the following reply in place.

Hello @cc333,

Sometimes while you are troubleshooting a difficult issue, you end up learning on the
fly, and oftentimes this new knowledge would have been really helpful back at the beginning.

For what it's worth, Sabinoerc is giving you solid advice, as seen above.

In order to help the remote troubleshooters reading this thread, I just want to make
sure that we all understand/share the definitions for the following terms:

* OPEN LOOP: Whenever the signals aren't available from the O2 sensors, the computer
will rely upon the (analog) signals from the remaining sensors in order to properly index
into the fueling and spark advance tables, allowing you to immediately drive a cold vehicle
smoothly and with the minimum emissions possible.

On a healthy GMT400, every time you drive the vehicle from a cold start, you *will* be in
Open Loop mode until the coolant temperature rises above a preset threshold and the O2
sensors start switching up & down around the ideal chemically-correct Air/Fuel ratio.
(Stoichiometric)

If your vehicle then switches successfully into Closed Loop mode at that time, then
the way your truck drives will be determined to a large degree upon how responsive/
how accurately/how poorly the O2 sensors measure the amount of oxygen in the exhaust
after the engine has attempted to burn the combustible mixture in each individual cylinder firing.

* CLOSED LOOP: Once the O2 sensors have been warmed up enough (by the hot exhaust
stream only on the old single-wire sensors, warmed up much faster by the internal
heater circuit on the newer 3-wire O2 sensors) ...the base fueling & timing values in
the computer tables are fine-tuned in realtime by the feedback of your Oxygen sensors.

IF you are getting fast/accurate/reliable feedback from your O2 sensors, the way your
engine drives will be improved.

On the other hand, it is quite possible for a mechanically sound engine + a correctly
operating computer to run just fine by themselves (Open Loop) ...but once the
engine heats up and the aged out of spec/outright bad oxygen sensors are switched in,
*now* the engine starts running rough/lethargic/stanky/etc.

****

Given the above, it can be extremely helpful if the troubleshooter of one of our
GMT400 vehicles can determine whether or not the engine bay is operating
in either Open Loop vs Closed Loop operation. And 29+ years ago on the
OBD1 computers, the engineers added in this slick feature to let the mechanics
who understand all of the above (normally the Regional Reps/support-level Field
troubleshooters supporting the GM dealerships) to be able to access this diagnostic
hint with nothing more than a neat little shorting plug. (Or jumper wire, or even
paper clip for the roadside MacGyver types :0)

NOTE: In the perfect world, GM would have placed a tiny pushbutton somewhere in
the cab where I could watch the flashing SES light any time I wanted. But they
decided against that, and I actually understand why. Instead, they made these 2
pins on the ALDL plug perform *2* separate functions:

1) KOEO mode: Key On Engine Off, jumper 2 pins together and retrieve any stored
error codes. (And if there are no codes, we see a Flashing 12 repeated 3 times so
that the mechanic doesn't have to wonder if the 'error remembering/reporting
circuit' is working as advertised or broken.

2) Engine running mode: IF you wait until the engine starts to jumper these 2 pins,
*now* you will get 2 different flashing speeds:

* FAST (2.5 times per second) flashing. Attention! Attention! You are in Open Loop mode,
does this make sense at your current coolant temperature and time since engine start?

* SLOW (Once per second) flashing. You are in closed loop mode. If the engine is responsive/running
like new, you are good to go. On the other hand, IF your engine was running fine in
Open Loop, and then went bad shortly after the system went Closed Loop, then
this makes a big difference in our troubleshooting approach. (!)

****

Good feedback loops can help (within limits) to tune around fuel & ignition components that
are aging/starting to drift away from their original calibrations. Relying upon a error-correcting
feedback loop like this allows our engines to run 10s of thousands of miles with clean emissions
and like-new driveability, instead of running purely Open Loop a few thousand miles before the next
widget ages just a little out of specification, goofs up the driving experience, and the car owner
ends up back in the repair shop waiting room. (!)

But given this same authority, a Bad feedback loop can take a good engine/good computer combo and
make it run poorly.

*This* is why it would help us a lot with remotely troubleshooting your truck if you can
report this Open Loop / Closed Loop status to us.

OK, so I'm running the truck with the two pins shorted for open loop I think, and Service Engine Soon is flashing quickly. Is that normal?

c

Hopefully the above clarifies how the 'post engine running' shorting of these
2 ALDL pins gives you some additional insight into what mode of operation
that your computer is in. And why this was referred to as "Field Service" mode.

So if you could go back to reply #9, reread that page out of the service manual,
and then monitor your truck and let us know when it is running well vs running poorly
(and if this change occurs when the computer switches from OPEN to CLOSED loop,
then we can use this info to refine our recommendations, and hopefully get this
sorted out with the lowest number of repair parts possible.

NOTE: If you figure out how to use this to your advantage, then you will be
troubleshooting at the old "Regional Rep" level. And of course, all of the above
is just a peek under the covers -- if you really want to set yourself up for full
Situational Awareness, at some point down the road you will want to invest in
some gear that will allow you to see ALL of the realtime data streams.

I tried to keep this explanation as straightforward as possible. But if I missed
the mark and you now have additional questions, please don't hesitate to post
them here.

Best of luck getting your GMT400 vehicle running as right as rain.

Cheers --

 

[cc333]

I think I understand now!

So it was flashing fast for the first few minutes, then it slowed down, so that must mean it entered closed loop. It is at this point that the problem usually happens (though not always; it seems less common in warmer weather). The interesting thing I noticed is that it seemed to bounce between open and closed loop a few times (the light would flash slowly, then it would speed up for a bit, then slow down a bit). It seems to me that once it enters closed loop, it should stay in closed loop, yes?

Also, given that it seems to do the right things regarding warming up and switching to closed loop when it should, the temp sensor is probably fine, I would think?

At any rate, based on my limited understanding, given that the idle problems begin when it goes into closed loop, thereby involving the O2 sensors, it seems likely that one or more of those sensors are falling out of tolerance and beginning to fail, agreed?

I say beginning to fail, given that it doesn't exhibit these problems in warmer weather and/or after the truck is fully warmed up.

How much would a shop charge for replacing one or more O2 sensors? (since they're all problably the same age, does it make sense to replace them all in one shot?)

c

 

[Schurkey]

it was flashing fast for the first few minutes, then it slowed down, so that must mean it entered closed loop.

Yup.

The interesting thing I noticed is that it seemed to bounce between open and closed loop a few times (the light would flash slowly, then it would speed up for a bit, then slow down a bit). It seems to me that once it enters closed loop, it should stay in closed loop, yes?

A failing O2 sensor is the usual cause for dropping-out of closed-loop operation, typically at idle. The reduced exhaust flow allows the sensor to cool-off too much to provide useful data. Misfire would make that worse.

I'm not saying that's the only cause...just the most-common.

Also, given that it seems to do the right things regarding warming up and switching to closed loop when it should, the temp sensor is probably fine, I would think?

I would want to verify it with a proper scan tool. However, it's obviously sending a "hot enough" signal to get the computer into closed-loop operation at least some of the time.

At any rate, based on my limited understanding, given that the idle problems begin when it goes into closed loop, thereby involving the O2 sensors, it seems likely that one or more of those sensors are falling out of tolerance and beginning to fail, agreed?

Not guaranteed, but very likely. Looking at the sensor waveform, or at least checking the O2 sensor cross-counts with the scan tool would be helpful.

For example, a faulty ignition coil could naturally warm-up in use, and create problems with ignition voltage that have NOTHING to do with the O2 sensor, but might only happen as the engine has been run long enough to enter closed-loop.

You still have diagnosis to do.

How much would a shop charge for replacing one or more O2 sensors? (since they're all problably the same age, does it make sense to replace them all in one shot?)

I replaced all four on my '97 K2500 soon after I bought it. They may have been originals at ~170K miles. I had multiple codes for the O2 sensors, anyway.

Check your service manual. Moderate-to-good chance a '94 has only one sensor; and almost certainly not more than two. None behind the catalyst in OBD1, but MAYBE one for each cylinder bank. The service manual will specify.

Probably best to replace it yourself; I can imagine a shop charging an hour's labor at maybe $100+ per hour, and also charging you double what the sensor actually can be had for if you buy it yourself instead of through the shop.

OTOH, O2 sensors are supposed to be installed with anti-seize on the threads...but they can sometimes still be seized in place and very difficult to remove. In that case, an hour's labor (or longer) is entirely reasonable. You won't know until the sensor is removed.

If your O2 sensor is the one-wire kind, they can be very inexpensive even for the "good" brands. Again, the service manual will tell you if it's a one-wire or 3-wire (heated) sensor.