VERY lean on acceleration

[Kudzupatch]

93 suburban 350, automatic, 185K Sorted out a couple of issues recently. New distributor, egr and just put in O2 sensor. It was running good with an occasional hesitation or roughness from start up. So I knew I didn't have everything fixed but I had the big issue I was having fixed.

Our first cold morning it no lost power on light acceleration, MOST TIMES. I can feel it running rough, probably misfiring. Light acceleration from start, hitting an incline with cruise on or just speeding up, per the o2 sensor reading it goes EXTREMELY lean. I can watch the knock count jump at the same time. Cruising down the road it bounces around as it should. But on acceleration it drops to near 0 and stays there.

I have Tuner Pro but I have been using ALDroid and it doesn't give O2 voltages just a 0-100 scale with 0 being bottom end lean and 100 top rich end. 02 sensor reading drops to nearly 0 or fully lean. I replaced the o2 sensor just to verify this was right and same readings from both. Or sense was very white too.

It runs fine in open loop. My fuel pressure gage went out, new one will be in tomorrow and I will verify fuel pressure. But it runs fine in open loop and under harder acceleration which doesn't sound like low fuel pressure to me.

Engine temp is showing 180-185 degrees on the computer, not my dash gage. MAP reading seem to be normal. Nothing stands out to me.

I will check again for a vacuum leak (done recently). Other than that I am rather stumped. Assuming the gage doesn't show fuel pressure problem.

[ChevyTech]

If you put in a Bosch Oxygen sensor , I would put the old sensor back in.

If you used an EGR that is not OEM, I would put the old EGR back in. These trucks are very picky about the EGR operation.

Check your fuel pressure when you gauge comes.

I would disconnect an plug the EGR hose, and take it for a test drive and see how it runs.

[speedygonzales]

If 2 different sensors gave you the same reading near 0 I'd bet your O2 sensor wire is shorted to ground. If you unplug the sensor and the reading stays the same, it's definitely shorted. The PCM is only trying to respond to the false reading. This is indicated by it running correctly in open loop. If you were truly running lean, it wouldn't matter if it was in open or closed loop. Understand?

[Kudzupatch]

New GM EGR, learned that lesson! But I was thinking about plugging it since it is the newest item and just seeing what it did.

Funny, it is a Bosch 02 sensor. I take it they have a bed reputation. As bad as aftermarket EGRs?

Speedy, O2 reading bounce around normal, except on light acceleration. Runs fine down the road. When it looses power you see the lean condition from the O2 sensor and that is only under light to moderate acceleration and then it not 100% of the time.

[ChevyTech]

Originally Posted by Kudzupatch

....Funny, it is a Bosch 02 sensor. I take it they have a bed reputation. As bad as aftermarket EGRs?

Worse then aftermarket EGR valves.

A lean oxygen sensor reading, is a low voltage oxygen sensor reading, as speedy points out.

[Kudzupatch]

Found the voltage option in the ALDL Droid. It drops to as low as 4.4 -8.0 (millivolts?) when accelerating. on cruise, level ground it bounces around 400-700.

Pluged the EGR and no difference.

Fuel pressure is 12-12.5 psi at idle. Rev the engine and still maintains 11.5 lbs.

However in replacing the gauge I did discover I have injector issues. One is 'dripping'. It making a cone but not like it should. Other one looks better but I suspect that neither is working correctly. Should I just replace them or service them myself?

[ChevyTech]

I would swap out the injectors.

Injectors are only $4 each at the local U-Pull-It so I keep a collection of them.

New injectors are ridiculously expensive at the dealer, but you can buy injectors online fairly cheap.

Make sure you lube the o-rings the injector slide into. I you install them dry you can crack the TBI housing. The housing is very thin.

A oxygen sensor voltage that only moves from .4 to .7 volt is a problem. It should swing farther but many scan tools are not great at sampling often enough to show an accurate voltage swing.

[Kudzupatch]

Thanks, I just ordered a set of rebuilt injectors. I kept thinking about and decided to go this way. I am going to take the old ones apart and see what I find. We don't have a pick-a-part anywhere near us or I would have a spare TBI unit.

[Kudzupatch]

OK, I have some TunerPro data saved. Better than the ALDL Droid data. Not seeing anything obvious but I could be missing something.

If anyone has suggestions what to look for, let me know.

Injectors will be here in the next couple of days, but I am skeptical.

[ChevyTech]

Here are some guidelines for looking at TBI sensor data.

Normal readings depend on the application, but here is some basics

IAC count 1-50 This should be checked with engine fully warmed up at idle without the A/C or any accessories on. A low number is better but you don't want it to get down to zero. 20 is ideal.

CT - temp of coolant should go up to the thermostat temp (195F if stock)

MAP 1-2 volts @ operating temp at idle (varies with barometric Pressure and engine vacuum)

The knock counts in winaldl is a count of all knocks since you hit the clear button last. Don’t think of it as knocks per any specific time frame.

TPS voltage:
It should be between .5 volt and 1.25 volts at idle and increase smoothly as you slowly open the throttle to at least 4 volts at wide open throttle. Some spots give a different allowable voltage reading at idle in GM manuals

[Kudzupatch]

IAC count 1-50

Mine is at 8.

CT - temp of coolant should go up to the thermostat temp (195F if stock)

This puzzles me. 195 Thermostat, I put it in. Shows 177.

MAP 1-2 volts @ operating temp at idle (varies with barometric Pressure and engine vacuum)

TunerPro doesn't give voltage, will check that manually.
Shows 31-32 Kpa at idle.

The knock counts in winaldl is a count of all knocks since you hit the clear button last. Don’t think of it as knocks per any specific time frame.

I have a high knock count. I get a 150+ starting the engine. I have noticed they increase during the lean conditions. I though there was something else knocking but after driving around and watching the data I think they are real knocks, from lean condition. I will have to look closes at some old logs.

TPS voltage:
It should be between .5 volt and 1.25 volts at idle and increase smoothly as you slowly open the throttle to at least 4 volts at wide open throttle. Some spots give a different allowable voltage reading at idle in GM manuals

.5 volts at idle. I will have to check WOT, I only had a 55% throttle on this run but it was 2.5 volts. Smooth readings all through the range. I checked that before I started recording but don't remember WOT voltage.

Thanks for the numbers. Everything has always looked good and that is what stumps me.

[ChevyTech]

At 177 degrees the computer will not run some self tests to set trouble codes because the engine is too far below normal temperature.

There are many thermostats being sold lately that are no good. I prefer STANT brand.


If the engine were running warmer the IAC count would likely go lower. You never want the IAC count to go to zero. You should retest this after your repairs are finished.

The MAP reading is ½ of what it should be. That would be about 9 inches of vacuum.

Check the voltage. 31-32 KPa would give around 3.3 volts

18 inches of vacuum would give around 1.7 volts.

I would put a vacuum gauge on it and see what the vacuum is at idle. Test the vacuum that is getting to the MAP sensor.

The low TPS voltage fits with the low IAC count. You may end up needing to adjust the minimum air rate (throttle stop screw), but before you do this lets discuss it more.

[Kudzupatch]

I have been reading EVERYTHING I can find and you are hitting on the items I suspected, but just didn't have the knowledge on.

I have been wondering about the MAP sensor. I have a vacuum gage and will check the voltages too.

I looked back at some Data Logs from 6 months or so ago when I first started to try to understand the ins and outs of Fuel Injection. I see that I had a lean condition then too but I didn't realize it. So what ever it is, is getting worse.

[ChevyTech]

I screwed up. The low TPS does not fit with the low IAC count. It would be a inverse correlation.
Low IAC count means the Computer wants less air in at idle. That would equal throttle plates closed farther which would give a lower TPS voltage.
High IAC count can be caused by vacuum leaking in anywhere it can. If someone backed out the throttle stop screw, trying to fix a Low IAC count, that would make the TPS voltage low.

Look for vacuum leaks. That could mess up the MAP reading. Look again at the MAP vacuum line, and the TBI base gasket.

[Kudzupatch]

My gage shows 19-20" vacuum. Voltage was 1.05 volts. Unless I missing something that MAP sensor is reading correctly?

I did swap vacuum lines so that I was using the same one to test with the gage and the computer. But no leak in the factory hoses. No carb. cleaner so I can't check the gaskets but the base gasket was replaced a few weeks ago.

Now the confusing part. TunerPro and ALDLdroid are both showing 30 kPa instead of 70 they should show. So the computer is providing the wrong information??

I am confused now.

[Kudzupatch]

FOLLOUP- Not sure what these are measuring vacuum in but I realized that the higher the vacuum the lower the number which is backward to a kPa scale. So both softwares may be reading correctly and it is not kPa they read in. But TunerPro say kPa.

Also, forced it into OPEN loop and it runs just as bad as it does in CLOSED loop.

[ChevyTech]

The computers do not go bad often, but when they do I have found many times that the data does not match what the sensors are reading.

When a vehicle does not run well in open loop, look everywhere else for the problem except the computer.

Any parts you have replaced are suspect – it is sad how bad the quality of parts is now.

Test things like engine compression, exhaust back-pressure, ignition output (weak spark)

Double & triple check the firing order – you worked there.

I have no experience or opinions from anyone of ALDLdroid, but TunerPro has a good reputation.

When you do remove the injectors, look at the screens on the bottom of the injector to see if they look dirty.

[Kudzupatch]

Found this thread on TunerPro and kPa conversions. Not sure it is right but the numbers do match up to what I see in a road test. Coasting down a hill and drop it down into a lower gear to increase the vacuum you see a lower kPa number instead of higher as I would expect. That matches up with the chart and idle vacuum I measured.

http://ls1tech.com/forums/pcm-diagno...i-kpa-bar.html

Waiting on the Brown truck to bring my injectors. Old ones have clean screens. All passages on the body are open. Very clean, just a little carbon build up in one passage.

I have been fooilng with this one for a while now. I have checked everything on you list except the Cat. I read that you can pull the o2 sensor as a way to check. Is that right? I need to remove that Bosch I bought anyway.

[ChevyTech]

Scan tools, and software, reading things backwards of what you expect, happens more then it should. If the voltage reads right, and you are comfortable with that is the way TunerPro reads, we will call it good.

Some people remove the oxygen sensor and use a gauge test for pressure in the exhaust.

I have always put a vacuum gauge on the intake and raised the RPMs to test for restricted exhaust. If the exhaust is restricted the vacuum will go lower as the throttle is held at a higher RPM.

[Kudzupatch]

Good information and an easy check!

TBI back together. Put on the car this morning cleaned and with rebuilt injectors. Turn on the switch to build some fuel pressure and check for leaks..... I've got one. One of the rebuilt injectors is dripping fuel!

Just to be sure I swapped sides to make sure it just wasn't the o-ring. Swapped that out too and drip, drip, drip.

What is going on in the world? It's getting to where you need to buy extras to make sure you get a good ones! Good thing is I got them Summit and they will have a replacement tomorrow. But mean time I am stuck with no way around.

[ChevyTech]

If they want the old injectors back, (for a core) I would keep them unless you are sure the old injectors are bad.

So many new parts are bad out of the box that I only replace what is necessary, and don't replace parts because they are old.

I'll take a good used OEM part, over the parts that are getting made today. This goes for after market or OEM.

It used to be safe to buy NAPA's top line parts. I have been sold so many bad parts from them, that I urge everyone to be suspect of parts you buy.

When I read on the boards about guys replacing parts, without doing any testing, the first thought I have is, now there are going to be so many bad parts, and problems, that they will never figure out how to get it fixed.

I have many parts, from the local U-Pull-It that I got for dirt cheap, are in great condition, I would trust over new parts.

[Kudzupatch]

I think the old ones are probably just dirty with varnish. Everythign else looks good, screens are clean. Previous owner took great cosmetic care of the car and appears to have oil changed and plugs put in regular. Otherwise seems if it ran he didn't mess with it. It had factory cap, wires and rotor at 160K. I swear it gained 20 hp just changing those.Had a few minor issues like that I had to take are of.

I am thinking of keeping the injectors and soaking them in something like Sea Foam to see if I can clean them. They work but there is a big difference in the two. One has a really bad spray pattern with droplets instead of a cone. There is a noticeable difference in the two and appears one has a lot more volume than the other.

Should know tomorrow if this was my problem.

[Kudzupatch]

HALLELUJAH!! It seems to be fixed!!

Just got the replacement injector installed and took if for a drive of about 5 miles. Before it could barely get out of it's own way. Now it runs like it did when I bought it! When I step hard down on it the o2 levels drop low, not nearly as low, and sits there for second and then it adjusts and it show high levels and starts bouncing around.

More normal acceleration, it just bounces around like I would expect. Obviously the injectors were an issue. Varnish build up or weak solenoids or who knows what? But they were in bad shape.

I just has my phone connected to it and I don't have it set up to show me all the data. So I still want to connect TunerPro and get all the readings. I did go dead sitting in the drive. There seems to be an idle issue but whatever it is can't be much. I need to pick up another thermostat tomorrow.

[ChevyTech]

Glad you are making progress!

The scan tools usually put the ECM in the 10K mode.
See if your software talks about:
Data link mode (also called 10K or special)

In the 10k mode, data is readable, but it alters:
Closed loop timers, Distributor ignition Timing, IAC control speed...

Go drive the truck with nothing attached to the ALDL.

Also drive it at highway speeds, or at least 45 MPH cruising speeds, so it will invoke relearn of the IAC.

[Kudzupatch]

I couldn't stand it, before I read your reply I just went 'cruising' around town and other side of the river. I drove it around for probably 10-15 miles on the highway and in town. Still had a somewhat irregular idle and it has gone dead a couple of time at idle. BUT, it runs so much better! When I got back home it idled just fine in the garage in gear.

But the throttle response is just so smooth! I would step on the gas and expect that hesitation and nothing much to happen. Now it goes! Obviously the injectors were the main issue.

Tomorrow morning I will pickup and install another thermostat. After I drive it a bit, if it still dies at idle after that I will put the laptop on it and get the numbers.

I have said it before but a big thank you of the help!

BTW, I am guessing you are snowed in? We rarely get snow so it all over the news and I can't imagine that much snow!!!