2.3L ('83-'97) Bucking under load, no power, poor MPG

[cstarbard]

I do not have a hand vacuum pump with which to test the EGR valve so I am going to get an EGR block off plate to see what difference removing the EGR valve from the equation makes, if any.

I did spray a little carb cleaner into the small holes in the EGR valve housing and my RPMS did not increase so I guess at least it's not stuck open/ the diaphragm isn't leaking?

I tried using propane to find vacuum leaks again, especially near cyls 3 and 4 while my wife watched the fuel trims on my scanner. Again, I did not find any leaks

I'm thinking about getting an ECU from a salvage yard for kicks but I feel like thats probably not the problem

 

[cstarbard]

I had to drive my ranger to work today, so I tried to make some observations about the bucking and I realized a few things.

I got to work at 9AM. 4 hours later (35degrees F today) I got in my truck and hit the road. I wanted to see if the buck behaved differently in closed loop/cold as opposed to warmed up. I dont have my scanner so I just decided to see how it behaved immediately after starting cold. I accelerated in a variety of ways, light, hard etc. Turns out, when the truck is just started, the buck is barely there regardless of load. I got two tiny jerks around 1500 rpm two different times. It felt a little more peppy and it accelerated much better than normal. The two bucks were super gentle by comparison to the normal experience. It all felt pretty normal, honestly.

After driving 15 minutes (truck seemed to be warmed up at this point), the bucking was back. Maybe a little worse than normal, but overall very consistent with how it normally is. It kept happening between 2k-3k rpms. It seemed worst when I was accelerating lightly/moderately... It would buck and misfire a lot. If I really stepped on the gas it actually settled down quite a bit (the bucking pretty much resolved, accelerated decently)

Just wanted to update my observations of the symptoms, hopefully these mean something to someone.

 

[RonD]

The computer runs the injectors from a table in memory until the O2 sensors are warmed up above 650deg
So it does read like once computer is doing its fuel calculations "on the fly" with feed back from the O2 sensor is when it the problem

It could be sensor input data is off a bit or computer calculator is
Did you unplug the MAF sensor and drive it that way AND get a code for no MAF sensor
Same for IAT(air temp sensor), and Cam sensor
One at a time of course, engine should run OK, maybe a bit of power loss, but should get a code as well, sanity check on computer

 

[BrHotte]

Cstarbard, I messed around with my egr as well last weekend. I took it off and it was pretty clean. It seemed to function properly and did not have any vacuum leak on the diaphragm. Put it all back together and it ran the same. I did unplug the vacuum line from the egr and drove it about 25 miles. Had to plug it back in because it seemed worse. Good luck.

 

[cstarbard]

Thanks for the update BrHotte. I need to look at my EGR for sure.

Ron I really appreciate the help more than you know. I suspect there are a few problems going on, but as you are saying, I at least suspect something between 02 sensor and ecu since the problem is so much worse in closed loop. Are there multimeter tests I could perform on the upstream 02? It's values on live data always look ok

Way back in the spring when these problems first began, I unplugged the maf and took a drive and there was no change in symptoms. I believe I got a CEL from it being unplugged. I'll try that test again this weekend.

I'll try the same test with the IAT and CPS. Are there multimeter tests I could perform on the crank position sensor and the cam position sensor as well?

 

[cstarbard]

Cstarbard, I messed around with my egr as well last weekend. I took it off and it was pretty clean. It seemed to function properly and did not have any vacuum leak on the diaphragm. Put it all back together and it ran the same. I did unplug the vacuum line from the egr and drove it about 25 miles. Had to plug it back in because it seemed worse. Good luck.

That sort of makes sense to me since, if operating correctly, the egr opening should lower cylinder temps and make a lean/condition, knock less likely. And our (or definitely mine, as far as I understand) problem is centered around a lean condition

 

[RonD]

Exhaust gases slow the burning of a Rich mix, which causes lower cylinder temps vs the high spike in temp you would normally get, and that spike causes a spike in NOx gases which are toxic

An actual lean mix would cause pinging/knocking and high cylinder temps

A Lean O2 sensor code means too much oxygen was detected in the exhaust on that bank, this code means computer is having to add more fuel than it calculated, it doesn't mean engine is actually running lean, pinging would mean engine is actually running lean

The "more fuel than calculated" can mean a few things
Computer is programmed for the engine its running, so it already knows a 2.3 LITER engine will pull in XXX amount of air at XXX RPM and XX throttle position, its just Math
But since air/fuel ratio is by WEIGHT it needs the weight of the in coming air, thats what MAF sensor is for
Computer is also programmed for fuel pressure, 35psi, if pressure is actually 20psi then less fuel flows out of each injector, so calculation is off

If there is a vacuum leak then not all the air is passing thru the MAF sensor so its "weight of air" would be off

An exhaust manifold leak sucks IN air, so O2 sees that air/oxygen as a False Lean
O2 sensors run out of chemicals to detect oxygen after 125k miles or 10 years, and when that happens they start to show Lean, so a false lean

 

[cstarbard]

Thank you for an abundance of helpful info as always Ron. That was educational.

Last night I unplugged the Maf and toom a drive. It set a CEL and I think it may have run a little worse than it normally does in open loop.

Disconnecting the IAT also set a CEL, and also seemed to make it a run a little worse than normal

I'll try the CPS this weekend

Do these results so far this tell us anything about the ECU?

 

[cstarbard]

I also changed the air filter last night which was one of the last tune up items to do, and I poured a bottle of Techron injector cleaner in the tank since I have wanted to do that anyway.

I have a can of seafoam for the intake so I'll use that this weekend as well. Not expecting much in the way of results from these things but seemed like a good time to cross off some general maintenance items anyway

 

[cstarbard]

What us very amusing to me is that when I got the truck ~50k miles ago, the carayltic converter was dangling from the manifold, and the upstream 02 sensor was physically broken, AND someone had connected it to a harness from,the t case instead of the right harness coming down from the firewall area... Despite all these things it ran fine somehow.

50k seems premature for an 02 sensor failure doesnt it? It was a Bosch or NGK

 

[RonD]

Yes, 50k is early for O2 sensor issue, but it depends on previous engine issues as well, i.e. rich running, burning oil or coolant shortens O2 sensor life

 

[cstarbard]

Good to know. I don't lose coolant or burn any oil to speak of.

Today I smoked out the intake to see if I could find any vacuum leaks that way. After doing that for quite some time, I highly doubt there are any leaks... I didn't see even a wisp of smoke anywhere except for what eventually found its way out through the intake snorkel in the fender well

I seafoamed the gas tank and the intake. After seafoaming the intake I was "driving aggressively" like the instructions say to and I noticed that under hard acceleration it never bucks. So it doesn't buck under light or hard acceleration... Just dead in the middle between the two

 

[cstarbard]

Also I checked my exhaust for looseness/leaks. The manifold seems good and tight and I put a new gasket in there a few years ago. Joint of manifold to cat pipe is also good and tight. Everything looks good there

 

[cstarbard]

So I drove my truck to work again today and decided to keep the scanner plugged in the whole way home so I could read the live data. Specifically, I wanted to watch the upstream 02 sensor to see if we could identify it possibly reporting a false lean based on its behavior. Here's what I observed:

At steady throttle/load cruising in a gear, or coasting in a gear with no throttle, and at idle, the upstream 02 sensor generally hangs around 0.8xx-0.9. However, with no change in operating conditions, it will randomly drop to 0.04-0.2. It happens very frequently. It's like half the time it reads 0.8-0.9, and half the time its reading 0.04-0.2. It just goes up and down over and over. Once in a great while it would fall to around 0.5, but pretty infrequently by comparison.

When the truck is bucking, the upstream 02 sensor values seem to be in the 0.04-0.2 range more often than not, and behavior goes back to what I described above when the bucking settles.

The downstream 02 sensor has a similar behavior to the upstream sensor but it peaks around 0.6. It never got much higher than that.

Does this seem like normal upstream 02 sensor behavior?

Also my timing advance is like literally always 25 degrees. Is that weird? I'm going to take the plastic timing cover off and check the timing this weekend anyway, but I replaced the timing belt and tensioner only a few years ago so I would be surprised if there was an issue there

 

[RonD]

You had to be watching the downstream O2, B1S2 bank 1 sensor 2
It should run .7-.8 steadily and then drop lower once in a while

Upstream O2, B1S1, bank 1 sensor 1, should run .2-.7 and switch very quickly as computer adjusts air/fuel mixture

Not all OBD2 readers ID the O2's correctly

.1 is Lean, high oxygen in exhaust
.9 is Rich, low oxygen in exhaust

Downstream O2 is after the Cat converters, Cat converters burn up most of the oxygen as they burn up pollutants
So downstream O2 voltage will be .7 - .8 and stable

Yes, the spark timing should be changing under acceleration