Rough idle and low power, stalls

[BoTwentyone]

I could really use some help on my 1998 Mazda B2500. It has a rough idle, low power, stalls when throttle is revved then let off. It had a misfire in cyl 2. We changed spark plugs/wires/coils and then cleared the codes. The misfire then went to cyl 3. Brought it to a shop for diagnostics. They suspected bad computer or fuel injectors. Replaced injectors, then got a cyl 2 AND 4 misfire. Replaced computer. Misfires seems to have gone away at least but still runs rough, low power and stalls.

Only CEL that has show up were the engine misfires. Currently not showing CEL, or any pending codes


Things we have checked/replaced:

replaced:
plugs/wires/coils
IAC
TPS
injectors
computer
MAF

checked:
compression was around 160
fuel pressure was around 60
timing is around 10 at idle

When we read the live activity on our OBD II scanner we can't find anything that seems off.

Off the top of my head these are the measurements I remember:

MAF reads .5
TPS reads around 18%
timing was between 7-12 at idle BUT we did have one spike around 25
RPM was around 800


The computer was coded to my specific vehicle using my VIN. When we replaced it and started up the vehicle, it could barely idle. It would stall when we tried to drive it. It took a good 10 minutes or more for the computer to work out what was going on and keep the truck running, barely. That tells me its compensating for something really wrong. I suspect timing based on what I've researched, but we haven't torn into the front end to check this. I think this will be next on our list to rule out. Can a computer compensate enough for a timing belt that has jumped a cog or two? The timing seems correct at 10 at idle when we use a timing light, and it reads around 10 when using the OBD II scanner. Can the timing have jumped a cog or more and still read correct? Or would tearing apart the front end be a waste of time?

Thanks in advance for any help. This has been a year and $600+ of chasing gremlins in this motor.

 

[RonD]

Welcome to TRS

First eliminate a vacuum leak
Warm up engine to operating temp
let it idle
Unplug the IAC valve's 2 wire connector
IAC valve will close all the way and RPMs should drop down to 500 or engine may even stall, either is good it means no vacuum leaks

If idle stays above 700 then you have a leak.

Next use a Vacuum gauge, $20, to see if you have sticking valves or other issues
Read here: http://www.gregsengine.com/using-a-vacuum-gauge.html

While it is popular to replace sensors and controls these rarely fail, 90% of those parts that are replaced were working fine, just FYI

Most engine issues are what they always were, mechanical issues.
OBD codes are there to help you narrow down where the mechanical issue might be, the code means sensor is working, and sees a problem, changing the sensor is like shooting the messenger, lol, and while popular it doesn't fix the problem.

So don't approach engine repairs from an electronic point of view, engine was there first, electronics were added to make it more reliable, and it does, but it is still a self powered air pump at its core, and that's where you start your diagnosis.

Spark plugs and wires are good to change, not part of the electronics, they always needed changing since the first gasoline engine.

Compression at 160psi would indicate cam and crank are properly timed, so no belt slip
What did the old spark plug tips look like, they tell you about each cylinders condition.
Bright steamed clean white is a coolant leak, blown head gasket
Light tan is good burn
Darker tan going to black is rich running.
Blackish build up is leaking valve guide seal

O2 sensors wear out(they use a chemical reaction like a car battery), recommended change is 100k miles, but if there was an engine problem like Rich exhaust, it can shorten O2 life.
They also tend to "fail lean" so as the chemical wears out the computer will run the engine Richer and Richer, so MPG slowly drops and it costs you money.

Yes, anytime computer is cleared it can take a few drive cycles(heat up/cool down) for computer to relearn engine control settings

Set MAF to read in gm/sc, grams per second, should be about the same as engine displacement at warm idle. 2.5gm/sc
But the .5volt is about right, MAF is 0-5volt scale, as air flow increases so will voltage, you look for varying voltage at steady RPM, that would indicate MAF problem.
Same with TPS, 0-5volt range, under 1 volt throttle closed, above 4.5volt throttle wide open, engine off, press down on gas pedal and you should see a steady increase in volts, no jumping or dropping indicating a dead spot

 

[BoTwentyone]

Thank you for your informative reply. I'm learning as I go here, so I appreciate the help.

We have pulled the IAC at idle and it does stutter some, but does manage to catch itself and keep running. I will try again when I get home and watch the RPMs to see if they go back up to its normal 750 after it stutters. That would indicate a vacuum leak?

Hubby changed out the spark plugs, so I didn't get a chance to see them. I'll pull them when I get home to see if they give me any indication of what is going on. I'll report back on that later tonight.

Is there any way to pinpoint a failing O2 sensor without replacing them just to see if anything changes?

Thank you for verifying that you think the timing is correct. This one has bugged me enough to want to tear down the whole front end just to verify the marks line up just to make sure.

I will be looking into investing in a Vacuum Gauge if we can't find one to borrow. The link you included was interesting and a great way to verify working/non-working components.

 

[BoTwentyone]

Im posting a little feedback from our testing when I got home from work. I pulled the IAC wires and it stuttered and died almost instantly, so that rules out a vacuum leak?

We rechecked the compression on cyl 1-3 (4 is hard to reach so we didn't ch clean that one), compression was around 170-175.

Plugs were a nice golden color.

I will look into a vacuum gauge for this weekends project.

Can an O2 sensor fail and cause no CEL? I have no codes, no pending codes, and have never gotten anything other than misfire codes. So no lean codes or anything like that

 

[RonD]

It is good the warm engine stalled when IAC Valve was unplugged, yes, no vacuum leak

As O2 sensor gets older, engine will run Richer without setting any codes, because computer has nothing to compare O2 sensor data with.
Computer varies open time on fuel injectors changing air/fuel mix from lean to rich and back to lean quickly and constantly, this helps Cat converter work, Rich mix keeps it Hot, and it also tests O2 sensor, a good O2 sensor will switch voltage very fast as computer adjusts the fuel trim(open time), when O2 gets older it will slow down and not react as fast, computer will set a code for that, "slow switching".
But your MPG will already be down a bit at that point, so best to change O2 sensor at 100-125k miles to stay ahead of the MPG drop, which makes the O2 free because of the fuel savings, lol.

Vacuum gauge will tell you a few things, 19-21" vacuum at idle means good compression(which you know you have) and good spark timing which you have a question about.
Also clogged exhaust, partially blocked exhaust limits power and causes misfires because if exhaust can't get out then new air/fuel mix can't get in.

Another simple check is the electrical system, test battery voltage with engine running.
Just after starting it should show over 14volts, after running 5-10minutes 13.5-14.0volts
Low voltage mean low spark.

Sorry to say I don't have spark timing info for 2.5l, but..................
While 10-12deg BTDC is correct for "base spark" timing it isn't correct for computer controlled spark timing at idle.
Base spark timing is done with Vacuum Advance(old distributors) or Computer advance(SPOUT) disconnected.

Once that is set you then hook vacuum or computer back up and timing changes, with your system you can't unhook computer to set base spark timing.

The point of spark timing is to get spark plug to ignite the air/fuel mix and for that mix to "explode" when piston is about 5 to 10degs After TDC, so it has leverage to push down on the crank, and there is a delay from when spark is sent and when "full power explosion" occurs in the cylinder.
Spark travels at a constant velocity from coil to spark plug, so that is a fixed time delay
RPMs are not, so spark timing has to be changed to match speed of piston
Air/fuel mix burn rate is also not constant, Rich mix will "explode" faster than lean mix(unless its too lean then it self ignites, lol)

If spark timing is off by just a few degrees, then engine is sluggish and without power, or can get "pingy" when warmed up



Extreme long shot here, but this sets base spark timing for your engine:
The 1995 and up Lima engines(2.3l/2.5l) did have an odd setup on the Crank Position(CKP) sensor's Tone Wheel.
It wasn't keyed or locked to the crank, it was press fit onto the back of the crank pulley.
A tone wheel looks like a gear because it has teeth, one tooth every 10degs, 36 teeth, then one missing tooth to denote #1(#4) Top Dead Center(TDC), 35 teeth
As a tooth passes by the CKP sensor the sensor generates an AC Voltage pulse, computer bases spark timing and fuel injector timing on these pulses, and when #1 is at TDC.
I have only read of one indecent where the press fit tone ring had slipped, and it was almost a no start, certainly not driveable.
Image of that setup here: http://www.partsgeek.com/assets/dimage/full/1406937.jpg