To be PC, its an indication of (too much) oxygen reaching the O2 sensor.


Kind of. A misfire will send too much oxygen to the O2 sensor, while the leaky injector will not send enough oxygen, UNLESS, it causes a misfire which will result in too much oxygen.


If the sum of the short + long fuel trims exceeds ~+28% it will set the P0171/4 code. In other words it is running with too much oxygen in the exhaust and the PCM has added fuel and still can't approach the "perfect" mixture.


Likewise, if the sum of the short + long term fuel trims exceeds ~-28% it will set the P0172/5 codes. In other words the PCM has been subtracting fuel and still can't get to "perfect".

 

I think that is a typo.
It should be
LONGFT >+25%, SHRTFT >+5%


and
LONGFT <-25%, SHRTFT <-5%


In other words the first one should read "greater than +25%........."
and the second one should read "less than -25%.........."


And they are talking about an engine with lean malfunction (i.e, vacuum leak) or a rich malfunction (i.e, torn vacuum regulator diaphragm (which would cause it to suck fuel)).

 

@pdqford


NOT. I am Not trying to be a smart *** here. Just pursuing the ultimate objective of diagnosing "mis fires" by using OBDII data readily available by scan tool data - Long Term and Short Term Fuel Trims. All while trying to stay on topic for this thread, and hopefully not getting too deep to be helpful.


But I have always understood the symbol ">" to mean "greater than" and "<" to mean "less than" (both exactly like what you say in the above quote). Admittedly, I don't understand why you indicate their use in the document might be a typo. That part of the attached document, and the diagrams, seem to agree with what we are BOTH saying (at least, in the final analysis [*1]). And it is in Agreement with the thought you expressed in the last sentence of the abobe quote. I would think a few misfires or series of misfires (enough to flash the CEL) would produce the same result as a torn vacuum regular diaphram - indicated by the second diagram and the second entry in the Typical Malfunction table. ie (LONGFT <-25%, SHRTFT <-10%) The PCM trying to choke off excess Fuel Mass.


[*1] NOW: It seems we are BOTH saying the same thing in POST #31 - (except my incorrectly expressing the O2 sensing 'too much fuel' which it does not [directly] do. But doesn't the conditions of this quote ---
--- match the First and Second diagram of the attachment respectively? And wouldn't misfires, or a series of misfires, have the identical effect as the second example ?? - 'ultimately' driving LTFT into the negative territory?


What you say in Post #30 is indisputably true. Paraphrasing you - "The O2 sensor will see unused oxygen from a misfire from ANY cause (even a stuck off fuel injector, or for that matter a stuck open fuel injector".


Therefore, as you correctly say, the PCM's reaction to that would HAVE TO be "ADDING" fuel (positive STFT) - INITIALLY to STFT anyway. BUT the PCM (via O2 sensor readings) would VERY QUICKLY determine that to be the WRONG reaction. THEN - I believe, perhaps due to the unbelievable speed with which the PCM can sample / calculates and react, - it quickly determines (by the constant closed loop cycling of injector pulse widths and feedback sampling of O2 readings) that the PCM determines that 'subtracting fuel trim' is the better reaction. Then reduces LONGTFT in response to a Misfire or series of Misfires - whether spark related or flooding from a sticky injector.


Challenge / Concurrence / Comments welcome.

 

How would the PCM decide that to be the wrong direction?


If it misfires, the PCM will see unused oxygen and add fuel to use up the oxygen.


If it sees no oxygen, it will subtract fuel to get back to the ideal mixture.

 

I agree. Try moving the #4 injector to another hole and see if the misfire follows.
If it doesn't, then its time to do a compression test on that engine.


If the compression test checks out okay, I'd try and jury rig two noid lights in the cab, one to monitor the COP control wire and one to monitor the injector control wire. Watch the light when it misfires to determine if each circuit is being pulled to ground by the PCM and its corresponding circuit.

 

-------------


That doesn't sound very CLOSED LOOP-ish to me.


In my past life, before retiring, (writing machine language process control software), closed loop routines had to have good feedback algorithms built in - including hysteresis - or we would just wind up with a situation we referred to as "Tilt !".


Wouldn't the scenario you described above just ultimately result in all four cylinders on that bank misfire due to over-rich mixture and finally cylinder flooding - if the PCM wasn't able to determine from its feedback (via the O2) that it was going the wrong direction?

 

As far as I know, the PCM runs the fuel control at "tilt",
There is no degree of leanness or degree of richness.
It just knows the mixture is either RICH (i.e., no oxygen present) or the mixture is LEAN (oxygen present).


So for fuel control, hysteresis is ideal. The O2 sensor oscillates back and forth between rich and lean. The average then is the sweet spot.


I mean, when the O2 sensor senses oxygen, it doesn't know how much oxygen is present. It just calls it "lean" and adds more fuel.


When the O2 sensor senses NO oxygen, it doesn't know how much oxygen is absent. It just calls it "rich" and subtracts fuel.

 

@pdqford


Would you like to have a good laugh at my expense?
I was hell bent to show you that YOU were wrong! - but all I did was get egg all over my face.

I created a special dashboard screen for my Torque Pro App to display graphs for Upstream and Downstream O2 sensors, gauges for Short Term and Long Term fuel trims, and digital display for RPM, Mass Air flow grams/sec and Closed Loop ON/Off. (Then organized them with BANK 1 on right, Bank 2 on left).


I grabbed a screen capture at idle after the engine warmed up enough to go into closed loop and idle down to near normal around 600 RPM.


Then I reached over and "UnPlugged" COP # 1. (I planned to grab a few screen shots to SHOW YOUR *** a thing or two - like fuel trims will go real negative because of misfires!) LOL !


GUESS WHAT? That damn PCM is smarter than ME. As you can see from the attached sequence of screenshots... the PCM instantly detected the misfire all right! But it knew it was the result of 'primary circuit problem' with the COP - and immediately shut closed loop OFF and slammed all fuel trim readings and O2 reference readings to absolute ZERO! Probably shut off the fuel injector to cyl #1 also - but I have no way to verify that.


The First screen shot was taken as soon as Closed Loop came "ON". Short & Long term fuel trims normal - bobbing around + - .8 to .9. The Cat converts had not warmed up and started working yet - as indicated by the downstream O2s as screen bottom tracking the upstream O2 at top. (when warmed up, they settle down fairly level at about .6 volt to .7 volts.)


INTERESTINGLY: on Screenshot 2, Loop went OFF, O2 sensor reference voltage and Short Term Fuel trim all went to ZERO. (Notice Mass Air Flo went up a little bit trying to maintain idle RPM, but RPM went down from 611 to 598.


Screenshot 3 stayed flat as long as COP#1 was unplugged. After I plugged the COP back on, it took a few seconds (perhaps 10-15 seconds) and LOOP came true and voltage bias returned to the O2 Sensors as seen on Screenshot 4. In a few more seconds, the O2 signals resumed normal switching like Screenshot 1. The PCM handled the situation just about as perfect as can be imagined.


-------------------------------
Next - (Screenshot 5) I took starting fluid (Ether) and sprayed a generous dose into the intake air plenum.


THAT drove the O2 sensors on BOTH Banks NUTS - HIGH, and correspondingly the PCM shortened fuel trims on Both Banks immediately. That matches the "Oxygen Sensor Switching Voltage" chart from a magazine (Image 6) and sorta goes along with what we had both been saying in posts of this thread.


BUT I have NO idea how to simulate or prove my theory that random misfires will result negative fuel trims. ((( But I still think it will. Just can't prove it ))).





Startup after Closed Loop and Idle settled down.


Immediate when COP on Cyl #1 was unplugged


As long as COP #1 unpluged


When COP #1 pluged back in


Sprayed Starting Fluid (Ether) into air intake


O2 sensor switching voltage chart from Magazine

 

Nope.
But I just love those controlled experiments.
What a way to (help us all) learn.

 

Smarter than BOTH of us! Note sure what year the PCM became that smart. What year is your test mule?
You can verify the injector was turned off by back probing the ground side of the injector with a noid light or volt meter. When the PCM grounds the injector the voltage should drop from 12v to near 0 volts. So if the voltage stops pulsing, the PCM shut the injector down.


And Thanks for sharing.

 

Original Poster here. Sorry I've been so late to update. But I still have the misfire and rough idle. The rough idle seems to slowly becoming rougher, and when I got the misfire I still had the flashing CEL which disappeared after the missfire stopped, along with the Code P0304 come up. And now I have code "P0304 P"

EDIT to be a bit more descriptive I notice the miss fire and CEL at a higher RPM.

 

Sorry the annoying problem is still around. You have done all the normal / right things so far. Cause of intermittent misfires, even on a fixed cyl, can be hard to nail down. It's a shame to just throw parts at it, and the list of inexpensive tests is dwindling. But without detailed data from the computer it is hard to know what it knows that we don't.


I do not think it is a compression problem (because you report it occurs at higher RPMs). @Green96 may have the answer - (a bad/sticky/partially plugged injector). His had the same symptoms. And I've heard '05s are a little prone to injector failure. They are difficult to diagnose unless you have a scope as @pdqford suggests. But most of us don't have one. So we're sorta left with swapping them around or throwing a new one in. But - as I mentioned earlier the fact the misfire is on #4 does not by itself guarantee it would be the #4 injector. (If #1 - 2 or 3 was over active, the PCM will shorten injector pulse width (fuel trim) on all four injectors on bank 1. Number 4 might just be the first one that will misfire because of being leaned out under that scenario.) That's hard to isolate. ??you could remove plugs on that bank looking for the ones rich/lean - (black sooty - white chalky) ??.


You had asked earlier -- "Would I be able to check freeze frame data and fuel trim myself?". Yes, with a scanner having such capability, you can. I use Torque Pro on my android phone. It will do FAR more than the average do-it-yourselfer will ever need - all for about $25.00. It even reads / displays Mode $06 data (manufacturer specific diagnostics, although you have to do some research to interpret its results.) If a person is a little computerish inclined, there is practically NOTHING it can't do within its customizable parameters.


Mode 06 data could be informative in your situation. ie: misfire count, by cylinder, for the last 10 drive cycles. You might find out it is occurring on other cylinders - just not enough to trigger DTC's on them.

 

Thank for you the fast reply, and I'm gonna look into getting a code reader for my phone, as it's starting to turn into winter up here. And I don't feel like throwing parts at it. And who doesn't like buying tools?

I can see where you mean that maybe other fuel injectors are the problem and then they are compensating and thats why the misfire is only being detected on cylinder #4.