Does it have to learn to idle?

[TJBaker57]

How exactly do I duplicate the "eavesdropping" on my vehicle?

Are there more steps then shown in this screenshot?

To do what I did there requires more hardware and software.

Plus the knowledge of knowing what you are looking at! It is ALL hexadecimal !!

I had Car Scanner running as usual but I have a splitter on my OBD port. Like we used to do with cable tv coax!!

So I had 2 Bluetooth OBD2 adapters connected at the splitter, one device paired to Car Scanner and a second device paired to a second Android device.

On the second Android device I run a free Android app that is a basic terminal interface. After the initial configuration I set up the second OBD2 adapter to simply send whatever messages are on the data network to the screen.

Looks like this.... Car Scanner requesting Freeze Frame 0 at 18:07:12.392 and a response from the PCM at 18:07:12.434.



I have a thread somewhere here about this sort of thing.

 

[TJBaker57]

This is after five 10 minute morning warmups (with defrost on) at idle and 11 short trips at no more then 40 mph with the purge line plugged at the TB and the AC relay removed

There will likely be little to no significant improvement in the negative fuel trims with compressor OFF at idle (cells 19 and 21) unless the engine is allowed to run at idle with the compressor OFF for like 20 minutes. Only THEN can we see if the PCM adjusts those trim cells indicating the purge valve is a contributing factor.

The fuel trims were not negative with the defrost ON.

Idling with the defrost ON or the snowflake button ON somewhat defeats the purpose of testing with the purge valve disconnected, as far as seeing if rhe purge valve is contributing to the fuel trims going so negative under idle/no AC conditions.

 

[AmpOverload]

So now I just eavesdropped on Car Scanner while it requested freeze frame data and caught their error. I will report this as a bug to the developer.

Excellent catch, nice work! I don't even use Car Scanner, but it's good to see that your efforts should make it more reliable for all those users!

Hit the next icon to try for frames 1, 2, 3 etc. and they just come up blank. So the bug only affects the frame 0 for some reason.

If the GM trucks are anything like the Buicks I frequently test, it's probably because freeze frames other than "00" always come back as "NO DATA". (For those who might not be aware and are following along, "NO DATA" is what the scantool will reply with, after a configurable waiting period, if the vehicle doesn't reply at all to a command from the scantool.)

For example, from an old capture of mine on a 2005 Buick LeSabre, querying freeze frames 0, 1, 2, and 3:

>020200
48 6B 10 42 02 00 00 00 89
>020201
NO DATA
>020202
NO DATA
>020203
NO DATA

 

[Sqrly]

Today I pulled the TB and gave it a good inspection. The gears move and sound smooth when I manually move the butterfly. No noise when I shake it. The shaft doesn't have play in it.

It passed all the electrical tests. I followed this guide for that.
Electronic Throttle Body Tests (2002-2007 4.2L Chevrolet TrailBlazer)
The TAC motor has a resistance of 15 ohms

I replaced the fuel pressure regulator vacuum line and the 5/16" vacuum plug on the front inside of the intake just because the rubber looked old and felt hard.

I shorted the battery cables to reset things.

For curiosity I got the fuel trims after I hooked the battery back up but before I started it. Even without a translation those look reset to me.



Then I got them again after I let it idle for 3 minutes and shut it off. (It used cell 21 the entire time)

Note: It never seems to get all the way to 190 degress when just idling when it's cold out but I checked the hoses (when it was at 160), the top is hot the lower cold so the thermostat is working.



After turning it off for about 5 minutes, I started it and let it idle for 3+ minutes again, then turned it off again for about 10 minutes.

Then I took it for a drive which I logged. 2024-12-08 15-14-02.brc

There's several 0-65ish regular accell and varied cruising. One near full throttle pass and another full throttle accell from a stop. The vehicle does not lack power.

It seems to be logging a lot more sensors then what I selected in the live data setup. I only selected Engine RPM, MAP pressure, LTFT, STFT, Oxygen Sensor Voltage, Throttle Position, and Vehicle Speed.

Then I got the trims again after the drive.



The HVAC was on floor only with the fan on off the entire time. The AC relay has been reinstalled.

The CAT and EVAP systems have not passed emission testing yet. I assume that's normal.

Tomorrow morning I will run a log while I warm it up for 10 minutes with the defrost working normally. I usually shut it off for 15 minutes after warming it up. I'll log the trip to school with the defrost still on too. Then I'll get the trims again after that.

 

[TJBaker57]

Then I got them again after I let it idle for 3 minutes and shut it off. (It used cell 21 the entire time)

As I have said, and you have now witnessed, 3 minutes is insufficient time for any updating of the stored fuel trim values.

It took 7 or 8 minutes at idle with A/C OFF for mine to even begin to adjust the stored fuel trim when idling the entire time with the A/C OFF.

It took another ten minutes afterwards to complete the adjustments.

So that amounts to about 18 minutes idling with A/C AND defrost OFF the entire time.

Even without a translation those look reset to me.

Yes. Values of "80" equate to a fuel trim of zero.


In your last set of values short term cells 0, 2, 5, 6, 7, 10, 14, 16, and 22 are all at -0.78125%.

Long term cells are naturally essentially the same as short term (since short terms are mostly copied from long terms) excepting long term cell #20 where that one is also at -0.78125%.

 

[Sqrly]

As I have said, and you have now witnessed, 3 minutes is insufficient time for any updating of the stored fuel trim values.

It took 7 or 8 minutes at idle with A/C OFF for mine to even begin to adjust the stored fuel trim when idling the entire time with the A/C OFF.

It took another ten minutes afterwards to complete the adjustments.

So that amounts to about 18 minutes idling with A/C AND defrost OFF the entire time.






Yes. Values of "80" equate to a fuel trim of zero.


In your last set of values short term cells 0, 2, 5, 6, 7, 10, 14, 16, and 22 are all at -0.78125%.

Long term cells are naturally essentially the same as short term (since short terms are mostly copied from long terms) excepting long term cell #20 where that one is also at -0.78125%.

What are 18,19,20,21 at?

 

[TJBaker57]

What are 18,19,20,21 at?

Unchanged, all at 0.0 trim.

And they most likely will stay there until you let it idle for no less than say 20 minutes or so.

 

[Sqrly]

Unchanged, all at 0.0 trim.

And they most likely will stay there until you let it idle for no less than say 20 minutes or so.

@TJBaker57

If this was your vehicle with this history, how would you do a "relearn" procedure after resetting everything by shorting the battery cables?

I've been starting it and letting it idle for about 5 minutes, shut it off 5 minutes, start it again and let it idle it another 5 minutes, HVAC off and never touching the throttle. Then I drive it for about 15 minutes with speeds up 65, still HVAC off. Then after letting it completely cool down I start it and let it idle for 10 minutes with the HVAC on defrost then drive it 10 minutes with the HVAC on defrost.

FYI: This morning, it idles and drives great, very smooth. How long it will stay this way .....


I ran a log while warming it up this morning then got the trim cells after shutting it off.

Then I ran another log on the morning trip to drop a kid off at school and got the trim cells again after that.

Glancing at the screen while driving the trims do seem to be moving into the negative.

 

[budwich]

hopefully, "forever"... note that you did change a "few parts" dealing with vacuum.

 

[Sqrly]

I don't understand the "A/C Hi Pres" sensor data. In the "Morning Warm up" log from post #168 it's going up and down between 55 psi - 85 psi on average. Shouldn't the pressure be much higher?

 

[Sqrly]

hopefully, "forever"... note that you did change a "few parts" dealing with vacuum.

I just changed them because I was there. Previously, I sprayed ether on those locations while the vehicle was idling and there was no change in rpm.

 

[Sqrly]

Once upon a time when I was young my Dad had a Sun Tune Up Tester in the garage. It was a big thing on a cart. I could basically watch my plugs firing on the scope and see if the sparks were equal.

Is there a way to do this with these OBD2 scanner apps?

I got this vehicle used and I can see at least two of the coils have been replaced, each with a different aftermarket brand.

I was getting a lot of misfires in the live data at first but last I looked they had gone away after a few tanks of the Chevron w/Techron gas I use.

 

[budwich]

I don't understand the "A/C Hi Pres" sensor data. In the "Morning Warm up" log from post #168 it's going up and down between 55 psi - 85 psi on average. Shouldn't the pressure be much higher?

I believe that's because your ambient temp (outside) is low so the system doesn't "krank up" the compression. There is not a lot of cooling happening at low temperatures.

 

[TJBaker57]

I don't understand the "A/C Hi Pres" sensor data. In the "Morning Warm up" log from post #168 it's going up and down between 55 psi - 85 psi on average. Shouldn't the pressure be much higher?

The A/C compressor cycles off and on in response to the LOW pressure. That switch we generally call the low pressure switch is frequently called the cycle switch in some service literature.

The high side pressure is largely a function of the ambient (outside) temperature and the amount of air passing through the condenser.

 

[TJBaker57]

Then I ran another log on the morning trip to drop a kid off at school and got the trim cells again after that.

I notice in this log the field for "Fuel Trim Cell" seems blank.

Does this not usually display a value from 0 to 22 ??

I also see that while you are idling, waiting for your passenger to depart, your short term fuel trim goes a good deal negative. Enough so that the long term trim begins to adjust downward as well.

 

[TJBaker57]

Once upon a time when I was young my Dad had a Sun Tune Up Tester in the garage. It was a big thing on a cart. I could basically watch my plugs firing on the scope and see if the sparks were equal.

In 1974 &75 I was one of those high school motor heads, spent a great deal of time at two high schools auto shops where we had the entire gamut of Sun Diagnostic equipment. It was a very well equipt shop. We even had a machine to grind your new brake shoes to match the measured diameter or your newly turned brake drums!!

 

[TJBaker57]

And I forgot....These 2 are essentially the same data. You could drop the "STFT" off the logging.

 

[Sqrly]

And I forgot....These 2 are essentially the same data. You could drop the "STFT" off the logging.

View attachment 115336

I did not check the box "Oxygen sensor 1 Bank 1 Short term fuel trim" in the live data settings before starting the log.

I mentioned earlier that when I view a log after the fact, there seems to be many sensors I did not choose to log.

For example, in this capture the items with a red line were not selected for logging.

Does this indicate an issue with the app?

 

[Sqrly]

So, 2 days after a reset I am once again seeing a huge drop in rpm's when the AC compressor turns off when idling in park. It's not stalling yet but I'm certain in the next few days it will.

I got these values this morning after the usual warm up and a trip to the school.

 

[TJBaker57]

mentioned earlier that when I view a log after the fact, there seems to be many sensors I did not choose to log.

I too have seen some sensor data appear in logfiles, though not any O2 sensor data.

I only get things like distance travelled, fuel consumption items, that sort of thing.

Oddly, even though I know I have seen these things I cannot find any examples right now !

I generally begin by pressing the deselect all button when beginning the sensor selection process.

Does this indicate an issue with the app?

I just had a look to see if there was some checkbox somewhere to disable this but found none. I may suggest to the developer that they add such a checkbox.

 

[TJBaker57]

So I have looked over your latest trim cell values. And I have compared them against other values you have posted. I have a question.

All the posted values are in the morning, either before or after the school trip. Are there other times you drive the vehicle?

There is evidence of a rich condition not only when idling but as you drive as well.

Between yesterday after the AM school trip and today at the same time there has been a very large shift to the negative for fuel trims, not only at idle but during your drive.

Was there any driving between those two times??

I am posting two screenshots here. The first is from yesterday morning after the school trip.

Column K is the raw saved cell value as decimal, column M is the same value expressed as a fuel trim percentage, and column N is the change in fuel trim cells between that time and the most recent posted values prior.

Rows 1 thru 23 are the short term values.

Rows 24 thru 44 are the long term values.

Cell numbers are contained within the text of column I.

The biggest change was in the short term/long term pair for cell 5. This is a driving down the road cell, not an idling cell.

Cell 16 also shows a negative trend, having dropped 3.13% since the previous values were posted.

Cell 16 use I generally see right after I have coasted for a while and am just easing the throttle back open. Again, NOT an idle cell.

Other than cell 5 and cell 16 everything else looks pretty reasonable.





Now we come to this mornings' posted values after the school trip.

Something significant has happened.

Column N here displays the change between these values and the values posted in the first screenshot. So just since yesterday AM after the school trip and this morning after the school trip.

There are negative changes in many cells.

I am thinking these values were collected while the engine was still running after the school trip??Before the ignition switch was turned back from RUN to ACC??

 

[TJBaker57]

@Sqrly

In your dieseling post you mentioned getting a P0302 code if you let it idle more than 5 minutes or so. Does this still happen? If yes, that sort of information very likely applies here!!

 

[TJBaker57]

@Sqrly

Well crap!! I totally spaced on your having mentioned a lot of misfires on cylinders 2 and 3 back in late October.

Misfires will absolutely make the O2 sensor see a rich condition !!

Were these misfires addressed? How many miles on the slark plugs?

 

[Sqrly]

Between yesterday after the AM school trip and today at the same time there has been a very large shift to the negative for fuel trims, not only at idle but during your drive.

Was there any driving between those two times??

Yes, the same short tips to pick up the kids from school. On Mondays and Tuesdays there are 2 kids to transport to different schools at different times, so 2 separate trips in the morning, and 2 more in the afternoon.

In your dieseling post you mentioned getting a P0302 code if you let it idle more than 5 minutes or so. Does this still happen? If yes, that sort of information very likely applies here!!

"Sounds" like a diesel. The valve train is very noisy. I'm used to mechanical lifters and know what they sound like when happy, but iirc, hydraulic lifters shouldn't be making any noise.

The constant P302 codes where when I first got the vehicle. Also, it appears 4 of my coils are original and 2 are aftermarket, 2 different brands. I did previously try swapping coils 2 and 3 and still got the P302.

Note: I was chasing the parasitic battery drain at the same too so the battery had been disconnected many times. Since then I have replaced the plugs, all the fluids and filters, and most likely been using better gas then the previous owner did.

The P302 went away after the new plugs and a few tanks of Chevron w/Techron and didn't show up over the summer even with 30 minutes of idling with the AC running 5 days a week.This stalling did not start until winter did.

Actually, I just got a P302 today after idling for a long time in front of the school while looking at some other things in the scanner app. First one in about 6 months.

No I haven't checked the oil pressure with a real gauge, yet and I've yet to pull and inspect the VVT solenoid too.
~~~~~~~~~~~~~~~

My O2S2 looks like it's working normally, that is it fluctuates slightly but, it is averaging 0.8v.

I also noticed this when tinkering with the app.

 

[TJBaker57]

Yes, the same short tips to pick up the kids from school. On mondays and tuesdays there are 2 kids to transpost to different schools at different times, so 2 seperate trips in the morning, and 2 more in the afternoon.

So if I understand it correctly there were like 2 or 3 trips between the two most recent sets of fuel trim cell values??

The constant P302 codes where when I first got the vehicle. Also, it appears 4 of my coils are original and 2 are aftermarket, 2 different brands. I did previously try swapping coils 2 and 3 and still got the P302

Same here. I have 4 original coils and 2 I have replaced. Personally I prefer original coils over todays replacement parts. I would wager my original coils are actually better than what is sold today.

Since then I have replaced the plugs,

This is what I wanted to know.

just got a P302 today after idling for a long time in front of the school

So there is still a little something there. Now the question is,

" Is a misfire causing the trouble?"

or

"Is the lean fuel mixture causing the misfire?"

My O2S2 looks like it's working normally but itis averaging 0.8v.

The #2 O2 sensor is downstream of the catalytic converter and when everything is working as it should that voltage should hold fairly steady in the range you are seeing it.

At first startup it will oscillate like the number 1 O2 sensor upstream of the converter. But within a fairly short time frame the number 2 sensor should stay high.

There may be times the number 2 sensor will change its behavior. Like under heavy acceleration or maybe even under prolonged idling.

About the ECU identifier stuff I am uncertain what you are specifically pointing out. The protocol and baud rate are correct. And the jibberish characters seen near the end is something I have seen on several different 4.2 P10 PCMs.

And the displayed VIN is a little weird as it is missing 4 characters.


I see similar anomolies here and I can see the actual data transmitted is free of such jibberish.

 

[Sqrly]

So if I understand it correctly there were like 2 or 3 trips between the two most recent sets of fuel trim cell values??

Correct. All only a few miles. The only warm up is before the first trip in the morning.

About the ECU identifier stuff

The ECU protocol, SEA J1850 VPW

Would his not indicate I should select the "GMC OBD-II + SAE J1850 (Duramax VPW)" connection profile in the app? The one that already has many of the sensors one would have to manully add to the basic OBD-II connection profile.

The jibberish is just character codes it does not recognize, I've seen this in windows.

I didn't notice the VIN, that's odd.

At first startup it will oscillate like the number 1 O2 sensor upstream of the converter. But within a fairly short time frame the number 2 sensor should stay high.

I thought the target was 0.45v?

 

[TJBaker57]

I thought the target was 0.45v?

Higher is better I believe.

The jibberish is just character codes it does not recognize, I've seen this in windows.

Well the second screenshot, the one in black and green is the raw data. And the VIN is clearly displayed in full as ASCII characters, no "jibberish" there. Very easily converted to the alpha_numeric characters.

And the Calibration numbers are also there in ASCII as well. Clear as day.

So the garbled stuff is happening in the app.

Would his not indicate I should select the "GMC OBD-II + SAE J1850 (Duramax VPW)" connection profile in the app? The one that already has many of the sensors one would have to manully add to the basic OBD-II connection profile.

We have covered this before but if you want to why not check it out for yourself?? You can make multiple "vehicles" in the app and see what works for you best.

I have all manner of "vehicles" there for my testing of various PCMs, TCCMs and so on.

 

[AmpOverload]

Jumping in again because of some discussion of things that I'm intimately familiar with, but hoping to not distract from the main thrust here....

It looks like the author of Car Scanner isn't parsing the vehicle's replies containing the Calibration Verification Numbers (CVNs) at all -- they're shown as a raw, hexadecimal-form dump of all 6 parts of the reply, including the checksum bytes!

Even worse, though, is the badly broken parsing of the Calibration IDs. There, it's clear that, much like the problem @TJBaker57 found with the freeze-frame DTC seemingly mis-using the message checksum as part of the DTC, the checksum values from these Calibration ID vehicle replies are appearing (sometimes as a human-readable character, sometimes jibberish) mixed in with the other mis-parsed data -- a real visual mess!

The VIN error is probably just another parsing bug.

The Car Scanner author clearly has some work to do on the parsing routines!

@TJBaker57: BTW, I parsed vehicle replies with CVNs and Calibration IDs for a long time before realizing, while testing on a 2005 Chevy Avalanche, that vehicles existed with multiple CVNs and Calibration IDs per node. Even now, I know of at least one app (OBDwiz) that cannot handle more than 7 CVNs/CalIDs per node. But the 2005 Chevy Avalanche has 8 of them reported by the PCM node!

I did not check the box "Oxygen sensor 1 Bank 1 Short term fuel trim" in the live data settings before starting the log.

I mentioned earlier that when I view a log after the fact, there seems to be many sensors I did not choose to log.

For example, in this capture the items with a red line were not selected for logging.

Does this indicate an issue with the app?

It's a bit hard to explain, and I admit that this is speculation on my part, but I don't think there's anything wrong with the app.

Even though I'm unable to use your Car Scanner recording files to view the data, it looks to me like the app is offering these extra data points for one of 2 (or 3) reasons:

1. it's a "composite" or "synthesis" of other data that is being recorded
2. it's another component of a PID that is being recorded

An example of the 1st case would be "synthesizing" a "fuel efficiency" (miles per gallon) data point (a pseudo-"PID") from the vehicle speed sensor (VSS) and the Mass Air Flow (MAF).

An example of the 2nd case would be where you request recording of O2 sensor voltage on Bank 1, Sensor 1. Well, included in the vehicle's reply will be the STFT associated with that sensor. The extra component "comes along for the ride" -- it's essentially "free" (no extra speed penalty for recording it). So there's little reason for the app not to allow the user to view those unrequested components when reviewing the recording.

A 3rd possibility is when apps offer an "acceleration" "PID" which is really just reading the internal accelerometer in a phone/tablet. I think Torque might do that, but I cannot recall because I don't use it.

 

[budwich]

Actually, I just got a P302 today after idling for a long time in front of the school while looking at some other things in the scanner app. First one in about 6 months.

You need to look at the pids for misfires as the code is somewhat "too late". The coding basically indicates a "servere problem"... BUT there can be on going misfire counts that do not cause coding OR flash the check engine light. Hence, you need to look at both current and history counts for each cylinder to ensure that you are not operating with that type of issue which may be causing an overall richness... maybe.

 

[AmpOverload]

Hence, you need to look at both current and history counts for each cylinder [...]

Looking for misfires is definitely wise, especially when you've seen recent, related DTCs. But be sure to record the data. Don't just try to monitor it "in realtime". Otherwise, misfire events are quite easy to miss. Please see my graph in this post, which demonstrates how "ephemeral" the live misfire data really is, even for the "historic misfires" PID!

Ideally, you'd be using what's known as OBD2 "Mode $06" ("Request On-Board Monitoring Test Results for Specific Monitored Systems") to view aggregated misfire data. That would preclude the need to be vigilantly monitoring and recording the "current misfires" and "historic misfires" PIDs in Car Scanner. Unfortunately, GM vehicles of that era (pre-CAN, generally pre-2008) are somewhat pathetic in their Mode $06 capabilities, especially compared to Ford vehicles of the same era.

 

[Sqrly]

What happens at 160f?

More specifically, what changes occur as the temperature rises?

This morning while warming it up with the defrost going I could not 'hear' any change in the idle. If I watched the tach on the IPC I could see it fluctuate an expected amount when the compressor turned on and off though.

Then just before I went to turn it off, I saw a, *larger then expected, drop in rpm. I glanced at the temp gauge on the dash it was at 160. (It reads about 10° over what the app says the coolant temp is).

Anyways, I'm not seeing rpm drops until a certain temperature and I'm just curious what things are happening at that temperature.

Note: On the morning trip I saw the temp guage get up past 210 then drop to just under 210. So it looks like the thermostat is opening at the right temp. It's cold and humid here right now. It's normal for our vehicles to not reach full temp just sitting and idling for 10 minutes this time of year.

*It hasn't actually stalled since the last reset but the rpm drops always start small and grow each time the compressor turns off until finally reaching zero and only after the coolant temp is high enough.

@AmpOverload @budwich

We have a 4-H thing this afternoon, it's about 5 miles from here. I'll record a log of misfires and histories and see what I get.

I believe a P30X is set when the misfire count exceeds a certian value over a given time.

Since the P302 is not persistant, I've not decided if it is a cause or a symptom. Since the engine appears to be running rich, the injectors are a suspect. However, the fuel system passed a leak down test. The pressure held at 50 psi for 10 minutes after shutting the vehicle off.

I'm considering buying an injector O-Ring set and pulling the injectors for a visual inspection. Other then needing to use a step stool to reach it, this engine is actually quite easy to work on.

I'm also going to start paying closer attention to fuel levels. I haven't looked at whether things are different when the tank is full vs near empty.

 

[Sqrly]

Even though I'm unable to use your Car Scanner recording files to view the data, it looks to me like the app is offering these extra data points for one of 2 (or 3) reasons:

1. it's a "composite" or "synthesis" of other data that is being recorded
2. it's another component of a PID that is being recorded

I had the same thought. Maybe sensor A data is the result of a calculation using data from sensors B and C.

The odd thing is it doesn't always do it. Sometimes the only things I can view in a log are the things I chose before recording it.

I can share the logs in a different format but I don't have a template to plug it into.
2024-11-20 12-21-41.csv

 

[AmpOverload]

I had the same thought. Maybe sensor A data is the result of a calculation using data from sensors B and C.

The odd thing is it doesn't always do it. Sometimes the only things I can view in a log are the things I chose before recording it.

It's hard for me to speak with any certainty given that I can't experience it 1st-hand. However, many PIDs (meaning, essentially, a single reply from the vehicle) consist of a single component (e.g. Engine Coolant Temperature). So if you've selected nothing but 1-component PIDs for recording, then (presumably) all you'd see is each of those components, unless (again, presumably) you've just happened to select certain PIDs that can be "synthesized" into a "pseudo-PID".

I can share the logs in a different format but I don't have a template to plug it into.
2024-11-20 12-21-41.csv

The CSV-format data is certainly more useful to someone like me than the native-format (*.brc) Car Scanner recording but, unfortunately, it's of no real help in knowing if our theory about synthesized PIDs is correct.

 

[Sqrly]

From everything I've read so far a downstream O2 sensor voltage of 0.45 would be perfect and anything above a steady .65v is too rich. (lack of O2)

I just ran a short log while warming it up with the HVAC off. You can see that when the downstream O2 sensor warms up it screams rich and the STFT goes lean.

The misfire history was 0 for all cylinders. (which apparently means squat, I have to log actual misfires, see post 197)

PS I don't have an IR Thermometer, yet! I'm going to pick one up and check the temps at the in and out sides of the cat.

 

[Mooseman]

How's power at WOT? Does it seem to bog down or has no guts? If that's the case, maybe your cat is clogged and throwing your upstream O2 readings off. If the cat seems fine, read on.

I'd be throwing an upstream O2 sensor at it at this point. At its age and mileage, it can't hurt. (yeah, yeah, I know, parts shotgun...). The O2 readings once it lights off seems to support a bad upstream O2 sensor.

 

[Sqrly]

How's power at WOT? Does it seem to bog down or has no guts? If that's the case, maybe your cat is clogged and throwing your upstream O2 readings off. If the cat seems fine, read on.

I'd be throwing an upstream O2 sensor at it at this point. At its age and mileage, it can't hurt. (yeah, yeah, I know, parts shotgun...). The O2 readings once it lights off seems to support a bad upstream O2 sensor.

It has plenty of power.

I got a thermometer and just checked the cat temps when I got home which included a few miles on the freeway

Upstream weld ring 470, downstream weld ring 670. So it's working, probably more then it should.

I am seeing missing on #2 again but only when idling. It's not enough to set a code unless I let it idle for a long time.

I recently watched a vid where people spent over $4,000 over 3 years because "At its age and mileage, it can't hurt." and it turned out all it needed was a valve adjustment.

 

[Sqrly]

It's apparently misfiring a lot more then I was aware of because it's almost never enough to set a code.

Note the scales are 0-1, 0-3 except #2 which is 0-100

I edited the screenshots to reorder the graphs and stretch them horizontally 200%

 

[Sqrly]

@Mooseman

Check out this video. No offense intended! I watched both parts and it made me laugh more then once.

The ironic thing is I could have nailed this one in minutes just by listening to it run as I have a lot of experience with Hondas.

 

[Mooseman]

I'm looking at it differently. We've eliminated most things except maybe a bad fuel injector, which can only be verified with an injector balance test. We saw that it goes into negative fuel trims once the O2 sensor lights off and goes into closed loop. This might be inducing an actual lean condition and misfires from it because of possible erroneous sensor data for a false rich condition.

How about this for a test. Reconnect and return everything back to normal, reset codes and then disconnect the upstream O2 sensor. See if the stalling and misfires go away.

 

[budwich]

I would address the most "heavy" count first to see if you clean that and it "helps" the overall operation there after.
I have read in "generic circles" that some systems also use the downstream o2 to "help" things when the system gets in "trouble" (ie. is having problems getting things to respond "normally" to adjustments). Without knowing the full internal system design semantics, it should not be ruled out as a possibility.