Help us help you. By posting the year, make, model and engine near the beginning of your help request, followed by the symptoms (no start, high idle, misfire etc.) Along with any prevalent Diagnostic Trouble Codes, aka DTCs, other forum members will be able to help you get to a solution more quickly and easily!

02 WRX, Front AFR sensor plugged in = misfires on all cylinders

More
4 years 10 months ago - 4 years 10 months ago #29299 by lottadro
Guafa, Haha oh no, not you too!

The idle smooths out, wideband shows rich at 13s or so, this may or may not be safe mode engaged by ECM due to the fact that although the sensor is manually ground, it is not seeing the ground signal coming in on D4 D5 from pin 1. Therefore, the computer thinks the sensor is doneso (pun), and safe modes the car; it acts exactly the same if I were to just simply unplug the sensor completely.

Also, a thing to remember, I already tried a spare ECM in there, and that did not fix it. Wires check low for resistance. Power supply is B+ at the power pin (2) and main relay, and SBF-5 fuse. The grounding pins D8 D9 which ground the front and rear sensors check good with low resistance and continuity through the harness all the way to their engine ground on top of the intake manifold.

Its hard to believe both ECMs would have the exact same issue. It is possible, probable, my OEM sensor is borked. It is also possible the Denso, even though they are the producers of the OEM component, is not meeting this specific car's specs and it needs OEM (this has been noted in prior posts on NASIOC). However, I have found nothing that really matches what I am seeing in my case.

Is it also possible, even though from what I can tell from the wiring diagram the o2 sensors are pretty much in their own circuits, that there is another component/ sensor in the engine bay that is causing "noise" within the system? I don't know if that is a good direction to take because it opens the field wide open again. Is the closed loop issue only a symptom of the actual problem?

I really appreciate the both of you taking time to help me with this. Thanks.
Last edit: 4 years 10 months ago by lottadro.

Please Log in or Create an account to join the conversation.

More
4 years 10 months ago #29303 by Tyler
I agree with your thoughts about the heater circuit. I'd like to get it sorted out so the car will allow other codes to set, as something like a heater code will often prevent other codes from popping up. Make sense?

The 13:1 reading in open loop means the car is capable of supplying fuel, which suggests a skewed input. MAF is always a big one, so that'd be my first check.

We know the reading will be all over the place with the rough idle, so how about the g/s reading at hot idle with no electrical loads on in open loop? Also a reading from the MAP sensor KOEO. Should match barometric pressure.

Please Log in or Create an account to join the conversation.

More
4 years 10 months ago #29304 by guafa
Ok guys, now i'm with you again. I had not noticed AFR when manually grounded. 13:1 AFR is telling us pcm is in open loop.

To Tyler suggestions, i would add a PCM ground test. I know you tested ground resistance, but it's better a voltage drop test. Check voltage between battery (-) and PCM ground when engine running.

Please Log in or Create an account to join the conversation.

More
4 years 10 months ago - 4 years 10 months ago #29309 by lottadro
Good morning, I checked the barometric pressure from a weather service source online for my area, which gave a reading of 29.84 inHg = 14.6 psi.

-I then proceeded to test, results are as follows-
g/s at idle engine warm and open loop, reading from Cobb Accessport (AP): steady 3 g/s and 1.2/1.3 MAF voltage.
MAP sensor KOEO, read from AP: 14.0 psi
Barometric pressure from ECM, read from AP KOEO: 13.8-13.9

*Here is where it gets a little strange* I then proceed to guafa's test. I ran a lead from the negative post of battery with alligator clip, I back probed ECM at D8 D9, I then connected multimeter positive lead to clip and negative to back probe. I got a reading of 10Vs, pretty steady. I was surprised by this, so I checked with my original (lacking 10 amp fuse), albiet cheaper meter, and it showed 0Vs.

My original meter was a Woods DMMW1.
My current meter is Innova 3320, which is "auto-ranging".

Also, it has started to pour here again, it seems as though I wont be able to test again today. I don't fully trust that reading of 10Vs. I am not completely certain my reading was accurate, I want to test again, and makes sure it doesn't say mV on the display. To be honest, I almost trust the manual reading from the Woods more so than the "auto-ranging". I am still not really familiar with new Innova, and I may have had it on the wrong setting.
Last edit: 4 years 10 months ago by lottadro. Reason: clarity

Please Log in or Create an account to join the conversation.

More
4 years 10 months ago #29312 by guafa
10 volts? Wow. Yes, you better check again. Anyway the old one is reading 0 volts (which is good. No more than 100 mv).

Just to be clear. Com lead of DVM at battery (-) post and positive lead at D8/D9. When engine running and D8/D9 connected to ground.

Never unplugged D8/D9 from ground. I have fried circuits just by unplugging one of the circuit grounds.

Please Log in or Create an account to join the conversation.

More
4 years 10 months ago - 4 years 10 months ago #29313 by lottadro
Ok, retested, 140-155mVs on both D8 D9. It cycles rather quickly.
Last edit: 4 years 10 months ago by lottadro.

Please Log in or Create an account to join the conversation.

More
4 years 10 months ago #29315 by guafa
Ok, with DVM when you see 140 - 155 mV, you don't see peaks (which could be up to 5 volts).

I can't tell you that is te root cause, but knowing how japanesse cars take care of electronic noises, i would decrease that voltage. That could be by re-doing all grounds.

I have decreased before also that voltage, by adding wires to the existing ones.

Please Log in or Create an account to join the conversation.

More
4 years 10 months ago #29321 by Tyler

guafa wrote: I have decreased before also that voltage, by adding wires to the existing ones.


Good thinking! But I believe he's already had someone over at NASIOC suggesting the same thing? lottadro did so and didn't see any improvement. lottadro, correct me if I'm wrong! :cheer:

I checked the barometric pressure from a weather service source online for my area, which gave a reading of 29.84 inHg = 14.6 psi.

...

MAP sensor KOEO, read from AP: 14.0 psi
Barometric pressure from ECM, read from AP KOEO: 13.8-13.9


Interesting. :huh: A bit further from each other than I'd like. My only problem with a skewed MAP theory is that this is a MAF engine, and I don't know how much weight the MAP carries in the fuel calculation. :silly:

g/s at idle engine warm and open loop, reading from Cobb Accessport (AP): steady 3 g/s and 1.2/1.3 MAF voltage.


Not the problem, unfortunately. :( Overreporting would cause a rich condition, not lean. Thank you for checking!

Please Log in or Create an account to join the conversation.

More
4 years 10 months ago - 4 years 10 months ago #29322 by lottadro
Indeed, I had added some straps and nothing had changed. Ill reattach them and redo the test to see if anything is different in the mV reading.

I am at a loss for what to check for next. Do you think I should bite the bullet and get an OEM sensor?

Tyler, attached are a slice of the datalogs I had sent to the tuner. I couldn't upload the whole excel file because of format restrictions on site. This is the car at idle. One picture you can see the title of the columns and the other a slice from farther down. You can see the hunting AFR and the fuel correction going wild.
Attachments:
Last edit: 4 years 10 months ago by lottadro.

Please Log in or Create an account to join the conversation.

More
4 years 10 months ago #29324 by guafa
If you decide redo the test, clear DTC's before. If you did it with DTC's cleared, i think it doesn't make any sense.

Another question. Have you checked heater operation once you clear DTC's and before ecm sets it?

Please Log in or Create an account to join the conversation.

More
4 years 10 months ago #29327 by Tyler

lottadro wrote: Tyler, attached are a slice of the datalogs I had sent to the tuner. I couldn't upload the whole excel file because of format restrictions on site. This is the car at idle. One picture you can see the title of the columns and the other a slice from farther down. You can see the hunting AFR and the fuel correction going wild.


I didn't expect to see that. :silly: :blink: If I'm reading it right, the trims are going negative at the same time the AFR is showing lean. Why would the PCM see a lean mixture, then take more fuel away?

Would you mind sending me the whole file? wrenchturnsyou@gmail.com Does my interpretation of the data make sense to anyone else?

I know you disconnected the downstream O2 sensor and installed an AFR, but does the scan data still show a PID for the downstream O2 sensor signal?

Please Log in or Create an account to join the conversation.

More
4 years 10 months ago #29332 by Tyler
Got your e-mail, thank you. :cheer:

I see the fuel trims and AFR is actually swinging both rich AND lean, but I'm guessing it runs the worst when the trims are negative. Also definitely seems to be happier off idle. Much happier off idle. :huh:

You can also 'see' the AFR cycling back and forth in the data, just like in your video. Like the PCM can't get into fuel control. AFR systems don't typically chase their tails in this way.

You've probably done this already, but any intake leaks? Intercooler hoses, PCV hoses, anything behind the MAF. I'll be thrilled if a Suby AFR fixes your issue, but I'm not yet convinced it will. :silly: No insult meant towards IATuning or NASIOC.

Please Log in or Create an account to join the conversation.

More
4 years 10 months ago #29339 by lottadro
Tyler, yes, I have pressure tested and smoke tested. No leaks. I have changed plugs, coil packs, maf, map, IACV, crank, cam, timing belt, and ecu. If new sensor doesnt work, I am going to look into the injectors, I have a spare set here. Nothing seems to change or fix this issue, hence why I started looking into wiring and electrical. I had noticed the initial start up it idled fine, and light driving was fine, and unplugging the AFR sensor, forcing open loop also cleared the symptoms. So, like I said I started going through the different diagnostics of the front AFR sensor and that is when I stumbled upon the D4 D5 pin discrepancy.

The theory right now, is that the OEM and Denso sensors, even though the are both produced by Denso, are of different calibrations. It should arrive by Wednesday, so I'll update once that comes in and is installed. If not, I still think there is something strange with D4 D5.

What are your thoughts?

Please Log in or Create an account to join the conversation.

More
4 years 10 months ago #29349 by chief eaglebear
hi guys there are some TSBs having to do with the poo30 not sure if that will help interesting problem probably connector

Please Log in or Create an account to join the conversation.

More
4 years 10 months ago #29354 by guafa
Guys, i am stuck in the point if all this crazyness is happening in open or closed loop.

For me that changes everything. Open loop drives me to one direction and closed loop to another.

Ok. I can wait the new sensor and good news.

Please Log in or Create an account to join the conversation.

More
4 years 10 months ago - 4 years 10 months ago #29364 by Tyler

guafa wrote: Guys, i am stuck in the point if all this crazyness is happening in open or closed loop.


Underlined for emphasis. :lol: It IS craziness!

From looking at the data file, I can see this issue happens in closed loop only. Open loop makes the engine happy. When the issue is happening at idle, the AFR says lean, but the trims are negative. This is an inverse reaction, and suggests the PCM isn't able to control fuel correctly.

Why? Is the AFR not measuring it correctly? Based on the first video lottadro posted, I'm saying no. His added downstream AFR agreed with the stock upstream AFR, more or less.

The only way this makes sense (to me, anyway) is a false rich condition. PCM sees rich, takes fuel away. Only, there is no rich condition, and now the engine doesn't have enough fuel to run correctly. :angry:

This is my understanding of the situation, anyway. :silly: I welcome any criticism you or anyone else may have.
Last edit: 4 years 10 months ago by Tyler.
The following user(s) said Thank You: lottadro

Please Log in or Create an account to join the conversation.

More
4 years 10 months ago #29371 by lottadro

Tyler wrote: [When the issue is happening at idle, the AFR says lean, but the trims are negative. This is an inverse reaction, and suggests the PCM isn't able to control fuel correctly.

Why? Is the AFR not measuring it correctly? Based on the first video lottadro posted, I'm saying no. His added downstream AFR agreed with the stock upstream AFR, more or less.

The only way this makes sense (to me, anyway) is a false rich condition. PCM sees rich, takes fuel away. Only, there is no rich condition, and now the engine doesn't have enough fuel to run correctly. :angry:


Leaky injector? I honestly have no idea, I feel way over my head on this. The trim starts a negative trend almost immediately and doesnt really recover, if an injector was leaking fuel would that be enough for the sensor to be constantly needing to reduce the trim but not being able to? Thus creating a lean condition within its AFR but seeing the fuel and reducing trim? Does that make sense?

Please Log in or Create an account to join the conversation.

More
4 years 10 months ago #29407 by lottadro
Ok, new OEM AFR sensor did not work.

Please Log in or Create an account to join the conversation.

More
4 years 10 months ago - 4 years 10 months ago #29409 by guafa
It's a shame to hear that. Let's connect some points with information that we already know.

I am just trying to think out of the box.

Can you please post datalog tables in the following conditions?
1. Everything connected, DTC's cleared, engine running, closed loop, heater ground pin activity (voltage) and ecu activity.
2. DTC's cleared, engine running, open loop (upstream o2 sensor unplugged) and ecu activity.
3. DTC's cleared, engine running, open loop (heater manually grounded) and ecu activity.

In order to not generate DTC's with heater manually grounded, use two o2 sensor at the same time like this:

One installed in pipe (reading exhaust gases), but heated manually. The other one in the air, but being heated by PCM circuit. Is it clear?
Last edit: 4 years 10 months ago by guafa.
The following user(s) said Thank You: lottadro

Please Log in or Create an account to join the conversation.

More
4 years 10 months ago #29420 by lottadro
FREE AT LAST FREE AT LAST, THANK GOD ALMIGHTY!

Replaced injectors 1 & 3, boom. End of the epic. I think my hypothesis was correct, one of those had been leaking, my guess is 3 as when looking at the plugs today it was black. The idle is steady, the trims look good and AFR is back to normal.

I want to thank you Guafa and Tyler for sticking it out with me. Everyday I would look forward to your replies and I honestly learned a lot through all of this process. I feel like I just won an Oscar or something. If you guys hadnt stuck with me, I'm not sure I would have even figured it out, it was with your contributions that we slowly worked our way to today.

Thank you.

P.S. I'll update in a few days time once finished with tuning and let you know how everything goes.
The following user(s) said Thank You: tmcquinn

Please Log in or Create an account to join the conversation.

Time to create page: 0.276 seconds