Lean/Rich issue

Background: Gen 3 5.3L LM7 out of 2005 silverado in a 2007 Jeep JK Wrangler. I didn't do the swap so I may not have some details on the components. Conversion is using a BP Automotive harness.

This thing has been doing this for some time. I thought I found the issues. NOTE: That this is my first attempt at tuning, but I thought I had it in decent shape yesterday.

Issue 1 - the MAP sensor was wrong and only reading in the 80's key on, engine off - FIXED
Issue 2 - The swap was using an LS7 MAF with integrated IAT and the stock LS1 MAF tune - (Tuned it for the LS7 MAF and used the IAT settings off the forums)
Also, ignore the speed in the Attachment 145475tune I had not set up the gearing yet.

I tuned this over the past weekend and drove it for two days and everything was running great. The VE table needed some more work, but I just did a couple passes on it this weekend.

Drove it to work this morning and the O2 sensors and fuel trims are acting crazy. Bank 1 is showing crazy lean and LTFT's on that bank maxed out. Bank 2 is still fluctuating but running rich. Log file attached (Drive to work).

Later that day I go to lunch, everything is fine again but pulling a lot of fuel out (-12) in the 800 rpm range (log LUNCH 1). So I took some out there and did another test drive. All the LTFT's came in -5 to -1. (Lunch 2 logfile)

On the drive home, bank 1 goes crazy lean again. (Logfile Drive Home) I managed to get it home and changed out BOTH O2 sensors. Also when I did that I unplugged the passenger side one and ensured that the pass side went to 450 mv to rule out the wiring being backwards. The driver's side was fouled out, so the plugs probably are too.

It didn't seem to help anyway though, see logfile (both o2 swapped). I checked the fuel pressure (this has the return line system) and it is reading 48 and when I pull the vac line off of it it jumps to 54 psi. (That seems low so I'll see if I can find out what is going on there. Going to test it again in the morning)

Also note that I hooked my Wideband up when I got home (stuck it in the tailpipe).

Note that I did drive it on the interstate to work and back and none of my tuning was on the interstate. So maybe that is it, but this problem has been going on for months... I really need some help.

Sounds like a cylinder (at least 1) on bank 1, which is driver's side, is not firing. The unburned air makes the o2 sensor read lean, which richens the rest of the cylinders, making things even worse.

Looking at the third log, if you have new upstream O2s why do you have pending heater circuit codes for both banks? P0135 & P0155. Did you use sensors looked up for whatever the year/make/model the PCM OS is from or is this a mix-n-match with whatever fits best or some website says is 'the' sensor to use in swaps, without knowing what they're talking about? Are they ACDelco/NTK/Denso?

The sensors are GM Genuine Parts 213-1702 Heated Oxygen Sensor from amazon, so a chance of them being fake I suppose.

The heater codes are because the BP Automotive wiring harness is wired with key-on power to the heaters instead of using the PCM. Someone once told me that the heaters are PWM controlled and that is the problem, but I'm skeptical since this company does these harnesses all the time like this without any issues.


On the cylinder misfiring, I'll see if I can verify that. I'm going to re-install the wideband and do some additional VE tuning but I highly suspect there's still an electircal or mechanical issue.

Keep in mind that this thing will run perfect for 20-30 minutes then go south usually

And it's always the same. One o2 goes lean and it starts dumping fuel so the other one goes rich. The wideband seemed to support the lean condition.

With the fuel pressure being low could the driver's side be pulling most of the available fuel? The gauge I'm using is a cheap Amazon gauge but it seemed to read my v10 engine correctly. I'll check fuel pressure prior to the regulator this afternoon.

Definitely verify fuel pressure. Someone had a similar problem about a year ago. They traced it down to incorrect injector wiring throwing off bank 1 and 2 assignment.

I don't see anything abbie normal in any of the data logs. Do you have a log capturing the event you're talking about? The "Bank 1 goes crazy lean"?

I'll go through some of the logs tonight and see if I can catch a good example.

What about the O2 cycling in the logs. Does it look normal? I was expecting more rapid swings.

The other thing I can do is install the wide band in the passenger side and run open loop.

The O2s and the trims both are incredibly slow. What's the exhaust setup on it? Headers, manifolds, etc. and how far are the O2 sensors from the cylinder head?

I'd address the heater codes. If the o2 isn't heating up properly there is a chance it might run the trims up or down before it actually warms up. This looks like the case here.

The heater codes are due to the waiting for the heaters being run straight to key on ignition vs the PCM. That's the way the BP Automotive wiring harness is set up. I've confirmed that they are getting power and ground. The code is showing because the PCM is expecting to be powering the heater and it isn't. Just to be safe I could rewire the heaters to the PCM.

The O2 sensors are down where the down pipe levels out under the vehicle. The exhaust was a stock exhaust manifold and a custom Y pipe

Quote:

---

Originally Posted by Rescu2000

The O2 sensors are down where the down pipe levels out under the vehicle. The exhaust was a stock exhaust manifold and a custom Y pipe

---

Can you be more specific? How far are they from the cylinder head? Can you put them closer to the manifolds?

The info in the logs (can't see the tune file, because it wasn't posted) shows as 2004 Sierra LM7. Application matters on these things. I mean, if you want the stuff to work like it's supposed to.

Attachment 145501

Power comes from a fuse, on all of them, everything, always has. Ground is PCM-controlled. Just because there's a company that does it wrong on all their harnesses doesn't make it right. Either fix it by wiring the sensors right or turn off the heater codes. There's actions the PCM takes when diagnostics fail, what those actions might be are complete unknowns in this case since the published documentation - the factory service manuals - only covers how it works when someone didn't decide to invent their own novel O2 circuit.

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

Bank-to-bank issues can be caused by either O2s or injectors (though if they're both swapped at the same time it'll work normal). The PCM just says 'if this sensor is rich, which half of the engine should I subtract fuel from?' and injector wiring can flip it just the same as O2 wiring.

Thanks for the replies. I'll double check this afternoon but I think the ground on this is just grounded. I didn't think it ran the heater ground back to the PCM but I will verify 100% tonight. I'll turn the codes off first and see what that does. I have the pins if I need to add the pins to the ecm (I think they were just plugged when I checked before). I have verified that the O2 sensors are on the right sides, but I've not verified that the injector wiring is correct from side to side. I assumed it wouldn't run if it was backwards, but I shouldn't assume.

As for the engine, I bought this already swapped. I was told it was out of a 2005, the ECU flash says O4. I'm pretty confident it's the LM7 though.


I'll do these actions and report back.

Thanks for the replies and I'll be applying every opinion. I'm not going to ask for help and then not try what's recommended.

Ok, some follow ups.

BlindSquirrel - Thanks and that is a great point I killed the codes for now and did a log (attached) and it looks good. HOWEVER, I've had a bunch of logs look good and then go on a longer drive and get this same issue again and again. I did not do any tuning on the interstate though. So maybe I need to touch up my VE map or MAF scaling. Also note that this engine does have a pretty bad oil leak so oil may be getting on wiring underneath. The O2 sensor wires look pretty clean. I think the leak is that rear seal and I'll probably just take it somewhere for that. I would like that fixed 100% correctly.

I also uploaded the tune below if that helps any. This is after removing the codes.


EdcMat - You asked where it was acting crazy. Open the "drive to work" log. There are times it just goes rich for several seconds. Like at 6:35:14 through 6:35:19. Maybe that is normal? I'm new at this. But then at 6:37:29 it does it for about 20 seconds and then continues pretty flat or slow until 6:38:11.

Then O2B1 just stops oscillating almost completely and starts dropping around 6:45:28 and just starts slowly dropping. LTB1 goes to 17.2 (even though it is reading 608mv) then it climbs to 25 (at 6:45:49) STFT's on bank 1 are at 53ish. O2 Bank 1 then drops to below 200 and continues down while B2 is fluctuating.

I'll get the measurements of how far the sensors are from the manifold.

Note that at this time I'm doing around 75 MPH on the interstate (Speedo is wrong on there)

O2BA stays way low, dropping below 100 and down to 3 sometimes. o2B2 goes to 800+ and stays there. The bank 1 O2 Sensor is the one that was black when I pulled it out to swap them. So I think it is a fake lean condition. I also hooked my Wideband up and stuck it in the tailpipe today and noticed it is black and reading 1.5 so I'm pretty sure it's shot unless I can clean it. I'll order a new one either way.

I'm thinking at this point maybe I run new wires to the O2 Sensor on the Driver side if this happens again. When I was tuning with the wideband I had the Bank 1 sensor removed and had the wideband in that hole.

If O2s or injectors are swapped bank-to-bank it creates a feedback loop. As long as everything is good, everything is good. But if either side starts to go away from zero then it just keeps going 'till it cain't go no more, because its corrections get applied to the wrong side. You need to verify the pinouts for injectors AND O2s.

Ok. I'll do that tomorrow afternoon.

Quote:

---

Originally Posted by Rescu2000

Ok, some follow ups.

BlindSquirrel - Thanks and that is a great point I killed the codes for now and did a log (attached) and it looks good. HOWEVER, I've had a bunch of logs look good and then go on a longer drive and get this same issue again and again. I did not do any tuning on the interstate though. So maybe I need to touch up my VE map or MAF scaling. Also note that this engine does have a pretty bad oil leak so oil may be getting on wiring underneath. The O2 sensor wires look pretty clean. I think the leak is that rear seal and I'll probably just take it somewhere for that. I would like that fixed 100% correctly.

I also uploaded the tune below if that helps any. This is after removing the codes.


EdcMat - You asked where it was acting crazy. Open the "drive to work" log. There are times it just goes rich for several seconds. Like at 6:35:14 through 6:35:19. Maybe that is normal? I'm new at this. But then at 6:37:29 it does it for about 20 seconds and then continues pretty flat or slow until 6:38:11.

Then O2B1 just stops oscillating almost completely and starts dropping around 6:45:28 and just starts slowly dropping. LTB1 goes to 17.2 (even though it is reading 608mv) then it climbs to 25 (at 6:45:49) STFT's on bank 1 are at 53ish. O2 Bank 1 then drops to below 200 and continues down while B2 is fluctuating.

I'll get the measurements of how far the sensors are from the manifold.

Note that at this time I'm doing around 75 MPH on the interstate (Speedo is wrong on there)

O2BA stays way low, dropping below 100 and down to 3 sometimes. o2B2 goes to 800+ and stays there. The bank 1 O2 Sensor is the one that was black when I pulled it out to swap them. So I think it is a fake lean condition. I also hooked my Wideband up and stuck it in the tailpipe today and noticed it is black and reading 1.5 so I'm pretty sure it's shot unless I can clean it. I'll order a new one either way.

I'm thinking at this point maybe I run new wires to the O2 Sensor on the Driver side if this happens again. When I was tuning with the wideband I had the Bank 1 sensor removed and had the wideband in that hole.

---

Scrolling through the log in this post everything looks normal. I just scrolled through quickly looking at the O2s and trims so could have missed something else.

Rear main seals rarely leak. It's almost always the rear cover isn't pulled down to the pan and everyone thinks it's the seal. So if you take it somewhere and all they do is throw a seal in it you'll end up with the same thing. All the cover bolts need to be loosened up and then you use the bolts through the oil pan to pull the rear cover tight to the oil pan, then tighten the cover bolts.

The O2s don't look swapped.

EDCMatt - Do you think it is possible I just didn't get the tables tuned well enough since I didn't spend time on the interstate?

Also, what about the fuel pressure being low? I can try a second gauge to confirm the first gauge readings.

I can confirm the O2's are not swapped. If I unplug one, the correct one shows faulted on the scanner.

I did remove the error codes that squirrel pointed out, that might help.

And I am, at this moment, confirming the wiring vs the ECM.

This is what they look like with injectors or O2s bank-swapped

Attachment 145573