Results 1 to 10 of 10

Thread: E38 - diagnosing the cause of detonation

  1. #1
    Junior Member
    Join Date
    May 2019
    Posts
    29

    Default E38 - diagnosing the cause of detonation

    Hello, guys. Vehicle is a 2008 Yukon XL with the 6.2L motor, AWD, 6L80E transmission. I have been using 93 octane fuel with no ethanol. The engine was rebuilt about 9500 miles ago. The compression ratio did increase a tiny bit, by approximately 0.1.

    I'm having some trouble finding the cause of detonation. The problem began within the past few weeks. I can hear the detonation when going up a hill, under what I'd consider moderate load. I logged some PIDs using version 8.3.22. Would someone mind looking at this log to point me towards the cause of detonation? Or perhaps I need to log additional PIDs or perform other checks?

    Thank you for any help.

    detonation (moderate load) April 2022 batch 3.pld

  2. #2
    Junior Member
    Join Date
    May 2019
    Posts
    29

    Default

    I thought it would be helpful to make it easier to see the PIDs that I logged.

    • Engine RPM
    • Throttle Position
    • Fuel System Status
    • Fuel System 2 Status
    • Fuel System 1 Status
    • Vehicle Speed Sensor
    • Engine Coolant Temperature
    • Intake Air Temperature
    • Heated O2 Sensor Voltage Bank 1 - Sensor 1
    • Heated O2 Sensor Voltage Bank 2 - Sensor 1
    • Manifold Absolute Pressure
    • Mass Air Flow
    • Ignition Timing Advance for Cylinder #1
    • Retard Due to Knock
    • Knock Retard
    • Cylinders for Which Knock is Detected
    • Knock detected for cylinder 1 (doesn't seem useful)
    • Knock detected for cylinder 2 (doesn't seem useful)
    • Knock detected for cylinder 3 (doesn't seem useful)
    • Knock detected for cylinder 4 (doesn't seem useful)
    • Knock detected for cylinder 5 (doesn't seem useful)
    • Knock detected for cylinder 6 (doesn't seem useful)
    • Knock detected for cylinder 7 (doesn't seem useful)
    • Knock detected for cylinder 8 (doesn't seem useful)
    • Long Term Fuel Trim - Bank 1
    • Long Term Fuel Trim - Bank 2
    • Short Term Fuel Trim - Bank 1
    • Short Term Fuel Trim - Bank 2

  3. #3
    Lifetime Member
    Join Date
    Mar 2006
    Posts
    570

    Default

    Quote Originally Posted by PLohr View Post
    I thought it would be helpful to make it easier to see the PIDs that I logged.

    • Engine RPM
    • Throttle Position
    • Fuel System Status
    • Fuel System 2 Status
    • Fuel System 1 Status
    • Vehicle Speed Sensor
    • Engine Coolant Temperature
    • Intake Air Temperature
    • Heated O2 Sensor Voltage Bank 1 - Sensor 1
    • Heated O2 Sensor Voltage Bank 2 - Sensor 1
    • Manifold Absolute Pressure
    • Mass Air Flow
    • Ignition Timing Advance for Cylinder #1
    • Retard Due to Knock
    • Knock Retard
    • Cylinders for Which Knock is Detected
    • Knock detected for cylinder 1 (doesn't seem useful)
    • Knock detected for cylinder 2 (doesn't seem useful)
    • Knock detected for cylinder 3 (doesn't seem useful)
    • Knock detected for cylinder 4 (doesn't seem useful)
    • Knock detected for cylinder 5 (doesn't seem useful)
    • Knock detected for cylinder 6 (doesn't seem useful)
    • Knock detected for cylinder 7 (doesn't seem useful)
    • Knock detected for cylinder 8 (doesn't seem useful)
    • Long Term Fuel Trim - Bank 1
    • Long Term Fuel Trim - Bank 2
    • Short Term Fuel Trim - Bank 1
    • Short Term Fuel Trim - Bank 2




    Remove some timing in those cells where it is detonating.

  4. #4
    Junior Member
    Join Date
    May 2019
    Posts
    29

    Default

    The tune is stock so far, I haven't purchased a license to tune. I would like to make sure the mixture is correct and sensors are functioning properly before taking timing out.

    I noticed that the oxygen sensors are reporting values around 100 mV at the low end. Shouldn't the low values stay at or above 200 mV? Would 100mV be considered a lean mixture? See the graphic below. I am assuming the oxygen sensors are working correctly but that has not been verified. These are factory oxygen sensors, not widebands.

    Click image for larger version. 

Name:	oxygen sensors possibly showing lean.png 
Views:	90 
Size:	47.8 KB 
ID:	24122

    Thank you for any help.

  5. #5
    Junior Member
    Join Date
    Aug 2018
    Posts
    44

    Default

    Quote Originally Posted by PLohr View Post
    Hello, guys. Vehicle is a 2008 Yukon XL with the 6.2L motor, AWD, 6L80E transmission. I have been using 93 octane fuel with no ethanol. The engine was rebuilt about 9500 miles ago. The compression ratio did increase a tiny bit, by approximately 0.1.

    I'm having some trouble finding the cause of detonation. The problem began within the past few weeks. I can hear the detonation when going up a hill, under what I'd consider moderate load. I logged some PIDs using version 8.3.22. Would someone mind looking at this log to point me towards the cause of detonation? Or perhaps I need to log additional PIDs or perform other checks?

    Thank you for any help.

    detonation (moderate load) April 2022 batch 3.pld

    The stock files come with a bit too much commanded timing for these IMO.
    I opened up a stock file to look at it, and the area in which you are under a large amount of load has a bit too much timing IMO.
    You aren't datalogging the correct PID for the g/cyl that the E38 needs, so I can't tell for certain where under the load axis you are at, but just looking at the file....I'd remove some timing there

    I'd also zero out B5112 so as to not have any adders of spark based upon commanded fuel.

    On your datalog I see 37 degrees at one point (after KR). That seems awfully excessive I bet you drop that down to 35 and you won't even notice the difference...actually, it would probably feel better since you aren't relying on the knock strategy and your flame will be nicer.

    02's look like they are reading properly.

    There are ways to combat knock as well, mechanically.

  6. #6
    Junior Member
    Join Date
    May 2019
    Posts
    29

    Default

    Quote Originally Posted by SOMFormula View Post
    The stock files come with a bit too much commanded timing for these IMO.
    I opened up a stock file to look at it, and the area in which you are under a large amount of load has a bit too much timing IMO.
    You aren't datalogging the correct PID for the g/cyl that the E38 needs, so I can't tell for certain where under the load axis you are at, but just looking at the file....I'd remove some timing there

    I'd also zero out B5112 so as to not have any adders of spark based upon commanded fuel.

    On your datalog I see 37 degrees at one point (after KR). That seems awfully excessive I bet you drop that down to 35 and you won't even notice the difference...actually, it would probably feel better since you aren't relying on the knock strategy and your flame will be nicer.

    02's look like they are reading properly.

    There are ways to combat knock as well, mechanically.
    SOMForumla, thank you for your time and effort on this. I see what you mean about the timing. I'll dig into this more then report back.

  7. #7
    Junior Member
    Join Date
    May 2019
    Posts
    29

    Default

    Is APCYL_M the correct PID to log in order to determine where the engine is in terms of load, to make timing adjustments where I'm seeing knock?

    I think I need to be looking at B5101 (High-Octane Spark Table) and B5102 (Low-Octane Spark Table) as a way of taking out timing to get out of detonation. Is that correct?

    Would someone mind helping with a theory question about B5112 (Fuel Mixture Spark Correction)?

    To check if I understand correctly, is it the case that 0.80 lambda is rich and 1.2 lambda is lean?

    Is it the case that somewhere around 0.86 lambda (12.6 air fuel ratio), flame speed is at its highest?

    If it turns out that the above questions are true, why does GM add timing in the B5112 table for mixtures richer than 0.86 lambda where flame speed is as fast as it gets?

    I am wanting to learn so please feel free to correct me on any of this.

    Here is a screen capture of B5112.
    Click image for larger version. 

Name:	B5112 table (fuel mixture spark correction).PNG 
Views:	80 
Size:	58.0 KB 
ID:	24126

  8. #8
    Lifetime Member
    Join Date
    Jul 2003
    Posts
    1,897

    Default

    Most of the knock in these is not during power enrichment. it is mostly on part throttle. they do have a high cylinder pressure when the cam is fully advanced. They add a lot of timing also when VCT is active.

    Thanks
    "All that is needed for the triumph of evil is that good men do nothing..."

  9. #9
    Junior Member
    Join Date
    May 2019
    Posts
    29

    Default

    Guys, thanks for your help. You've taught me a great deal. Because of that there has been a little progress in removing detonation.

    • I logged the APCYL_M PID to get the needed airflow per cylinder values.
    • I did more extensive logging to see if knock was showing up under a broader range of conditions.
    • After looking over the log, I removed approximately 2 degrees from the High-Octane Spark Table (B5101), in the cells where it made sense then loaded that calibration to the ECM. Also, I smoothed the adjacent timing values into the cells that were changed. See the screen capture for the changed cells.
    • I am seeing less knock in the log but there is still knock happening under light to moderate load. I can no longer hear knock but the KR PID in the log still showed some knock under the same conditions so I think there are more adjustments needed.


    Log before changing timing values:
    Attachment 24131

    Here are the changed timing values:
    Click image for larger version. 

Name:	B5101 High-Octane Spark Table changes.PNG 
Views:	74 
Size:	54.8 KB 
ID:	24134

    Here is the revised calibration:
    Attachment 24133

    Log after changing timing values:
    Attachment 24132

    What I have not done yet:
    • I have not yet zeroed out the B5112 table (Fuel Mixture Spark Correction). The AFRATIO_M:1 PID always showed 14.68:1 throughout the KR events. I could be misunderstanding, but I was not sure it made sense to change these values if lambda never goes below ~1.0 (~14.7:1) during KR events. But even so, as a test, I plan to zero out those values inside of an otherwise OE calibration, then log for knock. I am sure there's a good reason SOMFormula suggested the idea and I think it is worth trying.


    Any other suggestions for the next step?

    Thank you for any help.

  10. #10
    Junior Member
    Join Date
    May 2019
    Posts
    29

    Default

    Hello, guys. The knock had become inaudible but has come back. I still need to try zeroing out the B5112 table.

    I have another question and theory based on something that has been going on. So as we've seen above, there has been detonation at light loads. I did not mention this before, but at cold start the exhaust has a strong smell which I think is from an overly rich mixture. After driving the vehicle, the overly rich exhaust smell is gone. My theory: the overly rich mixture is causing excessive carbon deposits which are inducing detonation. Am I on the right track here?

Similar Threads

  1. Early E38 vs. New E38.....
    By Redline Motorsports in forum E37, E38 & E67 PFI ECM's
    Replies: 13
    Last Post: April 3rd, 2010, 07:50 PM
  2. Detonation After Tramsmission Tweaks
    By therock in forum General (Petrol, Gas, Ethanol)
    Replies: 0
    Last Post: June 6th, 2009, 09:21 PM
  3. Lean Cruise without detonation
    By cmitchell17 in forum Tuning For Economy
    Replies: 6
    Last Post: April 27th, 2008, 08:38 AM
  4. i hear detonation but can't see it
    By ASSAMI in forum General (Petrol, Gas, Ethanol)
    Replies: 20
    Last Post: February 9th, 2007, 10:56 PM

Tags for this Thread

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •