2008 GMC 6.2L - Detonation at Idle (Caused by excessive load at idle).. Why?
In a previous post an engine that was excessively loaded at an idle had a symptom of knocking at idle. Why do you think it did this? Understanding why may have value if addressing other knock related issues in the future. This engine was pushed outside of its normal operating parameters, but the issue of knock that it presented may be noted in other engines/programs etc. under odd operating conditions. Having more tools in the toolbox never hurts. A trick used on that vehicle during knock was that propane was added which lessened the knock during experimentation.
I believe the program for this vehicle was sloppy? I hesitate to say that because it was operating under abnormal load conditions. There are layers that effect ignition timing. The layers work in concert with each other. One layer that I can think of is the density of the air fuel charge within the cylinder and how much that charge is compressed. Essentially this is the mechanical aspect of…
Hi Bill, For reference, here is the video I recorded where I demonstrated the vehicle’s idle condition in gear with excessive load and detonation. Here is the data log showing how the knock went away when propane was added. In hindsight, I wish I had demonstrated the propane scenario.
Why couldn't this be caused by excessive carbon deposits in this cylinder?
In my opinion the symptom was too specific to the condition on multiple cylinders to suggest a pre ignition caused by carbon deposits.
Thanks Bill I need to hang out here more often. The diagnostic stuff you guys are doing and equipment you are using are a million miles past my experience and training
You're welcome David, In this situation you could download EFI live or HPtuners for free and "play" with files. I know from experience HP academy, and calibrated success are great resources for education on this subject. There are some talented professionals in here for sure. I am always learning something here.
Hi Bill, After some late-night reading and thinking about what was taking place here, I believe that the article abstract linked below provides a plausible scenario. The “End Gas” found in the compressed air-fuel charge is igniting ahead the normal, spark-driven controlled flame front. aip.scitation.org/doi/10.1063/1…
Thanks Scott, I agree with that, the explanation of end gas effect is similar to explanations of spontaneous combustion that occurs with the controlled flame front when excessive pressures are caused in the cylinder by excessive ignition timing. I wonder what grade fuel they were using at the time of testing?
Hi Scott, Isn't that called pre-ignition? I'm not an expert on this, but my understanding is pre-ignition always occurs before the actual ignition event. The effects of pre-ignition are usually catastrophic (holes in pistons...) and for the most part are silent. Also things go bad fairly rapidly. This could be a glowing chunk of carbon that ignites the mixture usually close to 180* BTDC, or…
This was my thinking as well, that it was a heat-induced pre-ignition. When pressure goes up, temperature goes up, and if the engine is under significant low-speed load, the heat is going to have more time to build in the cylinder. I think the reduced detonation that Scott perceived when he introduced propane supports that idea. Auto-ignition temperature of gasoline is somewhere around 280C…
Interesting theory Nathaniel, Can you expand on the details of why the heat is going to have more time to build in the cylinder? I interpret that a couple different ways. I want to make sure I am on the same page as you.
Well let's imagine two single cylinder engines doing the same work. One is working at 3000rpm, the other at 600rpm. At 3000rpm, the power stroke is going to last 40ms. At 600rpm, the power stroke will last 200ms. If the engines are outputting the same work, they must be ingesting the same BTU value of fuel, right? Now if we accept that any heat transfer, whether convective or conductive, takes…
Nathan, I like the way you think. You don't believe retarding the spark timing from where it was would reduce the cylinder pressure, temperature and heat enough to prevent the knock?
Absolutely I agree that retarding ignition timing will reduce cylinder pressures, but we're talking about an engine that was operating well outside of its design parameters. At some point with any load/rpm relationship, you can only retard timing so far before you have engine stall. Could the PCM have a bit more room to pull timing than it's tables would allow, and still keep the engine…
Nathaniel, I believe Scott mentioned he pulled it back to 8 degrees. Likely there would be no conclusion unless someone with an engine dyno has played with this part of the program before in this engine. I have reached out to a business that would have valuable input if they wish to participate. I am going to reach out to another as well. I really enjoy these discussions. It helps me gain a…
Right, so that answers the first question. Yes, there was room to further retard the timing without engine stall. He also stated that it made no difference with the detonation as far as he could tell. Possibly he could have pulled it back more, but I'm going place my bet and say that timing control alone wouldn't have eliminated detonation in this case.. I think the combustion chambers/spark…
Nathaniel, 100%. That would be a very valid data gathering. Genius. It would take time though, to go from the beginning of the flame propagation/pre ignition to detonation. That may make that method inconclusive depending on how much time difference there was. If the knock were prior to ignition then absolutely we would know the timing was NOT the issue.
Bill, For sure, interpreting the knock signal could be tricky. Disabling one cylinder might give you the knock signal gap needed to determine the spark-to-knock relationship.
Yes, this is a real brain buster. I enjoy all the input and different critical reasoning that comes out of these discussions. I am hopeful someone may contribute that has first hand experience with resolving this issue.
Hi Nathaniel, Nice thinking! Another thing to add to this would the amount of fuel that is being added for the increase in load to keep the engine running with this fault. Added fuel = Increased Heat. Let's say my numbers from the vehicle I experienced this fault on are accurate and the pulse width went from 4.2 to 11.2. That's an increase of 2.67 times the amount of fuel for a normal…
Right. And we could see that the PCM was clearly adding fuel. I don't recall combined fuel trims, but one was at least +10.
Although looking at fuel trim numbers in this case possibly doesn't give you an answer to how much additional fuel is going into the engine. From the post Scott put on his original query I put up some numbers from a vehicle I experienced the same fault. I wish I could have found the snapshot files so I can reference the data, but I did document it on the RO. It was also a different engine and…
No, that's true. Fuel trims won't give you any indication of fuel volume, but they will tell you about operating conditions at a given load cell. And what we can see here is that even the closest load calculation the PCM had available wasn't enough fuel for these conditions.
Scott, I agree with this 100% "The increase in fuel consumption, which produces more heat, is what I believe was causing the detonation". I think that explanation would also be accurate if the engine was knocking at WOT and 3500RPM.
Hi Scott, Could be, pinging/spark-knock/detonation/pre-ignition I believe is misunderstood and engineers and scientists are still looking for answers. I began reading papers on this many years ago and although I’m not a scientist or engineer, my perspective has changed. I’ve learned to believe in the way something works until proved wrong. I wish I had an adapter for an in-cylinder pressure…
Hi Scott, I read the link you posted above, at least what I could see without purchasing access to the full paper. I read this one last night - sci.utah.edu/publications/p….pdf I'll need to go through it again when I have more time, but you're possibly correct that what has been taught is not completely accurate.
Looks like a great article. I’ll be sure to read this at least a couple of times. Thanks for the link!
Yeah, I was thinking about that one too. Since the "pre" in pre-ignition, means it exploded BEFORE the spark occurred, the simplest explanation is that the ECM just could not figure-out that it should have advanced timing, because the load came from an unexpected (broken) condition, that it just couldn't comprehend or recognize. The PCM did not believe it was under load, even though we could see…
Geoff, Have a look at Scott's timing table. It shows the boxes populated with advance angle in degrees for cyl air/RPM with cyl air being a representation of load. Those numbers are able to be changed through programming.
I'm sure they could be, but the truck was actually broken, so wouldn't fixing it be better than reprogramming? Or maybe I misunderstood you?
Geoff, 100%. That vehicle needs the load repaired. For all intents and purposes on this vehicle there was no need to dig more. I just wanted to dive deeper. I have noticed a trend in the programs for this particular engine to have knock at low RPM tip in. I would imagine there are other engines as well that have similar quirks. I was trying to shed light on programs with their "quirks", and…
OK, I understand you now. On a similar note, I used to work on a fleet of 1997 Ford E350, and ALL of them pinged under light load. It would go away if you ran Premium, but they were all 87 octane spec. I figured it was Ford engineering's way of running it right on the edge, and to keep the KS busy. Actually, no, I just figured the vans sucked....LOL...I was only 25 then. :-)
Hi Bill, I recorded another short video showing data from the vehicle after the transmission was repaired.