When you click on links to various merchants on this site and make a purchase, this can result in this site earning a commission. Affiliate programs and affiliations include, but are not limited to, the eBay Partner Network.
The Timing is advanced or retarded based on knock sensed correct? So is it sensing knock or pre ignition/detonation when reducing timing?
Timing is retarded during knock pre-ignition. Reducing KCLV, thus reducing the amount of ignition advance applied.
Also with no Tune including a/f adjustments for the Headers, would it not be lean with Headers, thus knock and pulling timing?
No, we have wideband lambda sensors that constantly provide fueling feedback. It will not run lean unless the headers magically increased *airflow* SIGNIFICANTLY beyond ECM compensation values.
Knock/ pre-ignition is not always caused by a lean condition.
Aftermarket tuning will simply optimizes fueling and ignition timing for maximum performance.
Last edited by viprez586; Dec 25, 2024 at 12:31 PM.
Timing is retarded during knock pre-ignition. Reducing KCLV, thus reducing the amount of ignition advance applied.
No, we have wideband lambda sensors that constantly provide fueling feedback. It will not run lean unless the headers magically increased *airflow* SIGNIFICANTLY beyond ECM compensation values.
Knock/ pre-ignition is not always caused by a lean condition.
Aftermarket tuning will simply optimizes fueling and ignition timing for maximum performance.
With Headers and Intake, on the stock Non tuned ECU, typically that will cause a lean condition, especially a WOT.
With Headers and Intake, on the stock Non tuned ECU, typically that will cause a lean condition, especially a WOT.
That's not the problem we are experiencing. Plus, we're all running a tune anyway, and monitoring STFT and LTFT along with knock feedback value and knock correction learned value in real time.
The more I think about it, the more I think it's a reversion problem that the OEM headers don't have. The other piece is, they do EGR with cam timing, and it seems likely to me this is not working as intended with aftermarket headers. I can see how cam timing in the low rpm range could be completely wrong with aftermarket headers because the factory tune expects a certain amount of restriction from the OEM cats that is not there when they're replaced or not there.
I drive with OBDLink running all the time, and yes, I've learned to game the system by backing out of the throttle when KFV goes super negative. It always drops back to -3.0, and often will correct itself just by letting off the pedal.
A member with a GSF with stock exhaust manifolds recently posted about experiencing this issue too though, so there has to be something more to it.
I finally made time to get the extenders replaced. Paul sent them to me ages ago, but I've been busy with other stuff and I knew without any doubt I was going to be dropping the front subframe to install these things. I'm not finished with this yet, but I will be tomorrow. Of significant note, it is possible to connect the left side Denso O2 sensor without any extender at all. The right side is really close, but I didn't like the tension on the wires when connected, and I decided to put in the new extender Paul sent me with gold plated pins and 6 inch length. I was able to use the OEM wiring supports as well, so I have confidence the O2 sensors will be working as designed.
I also found out it is possible to lower the engine even more than what the swaybar permits by removing the swaybar and dropping it about 2 more inches. Yes, you need to release some of the wiring supports on top of the engine and disconnect the intake tube entirely, but it got me much easier access to the top rear bolts when removing/installing the right side header. I didn't need to remove the left side header at all. Also, I have PPE catted headers, so room to work is pretty tight, and removing the heat shields above the cats is mandatory for anyone with XL glove sized hands.
Last edited by lobuxracer; Jan 17, 2025 at 08:24 PM.
After driving today, it does not appear the O2 extenders are the root cause of the problem. As I mentioned in the previous post, I have the left side sensor plugged directly into the harness - no extension - and the right side on a very short extension with gold plated contacts. The tune is behaving exactly the same as it did before I removed the long extensions.
Next up is PCV valve replacement (they finally showed up!), and if that has no impact, I have a new fuel pressure sensor and gasket to install. I am monitoring fuel pressure and it is stable at idle, but I am not convinced everything is happy there. STFT and LTFT are wildly variable, and I am getting DTCs for too rich while the ECM is trimming -30% from STFT and LTFT at the same time. More to try now.
PCV valve replaced today. Does not appear to have made any difference either. Important note - it's a really bad idea to overtighten the PCV valve. It will break pretty easily. Getting the broken part out without removing the intake manifold isn't fun. I ended up using the classic jam a flat blade screwdriver into the remaining piece and get it to turn out. That wasn't the first or even fifth method I attempted, but it's the one that worked.
New fuel pressure sensor is next, but while I am there I will be replacing all the knock sensors (they're less than $20 each from Amayama and Megazip), and the PCV case. I'm also doing a MAF clean (again) this time with soapy water to get the dirt removed, not just the oils like the spray cleaners do. It's really getting frustrating that this tuning issue seems to be unrelated to the sensors working correctly.
**UPDATE** Cleaning the MAF made zero difference.
Last edited by lobuxracer; Jan 27, 2025 at 10:21 AM.
As I'm preparing to install headers on my car I've been spending way too much time thinking about this problem. It does seem that reversion, or maybe weaker exhaust gas scavenging at low rpm could be the culprit here. There's a few YouTube videos explaining how aftermarket headers are designed to maximize scavenging at high rpm where we want peak torque but it's at the cost of weaker than OEM exhaust scavenging at low rpm. And the more free flowing the entire length of the exhaust system, the more pronounced this effect is. This would explain why viprez586 noted less low rpm knock correction with the exhaust valves closed. This guy explains it well at minute 7:42
Any combustion byproducts remaining in the cylinder will raise cylinder temps and possibly contain ignition sources like glowing soot. At the same time, at low rpm cruising, afr is as lean as the ECU will allow for fuel economy. You would expect those conditions would be quite conducive to detonation or pre-ignition.
I still think oil detergents contribute to the problem. Detergent concentration is highest when the oil is fresh, so I suppose there is a cleaning cycle after each oil change that removes soot and sludge that could be making its way in suspension to the cylinders before it gets picked up by the filter.
My engine and exhaust are completely stock except for a catch can and I've never experienced any of these issues (or so I thought) but out of curiosity I've been monitoring kclv, kfv, fuel trims and lambda for the last couple months prior to oil change. Kclv was sitting happy at 23.3 steady. Then 1 day after oil change, while cruising at steady speed on the freeway, my kfv dropped to -5 and kclv to 4.2 within a couple minutes! But, I would have been completely oblivious to it had I not been monitoring. I suspect this always happens after an oil change and is probably more common than most realize. There was no discernable effect on performance or drivability, mild acceleration felt normal but when I decided to floor it to see what happens I did notice it's not pulling as hard as I'd expect, though after a few seconds it started picking up power and kclv until I had to let off before I'd get arrested. Kept cruising at variable speeds and kclv recovered slowly to around 15-18. Throughout the drive fuel trims were relatively steady and lambda hovered around 1.0 so I'm guessing my ECU successfully arrested knock by retarding timing. Is it possible that if the knock was severe enough the ECU would also dump fuel as some are experiencing? Especially in the case of pre-ignition, where pulling timing doesn't help?
There's an element you're missing. The variable cam timing creates EGR without having the usual EGR plumbing, so there's exhaust gas being retained in the cylinder intentionally. The cam timing has been set to enable this with the OEM headers, but I suspect this isn't working properly with aftermarket headers because the primaries are longer and larger than OEM.
I've also noted that the problem with rapid KFV drops is most prevalent once the oil is at normal operating temperature (3 bars on the 2008). It is possible to drive around this, but it's a PITA. I have not had the "new oil" problem ever.
So far, I've replaced the PCV valve and removed the driver's side (LHD) extender entirely and used a 6" extender on the passenger's side. It has made no appreciable difference. I have a new MAF sensor, 4 new knock sensors, a new fuel pressure sensor, and a new PCV separator case I plan to install shortly after I fix the valley plate leak in the wife's GS F. I am not especially hopeful this will fix the problem. I'm most likely to think fixing the cam timing will be the real answer. The video kind of touches on this, but there's still a lot missing. It's a good primer for someone who is just starting out, but the problem we're having here is part of the advanced course, not the 101 in the video.
Last edited by lobuxracer; Feb 4, 2025 at 12:20 PM.
IIRC the fuel pressure sensor has its own part number??? I can only find it w/ the rail assembly. I'm about to do the same thing, including the sub harness.
I was also kind of wondering what issues might arise if the gasket/grommet from the pcv system going into the block fails. Looks like an absolute nightmare to replace
