Mysterious Engine Trouble...knocking and power loss.
Is it possible the fuel rail may have been damaged/clogged in some manor during fuel injector swaps? Faulty fuel pressure regulator? Clogged fuel filter? Someone want to stick a fuel pressure gauge on there?
you would see that immediately on the wideband. I'm sure Bryan would have noticed any changes in the fueling that he did not make.
I think Kooldino tried det cans but not sure if he unhooked the knock sensor.
It was TTP that had a car on their dyno that only put out like 420-430 crank HP on a FP Red, Cosworth cams, Meth setup and 26 psi with 11* timing was the max they could get out of it. Turns out that the timing chain was advanced one tooth and caused the power loss and lower boost threshold due to a faulty installation.
https://www.evolutionm.net/forums/ev...not-right.html
While that theory would make sense, as far as I know none of the problem cars have touched their cams. I have no idea if it's possible for the timing chain to jump a tooth like that on it's own but it seems like it would be worth veryfing the timing marks on the chain are lined up properly before tearing off the head.
It was TTP that had a car on their dyno that only put out like 420-430 crank HP on a FP Red, Cosworth cams, Meth setup and 26 psi with 11* timing was the max they could get out of it. Turns out that the timing chain was advanced one tooth and caused the power loss and lower boost threshold due to a faulty installation.
https://www.evolutionm.net/forums/ev...not-right.html
While that theory would make sense, as far as I know none of the problem cars have touched their cams. I have no idea if it's possible for the timing chain to jump a tooth like that on it's own but it seems like it would be worth veryfing the timing marks on the chain are lined up properly before tearing off the head.
I do think it is possible that something is going on in the valve train, this is a new design after all. I posted up in the thread on the other site to see if anyone having these issues had checked the cam lobe to bucket clearances. If the lobes or buckets are wearing over time and under stress it could help explain some of this behavior.
Well boys...looks like I'm at the end of my rope. I'm working with the guys at GST now on a plan for pulling the head and taking a real look inside the motor.
Probably going to take place within the next few weeks...but I'll be sure to report back on our findings w/pics.
Cross your collective fingers....
Probably going to take place within the next few weeks...but I'll be sure to report back on our findings w/pics.
Cross your collective fingers....
More on my friend's car.
Part two of this story is interesting. After the race gas didn't work, I had looked at GST's post and read something about reducing boost in Boostd4's car took away the knock. The prior tune was perfect for the car, but it was professionally (shop) done in EcuTeK, and I have Ecuflash. I didn't want to flash and enable Mode23 in his car in risk of disrupting or losing his EcuTeK license, so good logging was out. But at least I could see a couple things like RPM and knocksum.
With the race gas all run out of the car, I flashed in a stock map with his timing and fueling values. The knock remained. I tried a couple other little combinations and tweaks with the map, but in the end, I put a 100% stock ROM in the car, and the knock reduced substantially, to the tune of 1 and 2 counts here and there, but rarely. At least he was able to pull the car through the rev range now!
But on the combi maps, I noticed the oddest thing. I let the logger run, and even though on the original mapping it would knock everytime past 3500rpm to the point we had to lift out immediately, on the combination mapping just cruising around, I saw really high knocksum just off idle! No boost, no nothing. He pulled out of a parking lot completely normally, and it threw 7 and 8 counts for a couple seconds. This was 1500rpm, super light throttle, unexplainable. I saw this same activity a couple of times on decel as well, but higher RPM. He had already been off the throttle a couple seconds. Also, the KS was variable. It was as high as 15 at points, but often half that. Really weird.
While I know people were leaning toward the ECU, I can't subscribe to that because of the sound. The owner thought the sound had gone away, but I could still hear it. It does sound like knock, but instead of a solid state pinging, you'll hear one click, then hear nothing else. Almost makes me wonder if the noise could be in response to something else, but the ECU is seeing it as knock when it really isn't. Mivec? I have nothing but speculations. One thing I do know, the car will not throw a limp mode, but I took a turn at the wheel, and at onset when the knock starts, the power goes away completely. This did not feel like an Evo pulling a couple degrees of timing, it felt like the compressor stopped, or like 10 degrees of timing was pulled. Boost stayed up, but the power loss was substantial.
The only way the knock would go away was a fully stock rom, just flooding the car in fuel. Funny thing is, the timing values on the stock rom are just crazy high, but the car didn't mind them. Its the fueling it doesn't like. This leads us toward a bad injector, but he and the shop were unable to see a change in tracking on the wideband in the thousand or so RPM they could run it.
Deep down, I just don't think its real knock. The high timing on the stock map didn't induce knock, and the race gas in the tank did not quell it on the standard tune. I am at the end of my rope as far as what to do now, and his car remains on a stock ROM. I would love to know if this is the same problem seen in Boostd4's and Dino's cars, and if so, we need to get to the bottom of it.
As aforementioned, we swapped relays, that is not the issue here. Thanks for reading.
Part two of this story is interesting. After the race gas didn't work, I had looked at GST's post and read something about reducing boost in Boostd4's car took away the knock. The prior tune was perfect for the car, but it was professionally (shop) done in EcuTeK, and I have Ecuflash. I didn't want to flash and enable Mode23 in his car in risk of disrupting or losing his EcuTeK license, so good logging was out. But at least I could see a couple things like RPM and knocksum.
With the race gas all run out of the car, I flashed in a stock map with his timing and fueling values. The knock remained. I tried a couple other little combinations and tweaks with the map, but in the end, I put a 100% stock ROM in the car, and the knock reduced substantially, to the tune of 1 and 2 counts here and there, but rarely. At least he was able to pull the car through the rev range now!
But on the combi maps, I noticed the oddest thing. I let the logger run, and even though on the original mapping it would knock everytime past 3500rpm to the point we had to lift out immediately, on the combination mapping just cruising around, I saw really high knocksum just off idle! No boost, no nothing. He pulled out of a parking lot completely normally, and it threw 7 and 8 counts for a couple seconds. This was 1500rpm, super light throttle, unexplainable. I saw this same activity a couple of times on decel as well, but higher RPM. He had already been off the throttle a couple seconds. Also, the KS was variable. It was as high as 15 at points, but often half that. Really weird.
While I know people were leaning toward the ECU, I can't subscribe to that because of the sound. The owner thought the sound had gone away, but I could still hear it. It does sound like knock, but instead of a solid state pinging, you'll hear one click, then hear nothing else. Almost makes me wonder if the noise could be in response to something else, but the ECU is seeing it as knock when it really isn't. Mivec? I have nothing but speculations. One thing I do know, the car will not throw a limp mode, but I took a turn at the wheel, and at onset when the knock starts, the power goes away completely. This did not feel like an Evo pulling a couple degrees of timing, it felt like the compressor stopped, or like 10 degrees of timing was pulled. Boost stayed up, but the power loss was substantial.
The only way the knock would go away was a fully stock rom, just flooding the car in fuel. Funny thing is, the timing values on the stock rom are just crazy high, but the car didn't mind them. Its the fueling it doesn't like. This leads us toward a bad injector, but he and the shop were unable to see a change in tracking on the wideband in the thousand or so RPM they could run it.
Deep down, I just don't think its real knock. The high timing on the stock map didn't induce knock, and the race gas in the tank did not quell it on the standard tune. I am at the end of my rope as far as what to do now, and his car remains on a stock ROM. I would love to know if this is the same problem seen in Boostd4's and Dino's cars, and if so, we need to get to the bottom of it.
As aforementioned, we swapped relays, that is not the issue here. Thanks for reading.
Last edited by Noize; Sep 8, 2009 at 11:56 AM.
Good update, sorry that nothing definitive was found yet.
BTW using ECUFlash won't effect the EcuTek License in any way that I know of assuming you are only pasting in maps and not overwritting with a completely different rom and editing the immobilizer code somehow. At least that's what I've read and I have an EcuTek rom that's ben heavily modified to say the least.
BTW using ECUFlash won't effect the EcuTek License in any way that I know of assuming you are only pasting in maps and not overwritting with a completely different rom and editing the immobilizer code somehow. At least that's what I've read and I have an EcuTek rom that's ben heavily modified to say the least.
Evolved Member
Joined: Apr 2009
Posts: 536
Likes: 0
From: US
More on my friend's car.
But on the combi maps, I noticed the oddest thing. I let the logger run, and even though on the original mapping it would knock everytime past 3500rpm to the point we had to lift out immediately, on the combination mapping just cruising around, I saw really high knocksum just off idle! No boost, no nothing. He pulled out of a parking lot completely normally, and it threw 7 and 8 counts for a couple seconds. This was 1500rpm, super light throttle, unexplainable.
But on the combi maps, I noticed the oddest thing. I let the logger run, and even though on the original mapping it would knock everytime past 3500rpm to the point we had to lift out immediately, on the combination mapping just cruising around, I saw really high knocksum just off idle! No boost, no nothing. He pulled out of a parking lot completely normally, and it threw 7 and 8 counts for a couple seconds. This was 1500rpm, super light throttle, unexplainable.
Quick update...heart surgery scheduled for Sept. 28th. I'll be making step by step updates as work progresses...but we're basically going to inspect everything until the head comes off.
Wish me luck...
Wish me luck...
Quick update...things got a little backed up at the shop so I got pushed until probably tomorrow. I'll update as things progress...work should get started in the morning.
Keep hope alive!
Keep hope alive!