Had some detonation... what do you think has changed?
Thread Starter
VIP Member
iTrader: (2)
Joined: Nov 2002
Posts: 3,639
From: Longing for my ol' white '02 WRX :(
Car Info: 2016 Acura RDX ... meh. Um, nice subwoofer?
Had some detonation... what do you think has changed?
In early January and then just the other day I was running over 800C and detonating for the first, and now second, time. After the first round of detonations, I lowered my EBC from 16 to 15.5. I was on the road at the time and stopped at a Home Depot and dropped some Xylene into the tank. I drove real nice for about 50 miles to be sure the juice got through the lines. Life was better. When I got home I reset the ECU and DAMN the car felt more responsive. Over time it got more sluggish and now its detonating again.
What's confusing me is, say its sluggish because the ECU is pulling timing because it senses detonation. Then why would it eventually detonate worse, until its quite audible? What's exacerbating the problem?
performance mods: uppipe, catback, EBC (use to max at 16psi, now at 15.5psi), IC hoses.
possibly related mods: STi hoodscoop w/ stock WRX shroud. (less IC cooling? I know, long shot)
Why is it detonating? Here are my ideas:
Possibility #1 - bad gas. (not from me, from the gas pump). Both times it had been running on questionable gas. The first time I did get the gas from a 76 station, but it was out in freakin BFE. Yesterday I was running on some Chevron. Before these two events,and since my modification, I had always gotten gas from pretty much two stations in San Francisco. I've read the data folks have logged about gas performance.
Possbility #2 - runtime on the ECU. Lately I have had some of the longest times between needing to disconnect the battery and so the ECU has had hundreds of miles at a time to configure itself. I have noticed that when I reset the ECU the car is MUCH stronger than before, even without doing Shiv's quick-n-dirty throttle programming.
Possibility #3 - something is actually wrong... but what?
jason
What's confusing me is, say its sluggish because the ECU is pulling timing because it senses detonation. Then why would it eventually detonate worse, until its quite audible? What's exacerbating the problem?
performance mods: uppipe, catback, EBC (use to max at 16psi, now at 15.5psi), IC hoses.
possibly related mods: STi hoodscoop w/ stock WRX shroud. (less IC cooling? I know, long shot)
Why is it detonating? Here are my ideas:
Possibility #1 - bad gas. (not from me, from the gas pump). Both times it had been running on questionable gas. The first time I did get the gas from a 76 station, but it was out in freakin BFE. Yesterday I was running on some Chevron. Before these two events,and since my modification, I had always gotten gas from pretty much two stations in San Francisco. I've read the data folks have logged about gas performance.
Possbility #2 - runtime on the ECU. Lately I have had some of the longest times between needing to disconnect the battery and so the ECU has had hundreds of miles at a time to configure itself. I have noticed that when I reset the ECU the car is MUCH stronger than before, even without doing Shiv's quick-n-dirty throttle programming.
Possibility #3 - something is actually wrong... but what?
jason
Registered User
Joined: May 2003
Posts: 149
From: Pennsylvania
Car Info: 2002 WRX
Re: Had some detonation... what do you think has changed?
[i]Then why would it eventually detonate worse, until its quite audible? [/B]
If you can hear it detonating then I would suspect the knock sensor, or wiring to it.
Another possibility would be a vacuume leak, but under boost what would that do??? hummmm
Thread Starter
VIP Member
iTrader: (2)
Joined: Nov 2002
Posts: 3,639
From: Longing for my ol' white '02 WRX :(
Car Info: 2016 Acura RDX ... meh. Um, nice subwoofer?
You missed the point of the question heh... "If the knock sensory can hear it detonating (before I can) and makes corrections in the timing to compensite, why would the problem continue to get worse until its so bad that I can hear the detonation audibly?"
jason
jason
Thread Starter
VIP Member
iTrader: (2)
Joined: Nov 2002
Posts: 3,639
From: Longing for my ol' white '02 WRX :(
Car Info: 2016 Acura RDX ... meh. Um, nice subwoofer?
Oh, and its usually at full throttle, full boost, way up in the RPMs at higher gears ... so WOT and lotsa load.
jason
jason
Registered User
Joined: May 2003
Posts: 149
From: Pennsylvania
Car Info: 2002 WRX
mmboost.. I dont believe that the knock sensor is either 1-hearing it detonate or 2- not sending a signal to the ECU that it has dectected knock or 3- the ECU isn't adjusting timming..
If the system is working the way it should. You should NEVER hear a knock. I have never heard one. But I have "felt" the ecu pull timiming due to inaudiable knock.
How do you know that the knock sensor is actually making adjustments to compensate for detected knock? Have you recorded it with an OBDII scan tool?
I know I had some issues with boost pressure causing knock when the temps outside got colder. i.e letting in more air causing it to run leaner.
After rereading your posts . I would think that your running lean at the top end and it worked fine till the temps (outside air) got colder . Thus causing a lean condition.
You seem to have made the motor breath easier and upped the boost (jamming in still more air) but have not added more fuel....................
If the system is working the way it should. You should NEVER hear a knock. I have never heard one. But I have "felt" the ecu pull timiming due to inaudiable knock.
How do you know that the knock sensor is actually making adjustments to compensate for detected knock? Have you recorded it with an OBDII scan tool?
I know I had some issues with boost pressure causing knock when the temps outside got colder. i.e letting in more air causing it to run leaner.
After rereading your posts . I would think that your running lean at the top end and it worked fine till the temps (outside air) got colder . Thus causing a lean condition.
You seem to have made the motor breath easier and upped the boost (jamming in still more air) but have not added more fuel....................
Last edited by andyhidley; Feb 18, 2004 at 05:28 PM.
Thread Starter
VIP Member
iTrader: (2)
Joined: Nov 2002
Posts: 3,639
From: Longing for my ol' white '02 WRX :(
Car Info: 2016 Acura RDX ... meh. Um, nice subwoofer?
Well, that's what I'm talking about. I should never hear a knock. So why am I? I'm not claiming to have proof of knock being sensed... but you don't think the knock sensor "hears" better than I do? 
As far as cold, I dunno. I don't live in an area where the temps fluctuate often outside a 20F zone. But you might be right - the additional modifications are in their first "cold" season which, while even though temperate, mighjt be just enough push it over the edge.
Sigh... Kalifornia "gas". Where oh where has my 100 gone Hopefully, I'll have the UTEC soon.
jason
As far as cold, I dunno. I don't live in an area where the temps fluctuate often outside a 20F zone. But you might be right - the additional modifications are in their first "cold" season which, while even though temperate, mighjt be just enough push it over the edge.
Sigh... Kalifornia "gas". Where oh where has my 100 gone
Hopefully, I'll have the UTEC soon.jason
Registered User
iTrader: (14)
Joined: Nov 2001
Posts: 14,221
From: Peoples Republik of Kalifornia
Car Info: 05 H2 SUT, 45 GPW, 10 Murano, 13 Boss 302
Jason,
The ECU can only pull so much timing. After that if you're still knocking you will definitely hear it. I had some audible knock at a certain rpm range. The ECU was pulling timing and my knock correction was at zero. I had to lower the boost and pull some timing myself to make the hesitation go away. It could have been a bad tank of gas or the crappy Kalifornia winter formula since my previous tune was from the summer.
The ECU can only pull so much timing. After that if you're still knocking you will definitely hear it. I had some audible knock at a certain rpm range. The ECU was pulling timing and my knock correction was at zero. I had to lower the boost and pull some timing myself to make the hesitation go away. It could have been a bad tank of gas or the crappy Kalifornia winter formula since my previous tune was from the summer.
iClub Silver Vendor
iTrader: (12)
Joined: Mar 2003
Posts: 8,228
From: 631 Railroad Ave. Fairfield, CA
Car Info: A Laptop
Are you sure the knock is occuring at WOT? With an EBC you run the risk of running PTFB which would cause high EGT's and bad detonation.
If it is in fact at WOT, I would check all your fuel components... fuel filter and fuel pump specifically. You say it happens mostly in the higher rpm's, maybe your pump is crapping out?
-- Ed
If it is in fact at WOT, I would check all your fuel components... fuel filter and fuel pump specifically. You say it happens mostly in the higher rpm's, maybe your pump is crapping out?
-- Ed
VIP Member
iTrader: (9)
Joined: Nov 2002
Posts: 3,928
From: ...a craphole in No.County San Diego
Car Info: MMVI AW WRX STI
Fuel filter, dirty injectors, bad gas, plugs??? Or just it's colder combined with some subtle variant of the above? How many miles on your car?
I don't understand your possibility #2. The car feels stronger upon reset because it's running less safe timing/fuel settings, detecting knock and pulling back to safe levels as it learns. I would think that if the knock is within the car's ability to adjust, the longer you run the ECU (ie the more learning that takes place), the less likely you would be to hear knock.
Good luck!
I don't understand your possibility #2. The car feels stronger upon reset because it's running less safe timing/fuel settings, detecting knock and pulling back to safe levels as it learns. I would think that if the knock is within the car's ability to adjust, the longer you run the ECU (ie the more learning that takes place), the less likely you would be to hear knock.
Good luck!
VIP Member
iTrader: (9)
Joined: Nov 2002
Posts: 3,928
From: ...a craphole in No.County San Diego
Car Info: MMVI AW WRX STI
And the knock sensor doesn't do shizzle after ~6k rpm's anyway.
From this thread, from a really smart guy: http://forums.nasioc.com/forums/show...knock+high+rpm
From this thread, from a really smart guy: http://forums.nasioc.com/forums/show...knock+high+rpm
The stock ECU has a characteristic that makes knock more likely under a high load high gear, high rpm situation. The situation, you would see during a high speed run at WOT in 4th or 5th gear or pulling a hill in a higher gear.
The stock ECU is adaptive. If it senses knock it will pull back timing and or add fuel at that point in the maps to prevent knock. Over time it will gradually advance timing until it senses knock again. The result is that the engine runs very close to the knock limit, but destructive knock is rare and quickly stopped at lower rpms. You will seldom if ever hear low rpm knock.
The Stock ECU however does not listen for knock at rpms above about 5700. Theories are that given the stock turbo and such, knock is not likely at the higher rpms. The other common theory is that at high rpm the ECU cannot detect knock due to all the other engine noises.
Regardless of the cause, especially if you have modifications, you are most likely to run into knock at high rpm on the stock ECU. With aftermarket engine management of course, you can hear it at any rpm, and many tuners report having to pull back timing near max torque to eliminate knock in that rpm range.
A lot of folks have reported killing an engine due to a long hard high rpm run, or an overly agressive dyno session.
If you think your are hearing knock you should pull your plugs and look for little light colored specks on the electrodes. If you look at them under about 10X magnifaction, you will see the specks are small ***** of aluminum blown off the piston top during detonation.
The stock ECU is adaptive. If it senses knock it will pull back timing and or add fuel at that point in the maps to prevent knock. Over time it will gradually advance timing until it senses knock again. The result is that the engine runs very close to the knock limit, but destructive knock is rare and quickly stopped at lower rpms. You will seldom if ever hear low rpm knock.
The Stock ECU however does not listen for knock at rpms above about 5700. Theories are that given the stock turbo and such, knock is not likely at the higher rpms. The other common theory is that at high rpm the ECU cannot detect knock due to all the other engine noises.
Regardless of the cause, especially if you have modifications, you are most likely to run into knock at high rpm on the stock ECU. With aftermarket engine management of course, you can hear it at any rpm, and many tuners report having to pull back timing near max torque to eliminate knock in that rpm range.
A lot of folks have reported killing an engine due to a long hard high rpm run, or an overly agressive dyno session.
If you think your are hearing knock you should pull your plugs and look for little light colored specks on the electrodes. If you look at them under about 10X magnifaction, you will see the specks are small ***** of aluminum blown off the piston top during detonation.
Thread Starter
VIP Member
iTrader: (2)
Joined: Nov 2002
Posts: 3,639
From: Longing for my ol' white '02 WRX :(
Car Info: 2016 Acura RDX ... meh. Um, nice subwoofer?
mp & vaus, that pretty much describes it.
I'm just over 30k miles... and its winter (meaning gas content not temp)... I can see EGTs high with PTFB moments, but when I mash the pedal EGT will drops... but then creep until RPMs get close to 6000 and then it goes *dink* *clink* *dink*. Bt.w - things have gotten better since I flushed out whatever was in the tank before and got some overpriced 76, although I do think I heard a little detonation, but not as bad. Right now I'm running a gallon of Xylene with 76, so its technically a 93 octane mix. Haven't heard anything yet, but I haven't been pushing it.
So this weekend I'll change the fuel filter and plugs. Do those injector cleaners really work? How do you test the efficiency of the fuel pump?
jason
I'm just over 30k miles... and its winter (meaning gas content not temp)... I can see EGTs high with PTFB moments, but when I mash the pedal EGT will drops... but then creep until RPMs get close to 6000 and then it goes *dink* *clink* *dink*. Bt.w - things have gotten better since I flushed out whatever was in the tank before and got some overpriced 76, although I do think I heard a little detonation, but not as bad. Right now I'm running a gallon of Xylene with 76, so its technically a 93 octane mix. Haven't heard anything yet, but I haven't been pushing it.
So this weekend I'll change the fuel filter and plugs. Do those injector cleaners really work? How do you test the efficiency of the fuel pump?
jason
Registered User
iTrader: (12)
Joined: Nov 2002
Posts: 18,369
From: Reno, NV
Car Info: 1993/2000/2001 GF4 mostly red
Re: Had some detonation... what do you think has changed?
Originally posted by mmboost
In early January and then just the other day I was running over 800C and detonating for the first, and now second, time. After the first round of detonations, I lowered my EBC from 16 to 15.5. I was on the road at the time and stopped at a Home Depot and dropped some Xylene into the tank. I drove real nice for about 50 miles to be sure the juice got through the lines. Life was better. When I got home I reset the ECU and DAMN the car felt more responsive. Over time it got more sluggish and now its detonating again.
In early January and then just the other day I was running over 800C and detonating for the first, and now second, time. After the first round of detonations, I lowered my EBC from 16 to 15.5. I was on the road at the time and stopped at a Home Depot and dropped some Xylene into the tank. I drove real nice for about 50 miles to be sure the juice got through the lines. Life was better. When I got home I reset the ECU and DAMN the car felt more responsive. Over time it got more sluggish and now its detonating again.
iClub Silver Vendor
iTrader: (12)
Joined: Mar 2003
Posts: 8,228
From: 631 Railroad Ave. Fairfield, CA
Car Info: A Laptop
You say your EGT's are over 800. What are you actually seeing at 6k RPM? If the knock is occuring at 6k RPM+ the ecu is not listening and has no way of detecting it. How much boost are you running at 6k?
-- Ed
-- Ed