fed up with knock control
an old school trick to reduce sensor sensitivity was to take the knock sensor out, wrap the treads with teflon tape and put it back on. I haven't done it on my car but maybe worth a try.
i dont want to reduce the sensitivity of the knock sensor as it works fine during WOT pulls in the upper rev range >4k rpm. i just want to ignore certain areas of the rpm range
I just had my first occurance with shift knock at the track a few weeks ago and I could tell from the logs that it was trying to go from roughly 19* at peak rpm to roughly 8* at the beginning of the next gear in an instant. I followed the load curve and made it taper down to 8* much more smoothly and the shift knock is non-existant now.
Account Disabled
iTrader: (38)
Joined: Aug 2005
Posts: 9,319
Likes: 1
From: Virginia Beach, Virginia
I just had my first occurance with shift knock at the track a few weeks ago and I could tell from the logs that it was trying to go from roughly 19* at peak rpm to roughly 8* at the beginning of the next gear in an instant. I followed the load curve and made it taper down to 8* much more smoothly and the shift knock is non-existant now.
Thats sounds like a great addition to the factory knock control. I think that should pretty much satisfy everyone.
I agree. When I turn the A/C on, my downpipe vibrates all over sub-frame and never generates engine knock.
After I had my engine rebuilt the first time, there were many instances where I had 36 counts of knock while cruising. After a few months, I noticed the antifreeze level kept dropping but with no sign of where the antifreeze was going. Idle also kept getting worse and worse. Then a week later.....plehk, plehk ppppllleeeeccckkkk........nothing.......big cloud of white smoke.......headgasket is blown. The damage included a nice gash in the head and block, and crappy head shaving job indicated by a "competent" engine builder. The head shave looked like someone took a belt sander to it, compared to the shaves his diamond cutter produced.
Anyway, the point is:
1. choose a good machine shop
2. pay the money to get the job done right
3. Power Enterprise headgasket may not hold up well past 26lbs of boost
4. rediculous knock count was most likely caused by a perforated head gasket (36 count knock is gone after rebuild)
5. phantom knock might not be phantom knock.
Anyway, the point is:
1. choose a good machine shop
2. pay the money to get the job done right
3. Power Enterprise headgasket may not hold up well past 26lbs of boost
4. rediculous knock count was most likely caused by a perforated head gasket (36 count knock is gone after rebuild)
5. phantom knock might not be phantom knock.
Last edited by D-VO; Oct 21, 2008 at 08:59 PM. Reason: whoops
After I had my engine rebuilt the first time, there were many instances where I had 36 counts of knock while cruising. After a few months, I noticed the antifreeze level kept dropping but with no sign of where the antifreeze was going. Idle also kept getting worse and worse. Then a week later.....plehk, plehk ppppllleeeeccckkkk........nothing.......big cloud of white smoke.......headgasket is blown. The damage included a nice gash in the head and block, and crappy head shaving job indicated by a "competent" engine builder. The head shave looked like someone took a belt sander to it, compared to the shaves his diamond cutter produced.
Anyway, the point is:
1. choose a good machine shop
2. pay the money to get the job done right
3. Power Enterprise headgasket may not hold up well past 26lbs of boost
4. rediculous knock count was most likely caused by a perforated head gasket (36 count knock is gone after rebuild)
5. phantom knock might not be phantom knock.
Anyway, the point is:
1. choose a good machine shop
2. pay the money to get the job done right
3. Power Enterprise headgasket may not hold up well past 26lbs of boost
4. rediculous knock count was most likely caused by a perforated head gasket (36 count knock is gone after rebuild)
5. phantom knock might not be phantom knock.
Who was the builder?
For the first rebuild it was Random Jackass C. LOL! I can't remember his name. It's just some guy my brother recommended because of his low price. Low price = low quality in this case.
For the second rebuild I had Twister over here in Orlando do the machine work and Joe from Speed Clinic assemble the block. I did everything else. I put about 1200 miles on it since the rebuild and I've yet to have to add oil or anti-freeze. These guys do excellent work.
For the second rebuild I had Twister over here in Orlando do the machine work and Joe from Speed Clinic assemble the block. I did everything else. I put about 1200 miles on it since the rebuild and I've yet to have to add oil or anti-freeze. These guys do excellent work.
Last edited by D-VO; Oct 24, 2008 at 02:07 PM.
For the first rebuild it was Random Jackass C. LOL! I can't remember his name. It's just some guy my brother recommended because of his low price. Low price = low quality in this case.
For the second rebuild I had Twister over here in Orlando do the machine work and Joe from Speed Clinic assemble the block. I did everything else. I put about 1200 miles on it since the rebuild and I've yet to have to add oil or anti-freeze. These guys do excellent work.
For the second rebuild I had Twister over here in Orlando do the machine work and Joe from Speed Clinic assemble the block. I did everything else. I put about 1200 miles on it since the rebuild and I've yet to have to add oil or anti-freeze. These guys do excellent work.
Further to this - I have found (well it was fairly easy todo so with the disasm I have) the place where the knock sensor is read.
I am wondering if a simple patch that replaces this ADC_Read code with something like:
if (load > min_load_for_knock) {
ADC_Read
}
Would work..
OR would people prefer knock_sum to still be calculated, but not taken into account for timing reduction?
I am wondering if a simple patch that replaces this ADC_Read code with something like:
if (load > min_load_for_knock) {
ADC_Read
}
Would work..
OR would people prefer knock_sum to still be calculated, but not taken into account for timing reduction?
Tephra,
If your looking for a wishlist to help people deal with 2.0L motors that cooperate, 2.0L motors that don't, and stroker motors that don't cooperate, here would be my wishlist.
1. Ignore timing pull up till a user defined moment knock threshold. It should still be able to be logged. IE: Ignore knock up to say 3 counts.
2. At the point where knock is over the threshold, allow a user defined variable to determine how much timing to pull. For instance say the knock threshold is 3 and we want to pull 1 degree of timing for every count. At 6 counts of knock this would pull 3 degrees of timing. If you were to enter a 0.5 in the variable for timing pull above threshold the ecu would pull 1.5 degree instead of 3.
3. You may also want to allow the user to edit the knock sum decay timer so that knock will bleed off faster. At engines make more power the stock knock decay becomes intrusive because it doesn't decay faster.
4. Ignore timing pull below a certain load range with a user defined variable. A 9:1 compression stock evo motor just really isn't going to ping and need timing pull on premium fuel in low load areas. 90% of the time this low load knock is not real. Even if it was it will not do any engine damage.
What all this will allow is for built motors to have an increased threshold before timing is pulled. My friends 2.3L stroker car for instance makes 3-4 counts of knock at full throttle no matter what boost or what timing we run. Thus we can increase the threshold as the background noise is higher than a bone stock 2.0L.
Having the ability to change the amount of timing being pull per count of knock allows you to vary the amount of protection you want to put on certain engine setups. Drag race cars for example can handle substantially more knock than a road race car.
Well thats my wishlist. Oh I'd also like a Startup Enrichment table to show up in the next version of your Tephra mods so that we can better account for switching from gas to E85 but that's slightly off topic.
If your looking for a wishlist to help people deal with 2.0L motors that cooperate, 2.0L motors that don't, and stroker motors that don't cooperate, here would be my wishlist.
1. Ignore timing pull up till a user defined moment knock threshold. It should still be able to be logged. IE: Ignore knock up to say 3 counts.
2. At the point where knock is over the threshold, allow a user defined variable to determine how much timing to pull. For instance say the knock threshold is 3 and we want to pull 1 degree of timing for every count. At 6 counts of knock this would pull 3 degrees of timing. If you were to enter a 0.5 in the variable for timing pull above threshold the ecu would pull 1.5 degree instead of 3.
3. You may also want to allow the user to edit the knock sum decay timer so that knock will bleed off faster. At engines make more power the stock knock decay becomes intrusive because it doesn't decay faster.
4. Ignore timing pull below a certain load range with a user defined variable. A 9:1 compression stock evo motor just really isn't going to ping and need timing pull on premium fuel in low load areas. 90% of the time this low load knock is not real. Even if it was it will not do any engine damage.
What all this will allow is for built motors to have an increased threshold before timing is pulled. My friends 2.3L stroker car for instance makes 3-4 counts of knock at full throttle no matter what boost or what timing we run. Thus we can increase the threshold as the background noise is higher than a bone stock 2.0L.
Having the ability to change the amount of timing being pull per count of knock allows you to vary the amount of protection you want to put on certain engine setups. Drag race cars for example can handle substantially more knock than a road race car.
Well thats my wishlist. Oh I'd also like a Startup Enrichment table to show up in the next version of your Tephra mods so that we can better account for switching from gas to E85 but that's slightly off topic.