Lift off knock (cure?)
Thread Starter
Evolved Member
iTrader: (5)
Joined: Oct 2006
Posts: 3,805
Likes: 2
From: Sacramento, CA
Lift off knock (cure?)
So I've been doing many runs getting my AFR curve perfect. When I switch to the next gear quickly and accelerate i get a very small lean spot and a huge amount of knock which starts to fall off.
So the question is what would you do?
I'm only guessing here that the ecu isn't keeping up with the ramp up on load so when i get back on the throttle before the injectors fully kick in its lean. The way I'm going to initially try and deal with this is start adding fuel into the 140+ load areas up to the point where it doesn't happen.
The point of knock happens just after i get it in gear and mash the gas again, the engine rpm is just about to start a very rapid deceleration.
So the question is what would you do?
I'm only guessing here that the ecu isn't keeping up with the ramp up on load so when i get back on the throttle before the injectors fully kick in its lean. The way I'm going to initially try and deal with this is start adding fuel into the 140+ load areas up to the point where it doesn't happen.
The point of knock happens just after i get it in gear and mash the gas again, the engine rpm is just about to start a very rapid deceleration.
Mine has been on lift off, not on going back on the throttle. I could cure it only with retarding in the 140-220 load zones.
If it is going lean when going back on the throttle, then acceleration enrichment would be what I'd change, but I don't know that anyone understands the units in ECUflash.
If it is going lean when going back on the throttle, then acceleration enrichment would be what I'd change, but I don't know that anyone understands the units in ECUflash.
Thread Starter
Evolved Member
iTrader: (5)
Joined: Oct 2006
Posts: 3,805
Likes: 2
From: Sacramento, CA
Well the cure for me ended up blocking out my fuel map from 5500+ with the same 11.1 AFR fuel map numbers from load cells 120 up. Looks kinda funky but definitely keeps my AFR in the good between shifts.
Hi John, I think you're saying that off-throttle knock is real? I had to reduce my MIVEC timing to reduce knock, but there is still some knock left (running 91 octane). It seems that I almost have to take it down to stock to get rid of knock. Any comments you have would be very helpful. You da man!
It is real knock - you can hear it in the cabin if bad, or in det cans. It does respond to timing or octane changes. It improves at lower charge temperatures or lower boost levels. Some say it does no damage because the cylinder pressures are low but I don't like knock sums of 12 to 15 and hearing my engine knock in any circumstances. After the throttle shuts there is still some air/fuel mixture in the inlet manifold that will be ingested. The timing advances too quickly on lift off, ideally we'd change the attack rate of the timing on lift off.
Trending Topics
Back from the dead.... I did some logging at a recent autox. I was on pump gas with a safe detuned map. Occationally I get up to 6* of knock when going from 100% TPS to 85% TPS. This would only last for a split second.
Looking at AIRFLOW vs TPS you can see when I let off the throttle the turbo was still sending air into the motor at a high rate. I'm since I'm still at 85% TPS the car still gets the airflow.
My accelleration is completely clean but my partial throttle high load provides a blip on the radar screen when easing off the throttle. I wish we had decell enrichment!
Looking at AIRFLOW vs TPS you can see when I let off the throttle the turbo was still sending air into the motor at a high rate. I'm since I'm still at 85% TPS the car still gets the airflow.
My accelleration is completely clean but my partial throttle high load provides a blip on the radar screen when easing off the throttle. I wish we had decell enrichment!
Thread Starter
Evolved Member
iTrader: (5)
Joined: Oct 2006
Posts: 3,805
Likes: 2
From: Sacramento, CA
Back from the dead.... I did some logging at a recent autox. I was on pump gas with a safe detuned map. Occationally I get up to 6* of knock when going from 100% TPS to 85% TPS. This would only last for a split second.
Looking at AIRFLOW vs TPS you can see when I let off the throttle the turbo was still sending air into the motor at a high rate. I'm since I'm still at 85% TPS the car still gets the airflow.
My accelleration is completely clean but my partial throttle high load provides a blip on the radar screen when easing off the throttle. I wish we had decell enrichment!
Looking at AIRFLOW vs TPS you can see when I let off the throttle the turbo was still sending air into the motor at a high rate. I'm since I'm still at 85% TPS the car still gets the airflow.
My accelleration is completely clean but my partial throttle high load provides a blip on the radar screen when easing off the throttle. I wish we had decell enrichment!
Yeah, so I went way back to find this thread. This is kind of in answer to the "Knock on deacceleration" thread, which got me thinking about this. I ran into throttle lift knock a little while ago. I got it at high load levels around 3-3.5k rpm. Basically if I left it in 6th to pass someone on the expressway, as I let off the throttle, the turbo would still be spooled and the airflow would still be high, but the 2-byte load would follow the throttle and decrease. The timing would increase, and the car would knock.
Has anyone figured out a patch to fix this? I'm thinking no, since it requires disassembling either the load, timing, or fuel algorithm. I think this is actually a mistake by Mitsu engineers.
Hmmm... I wonder if we could lobby to get this fixed under a TSB.
Has anyone figured out a patch to fix this? I'm thinking no, since it requires disassembling either the load, timing, or fuel algorithm. I think this is actually a mistake by Mitsu engineers.
Hmmm... I wonder if we could lobby to get this fixed under a TSB.
I logged and logged and logged..... but couldn't narrow down the timing/fuel values to fix the lift off knock on pump gas. I know its real and not mechanical because it goes away with race gas.
I very seldom pound the car on pump gas so its not much of an issue just an agitation. Planned poundings are accompanied by 100 octane making knock a thing of the past.
I very seldom pound the car on pump gas so its not much of an issue just an agitation. Planned poundings are accompanied by 100 octane making knock a thing of the past.
It's most likely preignition then since it doesnt occur under load and is remedied with octane. Preignition happens from combustible gasses remaining after the stroke that ignite durring compression stroke. Excessive NOx, bad fueling/timing or excessive boost exh manifold psi can cause this.
Colder plugs will also stop preignition but not detonation.
Colder plugs will also stop preignition but not detonation.
You can drop the timing in the areas where the engine revs and load drop during shifts and you would never normally be anyway. This seems to make a difference on the cars I have tuned and had issues with (25 counts between shifts). Though this is octane related I have heard diff bushings clunk so loud that it registered as knock. I assume it was resonating through the driveline.
Something like this (Ignore the rescale as that was long ago and not the point here):
Something like this (Ignore the rescale as that was long ago and not the point here):
That's about where I went with it, although not to the extreme that map does. The biggest event I saw was about 8 counts before I pulled the timing. Other than that I would get 1-3 counts pretty regularly. Since I reduced the timing, I haven't seen another big event, but I still see the 1-3 counts occasionally (7-8 times in a 1-hour log).






