The bottom line is after a month of banging my head and trying every known fix from race gas, e 85 and changing the knock sensitivity ect the knock was gone in 1 run soon as we changed the exhaust mivec. I dont care what fixed it i just know im chomping at the bit to make some power now. I was very happy with the power levels from wastegate only. The erl kit is about to show us whats up. So far just the gains from e 85 and the stroker and head build are impressive. I was impressed to see the turbo still making power at 7500+ rpm. I think it will work out just about perfectly with my rpm limitations.

 

3dman1 do me one favor. Quit posting on the forums and get that sick and twisted tuner working his magic on that car.

Ya know I'm just bored at work and am ready to hear you make some power

 

Cleaned as requested. Gentlemen, let's please keep this one on topic and listen to those who actually resolved the issue and how they did it. There's no further need for speculation, she's all fixed and knock free!

 

Well, it turns out that we can still produce massive (presumably false) knock. It happens when going WOT at the rpm range for peak torque. There's little or no knock if starting above or below that RPM range. 3dman1 thinks he's hearing surge, but its not clear to me. Guess we've got some more investigating to do...

 

Any update to this?
I am facing the exact same problem on a built 2.0....

 

most likely you'll need to adjust the knock sensor filters to accommodate the additional noise levels of the new engine.. but before doing that, you should check for any loose parts that could also be causing the noises before doing that. loose heat shields, rubbing parts, etc.

before adjusting the sensor though, research into how the sensor actually functions so you get a better understanding what you're doing. jcsbanks has detailed information regarding the function of the sensor on the older evos, the x seems to use the same setup.

 

+1 for this. There more to it than just this though. With a tip from Hollywood_X, we found out that another contributor to the massive knock was having different high and low octane ignition tables. As counter-intuitive as it may sound, setting the low octane timing table to be the same as the high octane table makes a big reduction in knock. Do this before tweaking the knock sensor tables.

 

Yeh I reported that a while ago.. But I thought it was only for low-load situations... Not once you started getting into boost.

I will have to review the code - again

https://www.evolutionm.net/forums/ev...-knocksum.html

 

I call it perpetual retardation
(Just got pointed over here to explain)
What I've noticed happening is whenever you get any hint of knock during tuning and drop timing it starts a downward spiral. False knock - pull timing and add fuel - causes more knock - pull more timing and more fuel - so on and so on. These cars hate being too fat and despise too low of timing.

What I have been doing is keeping the tables below 100 load identical and only 1-3 degrees different on the others. Once you have a solid tune in place then you can create a larger difference between the maps and it will run normal.

 

I've been doing this for well over a year now. That is how I always set up my Low Octane maps.

Remember that the ECU only uses these tables when it sees knock for a while and starts to lower the "octane number" or "percent high octane timing" value. Which is very uncommon in a tuned car.

 

yeh "octanenumber" is pretty useless on our cars...

i mean sure you could get a bad tank of GAS, but with KnockCEL you will know about it straight away...

 

This is very good to know....guess it's time to revisit my LO timing table

Thanks!

 

And it takes a lot, and I mean a LOT of knocking before that octane number starts to go down. Any normal driving and it goes right back up to 100. It's really a last ditch effort to prevent total melt down of the car. If your octane number is less than 100 for longer than a minute, you are in a world of hurt. I doubt an engine would last more than a week under those conditions.


I'll also add this... The Low Octane tables combined with the Octane Number value are a sort of self fulfilling prophecy. Octane number goes down, ECU starts using the low octane maps, fuel increase, timing decreases, knock goes away, octane number rises, knock returns. Either way the car has issues.

So... moral of the story is that Low Octane maps should be identical to each other in the less that 100 load values and roughly 3 steps down in the greater than 100 load values. Just copy and paste over the high octane tables and select the greater than 100 load values and hit your ']' bracket a few times. Done.

 

2 successive counts of 6 or more will start to pull it down.

but it goes down SO SLOWLY, that its a pointless system.. the lowest I have every seen is like 94%, which if you interpolate 94% between low and high, you still get high..

 

I added to my post above. Not that it changes anything, just that you probably missed it. I noticed your reply after my edit was posted.