The Central Hub for DSM Community and Information

For 1990-1999 Mitsubishi Eclipse, Eagle Talon, Plymouth Laser, and Galant VR-4 Owners. This is where the DSM platform history is documented and archived. Log in to help us in our mission, and to remove most ads from the browsing experience.

KNOCK (Yes I know, it's getting old)

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

90 GSX

15+ Year Contributor
322
2
Jun 24, 2004
Fremont, California
I don't think my sig has been updated recently but I'm pretty sure it includes all pertinent mods. I have cut my timing down to 13* max and my a/f ratio is a steady 10.5:1 across the board. Even then, I am still getting anywhere from 0-20 counts of knock under full throttle. It's enough to limit my max timing to 10 or 11*. Sometimes I can pull without knock (or just very little) and sometimes it feels like superman is holding my car by the rear bumper. Keep in mind I am on 91 octane with 2g pistons, but I would think that I could get a little more out of it. I mean, I am tuned as conservative as I can get and it's still pulling timing. Any more timing pulled and the car won't move.

There is no magic on my car - its speed density. The computer only does what I tell it to do. Boost leaks, exhaust leaks, any of those problems are no longer an issue. The car has a steady a/f ratio at WOT and my timing is very conservative. Why I am knocking. I ws contemplating alky injection but I don't want to do that if I have a serious problem. Oh and all of this is about 20-22 psi.
 
You run 20-22 psi on a holset hx35 on piss water cali gas? WTF

Time for water/meth injection. You can also try to replace or test the knock sensor, or see if it is just a phantom knock.

With 91 you are limited, you can pull more timing and dont let the ecu add timing or find better gas.
 
Get some injection of something or go to a better fuel. Try maybe lower boost and see what happens. My car is very touchy with the 2g pistons. But I dont have timing control.
 
The best way to deal with knock is to start thinking about it in common sense terms. what causes knock, and is there anything you are doing that could cause knock?

Are you using good fuel?
Are the insides of the enigine is reasonable shape?
Excessive boost?
Excessive timing?
Bad AF ratio?

If you can eliminate those items, the only thing left is phantom knock. Then you have to decide how to deal with that. That part can be frustrating. At 18 years old, sometimes its near impossible to work with the stock system anymore, due to wear and age, and you may want to replace the knock circuit with something newer.
 
Take a look at your logs and see where the knock starts. Check the rpm, a/f ratio, and timing just before knock starts. Sometimes knock will start if timing is high, or a/f ratio is lean for about one second before knock starts. Check to see if the timing and a/f ratios are still in the middle of the maps, they might not be at the highest load level yet. If it is, you'll need to adjust the middle portion of the maps where knock starts.

Make sure you're not still in closed loop as boost starts. I know some eproms will have the 'enhanced high speed closed loop' feature added which can keep you in closed loop a little too long. You don't want the O2 volts cycling around .02 to .9 as boost hits.
 
How long have you been tuning with DSMap? Could your hard VE be way off? Maybe also try a colder heat range of plugs.
 
Take a look at your logs and see where the knock starts. Check the rpm, a/f ratio, and timing just before knock starts. Sometimes knock will start if timing is high, or a/f ratio is lean for about one second before knock starts. Check to see if the timing and a/f ratios are still in the middle of the maps, they might not be at the highest load level yet. If it is, you'll need to adjust the middle portion of the maps where knock starts.

Make sure you're not still in closed loop as boost starts. I know some eproms will have the 'enhanced high speed closed loop' feature added which can keep you in closed loop a little too long. You don't want the O2 volts cycling around .02 to .9 as boost hits.

Everything he just said ^^ is exactly my experiences.
I had the extended closed loop in my chip, and on the high way in 5th gear, with my setup, my little 16g is trying to build boost before I even go into open loop, I could literally be riding down the highway boosting 5psi the whole time and not even know it if I didnt look at my boost gauge, and it causes it to run into some nasty knock. I had to pull out a lot of timing and fuel in that area. Even though only at partial throttle, running down the highway at 80-85mph under 4-6lbs of boost for 2-3 miles at a time in closed loop isnt a good thing...

Also, even though you max timing sounds very conservative, If you run into some bad knock at say 4k rpm, it can chase you thru the whole pull and make your whole tune seem sloppy, when really it was just that one spot. Hell, it could even chase you into the next gear....Were exactly does the knock begin.
For awhile I thought I couldnt dial in any decent timing on my car due to knock, but later I found that i actually could get away with a LOT of pump gas timing ( 93oct though ) just by changing the shape of the curve.

Therefore I started my timing curve as low as 9 degs at 4k, BUT; by 7k Im already seeing 19-20 degs due to how smooth my curve is and how safe my afr is. Were as previously, starting with timing at say 12 degs would cause it to knock the whole pull even with less overall timing...
(I run about 23psi falling to 20 redline btw.)
Its all in the shape of the timing curve, you should naturally be able to increase timing more than you have now anyways due to the engine speed increasing,
and I think 10.5 afr is too rich; my car had rich knock issues before as well so dont take that out of the possibility.

Some other things that helped me dial in a little more out of my engine,
for one, make sure you dont have boost leaks, and that your not heatsoaking everything by not having a exhaust heatshield or etc .
I used the seafoam to clean up some bad carbon deposits I have, and I changed my spark plugs to the NGK BR7ES.
I had BRP6ES in it before and I mean changing plugs wont let you add timing or anything but it made a significant difference in the random knock counts.
So make sure your plugs are non projected ( projected means they have a P in the code)

If I were you, Id lower the boost to say 17-18 psi, revamp your entire timing curve, lean it out to a good 11.1-11.2 afr, and then do some logs. Then if the logs look good start slowly adding boost back in, even with the 2g pistons I think you should still be able to get out a good 21-22 psi of boost with 11.1 afr and up to 16-17 degs timing by redline on 91 piss fuel. And I wouldnt worry about anything under 9-10 counts of knock, as long as it never goes above that.

Well thats all I can think of, hope any of it helps.

CAn you post a log for us?


+1
 
I would still like to see a log...can you post us one???? really all we can do is stab in the dark without it and make you try EVERYTHING, if we had a log one of us could maybe spot the problem and get you going in a few hours one time instead of trying this and that with no results

+1 more vote for "extened closed loop" being dangerous. I'm always tuning to get the best economy i can and i played with letting closed loop function up until 30% throttle and one day on the expressway i was on a slight incline and only 27% throttle i hit over 15psi at 14.7:1 being targeted by the ECU, not a good AFR for that much boost on piss-gas 91 ( i was data logging the drive because i was setting up the closed loop )

Also, try going leaner to about 11 - 11.5:1 - I have gotten knock from being rich into the low 10's before and that was on my J&S not the dramatic stock KS

ONe way to tell if it's PK, piston slap or real knock would be to wait till you're super low on gas and throw in a few gallons of race gas. You can get 112 octane for about 5-6 bucks a gallon...C16 is rediculousely expensive compared to some other race fuels, or even go to the local small air strip and buy 10 gallons of low lead aviation fuel..i ran it all summer one year.

Then run the same settings and see if there's less knock, no knock or if it stays the same.

A.) less knock... well, real detonation is occuring and you need to address the common causes (AFR, timing, hot spots, heat soak etc..)

B.) stays the same..... PK, start hunting down things to dampen then KS or change the coding to filter it out some more (there was a post on here about it a few weeks ago)

*C.) Test: Free rev the motor with quick jabs of the throttle from idle, make sure you give it enough that the TPS will exceed 1.75v IIRC ..I use half throttle jabs up to about 4k... if you see knock come on and then die off as the revs slow down, then that's most likely piston slap. Now there does seem to be some cams that are noisier than others (from what i have read on here) and I've heard "rumor" that noisy lifters can set off the KS (but i have no solid proof) So those are some more things to take into consideration.

Hope you get it worked out ...AND POST US A LOG ASAP!!!!!!!!!!!!!!!!! :D
 
Add Value - Be Respectful - No Trolling - No Misinformation - Participate Often!
Support Vendors who Support the DSM Community

Build Thread Updates

Latest Classifieds

Back
Top