MMCD - lots of knock... suggestions?

[decathect]

Please take a look at the attached snippet of my mmcd log.

I'm getting lots of knock... on a pull up to 43 counts, and timing seems to be getting scaled back 10* or more. At that point O2-R seems a little low, what should it be at? I've been trying to diagnose an idle issue..

Any suggestions or help on this?

Thanks!

 

[rush2252]

Please take a look at the attached snippet of my mmcd log.

I'm getting lots of knock... on a pull up to 43 counts, and timing seems to be getting scaled back 10* or more. At that point O2-R seems a little low, what should it be at? I've been trying to diagnose an idle issue..

Any suggestions or help on this?

Thanks!

Do you have access to a scan tool? I see in your table that the TPS is @ 10% @ curb idle it should be at 0 +/- 2.

As for the knock it looks like its running rather rich when it hit 43 counts. Fix the TPS and then post another log

It would help to know the car/mods it has that your working on...

 

[decathect]

car is completely stock other than an axleback exhaust, stock boost, fuel system, etc. Why do I need a scan tool?

 

[rush2252]

car is completely stock other than an axleback exhaust. Why do I need a scan tool?

to adjust the TPS unless your data logger will read it as well. Im not familiar with what your running. Any rate set the TPS to zero at idle and then do another log post up. Are you running 91 octane?


the 1gs pull timing after 9 counts and you still havent stated which car and there isnt one listed on your profile...

 

[decathect]

shoot, sorry. It is late. the car is a 93 eagle talon TSi

I'm using MMCD on a palm pilot, with a cable to connect it to the data port.

I am running 91 octane only.

My logger does read TPS... that spreadsheet was taken from my logger output. I'll adjust the TPS down to 0 and post again. However I suspect that won't solve my knock issue.

 

[rush2252]

shoot, sorry. It is late. the car is a 93 eagle talon TSi

I'm using MMCD on a palm pilot, with a cable to connect it to the data port.

I am running 91 octane only.

My logger does read TPS... that spreadsheet was taken from my logger output. I'll adjust the TPS down to 0 and post again. However I suspect that won't solve my knock issue.

I know it wont solve the knock issue but its not helping that its reading 10% at idle.

but on the other hand....

Purpose
The Throttle Position Sensor (TPS) is a rotating type variable resistor. It senses the throttle valve angle.

Location
The TPS is located on the throttle chamber at the throttle shaft.

Operation
As the throttle shaft rotates, the output voltage of the TPS changes. The ECU determines the throttle valve opening based on the output voltage. The ECU computes throttle valve opening change (output voltage change) and judges the engine acceleration state and accordingly corrects fuel injection amount during acceleration - alldatapro.com

 

[decathect]

Yeah, I heard the same thing.

So that being said... what should I try? New O2 sensor? Can't hurt on a car with lots of miles on the current one.

 

[rush2252]

found this by googling the site... http://www.dsmtuners.com/forums/tun...3-counts-knock-extremely-high-fuel-trims.html
http://www.dsmtuners.com/forums/tun...157410-43-counts-knock-any-throttle-load.html

 

[rush2252]

okay this one bothered me so i did some reading. It appears that with the sensor unplugged you should pull 9 counts of knock. 43 is the highest the 1gs read @ the ECU.. possibly bad knock sensor? are you getting and CELs"?

 

[rush2252]

here this is the best link ive found yet... http://www.dsmtuners.com/forums/tun...ed-how-i-resolved-my-knock-phantom-knock.html

 

[decathect]

Seems to me that he just disabled his knock sensor by moving it to an improper location... thus not really addressing the knock issue.

 

[smokem]

Correct me if im wrong but im sure his tps is correct @ around 10% as he has a 1g, I believe that on 2G it should read 0%? I know on my logger i get 10% @ idle.

 

[rush2252]

yes he moved it.. but theres no possible way a car can get 43 counts of KP without you hearing it. And if you google it theres 400 DSM sites that people are experiencing the same problem> the odd thing your the only one thats still stock. I would check motor mounts loose brackets etc. this one has me stumped WTF


and yes the 2gs are set to 0 but so has every other car I worked on at idle... Mine is set to 2% because I have the 1g TB and I had to adjust mine in order to get the Idle set correctly.

 

[steve]

Do you have access to a scan tool? I see in your table that the TPS is @ 10% @ curb idle it should be at 0 +/- 2.

Any rate set the TPS to zero at idle and then do another log post up.

I know it wont solve the knock issue but its not helping that its reading 10% at idle.

Your confused, 10% is exactly what a stock 1G should read at idle. That's 0.5v out of the possible full scale +5v sensor voltage.

Correct me if im wrong but im sure his tps is correct @ around 10% as he has a 1g, I believe that on 2G it should read 0%? I know on my logger i get 10% @ idle.

This is 100% correct.

and yes the 2gs are set to 0 but so has every other car I worked on at idle... Mine is set to 2% because I have the 1g TB and I had to adjust mine in order to get the Idle set correctly.

You have a 2G and that's the difference. It's TPS is adjusted to where the IPS switches and the ECU normalizes the TPS output. On a 1G the TPS adjustment is for 0.48v to 0.52v output and the ECU outputs the raw values.

 

[decathect]

ah, ok. So my TPS is good. Lets move on... why am i getting so much knock?

 

[->PrOjEcTGS<-]

Phantom Knock. (Knock that's not real)

Lots of people have/get PK. A lot of people think it's their lifters. Pretty much your knock sensor is picking up a signal that you're detonating or preigniting your air/fuel mixture and sending it out to the logger. The reason I think it's PK and not real knock is that it's at partial throttle. I'd like to see a log of WOT from 2500rpm to Redline. PK at low load/partial throttle is not something to worry about. It's just annoying because it pulls your timing making your car feel like a dog!

Check your knock sensor. If the "goo" that is inside of it is flowing out and getting all over your block it's time to replace it. This makes your sensor more sensitive to sounds and vibrations.

Just so you know, if you ever decide to get a chip for your ECU to correct for injectors as well as other things, like one from Jeff at DSMchips.com, he has the ability to put a Phantom Knock code in the chip to ignore knock at low loads/before boost. It's an option if all else fails. I finally gave up and added it to my last bin and it works great!

 

[decathect]

sweet, thanks projectGS! I will look at the sensor, and look into getting a chip.

I'm also going to log a pull like you said, at WOT.

 

[habitatguy187]

I was getting pk and it went away after I ran seafoam through my intake mani. The seafoam quieted down my lifters which was evidently causing the pk. Might be worth a shot of your lifters are making noise.

 

[decathect]

Good call, I might do that today as well.

 

[->PrOjEcTGS<-]

While your at it I would also do a boost leak test to make sure you don't have any leaks. Also take a look at your fuel trims and see where they are. <100 the ECU is leaning the car out which means you probably have a boost leak. >100 the ECU is richening the car up which could indicate a post MAS pre Turbo Vacuum leak. Overall +/- 5-10% is good.

 

[rush2252]

Your confused, 10% is exactly what a stock 1G should read at idle. That's 0.5v out of the possible full scale +5v sensor voltage.



This is 100% correct.



You have a 2G and that's the difference. It's TPS is adjusted to where the IPS switches and the ECU normalizes the TPS output. On a 1G the TPS adjustment is for 0.48v to 0.52v output and the ECU outputs the raw values.

At the time of that post he hadnst stated what car he was working on just to clarify...

 

[decathect]

While your at it I would also do a boost leak test to make sure you don't have any leaks. Also take a look at your fuel trims and see where they are. <100 the ECU is leaning the car out which means you probably have a boost leak. >100 the ECU is richening the car up which could indicate a post MAS pre Turbo Vacuum leak. Overall +/- 5-10% is good.

I have boost leak tested the car a bunch, and fixed leaks. The only leak i have reamining is a leak on the turbo itself, haven't gotten around to taking it off and replacing the compressor gasket. Its not too bad though.

Which fuel trim should I be looking at? Low is 110.8%, Medium is 99.8%, High is also 99.8%. What is 'FT02'?

 

[habitatguy187]

Which fuel trim should I be looking at? Low is 110.8%, Medium is 99.8%, High is also 99.8%. What is 'FT02'?

Fuel trims are pretty good, and FTo2 is what the ecu is doing at that exact moment. Adding fuel or taking it away.

You need to read this The Essential Primer on the Diamond-Star Engine Control Units - Table of Contents

 

[decathect]

ooh, that looks like a good read. THanks!

 

[->PrOjEcTGS<-]

Those trims look fine.

From Pocketlogger.com:

O2 Trim:

The Low/Med/High Trims we see are long term trims i.e. they are 'dampened' quite a bit (move slowly) and maintained between starts. They are derived from the trend observed from the O2 trim which is a short term trim (moves quickly). The O2 trim is derived from comparing the actual O2v to what the ECU thinks the O2 should be (stoichiometric). The idea is to keep the O2 trim right. If there is a trend (lean or rich) in the O2 trim for a period of time, it is corrected via the appropriate (high, med, low) trim. Which trim is adjusted is based on the current airflow. This will allow the O2 trim to go back to its midpoint because one of the other trims is making the fuel delivery correction.