Trouble, IAM@0.875 was 1 10 days ago

[keithk831]

Thanks, is there any evidence that there might be boost creep ?

 

Its hitting 18.5 psi now near 3300 rpm in 3rd.

 

Thanks, seems like most of the knock is from idle and sitting in traffic....which the ECU should adapt to....

 

Yeah, it adapts by pulling timing and dropping your IAM...

[mwiener2]

Really, show me a log of that.

 

The higher the load, the higher the threshold the ECU needs to consider X db as knock.

 

I have never seen this, and the evidence points more towards the ECU seeing it is knock at low load, not WOT.

 

The knock sensor is a microphone that has been tuned to the engine and exhaust. It is extremely sensitive to the point that once you modify even the catback, it is no longer trustworthy. Sure, it will probably hear knock events, but it may also report false knock.

 

I'm sure YOU have never seen it, but do you work with as many as 4 cars per day?

 

 

If you don't believe me on this, do some research on Motec's testing of knock control systems and look up how much it costs to have a knock sensor properly programmed.

[LittleBlueGT]

The knock sensor is a microphone that has been tuned to the engine and exhaust. It is extremely sensitive to the point that once you modify even the catback, it is no longer trustworthy. Sure, it will probably hear knock events, but it may also report false knock.

 

I'm sure YOU have never seen it, but do you work with as many as 4 cars per day?

 

 

If you don't believe me on this, do some research on Motec's testing of knock control systems and look up how much it costs to have a knock sensor properly programmed.

 

You can say whatever you want, but you are still wrong.

 

The ECU applies a higher noise floor threshold the higher the engine load is. I have never ever ever seen any evidence that a bumpy road can cause false knock.

 

You can go on and on about how our knock sensor (a simple mic that the ECU filters for about 6400 Hz and then subtracts a varying noise floor from based on engine load) is not very good. But the facts are that a ton of tuners quite successfully use our knock sensor to tune all sorts of cars, and especially if the internals are stock, it works very very well.

 

I know you see lots of cars, but it is obvious you DON'T TUNE them.

[LittleBlueGT]

Yeah, it adapts by pulling timing and dropping your IAM...

 

 

That is true.

 

If the DAM (or IAM) is falling, then you need to pin-point why.

 

Our ECU has a number of tables that define at what load ranges and at what rpm what form of knock control strategy the ECU uses. W/o seeing the tune, and w/o seeing it in logs it can sometimes be hard to pin-point.

 

ie: if the load range that the ECU adjusts IAM is low enough, simple low load noise (in my case a noisy clutch disc) can drop IAM.

 

Here is the table I am talking about:

 

http://i248.photobucket.com/albums/gg194/littlebluegt/IAMloadrange.jpg

 

In RR it is all in one table, ATR uses two (one for the low end, one for the high end).

 

So in the ATR table, we see that if you get knock right around the 1.0 load area, it could affect your IAM (this is slightly over-simplified, as there is a bit more to it), in the RR table you can see that a change in IAM will happen above about 2.0 load.

[Ridgeracer]

I suspect that the table above still has stock values. What's interesting though is that during idle in the logs, the car randomly sees knock. Any LWFW or when was the last time you changed plugs? Do you have a nice smooth idle without misfires when warmed up? And FWIW, I had to change my knock sensor once however I got a CEL.

[mwiener2]

You can say whatever you want, but you are still wrong.

 

"A competent tuner needs to carry out a frequency analysis to determine the target frequency and knock module settings. "

 

That is directly from the MoTeC website. Every time you change the sound of the drivetrain, you need to recalibrate the knock sensor. Which is something maybe 1 person on this forum could do, maybe.

 

Something as simple as a cat-back will change the sound enough to require recalibration.

 

I'm not saying everyone should freak out that their sensor will never record knock properly, but don't trust it so closely.

[Deadleave]

I suspect that the table above still has stock values. What's interesting though is that during idle in the logs, the car randomly sees knock. Any LWFW or when was the last time you changed plugs? Do you have a nice smooth idle without misfires when warmed up? And FWIW, I had to change my knock sensor once however I got a CEL.

 

No light weight flywheel, And for spark plugs, not sure....only if it's part of the 30k service performed by subaru.

 

My idle is rough now in colder weather, and I do not think knock go away once warmed up. According to logs....even when car is up to temp, it still knocks at idle a lot.

 

Stock tables?, this is upsetting , after~$900 dollars just in tuning.

 

Sigh...

[mwiener2]

you need a new tuner. Closest one I know off the top of my head is Agile in MD

[Deadleave]

Well logged for 2 hours today....

 

I deleted most of the cell with no significant activity but here are the results.

 

For most stage 2 tunes, on stock VF40.....wher do you guys have your boost ceiling set?

Target boost (requested via tune) goes up to 18.44

 

And yet still LV stays pretty clean, but again I was driving pretty normal, and I average 19~20 mpg............today on fresh tank, reset trip computer its read 16.5 mpg. no idea what going on with fuel now.

 

Man when its good is good, but when its bad, its like abyss of problems.

 

romraiderlog_20101002_122556.csv

 

romraiderlog_20101002_123701.csv

 

romraiderlog_20101002_125636.csv

 

Thanks for everything guys.

 

Of all the parts purchased by this car, that $15 ebay vag-com cable has been the best thing ever.

[LittleBlueGT]

Why don't you diagnose the issue.

 

Where was there knock sum increments at higher load?

-that is where your concern should center

[Deadleave]

I am trying to.

 

I know ignoring "noise" at idle for now, there seems to to area of knock.

First being in 1600-1800 rpm area, at medium load.

2nd at higher load 3000-3200 rpm. It most likely knocks at higher rpm and load too, but I haven't pushed the car fear of damaging the engine even more.

And another thing I noticed was knock occurring when ever I lifted off the throttle.

 

Sorry if I came off as someone just looking for others to solve my problem. I am not, I am trying to get an idea of what is happening; to be able to understand more based on more experienced users. So when I do talk to my tuner, I can give him more useful information than "it doesn't work right", and also not come off as someone making statements about something he doesn't fully understand.

[Ridgeracer]

He should be able to understand the logs you have taken. That way he can pinpoint where the knock is in the timing map. The changes are simple. As for boost 18.44 psi peak on tri-state 93 is fine..

[LittleBlueGT]

I am trying to.

 

I know ignoring "noise" at idle for now, there seems to to area of knock.

First being in 1600-1800 rpm area, at medium load.

2nd at higher load 3000-3200 rpm. It most likely knocks at higher rpm and load too, but I haven't pushed the car fear of damaging the engine even more.

And another thing I noticed was knock occurring when ever I lifted off the throttle.

 

There is a start.

 

You can try adding torco, or some REAL octane booster, and see if it goes away. If you can get it to repeat easily on pump gas, but then it goes away on higher octane fuel you know that it was real knock.

 

The knock that I saw in your logs at about .90 load is something I would not worry about. It seemed to come when the throttle was in transition.

 

It honestly looks to me (from the logs you have sent) that there is maybe just one or two areas where your tuner was maybe running a touch too much timing, nothing major. I think your motor is fine.

 

FWIW I have yet to see a local stock car run a IAM of over 0.750.

 

Do some more WOT runs (start small if you are worried, like 2000-3500 rpm, then to 4000 etc....) and post up the logs.

 

Maybe do some 30% throttle logs first and post them up.

[Deadleave]

I think the engine is done, I thank everyone for the help and good knowledge for the future.

 

I am making boost, still knocking some, but now....I have bunch of new symptoms and problems.

 

1. Burning oil, 500 since last oil change (synth 5w30), and dip was on E hot, added a quart.

 

2. Driving 120 miles on fresh tank of gas, on mild tame driving, <3500 rpm.

120 miles and needle read 1/2 tank.

 

3. Exhaust is caked in black soot,

 

Still, no smoke, and no loud noise, going to Wednesday to have the engine looked at.

50k on the engine......

[Ridgeracer]

You should check your oil on a level surface after the engine has cooled.

[Infamous1]

I think the engine is done, I thank everyone for the help and good knowledge for the future.

 

I am making boost, still knocking some, but now....I have bunch of new symptoms and problems.

 

1. Burning oil, 500 since last oil change (synth 5w30), and dip was on E hot, added a quart.

 

2. Driving 120 miles on fresh tank of gas, on mild tame driving, <3500 rpm.

120 miles and needle read 1/2 tank.

 

3. Exhaust is caked in black soot,

 

Still, no smoke, and no loud noise, going to Wednesday to have the engine looked at.

50k on the engine......

That doesn't sound like a blown typical blown engine. If you were burning that much oil you'd see it.

[Deadleave]

Well back, and the shop said everything sounds fine. And its most likely all in my head.

 

Since the ECU reset,

 

MPG is back up to 20.x but I haven't been driving hard much...

I will check the oil, in about 200 miles to see how its holding, but burning oil is normal for turbo Subaru's, not as much as it does, and it maybe the turbo slowly dying. What he did was cold crank the car, and listened for pitch change to check that ringlands were ok, which was actually kind of cool.

 

As for the boost, its still hits max at 3200 rpm, but my butt no longer feels it. Car feels NA.

 

And lastly he says its normal for IAM to move around a bit, thats car adjusting to different driving conditions and driver.

 

Well, I guess next to install the following/buy whats missing.

PERRIN TMIC, have

245 AVO pump, have

injectors, need

BNR evo16g, need

e-tune, lined up

 

 

Again thanks for all the help, and advice, I learned a lot.

[Deadleave]

 

Sorry to bring up an old thread, do I have a bad 02 sensor ?

 

Not sure what the CEL code means in plain English.

[black318i]

That doesn't sound like a blown typical blown engine. If you were burning that much oil you'd see it.

 

 

I went through a qt in 200 miles and didn't notice any smoke and I was looking. It was clean at idle too. It wasnt until somone else drove my car when I noticed the smoke.

[mwiener2]

[ATTACH]92955[/ATTACH]

 

Sorry to bring up an old thread, do I have a bad 02 sensor ?

 

Not sure what the CEL code means in plain English.

 

 

 

with the problems you've had, probably... but that starter switch code is odd

[Deadleave]

Trying to get it taken care of, but someone told me something to check out.

 

Since it has happened since reflash (on TG day), that that learning view could have been taken while the diagnostic plug was enabled. The code hasnt reappeared again since, but my tuner said AF voltage was out of whack, so having it replaced to be safe right now.

[Deadleave]

I am so retarded.

 

The o2 sensor was fine..........

 

After the tech didn't see the CEL, and I tried to show him the LV screen....

 

Lesson learned

 

Don't take LV screen caps with those green diagnostics connectors connected.

I think only the 05-06 have these.

[McDowell Performance Tunin]

The newer cars' have them as well

[mwiener2]

they just don't need to be connected for flashing

[rock2534]

So just as a rule of thumb, what would be a real bad number for IAM and Knock Sum?