Need help with PK/ Mechanical Knock issues w/ DSMLink.

[project_tsi]

I'm getting knock everywhere. Up and down, left and right, round and round. I honestly believe its a mechanical (noise) knock that the sensor is picking up but I desperately need help.

First off, the knock sensor is about 1 year old, looks brand new and I have modified it, by welding the "ear" shut. Its snugged in and not overtight.

I have the knock control set at 3319 Rpm and at 50%TPS.

In any gear, past what my knock control is set at, anytime I am approaching boost the car just pegs the knock guage and pulls timing. I can even get the car to give a flash of full knock (43 cts) while FREE Revving. At about 4300-4700 or so, somewhere in there the car will flash the knock and give me 43 counts.

Take a look at the logs. A/F is good, Airflow is good, I've tried everything for timing from 12deg to 22deg.

The only thing I can think of this has to be some sort of mechanical knock, but the engine makes no actuall knocking sounds. I.E. Rod knock, etc. Its a new built 2.3L engine.

Is it possible the knock sensor has taken a shit and giving faulty readings?

Anyone have any advice as what to try next? I can NOT tune the car at all this way unless I dissable the knock senson and I don't want to do that.

FYI, the car did this last year also with the old engine so I want to rule out a noise comming from the engine itself.

With the logs to see the right A/F you may need to change your A/F ratio props to 950's w/ 330DT.

 

[scottsee]

Wow, I don't see anything that would attribute to those knock values. I don't reamber if the 1g's have the Rawknock 1 through 4 values available for logging. That would be extremal helpfully diagnosing your WOT. Nothing looks out of the ordinary with the exception of your 11.0AFR.dat file where AirflowPerRev, boost, Injon flatline. You're 100% possitive the knock sensor isn't over toqured? There's not a code in DTC box? Idle and cruise are not effect? I don't know what kind of time you have but changing out that knock sensor would be where I would start. Being nothing obvious points to a cause. Wish I knew the answer on that one.

 

[andymoraitis]

Dan,

Since there's nothing wrong with your settings in Link as far as we can tell, this would certainly point to a knock sensor as well. The fact that you can free rev it to 43 counts of knock is a clear indicator that something's wrong there. Having listened to the motor at idle and driven in the car (thanks a bunch for a great day!) I didn't hear any excessive lifter clatter or drivetrain noise that would make me think that your knock is real.

I'd swap the sensor and see if that works. Matter of fact, I'll pay for the new one so you can try it out and see what happens.

Andy

 

[project_tsi]

Wow, I don't see anything that would attribute to those knock values. I don't reamber if the 1g's have the Rawknock 1 through 4 values available for logging. That would be extremal helpfully diagnosing your WOT. Nothing looks out of the ordinary with the exception of your 11.0AFR.dat file where AirflowPerRev, boost, Injon flatline. You're 100% possitive the knock sensor isn't over toqured? There's not a code in DTC box? Idle and cruise are not effect? I don't know what kind of time you have but changing out that knock sensor would be where I would start. Being nothing obvious points to a cause. Wish I knew the answer on that one.

Yes, I am positive the knock sensor is not over tightened. I took it out early this week, welded the ear shut and put it back in finger tight, then snugged it up about 1/8th of a turn. However, I'm wondering if maybe the knock sensor was overtorqued previously? Could a knock sensor be damaged if previously overtorqued? I ask b/c this knock sensor I have now I pulled it off another car when I bought it as a parts car put the previous owner stated it was nearly brand new, and it does look to be. Oh and no, the car is throwing no DTC's at all. The car won't knock at idle unless I free rev it then it flashes a full count of knock for a split second at about 4500 and then dissapers while free revving. Cruise however will knock like its PK, which is why I have my link knock control setttings set to 3319RPM and 50%TPS. This means the ecu ignores any knock below 3319RPM's and also any knock when the TPS is at 50% or lower. If I change the setting back down, yes the car is affected durring part throttle like PK acts.

Thank you for your help so far.

 

[project_tsi]

Dan,

Since there's nothing wrong with your settings in Link as far as we can tell, this would certainly point to a knock sensor as well. The fact that you can free rev it to 43 counts of knock is a clear indicator that something's wrong there. Having listened to the motor at idle and driven in the car (thanks a bunch for a great day!) I didn't hear any excessive lifter clatter or drivetrain noise that would make me think that your knock is real.

I'd swap the sensor and see if that works. Matter of fact, I'll pay for the new one so you can try it out and see what happens.

Andy

Andy, its very weird how I can free rev it and it will show knock. I'll have to give credit to Jeremy for this. He said why not free rev it and see if it knocks with no boost load and so I did while logging in Link. Sure enough the car has no knock until about 44,4500 or so and then it flashed a nearly full count of knock on me for a split second then dropped back to 0. So somethings defineatly wrong.

Also, thats very nice of offering to pay for a knock sensor, but you don't have to do that. It is my car and I should buy the parts for it but thank you. But if you insist on it then I suppose there is nothing I can to do stop you.

Ok I'm off to work so I'll talk to you later bud.

 

[spyderturbo007]

I don't have DSMLink, so I can't look at the logs. But the easiest way for you to see what is going on is with a tank of race gas.

These days, with gas prices the way the are, you might have to give them your left arm, but it will tell you what's going on.

Also, how old are your motor mounts?

 

[andymoraitis]

PM me your PayPal and I'll get this done right away.

 

[andymoraitis]

I don't have DSMLink, so I can't look at the logs. But the easiest way for you to see what is going on is with a tank of race gas.

These days, with gas prices the way the are, you might have to give them your left arm, but it will tell you what's going on.

Also, how old are your motor mounts?

Nathan,

We tried higher octane and knock persisted so I'm really leaning towards a knock sensor issue. Also, the motor mounts are new prothanes. I know they transmit some vibration and I wonder if that isn't having an effect. I doubt it, but it's possible.

Andy

 

[spyderturbo007]

Nathan,

We tried higher octane and knock persisted so I'm really leaning towards a knock sensor issue. Also, the motor mounts are new prothanes. I know they transmit some vibration and I wonder if that isn't having an effect. I doubt it, but it's possible.

Andy

Ah, I didn't know that. If nothing changed when you used the higher octane fuel, then you are most likely dealing with mechanical knock or a knock sensor problem, as you suspect.

Did the problem start after installation of the motor mounts? I have heard of people having problems after installing prothanes, but it doesn't happen very often. If not, were there any changes that coinsided with the knock, or has it been an ongoing problem?

Also, are you running an external dump on the wastegate? I'm too lazy to go back and check his vehicle profile.

 

[project_tsi]

Right, octane didn't help either. I put in a can of the good stuff that actually works and it didn't help. About the prothanes, they are not new, they were in the car last year also. In all actuality, it seems the knock problems started last year when the install of the new turbo and all other mods went in, in the old engine, which is why I'm leaning toward it not being a noise comming from the engine itself.

The knock has been an ongoing problem and its driving me insane. The fact that the car will knock while free revving and while on the onset of boost (while building) leads me to believe its a noise issue. Yes, I'm running an external dump, its load, but the car will knock before the WG even opens, so that rules that out.

 

[steve]

First when did this start happening. You say about a year ago but it's not clear what parts of the car are the same and what parts are different. It sounds like you have a different engine in the car when it first started. Are you using the same knock sensor or did you put a different one in with the new engine.

You asked me is a bad ECU could cause this and the answer is sure but that's easy to test, put yuor ECU in somebody else car or swap a known good on into yours.

Steve

 

[project_tsi]

The knock came on strong about the 2nd day the car was running, but it also knocked like this last year. Yes, the engine is different than last year, its a built 2.3L now. I'm using the same knock sensor as last year also. Last night tried a buddies knock sensor and it didn't change, ruling out a faulty knock sensor.

I can't put in someone else's ecu b/c simply no one around me has a turbo 1st gen at all, let alone a 1g eprom ecu.