Welcome to the North American Subaru Impreza Owners Club Monday September 22, 2014
Home Forums WikiNASIOC Products Store Modifications Upgrade Garage
NASIOC
Here you can view your subscribed threads, work with private messages and edit your profile and preferences Home Registration is free! Visit the NASIOC Store NASIOC Rules Search Find other members Frequently Asked Questions Calendar Archive NASIOC Upgrade Garage Logout
Go Back   NASIOC > NASIOC Technical > Factory 2.5L Turbo Powertrain

Welcome to NASIOC - The world's largest online community for Subaru enthusiasts!
Welcome to the NASIOC.com Subaru forum.

You are currently viewing our forum as a guest, which gives you limited access to view most discussions and access our other features. By joining our community, free of charge, you will have access to post topics, communicate privately with other members (PM), respond to polls, upload content and access many other special features. Registration is free, fast and simple, so please join our community today!

If you have any problems with the registration process or your account login, please contact us.
* Registered users of the site do not see these ads.
Reply
 
Thread Tools Display Modes
Old 12-27-2012, 12:08 PM   #1
Nickcrsx
Scooby Newbie
 
Member#: 196071
Join Date: Dec 2008
Default 2008 STI - KNOCK, KNOCK, KNOCK - Any Ideas?

edit: friends are was fixed.

Turns out it was a bad spark plug. thank god!
* Registered users of the site do not see these ads.

Last edited by Nickcrsx; 03-06-2013 at 04:23 PM.
Nickcrsx is offline   Reply With Quote
Old 12-27-2012, 01:12 PM   #2
roadtrip1098
Scooby Newbie
 
Member#: 185688
Join Date: Jul 2008
Chapter/Region: NESIC
Location: Maine
Vehicle:
2006 STi
WR Blue

Default

I would do the compression/leakdown test to be sure.
roadtrip1098 is offline   Reply With Quote
Old 12-27-2012, 01:13 PM   #3
roadtrip1098
Scooby Newbie
 
Member#: 185688
Join Date: Jul 2008
Chapter/Region: NESIC
Location: Maine
Vehicle:
2006 STi
WR Blue

Default

Also with it being a brand new long block, have you done the first few oil changes yet?
roadtrip1098 is offline   Reply With Quote
Old 12-27-2012, 02:09 PM   #4
NoGraffix
Scooby Specialist
 
Member#: 276209
Join Date: Mar 2011
Chapter/Region: MWSOC
Location: Windy City
Vehicle:
I am Rub
Function>Form

Default

Did you check the voltages on the knock sensor? It honestly sounds like a fail sensor more then anything else. I can't imagine why the car would knock @ 8* of timing, in vacuum, at stoich AFR, with <1 g/rev load values; it just doesn't make any sense.
NoGraffix is offline   Reply With Quote
Old 12-27-2012, 05:39 PM   #5
Nickcrsx
Scooby Newbie
 
Member#: 196071
Join Date: Dec 2008
Default

I will try other alternatives prior to the compression and leak down. I dont suspect there to be a problem.

-Yes, all changes were done properly at the right interval with no strange occurrences.

-I did not check the voltages. Something that I can do via Romraider logger or with an actual volt meter?
Nickcrsx is offline   Reply With Quote
Old 12-27-2012, 09:14 PM   #6
jason02wrx
Scooby Newbie
 
Member#: 222893
Join Date: Sep 2009
Chapter/Region: NESIC
Location: Rowley, MA
Vehicle:
2012 WRX Hatch SWP
2002 WRB Bugeye

Default

Knock Sensor sir!

My 02 had the same issue as you. Started out at first only when high humidity/rain. As soon as you touch the gas it pulls -12* and DAM goes to ****. There will be a CEL soon...
jason02wrx is offline   Reply With Quote
Old 12-27-2012, 09:58 PM   #7
Uncle Scotty
Scooby Guru
 
Member#: 16200
Join Date: Mar 2002
Vehicle:
OK Houston
we have an Uncle

Default

take the oil filler cap off with the engine running

does it blow smoke???

yes = its done, stick a fork in it

Quote:
Only knocks between 2700-3300rpm under
this is classic rod knock symptoms


and you dont yank out the injectors to do a compression check
Uncle Scotty is offline   Reply With Quote
Old 12-28-2012, 07:37 AM   #8
FRWRX
Scooby Newbie
 
Member#: 15306
Join Date: Feb 2002
Location: FRANCE
Vehicle:
2011 STi
White pearl

Default

OK, just few questions :
The long block has been changed so you have removed the knock sensor from the previous engine and fitted it onto the new engine. Are you sure you use the correct torque to fit the sensor ?
You didn't log the VVT. Did you make any modification in the maps in regards to VVT ?
What about timming belt ?
FRWRX is offline   Reply With Quote
Old 12-28-2012, 08:42 AM   #9
Ziggyrama
Scooby Specialist
 
Member#: 53665
Join Date: Jan 2004
Chapter/Region: NESIC
Location: Northboro, MA
Vehicle:
2011 STI Stg2 w/EBCS
2012 Tribeca Ltd.

Default

Some food for thought, stock STI does not listen for knock below 0.9 or 1.0 load, depending in which direction you're transitioning in or out of the listening range. What are your threshold set at? Before everybody jumps down my throat on how crappy the stock tune is and all that, the low load timing maps are not that bad. So, why Subaru would do that? It's possible that the sensor may be picking up lots of false knock under low load due to resonances in the engine bay and sounds being interpreted as knock. The knock sensor is nothing more than a microphone whose signal is fed into an analyzer that looks for knock signatures in the feedback. OP, search for false knock. You'll see many threads of people pulling their hair out about this. Low load, cruising situations, crazy FBKC. Makes no sense. Some anecdotal evidence, I tried lowering my listeing thresholds down and it produced exactly what you're describing. My logs started to show major knock events and ECU constantly pulling timing at loads <1.0. ECU would pull timing, more knock, more pull, more knock. That was a year ago. I went back to stock level thresholds as there seems to be more to it that meets the eye. I am actually willing to give Subaru the benefit of the doubt on this one. Now, I am not saying that you're fine. Just pointing out that it's possible that you're chasing a ghost. But, you should investigate to make sure you don't have a problem. If you have an EGT gauge, look at your temps during those events. If they're high, you may not have enough timing.
Ziggyrama is offline   Reply With Quote
Old 12-29-2012, 06:06 AM   #10
jaxscuby
Sammo Hung
Moderator
 
Member#: 10613
Join Date: Sep 2001
Chapter/Region: South East
Location: S Ga / Hell Rock
Vehicle:
2002 USDM WRX
WRB

Default

if its not done..blown up..piston failure due to a lack of fuel.

i would double check anything that the engine is bolted to
or bolted to it.
jaxscuby is offline   Reply With Quote
Old 12-29-2012, 11:12 AM   #11
Nickcrsx
Scooby Newbie
 
Member#: 196071
Join Date: Dec 2008
Default

Quote:
Originally Posted by jason02wrx View Post
Knock Sensor sir!

My 02 had the same issue as you. Started out at first only when high humidity/rain. As soon as you touch the gas it pulls -12* and DAM goes to ****. There will be a CEL soon...
Thanks for the post! The day I got in the car it was rainy as hell and super humid. I will take a look.

Quote:
Originally Posted by Uncle Scotty View Post
take the oil filler cap off with the engine running

does it blow smoke???

yes = its done, stick a fork in it



this is classic rod knock symptoms


and you dont yank out the injectors to do a compression check
Good simple test. And yes I know how to do a compression test, haha. I was simply referring to pulling the injector to make sure the o-ring was properly seated. Will do this first thing. Thanks Uncle!

Quote:
Originally Posted by FRWRX View Post
OK, just few questions :
The long block has been changed so you have removed the knock sensor from the previous engine and fitted it onto the new engine. Are you sure you use the correct torque to fit the sensor ?
You didn't log the VVT. Did you make any modification in the maps in regards to VVT ?
What about timming belt ?
Awesome question! VVT is stock and timing belt is brand new w/ all components. As for re-checking the knock sensor, I will take a look ASAP. You are correct in that the knock sensor is from the old block, and over time may have developed a problem from being removed and re-installed.

Quote:
Originally Posted by Ziggyrama View Post
Some food for thought, stock STI does not listen for knock below 0.9 or 1.0 load, depending in which direction you're transitioning in or out of the listening range. What are your threshold set at? Before everybody jumps down my throat on how crappy the stock tune is and all that, the low load timing maps are not that bad. So, why Subaru would do that? It's possible that the sensor may be picking up lots of false knock under low load due to resonances in the engine bay and sounds being interpreted as knock. The knock sensor is nothing more than a microphone whose signal is fed into an analyzer that looks for knock signatures in the feedback. OP, search for false knock. You'll see many threads of people pulling their hair out about this. Low load, cruising situations, crazy FBKC. Makes no sense. Some anecdotal evidence, I tried lowering my listeing thresholds down and it produced exactly what you're describing. My logs started to show major knock events and ECU constantly pulling timing at loads <1.0. ECU would pull timing, more knock, more pull, more knock. That was a year ago. I went back to stock level thresholds as there seems to be more to it that meets the eye. I am actually willing to give Subaru the benefit of the doubt on this one. Now, I am not saying that you're fine. Just pointing out that it's possible that you're chasing a ghost. But, you should investigate to make sure you don't have a problem. If you have an EGT gauge, look at your temps during those events. If they're high, you may not have enough timing.
This post is epic,thanks so much.

Few points:
-Knock that low in the G/rev range doesnt make sense. The problem is that I kept the thresholds at a stock level. Now the car has full torque solutions motor-mounts. Now they never produced a problem before I changed the injector o-ring so I dont suspect they are the issue. I have checked the engne bay many many times for something causing ambient noise levels that the knock sensor would pick up and produce knock like symptoms.

-As for EGT's being high, you bet! I guarantee that EGT's are massive due to IAM being at zero and running base timing right now. Just to be safe I am not going to disable the knock sensor under 4100rpm and run the risk of potentially melting a valve.

-Your post has me thinking. Chasing ghosts is typically what I do. Ha. I am going to take a triple check at whats going on in the engine bay and look for rattles, etc. From there I think I will take a look at the actually knock sensor and see what the hell is going on.

Quote:
Originally Posted by jaxscuby View Post
if its not done..blown up..piston failure due to a lack of fuel.

i would double check anything that the engine is bolted to
or bolted to it.
Thanks. And always a possibility.
Nickcrsx is offline   Reply With Quote
Old 12-29-2012, 02:40 PM   #12
Nickcrsx
Scooby Newbie
 
Member#: 196071
Join Date: Dec 2008
Default

Update- no smoke from oil fill cap as I expected. Noticed that I am seeing knock in lower and lower load sites. Really think this knock sensor is on its way out.
Nickcrsx is offline   Reply With Quote
Old 01-02-2013, 08:28 PM   #13
Nickcrsx
Scooby Newbie
 
Member#: 196071
Join Date: Dec 2008
Default

Some new data.

Decided to LOG pretty much everything. It all looks Kosher and perfectly normal. The bucking is getting worse. No oil consumption, no external strange engine noise, no smoke out the oil cap, or exhaust.

I just dont get it. The car hasn't thrown a CEL but is nearly undrivable under load. Knocking has pretty well stopped due to IAM at .650 (i assume bucking is due to lack of timing)

Havent tested the knock sensor....but I should see the problem in the data. I see f%$* all.

Experts.....replace knock sensor? Burn car to the ground? Fuel trims are complete INSANE!

https://docs.google.com/open?id=0B-D...2RqNFFIU0JZekU
https://docs.google.com/open?id=0B-D...EZreE1SZ2ZiSFk
https://docs.google.com/open?id=0B-D...GsxWEloT0NnX0k



Thanks,
Nick
Nickcrsx is offline   Reply With Quote
Old 01-02-2013, 10:00 PM   #14
WRXSteye20008
Scooby Specialist
 
Member#: 175565
Join Date: Mar 2008
Chapter/Region: MAIC
Location: Lancaster, PA
Vehicle:
2014 Sport Premium
Past: 351awhp Sti

Default

If you are stilling having the issues and nothing is showing up in logs, why not just replace the knock sensor and rule that out? They run around $100. Check the voltage first and proceed from there. Aside from your engine insides wanting to be on your outsides, my guess is the sensor as well.
WRXSteye20008 is offline   Reply With Quote
Old 01-09-2013, 01:14 PM   #15
STi*guy.kiev
Scooby Newbie
 
Member#: 117854
Join Date: Jun 2006
Location: KIEV UKRAINE
Vehicle:
2003 spec c

Default

nasioc is definately not the place to get some info. UNLESS you want to be missled)
/returns to romraider
STi*guy.kiev is offline   Reply With Quote
Old 01-09-2013, 01:35 PM   #16
Ziggyrama
Scooby Specialist
 
Member#: 53665
Join Date: Jan 2004
Chapter/Region: NESIC
Location: Northboro, MA
Vehicle:
2011 STI Stg2 w/EBCS
2012 Tribeca Ltd.

Default

Quote:
Originally Posted by STi*guy.kiev View Post
nasioc is definately not the place to get some info. UNLESS you want to be missled)
/returns to romraider
I know....because no one on RR forum ever comes to this site
Ziggyrama is offline   Reply With Quote
Old 01-09-2013, 02:49 PM   #17
bmx045
Scooby Newbie
 
Member#: 262684
Join Date: Nov 2010
Location: Wakefield, MA
Vehicle:
2011 MCS

Default

Quote:
Originally Posted by Nickcrsx View Post
Some new data.

Decided to LOG pretty much everything. It all looks Kosher and perfectly normal. The bucking is getting worse. No oil consumption, no external strange engine noise, no smoke out the oil cap, or exhaust.

I just dont get it. The car hasn't thrown a CEL but is nearly undrivable under load. Knocking has pretty well stopped due to IAM at .650 (i assume bucking is due to lack of timing)

Havent tested the knock sensor....but I should see the problem in the data. I see f%$* all.

Experts.....replace knock sensor? Burn car to the ground? Fuel trims are complete INSANE!

https://docs.google.com/open?id=0B-D...2RqNFFIU0JZekU
https://docs.google.com/open?id=0B-D...EZreE1SZ2ZiSFk
https://docs.google.com/open?id=0B-D...GsxWEloT0NnX0k



Thanks,
Nick
I'm here from RR forums

Sorry man, not sure what logs you're looking at but none of those are perfectly fine, actually they're the exact opposite. You're maxing out the FLKC threshold even at an iam of 0.625, which happens to also be the boost control cutoff point because your ecu is trying to protect the engine. Compression test it. Your fuel trims are showing that fuel is being pulled because the ecu is detecting an overly rich afr, which can happen with low compression as well. Do the compression/leak down, and boost leak test and come back, then we'll move onto other culprits.
bmx045 is offline   Reply With Quote
Old 01-09-2013, 04:13 PM   #18
NoGraffix
Scooby Specialist
 
Member#: 276209
Join Date: Mar 2011
Chapter/Region: MWSOC
Location: Windy City
Vehicle:
I am Rub
Function>Form

Default

Quote:
Originally Posted by Nickcrsx View Post


When you do that the accuracy of the data you see goes out the window, so get some logs with parameters that will actually help you evaluate how the car is running and leave a bunch of stuff off. I'm not sure what you mean by the second statement because everything likes completely wacky lol. You need to do what I suggested in my original post, check the voltage of the knock sensors (with a volt meter, sorry I didn't get back to you work got busy) and see if they are function properly, if not replace them. From what I can see this is your obvious problem and is completely distorting the tune on the car. With all the FLKC your timing is literally all over the place, which simply put is in no way good for your engine.
NoGraffix is offline   Reply With Quote
Old 01-09-2013, 04:18 PM   #19
Nickcrsx
Scooby Newbie
 
Member#: 196071
Join Date: Dec 2008
Default

Quote:
Originally Posted by bmx045 View Post
I'm here from RR forums

Sorry man, not sure what logs you're looking at but none of those are perfectly fine, actually they're the exact opposite. You're maxing out the FLKC threshold even at an iam of 0.625, which happens to also be the boost control cutoff point because your ecu is trying to protect the engine. Compression test it. Your fuel trims are showing that fuel is being pulled because the ecu is detecting an overly rich afr, which can happen with low compression as well. Do the compression/leak down, and boost leak test and come back, then we'll move onto other culprits.
I agree with your first sentence. The rest I would agree if it had shown symptoms of down compression. ie. excessive smoking, oil consumption, audible noises.

I feel like its strongly correlated to the injector in that cylinder. Somehow it could have gotten damaged or at that specific pule the spray pattern could be bad. Either way I will be pulling the plugs to analyze what they look like. If I see something unusual I will then yank the corresponding injector and send it off for analysis.

Heres my thought process - 1 cylinder isn't getting enough fuel (somehow) causing detonation while the others are over compensating to achieve a proper AFR. This is why I am seeing insane long term fuel trims but perfect AFR's.

of course I will do a compression test while in there as well.

If that doesn't solve the issue I will check out the knock sensor.

If that fails, ill set the damn thing on fire.
Nickcrsx is offline   Reply With Quote
Old 01-10-2013, 08:38 AM   #20
bmx045
Scooby Newbie
 
Member#: 262684
Join Date: Nov 2010
Location: Wakefield, MA
Vehicle:
2011 MCS

Default

the knock sensor isnt hard to get to, just pull the ic off. torque the sensor to spec, it's the easiest of the possible issues... and cyl 1 is furthest from the knock sensor. remember how detonation can occur- poor fuel quality, too much timing, too little fuel, poor compression, hot charge temps. your knock correction is massive, more than just the typical pesky knock. a false knock- like loose bolt on the engine somewhere could cause this. or poor engine grounds too.... check your engine grounds, there should be ground straps attaching to the left and right frame rail and to each head.
bmx045 is offline   Reply With Quote
Old 01-13-2013, 11:16 PM   #21
Nickcrsx
Scooby Newbie
 
Member#: 196071
Join Date: Dec 2008
Default

Does anyone know what the knock sensor torque specs are?

Removed the injectors tonight. Obviously looks great. Sending them off for inspection and to be flow tested tomorrow. Will remove the spark plugs tomorrow night and see what they look like. Then address the knock sensor and grounds.

Will also do a compression test while I have everything out. That should pretty well remove any possibility of a fuel delivery, spark to compression issue.
Nickcrsx is offline   Reply With Quote
Old 01-14-2013, 12:06 PM   #22
bmx045
Scooby Newbie
 
Member#: 262684
Join Date: Nov 2010
Location: Wakefield, MA
Vehicle:
2011 MCS

Default

17.4ft-lb
bmx045 is offline   Reply With Quote
Old 01-15-2013, 12:05 AM   #23
Nickcrsx
Scooby Newbie
 
Member#: 196071
Join Date: Dec 2008
Default

Injectors are out and off for testing tomorrow. What's interesting here is that all 4 plugs looked awesome. No sign of lean or rich mixture. I think I maybe chasing the wrong ghost. At any rate they need to be tested and cleaned to rule them out.

Wednesday I will do a compression test on the 3 cylinders I can get to just to make sure.

As for the knock sensor. This sounds been but I can't locate it. Anyone have some good pictures of its location to share?

Also, if I can't figure this out, my next guess is that possibly the timing belt slipped a tooth perhaps? Does this sound like the type of symptoms you'd see?

At any rate I has to bent the snot out of the bracket that surrounds the rails and the damn rails themselves to get the injectors out. I hope to god I didn't create more problems.

Any comments let r rip!

-nick
Nickcrsx is offline   Reply With Quote
Old 01-15-2013, 08:54 AM   #24
SoapBox
Scooby Specialist
 
Member#: 204578
Join Date: Mar 2009
Chapter/Region: MAIC
Location: Philadelphia
Vehicle:
S54 M3, N54 135i
In another life: REX8

Default

IIRC, isn't the sensor bolted to the block, under the intake manifold?

SoapBox is online now   Reply With Quote
Old 01-15-2013, 08:58 AM   #25
Nickcrsx
Scooby Newbie
 
Member#: 196071
Join Date: Dec 2008
Default

Thank you sir! looks easy enough to get with a socket.
Nickcrsx is offline   Reply With Quote
Reply

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Forum Jump


All times are GMT -4. The time now is 12:05 PM.


Powered by vBulletin® Version 3.7.0
Copyright ©2000 - 2014, Jelsoft Enterprises Ltd.
Powered by Searchlight © 2014 Axivo Inc.
Copyright ©1999 - 2014, North American Subaru Impreza Owners Club, Inc.