Subaru WRX Forum banner
1 - 14 of 14 Posts

·
Registered
Joined
·
31 Posts
Discussion Starter · #1 ·
After months of researching and troubleshooting this issue on my own, I feel like I've run out of options, so maybe someone here can point me in the right direction. I will try to be as detailed as I possibly can, but it's always difficult to adequately explain issues like this to someone who can't physically see what's going on with the car. I will include a full list of every modification already done to the car in regards to the engine and drivetrain, as well as a list of what has already been replaced to try and rectify this issue and maintenance that needed to be performed anyway. If I left anything out, feel free to ask.

Car: 2006 WRX
Mileage: 87k
Mods:
VF39
STI TMIC
Tomei turbo inlet
AEM cold-air intake
Cobb 3-port EBCS
Tomei equal length exhaust manifold
Grimmspeed EWG uppipe
Tial MV-S 38mm EWG
Invidia catless downpipe
TGV deletes
AEM 320 fuel pump
Perrin air-oil separator
Fluidampr crank pulley
Group N motor mounts
Perrin transmission mount
Perrin pitch stop mount

Currently tuned to 19 PSI by Eric Minehart at Torqued Performance

Replaced:
All timing belt pulleys, tensioner, belt, and water pump
All rear engine seals
Engine wiring harness
MAF sensor
Injector o-rings
Intake manifold and TGV gaskets
Intercooler y-pipe and BPV gaskets

I had a smoke test performed and my mechanic and I were able to confirm that there are no vacuum leaks. All grounding wires are good and snug. There are no counts of cylinder roughness during the knock events, AFRs are normal, no issue hitting target boost, and DAM stays at 1. No exhaust leaks. No CELs. Oil changes have always been done no later thank 3500 miles and only with high quality oil.

Now, onto the issue. For the past few months, I have been experiencing excessive feedback knock, with values ranging from your normal -1.4 to -2.8, as high (low?) as -11.8. These knock events only occur immediately following shifting from 2nd to 3rd gear, at approximately 25-40% throttle, and between 2200-2500 RPMs. Never above 3000 RPMs. If I shift slowly and roll onto the throttle, no knock event occurs. I've discussed this at length with my tuner and he says it is most likely false knock, but of course it still worries me. Additionally, I have been experiencing cold start issues, but I am unsure if they are related. In cold weather (>40 degrees), there is a knocking sound coming from the engine bay, but it never lasts for more than 30-45 seconds at the absolute most. I've seen quite a few people say that this noise can be attributed to piston slap, which from what I understand can be a common occurrence on older Subaru motors. My car is 12 years old, so that seems to logically follow. Apart from that, however, the car will sometimes sputter and almost stall on start-up, with choppy idle for about 60-90 seconds before it stabilizes. I have also noticed that even with the TGVs deleted, the car still runs lean on cold starts, about 16.5:1. And most recently, my car will occasionally experience a delay of about 1 to 2 seconds between turning the key in the ignition and actually turning over. There are no sounds whenever this occurs, just silence and no response. I replaced the starter in the event that perhaps the solenoid on my old one was going out, but it has still happened a couple of times since then. Once the car gets up to operating temperature, none of these issues persist except for the feedback knock

Thank you for reading. Any insight would be greatly appreciated.



Sent from my 2PS64 using Tapatalk
 

·
Super Moderator
Joined
·
3,376 Posts
I will allow someone more knowledgeable to answer on most of these issues. Supplying a datalog will help tremendously with us assisting you as to what the issue may be.

But I will add this, the knock sensor on these cars is famous for picking up false knock (loose heat shield rattling, a/c compressor kicking on and off, etc). But I had a similar issue with my Torqued Performance map on my last wrx. ***Now before I speak any further, I am not bashing Eric or TP in any way as many people are tuned by him and have had satisfying results.***

But, when I had my TP tune on my 11 wrx I was experiencing very similar issues with the feedback knock. It was literally all over the place. I ran the map for about a year and never had any real issues with the car other than the feedback knock issue. Eric at the time also wrote this off as false knock (which it may have very well been). But about a year after using his maps with a few different changes from Eric, I added some new parts and took the car to IAG for a protune. After the protune the feedback issues were gone.
 

·
Registered
Joined
·
31 Posts
Discussion Starter · #3 ·
I will allow someone more knowledgeable to answer on most of these issues. Supplying a datalog will help tremendously with us assisting you as to what the issue may be.

But I will add this, the knock sensor on these cars is famous for picking up false knock (loose heat shield rattling, a/c compressor kicking on and off, etc). But I had a similar issue with my Torqued Performance map on my last wrx. ***Now before I speak any further, I am not bashing Eric or TP in any way as many people are tuned by him and have had satisfying results.***

But, when I had my TP tune on my 11 wrx I was experiencing very similar issues with the feedback knock. It was literally all over the place. I ran the map for about a year and never had any real issues with the car other than the feedback knock issue. Eric at the time also wrote this off as false knock (which it may have very well been). But about a year after using his maps with a few different changes from Eric, I added some new parts and took the car to IAG for a protune. After the protune the feedback issues were gone.
You know, you're not the first person I've heard about who had issues with a TP map. Eric has been my tuner for about a year and a half now, and he's been great (quick turnaround time, answers questions really fast), but I'll admit I have found myself wondering if this is a tune issue. I'll see if I can find the last set of logs I did for this current map and post them here. If I can't find them then I'll do some tonight and have them posted tomorrow.

Edit: I've gone over everything in the engine bay (multiple times) even remotely close to the knock sensor that could be causing vibrational feedback but found that nothing was loose.

Sent from my 2PS64 using Tapatalk
 

·
Registered
Joined
·
31 Posts
Discussion Starter · #4 ·

·
Not a mod... or is he?
Joined
·
15,530 Posts
For some reason the logs won’t open for me.

There could be a few things at play. Generally when the engine is at low rpm and high load knocking is going to be more prevalent. This is typically normal especially right after a shift where the rpm is at the lowest. Coupled with the rough idle thing you may well have an issue somewhere.

You’ve checked vacuum and boost. Have you checked your compression and have you completed a leak down test? That may be enough to cause a misfire and wonky idle but it may be good enough you don’t notice under higher load.

If the tgvs are deleted that may be your idle problem. If not I would almost suspect a leak on one of them.

Then there is the false knock issue. That’s not unheard of for an ej. It will give you the results you are seeing and generally shouldn’t hurt anything.
 

·
Registered
Joined
·
124 Posts
If you are reading a knock of -11.8 you would hear it with your own ears. Radio off, do you actually hear the motor ping at all? If not, it is most likely a false knock of some sort.
I would agree with this. -11.8 you would most definitely hear. False knock is common with low engine loads. If you're getting out of first into 2nd, and don't add enough gas, you will see knock. I would only worry if your WOT and you're seeing knock events occur.
 

·
Registered
Joined
·
31 Posts
Discussion Starter · #8 ·
If you are reading a knock of -11.8 you would hear it with your own ears. Radio off, do you actually hear the motor ping at all? If not, it is most likely a false knock of some sort.
Nope, I've never actually heard any audible knock whenever the knock events occur. It's probably a little bit of false knock and a little bit of my motor is getting tired. Either way, I'm getting a motor build this summer.

Sent from my 2PS64 using Tapatalk
 

·
Registered
Joined
·
1 Posts
I know this is an old thread, but please tell me you found a solution to your problem.

I have a 2017 with the same issue, except when on a 93 Cobb map I get DAM drops and FKL. On 91 map I don’t.
2021 STI here, Cobb 91 map and low load does the same. Its the ecu doing its job honestly. Had a tuned WRX and I could feel a big difference from an aggresive tune to a full good etune, but tamer. If you see the numbers going back to 0 quickly, its just ghost knock.
 

·
Registered
Joined
·
24 Posts
I'm trying to figure out a similar situation currently. 2005 FXT with some STI parts (new pistons, 51k mile VF48, older STI TMIC), remachined B25 heads (guides were loose), new crank and bearings in a 13k mile 705 case, catless DP & UP, stock fuel system and airbox, heat rating 7 sparks, TGV delete. Compression ratio is a little higher due to heads being shaved twice now. I think the new STI pistons didn't change the compression with B25 heads unlike my 2007 FXT where new STI pistons raised compression by 0.4 on the W25 heads.

It is inaudible knock but I'm not 100% I would hear knock over the exhaust and the random rattles of the car on the broken up icy roads that we have right now. It does have a Magnaflow muffler and a resonator, but it's catless and the muffler has a patched hole in it so it's not good.

Piston slap: Not sure if this is part of it, but this engine might have more slap than most. The top ring gaps are normal values (0.018"), not Subaru tighter-than-every-other-car-made-to-make-sure-ringlands-break-as-much-as-possible 0.008". This might result in piston slap, but I can't hear it - it just sounds like a Subaru to me. Also, these pistons were in the car and it wasn't knocking prior to the STI turbo and TMIC upgrades, so I don't think the pistons are slapping to the point of causing knock or triggering the knock sensor.

So I've been trying to tune the knock out. Here is what I've done, though I've tried a LOT over the past couple days (just myself, no dyno, so it's slow-going).

I've lowered base timing quite a bit and still get inaudible knock at low load, mostly right after shifting into 2nd but also into 3rd sometimes. If I slow down in 2nd or 3rd to low RPM like 1800, then accelerate at similar throttle input, the knocks don't happen. For now it happens exclusively when shifting gears.
AFR is doing okay on the shift, not hunting around too much or like too much tip-in enrichment, but it does seem to be cutting fuel as soon as the knock occurs, probably because it's cutting timing while the MAF remains the same. The tip-in is still set to 2005 OEM levels.
I've raised the minimum RPM and load for FKC, which helps but it still starts detecting knock at like 0.80 engine load just like it was at 0.60, maybe not as much like -9 instead of -12. Here is a pic of the latest log when it knocked in a lazy 2nd gear acceleration, log from BtSSM unfortunately since RomRaider likes to fail to log after I use ECUFlash 1 time and don't reboot:
Font Rectangle Material property Parallel Pattern


That -6 FKC would have occurred sooner and probably -9 when min load was at 0.60, but this is with it at 0.8 so as soon as it hit 0.8 load, it started reporting knock and pulling timing. I don't think 30° of advance is too much to ask at this level. The OEM tables and most tables I find on the RomRaider forums would be trying at least 40° of advance right there.

A little later in that same drive, here is a 3rd gear knock that's basically doing the same thing:
Colorfulness Font Material property Rectangle Parallel


My feeling is that it's false knock, but I don't want to disable the low load FKC completely. Today I'm playing around with the AVCS tables, grasping at straws really, wondering if it has any impact on this. I'm going to zero the table out, like maybe the intake cams are opening too early and causing something with the combustion, though I highly doubt it.

Prior to the latest upgrades, this car had a TD04 and stock intercooler and would had very little knock with quick a bit more ignition advance. I could only get that turbo to hit 11psi though (3200' elevation), with a fresh rebuild even, while the tired 193k mile TD04 in my 07 FXT hits 16psi no problem.

I have tried to make it knock where I would normally expect it, like 67% throttle at 2600 RPM in 4th or 5th gear, but it has almost no knock there, just some random -1.4 readings as expected. So I'm not too worried about harm to the engine at this point.

I guess I'm looking for advice from someone that knows how to tune for knock on these cars or anyone that has been through a lot of knock like this and knows more about it. This is my first time dealing with it out of 6 turbo Subarus that I've tuned.

I'll keep trying stuff and researching, but I'm getting frustrated at this point and would love some help. Sadly, I haven't found any forums where tuners hang out and discuss topics with each other. It's like a mad science where everyone holds knowledge close to the vest and won't share it, which I understand if you are a tuner trying to set yourself above others to try and entice more customers. But really at this point, I thought the open-source tuning community would be thriving, especially after Cobb got wrecked by the EPA.
 

·
Registered
Joined
·
106 Posts
lightened flywheel or damper can cause false knock. the ecu monitors crank speed and if it does not match what it thinks it should be it assumes "knocks"
the lightened rotating assembly spins a bit faster than normal
i looked in your mod list and you did not mention the clutch/flywheel setup
 

·
Registered
Joined
·
24 Posts
Thanks for the reply.

The car has an OEM pulley and ACT Streetlite flywheel. It wasn't knocking like this with the TD04 and OEM TMIC, and with the not-very-sealing valves on the heads prior to them getting machined, at least I'm pretty sure it wasn't. When I logged data on it before, it would randomly throw small FKC like -2.8 at low speeds, probably when lugging it.

Next I'm going to check the torque on the knock sensor and swap it out for another one I have to see if there is any difference. I need to eliminate that as a possibility asap.

Earlier, the STI TMIC was basically touching the pitch mount brace, so I added some padding temporarily to stop any vibration there. I am going to hunt down random vibration sources and try to eliminate them. I am pretty sure I felt something rattle at low RPM a few times. This car has quite a few rattles compared to my other Forester, not really in the engine bay, but I need to dive into those.

I probably should do more troubleshooting before posting so much on here. :) I would have checked compression again if it wasn't such a PITA and my hoist being unavailable so I can't lift the engine up to make it easier. I'll have to lift it from below, which I know is tricky since there aren't many places to push up on.

It very well could be real knock and issues with rings/pistons/cylinders. I'm not being blindly optimistic, but am always curious what others might have to say.
 
1 - 14 of 14 Posts
Top