Massive detonation, help!

Hi Guys, searched some and can't seem to find anything relating to my issue..

 

My 08 GT has recently starting developing severe detonation when going into boost, it literally falls on its face and sounds like a can of BB's getting shaken! I know these cars have knock sensors that is supposed to catch an audible knock well before the human ear hears it, so what can cause this? Yes I run Premium 93 octane fuel and It does this with any brand fuel.

 

I did a quick log with my ELM327 OBD module and everything appears normal, no lights on and no codes found.

 

This is a stock car, no access port, nothing.

 

I just put new plugs (NGK) in it about 15k miles ago and it is always well maintained.

waaait a second, are you describing the sound of a failed turbo?

 

 

Probably should stop driving and check that turbo.

Yep turbo has shit the bed. my 06 Sound like the BBs In a can right now. My turbo is toast. So it parked till I get all the parts and to the shop for build.

Nope, absolutely not the turbo. It was replaced approximately 30k miles ago and boosts just fine. It is certainly detonation under boost.

Boosts just fine... but falls on it face?

 

How are you monitoring boost? Does the elm327 monitor feedback knock correction , fine knock correction or DAM? If so you should see timing being pulled when you hear the noise.

 

I’d take the time to pull the downpipe and check for shaft play.

Then, check your long term fuel trims. You may have a massive boost leak or something. I would still check the integrity of the turbo though.

I'd stop driving the car until you figure this out. A simple issue could turn into something much bigger if you're not careful.

My 08 GT has recently starting developing severe detonation when going into boost, it literally falls on its face and sounds like a can of BB's getting shaken!

 

Nope, absolutely not the turbo. It was replaced approximately 30k miles ago and boosts just fine.

 

Falling on its face = no boost. The lack of boost is probably being caused by the CHRA contacting the housing b/c it has failed, that can of BBs sound you mentioned.

 

The turbo isn't boosting, the ECU is operating based on various inputs it expects to be present, boost is one of them. Might want to check the turbo and def do not drive it.

Just to clarify, I'm a pretty competent mechanic (yea I know, many say that) but I've been around turbocharged cars for many years and have built several race engines over my lifetime. I know what detonation is and what it sounds like. My turbo is spooling up fine, and when I log boost it goes very easily to 8-9psi before the detonation starts, then timing is pulled and it falls on its face. Sorry if my OP wasn't very clear.

 

I really think its a faulty knock sensor in combination with a lean problem. I don't have an AFR gauge but I've been in cars many times that went lean under boost. I plan to look at my fuel pressure and the FPRM this weekend, after reading that whole thread, I'm curious if I've getting the voltage needed at the pump.

 

The car runs perfectly fine otherwise until you go into harder throttle. I actually bought the car cheap a few years ago because the turbo had failed, kid had no idea and thought it was trash so I certainly know what that failure mode is and acts like!

make sure that your boost control systems are not compromised. check the WG/BPV and BCS and associated vacuum lines. you dont need and AFR gauge, just a OBD2 scan tool monitoring the front O2 senor since it is a wide-band sensor. How are you monitoring boost? do you have a mechanical gauge or are you reading a PID?

OK I'm not a competent mechanic and let me point out a few things you can easily fine with searching.

 

Turbo gets replaced, 15-30k later new turbo also goes because no one pulled the banjo bolts.

 

If you're getting detonation in boost and you keep driving around because "it drives fine" especially if you aren't staying out of boost, you're going to get cylinder 4 ring land failure if you haven't already.

 

Are you so good at determining knock that you know it's predetonation and not a thrown crank bearing?

 

These vehicles and all subaru turbo post 04 except the sti run lean in boost during initial boost due to dumb open loop delay.

 

The knock sensor doesn't prevent knock, it leans heavy on cylinder 4 due to location and only pulls engine timing but if you're knocking more than it can keep up with it doesn't mean you won't blow your engine. If you REALLY want to know how the knock sensor works here ya go.

http://www.romraider.com/forum/topic1840.htm

 

You've been given the answers already of what your first steps should be. If you don't want to believe it or troubleshoot it that's up to you. If you really think it's a faulty knock sensor or a lean condition other than the already know open loop delay. Then you need to get some logging from btssm or romraider.

 

 

Sent from my iPhone using Tapatalk

its the fix and learn scenario in owning an aspirated Subaru.

make sure that your boost control systems are not compromised. check the WG/BPV and BCS and associated vacuum lines. you dont need and AFR gauge, just a OBD2 scan tool monitoring the front O2 senor since it is a wide-band sensor. How are you monitoring boost? do you have a mechanical gauge or are you reading a PID?

 

Yes, I'm using an OBD monitoring system with boost, MAF, timing and air/fuel PID's all simultaneously. I'm well aware of the open loop delay also and see that but I think its more than the delay on my car. Initial low boost, it stays in the 14's but mine even stays there once I'm over 8-9 psi.

 

I did a long log yesterday on my ride home and recorded it, as soon as I figure out how to post the video of the screen record off my phone on here I will show you guys so you can maybe better see whats going on.

 

Once again, turbo is working normal and is not the issue, banjo screens were removed when I put the new turbo in.

What are your long term fuel trims (A, B, C and D)?

If turbo is good, vacuum lines are good with no leak, I would definitely check fuel pressure....I've scratch my head a lot before realizing the issue was related to a pressure leak with the new fuel pump (o-ring not seated properly).

 

Good luck!

Ok fellas, not sure if this is allowed but I put my log video up on youtube if any of you want to take a peek. Hopefully it shows enough valuable info to make some suggestions. This was my 15min ride home so it is comprised of a few light throttle runs through the gears and a few 4th gear pulls very lightly to build boost and show the turbo is working and that it stays mostly lean.

 

OEM air intake?

 

Are you certain the turbo was replaced with a genuine IHI unit?

 

Are you positive the car has never been tuned? Modded and returned to stock?

 

With very limited history on the car, this could be a number of issues.

 

A video including the sound you describe could help as well.

 

Also as BonBon was asking, what do your fuel trims look like? Those are a better indicator than the A/F ratio at any given time.

OEM air intake?

 

Are you certain the turbo was replaced with a genuine IHI unit?

 

Are you positive the car has never been tuned? Modded and returned to stock?

 

With very limited history on the car, this could be a number of issues.

 

A video including the sound you describe could help as well.

 

Also as BonBon was asking, what do your fuel trims look like? Those are a better indicator than the A/F ratio at any given time.

 

-Yes, ALL OEM, no mods, stock daily driver

-I used to work for a turbo shop, so yes, it was certainly an IHI that I ordered and installed

- Positive car was never tuned, purchased from an older couple, 2nd owner

-I'll try and get a video of it and capture the sound and events as soon as I can.

 

Keep in mind this just started approximately a month or so ago, the car has ran great, with the exception of the common "stumble" these are known for. Never an issue from it in many years.

 

Thanks for your help.

The log you provided does not have enough parameters to help us troubleshoot the problem. I really want to see the long term fuel trims. It will give us a clue as to whether or not you are suffering from either a vacuum leak or boost leak.

In addition, it would be nice to log the OCV (or VVT) angles to make sure they are in sync.

 

 

You should look into the BtSsm app with appropriate hardware.

 

 

In the meantime, you can try this:

 

 

Most vacuum leaks originating between the MAF sensor and turbo inlet will lean out the mixture, as outside are is drawn into the system beyond the point of metering. While boost leaks after the turbocharger typically make a rich in boost as air leaks out. Boost and vacuum leaks can occur in the manifold or tgv area which generate lean fueling in low load (vacuum) and rich fueling when above atmospheric pressure. Leaks will also effect the ecu's function as it tries to trim fuel to meet demands based on its sensor feedback. Learned fuel trims will often be greater than +/- 8. The key to optimal performance and reliability is to regularly check for boost and vacuum leaks. This test can be very simple and accomplished without many tools or a compressor.

 

Leak Test:

You will need, a flat head screw driver and a plug for the intake 2.75” OD (aerosol can, spray paint etc will work). You remove your intake from your inlet tube, then insert the can into the coupler just in front of the inlet tube. Next you remove the vacuum source line from the by-pass or blow-off valve and blow into the source line (which connects to the manifold). Try not use above 5 psi because the inlet, pcv, and valve breathers will be under pressure and could pop off or generate a leak. The breathers normally do not see boost pressure. I typically blow into the line with my mouth (lungs) because its quiet and easier to hear air coming out of the system. Once all leaks are contained I blow into it and put my thumb over the hose for 3-4 seconds, then release my finger, and it should still contain some pressure and you can hear it hiss back out at you. A smoke test is another good way to locate leak sources but a smoke machine is needed.

 

Above text quoted from https://legacygt.com/forums/showpost.php?p=4491795&postcount=1

I'll take a log today with fuel trims included, thank you all!

Quick update for those that care...

 

I did some logging with my Fuel trims this weekend, LTFT averages -4.5-5.5% and STFT average +5-6%

 

Not too bad right? Not indicative of any major issues at least in my research.

 

Any way, I dove in and took the turbo inlet piping off (damn what a job) and found the boot where it connects to the turbo very soft and partially torn. I've suspected this for awhile but just can't justify the cost for the replacement. Ended up finding a Rev9 (Perrin knockoff) and ordered that. Once I get it all back together I'll reset KAM and pray it fixes it.

 

Thanks for the suggestions guys!

OEM inlet is like $90 at a dealership iirc.

 

Hope you checked the turbo while the inlet was off and the wheel was in good shape? Get your fingers on it and feel around, take a video with your camera and the flash double check.

OEM inlet is like $90 at a dealership iirc.

 

Hope you checked the turbo while the inlet was off and the wheel was in good shape? Get your fingers on it and feel around, take a video with your camera and the flash double check.

 

Yes sir, verified it was 100%, no noticeable shaft play, all fins on the compressor wheel look perfect and it was surprisingly clean considering I'm still running the stock CCV/PCV to the inlet.

 

Oh, I only paid $89 for the Rev9 inlet so all good and its purdy blue!

Curious on the fitment and quality of the rev 9.

Been hit and miss from everything I've seen.