low boost - 6psi or so?

[GTEASER]

Did you check the vac lines under the TMIC?

 

Do you have an AccessPort to see knock, MAF etc so you can get some clues?

 

Have you cleaned your MAF sensor?

[DrD123]

The dealer tech is contacting Subaru to open a techline and see how they want to proceed. I didn't go over the lines under the intercooler, but will ask if the tech did at the dealer (they have had it for almost 2 days) - all of the o-rings/seals are new, so it's unlikely that there is an issue, plus if it were a leak, I would expect to see evidence of that in the fuel trims, and they are normal (bouncing around close to zero). if there is a leak between the MAF and the turbo, I should be running lean with positive fuel trims, if there is a leak between the turbo and the throttle body, I should be running rich, and see negative fuel trims. The service manager verified that the system is detecting knock, and that's what's putting the car into limp mode.

 

I don't know what could be causing the knock (nor does the dealer) - it's conceivable that the same fuel truck delivered to both Chevron stations, and both have messed up gas, so it's bad gas, but I think that's unlikely.

 

Plugs are relatively new, and were working fine before. There really shouldn't be carbon deposits of any significant degree as the block is new. they verified there's no damage from the airbag install to the ECU or associated cables. For the knock sensor, they said it would throw a code if it were not operating properly.

 

versus asking us here and shooting down every opinion that is given to you.

I am not "shooting down opinions" - if a suggestion is offered and I have already done testing to verify that's not it, I am pointing that out. I don't have the OEM coupler which is known to leak - I have a Perrin silicone one, with t-bolt clamps on either end - it is seated completely on the intercooler and onto the throttle body and the clamps are tight. I appreciate the thoughts - that is what I am looking for - and also why I am trying to be as clear as possible with what I have done.

[Fabsx]

So if everything you said is true,

 

Here’s the most possible reasons to where the fault is.

Bpv/bov

Wastegate issues

Clogged cat

 

Bov/bov, replace with the aftermarket one you have.

 

Wastegate, test function with air compressor and see if it moves.

 

Cat, not sure, aftermarket downpipe with cat?

 

If fuel trims are good, thats the only thing I can really say

 

 

Sent from my iPhone using Tapatalk

[DrD123]

why would I lie?

 

 

Spoke with the tech, and they did a pretty thorough evaluation - all mechanical systems are fine. If you take the BCS out of the equation, it boosts right up to 15psi, and he didn't feel comfortable going higher. When it's connected properly, the wastegate opens properly (again - all mechanical components have been tested, leaks have been inspected for, etc.). I am not sure what your line of thinking is with the the BOV - they verified all of the mechanical aspects they could get their hands on are working properly. If it was leaking badly, it would have issues building boost, but without the engine keeping the BCS closed, it boosts just fine - also, after a computer reset, it boosts just fine. They also even brought in the local subaru tuning guru (Cam from CamTuning) - a plugged cat would have other symptoms, and would tend to make things bog down.

 

 

currently, they have opened a techline with Subaru, and when they hear back (Friday or Monday) will follow direction from Subaru. The only data that has come back out of whack is that the engine is detecting knock, and putting the car into limp mode. The only hypothesis on the table that hasn't been disproved conclusively is the fuel issue - I have a full tank and it's 100 degrees out, so they requested that I drive it a bit and run the fuel level down so that they can more readily drain and replace it with a known good. While it seems unlikely, maybe the filling truck messed up at multiple stations and put 87 in the 91 tank...

[Val Demar]

Bad gas is certainly a possibility - I had the issue on Friday, though, and then filled the tank on Saturday and it was still doing it Monday. I got gas at two different Chevron stations, and 15 days apart (I don't drive a whole lot), so the odds of it being bad gas are low, I think.

 

I get bad knock with Chevron gas on my Tribeca, well, got it two or three times from gas stations in different parts of California, I stopped using Chevron after that, haven't heard knock since.

[DrD123]

If it turns out to be the gas, I might just be done with the Chevrons in town! Thing is, probably 75% of my fill-ups have been from Chevron and I've never had an issue. Only place I have actually had a less than good experience from was Costco and that was a number of years back with my WRX (low power and lower fuel economy after trying them several times)

 

 

Waiting to hear back from the dealership as to when I need to bring the car back in.

[moral hazard]

How did it turn out ?

[DrD123]

Sadly, it has not turned out, yet. On Thursday the dealer indicated they had still not heard back from Subaru, so were contacting them again. I have been driving the car, and trying to stay out of boost. It drives just fine, and actually, after driving it for a while I can get the same effect as resetting the computer (so it lets me go to higher boost once or twice, then back to limp mode) - so it must be something related to while being at high boost. (it's detecting knock when under boost) - I am wondering if the knock sensor wasn't damaged when being moved over to the new block - maybe they overtorqued it and the signal from it is a mess? Still running down the gas (down at a quarter tank or so now) - I'll get it as low as I reasonably can, then add maybe 5 gallons at another station (not Chevron) to see if that changes the behavior...

[GTEASER]

I wonder if something is wrong with the fuel pressure regulator and it's not compensating for boost pressure and the injectors are leaning out and falling behind under boost causing your limp mode. I don't think it would necessarily cause a misfire to be thrown as a code. Do you have a way of watching your AFRs while in boost? Did you check the vac lines to the fuel pressure regulator?

[DrD123]

took a look at the fuel pressure regulator - the line coming off the intake manifold is in great shape - pliable, no signs of damage (it's protected with some plastic mesh/loom, too) - fasteners are tight holding the FPR in place. I was focusing mostly on fuel trims when boosting, to see if they went bananas - they stay pretty stable. If the engine were leaning out, it should show up as an increasingly large positive short term fuel trim, but I didn't see any trends or even starts of trends when I was boosting (while it was sitting at 7psi or so).

 

I don't have a wideband or anything like that, and wasn't watching AFR directly - I figured the trims would tell me something was amiss with fuelling. I am not sure how accurate it is from the OBD2 port, but can take a look. I have to believe the Subaru guy was looking at pretty much anything he could (I got the car back and they had been driving around a bunch - under boost most of the time as the car came back down a quarter tank with an mpg of 16.5!)

 

The bad gas is getting more credible to me... the fuel light came on, but I have a good 3 gallons in there, so I am going to drive it down a bit to a couple gallons then do a half tank from a different station and see what happens. Also contemplating replacing the knock sensor...

[GTEASER]

Remember that when in boost you go into open loop and fuel trims mean nothing.

 

Pull up the AFR on whatever you're using to view parameters. As soon as you hit 8 psi you should be pretty well pegged at 11.6 or lower AFR.

Edited July 14, 2019 by GTEASER

[DrD123]

Doh - you are correct!

 

 

Went out this evening (still 90 out...ugh) and did my pull in third - AFR leveled out at 11.1:1 or so (11.07, iirc) - I am guessing a little lower here since I am at elevation (6k feet) and it's really dry (18% RH according to "accuweather" on the phone) so quite a bit lower density than sea level.

 

 

I am about at the point I need gas... I think I have just about 2 gallons in there. It's just weird that the engine thinks it's knocking, etc., but you feel nothing, no odd changes to engine tone, etc.

[DrD123]

Ok - ran it down to under 2 gallons, then filled it up - went to a Phillips 66 this time - drove home and removed/reconnected the battery to reset everything - we'll see what happens tomorrow. (what a pain in the butt to find top tier that's no Chevron around me - seems like all of the Conoco, shell, etc. that were around got bought by Speedway which is not top tier)

 

 

So another thought I had - could the timing belt be off a tooth? Searching suggests that no, the engine would run like poo and throw a few codes...

[dgoodhue]

Ok - ran it down to under 2 gallons, then filled it up - went to a Phillips 66 this time - drove home and removed/reconnected the battery to reset everything - we'll see what happens tomorrow. (what a pain in the butt to find top tier that's no Chevron around me - seems like all of the Conoco, shell, etc. that were around got bought by Speedway which is not top tier)

 

 

So another thought I had - could the timing belt be off a tooth? Searching suggests that no, the engine would run like poo and throw a few codes...

 

In my experience, when I set one of my timing gears off one tooth is that it set a code after about 5 miles. My car seems like it was off, but it didn't run horrendously. It may vary depending on which cam gear was off. (mine was 2/4 exhaust gear)

 

Perhaps the engine wouldn't know if the crank was off by one tooth.

[DrD123]

well, the problem persists (thought it was better at first, then back to the way it was)

[Fabsx]

Still haven’t checked out the stuff I mentioned huh?

 

 

Sent from my iPhone using Tapatalk

[DrD123]

they verified the integrity of the intake and all of the mechanical systems attached to it. The only suggestion which you made that they didn't evaluate is a plugged cat, as the symptoms are not consistent with that.

[Fabsx]

Well if everytime you reset the ecu, your resetting everything.

When you go back into full boost again after, ecu detects something off and tries to correct itself.

Your dam/iam is probably dropping below 0.5 and disabling your boost controller and only allowing wastegate. I’ve seen this personally without a cel popping up.

 

I would say it’s fuel related, but you tried another station. Maybe try another and pour yourself inside the tank a octane booster. (Royal purple max boost)

See if that helps your problem.

 

Other wise, you might benefit to a cobb tuner to be able to monitor and do some logs to see what’s truly going on.

 

 

 

Sent from my iPhone using Tapatalk

[GTEASER]

DrD123, do you have a Cobb AccessPort?

[SchwarzeEwigkt]

It’s been a while since I’ve done it, but I seem to recall that it’s not super hard to pull off the outer timing covers. Why not take them off and see if all the timing marks line up? Shouldn’t take too long and would check something off your list.

 

 

Sent from my iPhone using Tapatalk

[dgoodhue]

Driver side timing cover is pretty easy to do. Passenger side requires the turbo to IC pipe, disconnecting the bypass valve and removal and the 5 hoses that go over the cover, plus their might need to be more disassembly for access.

[DrD123]

We've confirmed that the car is thinking there is knock under boost and pulling timing, going into limp mode (so the DAM drops however far it has to in order to go into limp mode) - resetting the computer resets it, as does taking it easy in the car for a while - after both events, the car will boost properly once or sometimes twice.

 

 

I am on the third tank of gas, each one from a different gas station (two different Chevron stations, and a Phillips 66 station - all premium (91), all top tier) - each time, I have run the tank down to 2 gallons or less before filling up.

 

 

The car does not have an AP installed on it.

 

 

When they last were poking around on it, they had the local subaru tuning guru working on it (Cam from Camntuning) as well as the person who did the shortblock (used to be at Boost inc.) - they used the subaru select monitor, as well as whatever you use for romraider (tactrix cable, I think?) and were not able to find anything wrong, other than the knock.

 

 

I am off tomorrow, so plan to bring it in and have a more frank discussion with the service folks.

 

 

 

I was reading about "false knock" - which seems to be usually something rattling around that shouldn't be, triggering the knock sensor. I can't find anything amiss, but might poke around a bit more tomorrow. (it's not like there are any rusty heat shields or anything like that on the car - they don't really use road salt here, and it hardly ever rains, so everything is more or less like new) - I am curious if the knock sensor is messed up, but the only "test" you can do is to ohm it out, which doesn't really assess performance.

[DrD123]

Well, it goes back in on Tuesday (they'll have a loaner for me). Hopefully they'll be able to sort it out this time.

[DrD123]

Still at the dealer... getting pretty annoying at this point. I have a loaner, but still - I would really like my car fixed! Got a message today that they are still trying to figure it out, and that it will probably not be done until next week (!)

[DrD123]

Really frustrated at this point. No progress yet - got an update telling me it's going into limp mode today and got a little annoyed since that was what was established back on the 3rd. They did try a known good knock sensor with no change. They are basically following direction from Subaru at this point, so I guess there will be a lot of repetition from what's been done thus far. This week they looked at the fuel pump, etc. as well as the intake, apparently... Basically, it sounds like they are asked to gather data, provide it to Subaru, then Subaru provides additional direction. Still in the loaner, so that's a huge help. Would really, really rather be in my car, though.