Trying to Figure out whats wrong with my SS

[SubiePrice88]

So Ive been having some issues with my car, its a 1992 Subaru Legacy Sport Sedan, 140,xxx miles

Engine: EJ22T
Tranny: 5 speed

No CEL or anything stored codes

Mods/Things done engine wise:
-3" Divorced DP
-2.5" straight back (with a cat, no muffler)
-Desnorked
-Kn'N panel filter
-Knock sensor was replaced sometime before I had it so between 120,xxx and 130,xxx
-Valve cover gaskets, Turbo oil outlet hose, Turbo up and down pipe gaskets, exhaust manifold gaskets @ 131,708
-Agency Power Light weight crank pulley @ 134,169
- NGK/NTK 02 Sensor @ 135,793
-Engine Ground mod w/ 8 gauge wire @ 135,837
-NGK plugs and wires and PCV valve @ 136, 629
- Swapped the Greddy EBC with a Halman MBC @ 137,841
-Replaced coolant Temp Sensor @ 140,397
-Oil replaced every 3-4k miles with Mobil1 10w-30 and Napa Gold filter since Ive had it


I believe thats it. Now the issues with the car

-The needle on my boost gauge flutters under boost
-Also flutters at idle between 15-16 sometimes 15-17 inHg
-Cold starts are fine but I was having problems with warm starts, replaced the Coolant Temp. Sensor and it seemed to fix that issue
- Gas mileage sucks! I average about 15-16 mpg
- When accelerating it hesitates... Like you accelerate and it just wont go and kind of surges and than once it hits 3k RPMs it throws you in the seat and takes off, almost like crazy turbo lag but its at full boost from 2250k
- Also the 90*elbow into the turbo cracked awhile ago, so I silicone it and than electrical taped it
-Also when accelerating it has a whistle or hiss and its not the turbo whistle, its different and louder
-Idle fluctuates nothing crazy but you can watch the needle go up and down as the boost gauge needle flutters
-Backfires (mostly because of the straight pipe exhaust )


Ive sprayed carb cleaner around vac lines and such and havent had any luck. I do have RevScan now, so I can run a log but Im not sure what I should be looking for.

I checked the Resistiance on my Coil Pack today just because as I was reading about intake/vac leaks and it mentioned a bad coil
- Primary Resistance: between 1-2: .9 ohms 2-3: jumped between .8 - (1.1) ohms
- Secondary Resistance: 1-2 = 21.7k ohms and 3-4 = 21.5k ohms

Oh and when I was messing with the RevScan my octane factor was at 100%, whatever that means

I believe thats it for now. I just would like some help diagnosing and finding out what the hell is wrong with my car. I want it to run like a champ, plus autocross is coming up!

Matheww044's car has 170+k miles, sat for a year we pretty much have the same mods and his car hauls balls..and mine doesnt. Id like to fix that! (I knew I shouldnt have sold that SS )

- Much Thanks In Advance

[Dynamic Entry]

I feel you man. Good of you to make such a detailed list. I basically have the same situation, but it is far less documented. Boost flutters, crazy hesitation, but not all the time.

Over the years I have found everything I do helps a bit, but I've never found 'the cure'.

[dscoobydoo]

Checked out the shaft play on the turbo and if it eats oil?
Also if it is at full boost by 2250, but you have a delay, then you may have a leak somewhere.
You could check on the coil issue by swapping with someone who has a good one.

[SubiePrice88]

Dynamic Entry - it honestly sucks. But what have you done thats helped a little?

dscooby - I actually was going to replace the vf-11 with another one I got from a board member, but when I got my turbo off it has ZERO shaft play, up down side to side in and out, there isnt any. So I kept my original on, but swapped the waste gate actuators and that didnt help. I was thinking theres a leak some where but I cant find it, maybe its a intake manifold gasket? I'll swap coils with someone who has a good one, in fact I have an extra. But I dont think a coil would have anything to do with flutter...but maybe it does.

Thanks guys

[SubiePrice88]

I have RevScan, so I could data log stuff and check parameters but Im not 100% sure what parameters I would need to pay attention to.

Also this weekend Im doing fuel filter and walbro 255, doubt that will do anything but it will be peace of mind

[BSOD2600]

Octane factor at 100% means the ECU has fully learned. No spark retardation or increased fuel. Not enough knock to make the ECU unhappy. all good things.

Your idle engine vacuum fluctuating isn't a good thing though. Heck even my old ass engine around 230K is a solid ~19 inHg. You done a compression test per chance? I have a hunch once you fix the vacuum leak, it'll solve a bit of your problems. That also means unmetered air is getting past the MAF, which is throwing off the ECU's calculations. Have a wideband?

Could log the 'manifold absolute pressure' parameter in revscan for a little bit to see what the MAP sensor is seeing for the fluctuations. Use Google to convert kPa to inHg. I assume you also figured out by now that logging a ton of parameters in revscan makes it go really slow -- this is a SSM-I protocol limitation, not revscan.

[SubiePrice88]

OK well the octane factor is a good thing. That makes me a 1% happier

Havent done a compression test yet. Ive been thinking about doing it, but I had a 86 F150 that I did a compression test on and come to find out the number 6 cylinder was gone...than I lost the number 5, I was running an inline 4. Took it all apart and I had a flatten cam...so Ive been kind of nervous to find out what my car is at. I'll check compression this weekend I suppose

I'll also log the map. What numbers would be normal?

Anything else I should log?

[BSOD2600]

MAP when idling should be reading around 33-35 kPa.

Suppose for starters you could log everything while idling in the drivway after the car has been warmed up. Have at least 10 entries to average things out. Logging granularity will be pitifully slow though.

[SubiePrice88]

Checked the MAF sensor, just in case it was dirty. It was clean, but it was missing a bolt from the maf to air box so I replaced that. Also swapped coils with the extra one I have. Cant tell if that made a difference.

Did a little data logging at idle, engine warm:

RPMs, Barometric Pressure (psi), Manifold Absolute Pressure (psi)

800,-0.8,-8.5
775,-0.8,-8.5
775,-0.8,-8.3
775,-0.8,-8.3
775,-0.8,-8.5
775,-0.8,-8.3
775,-0.8,-8.5
775,-0.8,-8.5
800,-0.8,-8.5
800,-0.8,-8.6
800,-0.8,-8.6
775,-0.8,-8.5
775,-0.8,-8.5
775,-0.8,-8.5
775,-0.8,-8.5
775,-0.8,-8.3
775,-0.8,-8.5
750,-0.8,-8.3
775,-0.8,-8.3
775,-0.8,-8.6
800,-0.8,-8.5
800,-0.8,-8.6
775,-0.8,-8.5
800,-0.8,-8.6
775,-0.8,-8.5
800,-0.8,-8.6
800,-0.8,-8.5
775,-0.8,-8.6
775,-0.8,-8.3
775,-0.8,-8.5
775,-0.8,-8.5
775,-0.8,-8.3
750,-0.8,-8.3
800,-0.8,-8.5
775,-0.8,-8.5
775,-0.8,-8.5
775,-0.8,-8.3
775,-0.8,-8.5
800,-0.8,-8.5

"MAP when idling should be reading around 33-35 kPa". .... so thats about 4.7 - 5.0 psi ....uh mines not


Heres another one:

RPM, MAF voltage, Fuel Trim (short), Fuel Trim (Long), IAC Duty Cycle, TPS Voltage

800,1.07,-0.8,3.9,16.0,4.53
775,1.07,-1.$,3.9,16.0,4.53
1775,1.07,0.8,3.9,16.0,4.53
800,1.07,0.8,3.9,16.0,4.53
800,1.09,-1.$,3.9,16.0,4.53
800,1.09,-0.8,3.9,16.0,4.53
800,1.09,-0.8,3.9,16.0,4.53
775,1.07,-0.8,3.9,16.0,4.53
775,1.09,-1.$,3.9,16.0,4.53
800,1.09,-1.$,3.9,16.0,4.53
800,1.09,0.8,3.9,16.0,4.53
800,1.09,0.8,3.9,16.0,4.53
775,1.09,-1.$,3.9,16.0,4.53
775,1.09,-0.8,3.9,16.0,4.53
775,1.07,-0.8,3.9,16.0,4.53
775,1.07,-1.$,3.9,16.0,4.53
775,1.07,0.8,3.9,16.0,4.53
800,1.07,0.8,3.9,16.0,4.53
800,1.07,1.6,3.9,16.0,4.53

Whats the $ sign mean?

[SubiePrice88]

The "new" coil seemed to make a little difference, it feels a little smoother

[BSOD2600]

Yea, so with your vacuum that low during idle, something isn't right. Could be a leak. Could be the timing is off (yes, happened to me too and low vacuum was a symptom), could be other things...

No idea about the logging S. What version of Revscan you using? I always open the logs into Excel and do additional formatting -- never seen 'em.

Sample logs from my car while idling last month:

Code: Select all

Time,"Engine Speed","Engine Speed (Coarse)","Vehicle Speed","Coolant Temperature","Engine Load (Stock MAF)","Engine Load","Airflow (Stock MAF)","MAF Voltage","Injector Pulse Width","Fuel Trim (Short Term)","Fuel Trim (Long Term)","Fuel Trim (Full Throttle)","Fuel Trim (Full Throttle, Power)","Oxygen Sensor","Spark Advance","Knock Retard","Spark Advance Trim","Knock Counts (Average)","Octane Factor","IAC Duty Cycle","Throttle Position","TPS Voltage","Barometric Pressure","Manifold Absolute Pressure","Boost Solenoid Duty Cycle","Battery Voltage"
hh:mm:ss.sss,RPM,RPM,mph,°F,g/rev,%,lb/hr,V,ms,%,%,%,%,mV,°BTDC,°BTDC,°BTDC, ,%,%,%,V,psig,psig,%,V
11:55:44.132,750,750.00,0,83,0.18,9.8,2.4,0.98,2.18,-0.8,12.5,0.0,0.0,40,15,0,0.0,0,0,16.0,0.0,4.61,102.2,32.4,1.6,14.1
11:55:48.235,750,750.00,0,83,0.17,9.4,2.4,0.96,2.18,0.0,12.5,0.0,0.0,770,15,0,0.0,0,0,15.6,0.0,4.61,102.2,34.5,1.6,14.1
11:55:52.416,725,700.00,0,83,0.20,10.5,3.2,1.04,2.43,-0.8,12.1,0.0,0.0,550,15,0,0.0,0,0,15.6,0.0,4.61,102.2,35.6,1.6,14.1
11:55:56.674,750,750.00,0,83,0.18,9.8,2.4,0.94,2.18,0.8,12.1,0.0,0.0,350,15,0,0.0,0,0,15.6,0.0,4.61,102.2,33.5,1.6,14.0
11:56:00.933,750,700.00,0,83,0.19,9.8,3.2,1.04,2.18,0.0,12.1,0.0,0.0,290,15,0,0.0,0,0,15.6,0.0,4.61,102.2,33.5,1.6,14.1
11:56:05.208,725,700.00,0,83,0.17,9.4,2.4,1.00,2.05,-1.6,12.1,0.0,0.0,30,15,0,0.0,0,0,15.6,0.0,4.61,102.2,33.5,1.6,14.1
11:56:09.466,725,700.00,0,83,0.19,10.2,3.2,1.02,2.43,-0.8,12.1,0.0,0.0,40,15,0,0.0,0,0,15.6,0.0,4.61,102.2,34.5,1.6,14.1

[SubiePrice88]

version 1.100.0.0


I'll check my timing as well than

[Dynamic Entry]

For example, when I put on the TD04 and top mount, the thing ran like a champion. For like 3 days, and then it went back to hesitating around 2700.

Then, when I put in the Walbro and changed fuel filter it pulled amazingly and (of course with natural turbo lag up to ~2500) made smooth and intense power all the way up the revs.

For a couple days, and then back to SOS (same old shit)

It has always been tempermental and correlated to ambient and internal temperatures. Also, it is acknowledged that my tests are the butt-dyno which could always be affected by the psyche


However, I must say that I found a hidden vacuum leak last month when I was swapping valve cover gaskets and since then it has been quite consistently good. However since I am on a bad tune (440s and Jecs but stock ECU) I can't properly explore the revs and load to know if this vacuum leak and/or the VCGs were the ultimate culprit.

BTW the vacuum leak was a broken plastic pipe that connects a breather from each valve cover to the block. The pipe is tucked behind the alternator and basically can't be seen without taking the alternator off. My point being not that yours may be broken in the same fashion, but that to really know if you have any leaks you have to be quite thorough.

To which the vets on the board say 'chyaaa'

[SubiePrice88]

Thats actually some good info thanks!

Plans for tomorrow:

-Check for vac leaks
-Install New fuel filter
-Install walbro 255
-Check Timing
-Check compression
-Finish install of EGT and fuel pressure gauge

[SubiePrice88]

could a crappy PCV system make a vac or boost leak??

When I replaced the oil outlet 10k miles ago and had the turbo off, I broke the plastic F pipe and replaced everything with "T"s, hose and vac line. It works, but Im thinking it leaks.

[BSOD2600]

Possibly...

I know that when I replaced the valve cover gaskets, my vacuum increased 0.5 to 1 inHg per the analog boost gauge (no data logs at that time). Although, I'm thinking your ~4 inHg leak is something a wee bit larger. If all your PCV tubes are baked solid, they could easily be leaking and time for replacement, since ziptieing them won't do any good at that point.

While on boost, do you hear whistling or air rushing? Yea hard to tell if thats all you've ever heard from your car and can't compare to another subaru.

[SubiePrice88]

Im honestly thinking about pulling the engine, because some oil seals leak and replacing those seals, gaskets, and all vac line and redoing everything so its legit. Now this wouldn't happen for awhile as Id need to save money first. But Ive been thinking about this a lot.

I also might pick up a smoke machine to help find the possible leak, thats what we used in the Auto Tech School

Now when boosting there is a noise, its not really a whistle its more of a "pshhhhhhhhh" with the turbo whistle in the back ground and the gauge flutters but still reaches boost. If that makes sense? Hard to explain a noise

[SubiePrice88]

My buddy matt (Mattheww044) has a car I can compare things with. His car runs like a champ with 40k more miles than me

[Mattheww044]

[SubiePrice88]

UPDATE:

-Replaced fuel pump with Walbro 255
-Replaced Fuel Filter
-Fixed gauges(which doesnt do anything for this thread)

-and swapped over "New" 90* intake piece from the intake box to silencer, replaced silencer with custom coffee cup mod, "new" 90* from coffee cup into turbo (all that is from Matts Car), Replaced BPV with Supra metal BPV and silicone BPV hoses.

According to my boost/vac gauge the vac is better, it sits at ~18inHg

Havent checked timing or compression yet though


Heres some data log numbers, compare these to the old ones:

"RPM", "Barometric Pressure", "Manifold Absolute Pressure"

775,-0.5,-8.8
825,-0.5,-8.6
800,-0.5,-8.8
825,-0.5,-8.9
850,-0.5,-8.9
825,-0.5,-8.8
825,-0.5,-8.9
800,-0.5,-8.9
800,-0.5,-8.8
800,-0.5,-8.8
825,-0.5,-8.8
800,-0.5,-8.6
825,-0.5,-8.9
825,-0.5,-8.9
825,-0.5,-8.9
825,-0.5,-8.8
825,-0.5,-8.9
800,-0.5,-8.8



Whyd the Barometric pressure go from -0.8 from the last datalog to -.05?

[BSOD2600]

Barometric Pressure is what the current atmosphere is. Look up what the Barometer reading is for your area on weather.gov or something. This number will change a bit through out the day with the weather.

[SubiePrice88]

Oh tricky.

Well I now know that I need better intake pieces, as mine have some pretty big cracks and it helped having good ones. The vac improved, but still flutters a little with the rpms

Its still sluggish and hesitant. Feels like its really restrictive or something

[SubiePrice88]

Had a fellow Subaru Guy drive my car today, he has a AA in Automotive Tech. and has done some crazy swaps into a 93 Impreza. He didn't see anything wrong with it but wants to meet up again when I have my laptop and RevScan. The only thing he said really was "I dunno it seems fine, its just slow" and I said Yeah thats the problem!

Considering taking it to PIA in April, since I'll be over that way for the Olympus Rally. Until than I still am going to check timing and compression *crosses fingers* PLEEEASE dont be low on compression*