Car shakes violently and boost limited @ 9 (RESOLVED: SHORT)

[free5ty1e]

OK so -- after I picked my '94 SS up today from it's new timing belt / water pump / oil pump / ecu / etc, it started and drove beautifully - until I decided to get on it.

Below around ~%40 throttle, everything runs great; Turbo is spooling up nicely, car is accelerating and driving smoothly, all is well. As soon as I give it some gas, no matter the gear or RPM, the car shakes rather violently and it feels as though power is being cut and restored rapidly.

What's odd is that, even though my MBC is still plumbed in as before between the compressor and wastegate, my boost ends up being limited RIGHT at the stock level of 9 PSI.

If I watch my (non-wideband) A/F gauge, it acts normal at lower throttles (searching under vacuum and rich under any level of boost) but when I give it more throttle and I experience the shaking and all, it's searching just like it was partial throttle. Even at WOT.

No check engine light, although when I attempted to flash codes with the D-connectors to see if it had any to show me, it seemed to just flash steady, fast, and endlessly. I thought I noticed an almost imperceptible pause after the count of '9' -- but I could have swore it would pause for an obviously-longer interval between codes as we had just done that earlier on the '94's ECU (my poor '91's ECU was swapped in). I also didn't see a code 9 on the list.

Now, the '91 ECU had recently witnessed pretty much total engine failure; the '91 lost compression in 3 of the 4 cylinders completely (0psi 120something psi 0psi 0psi) and we suspect a serious overheating caused it. It has also had no battery for at least a month. Would it still be pissed off about that, possibly - after having no power for so long? I didn't think so, but wanted to put it out there.

I'm going to make a serial cable for Vikash's b10 scan tool and see if it works for me; my poor old laptop that actually had a parallel port on it was stored under the driver's seat for several years and appears to be dead to the world now. Looks like it will also work for RevScan and possibly EvoScan as well, so this is a good cable to make.

Anyway, I'm just concerned I have a crap ECU again here. I'm going to go through the wiring as best I can and see if I can find any issues, maybe I'll work on my grounds. I'm at a loss here and pretty bummed. My crank pulley is also quite wobbly - he told me to keep an eye on it and look for a metal replacement. I'm gonna search the boards later for suggestions on that as well.

( >< )


...I'd appreciate any suggestions from the collective intelligence that is this forum, as usual, on my strange shaking and odd limiting to 9psi (and it does not feel powerful, half the time it feels as though fuel is being cut but in a very rapid switching fashion). I need to find junkyards with some turbo Legacys in them in the Portland area. Anyone know of any and possibly want to go? I'd like another ECU or two after I go through all my wiring and grounds...

[subaru_styles]

at least im my case I had a year of problems that all boiled down to dirty MAF(s) that weren't throwing a code

[free5ty1e]

Hmm.. I did spray maf cleaner thru mine until it ran clear but maybe it would be worth going in for a closer look. Im still worried I have some sort of electrical problem and i need to get familiar with my car again as ive done little more than oil changes and check fluids for several years now.

Sent from my SAMSUNG-SGH-I337 using Tapatalk 4 Beta

[free5ty1e]

Wait... I see a pin on Vikash's ECU pinout page marked "1990-1992 models only: "B48.16 Barometric pressure sensor 1N" -- does the 1N indicate it's only for 1990-1992 N/A models or would this possibly be an issue with me having swapped in a turbo ECU from a '91 to my '94?

Just a stab in the dark. Anyone know?

[Legacy777]

The barometric sensor is only on the non-turbo cars. The 91 turbo ECU is fine.

Does it feel like you're hitting fuel cut? Is the problem related to boost pressure or throttle %?

[free5ty1e]

Throttle % it seems... been up all night rewiring many poor wiring jobs under my hood and started the grounding mod... going to pass out now and test drivs tomorrow sans fcd and on stock boost control.

Edit - no fuel cut its far too rapid. Its a really bad constant stumble

[Legacy777]

I'd check all your ignition components and wiring, coil pack, wires, plugs, igniter, etc.

[dscoobydoo]

On my car with the PP6, the FCD will cause the car to do that (when I hit boost pressures) when it is not working right. It bucks like a pony and sometimes wont give off a code.
It almost sounds like when you are going from closed loop to open loop it is doing something. That would be an ECU issue.

But there are a lot of things that could do that.

[robertpaige]

The only Turbo Wagon I know of was at the South Pick N Pull off Foster Rd, it was an auto wagon. No engine components or ECU, but it was there a couple weeks ago. Still there I bet you.

Get that crank pulley fixed asap, you can take one off of any Legacy 2.2. I'd stick with 90-94 just to be safe, idk FOR SURE about the other years.

[free5ty1e]

Hmm... well, after all my efforts this weekend, the problem remains. I did make good progress though; I still had many "temporary" wires connected up to the throttle, airflow sensor, and pressure sensor from the days when I was experimenting with the LegacEBC. These were suspect so they had to go; I was, as usual, worried about how little slack some of the pressure sensor assembly's cabling was but this time I did something about it - I added a couple inches of slack to each of these cables by way of solder and heat shrink, and got rid of several ugly FCD wires that were half-assed at best. I removed the unused connectors everywhere else and really cleaned things up, which had to be done anyway before I could really troubleshoot anything else as they were suspect.

I also went ahead and removed my FCD, and went back to the stock boost control solenoid. Since the ECU was somehow forcing me to run 9psi anyway, I figured I might as well give control over the boost back.

So, it's not just throttle position; this is just how it's manifested thus far. I've had a slight throttle result in this violent stumbling, and then I've had a successful full-throttle run at 9psi that felt strong-ish. I'm wondering at this point if all that cranking without injector pulse has fouled one or more of my spark plugs - the engine was flooded bad there for a bit. Perhaps it is time for some new wires and NGK v-powers....I still heavily suspect the ECU but at this point I gotta try other things.

I'll see if the crank pulley on my '91 is in good enough shape to swap over. Having my crank pulley come apart on me on top of all this other crap is something I'd like to avoid.

[Legacy777]

On my car with the PP6, the FCD will cause the car to do that (when I hit boost pressures) when it is not working right. It bucks like a pony and sometimes wont give off a code.
It almost sounds like when you are going from closed loop to open loop it is doing something. That would be an ECU issue.

But there are a lot of things that could do that.

I experienced some weird bucking issues with the PP6 installed as well. Not sure why it happened....it was not consistent.

[free5ty1e]

It's even happening sometimes at idle. It kinda sounds and feels like a misfire... I've never experienced misfires on boost before, perhaps that's why it's been so worrying.

Knock sounds like ... like someone just threw quarters into the throttle, right? I haven't heard that. (I've actually never heard that... perhaps I should look up a Youtube video of engine detonation. And after all the cool videos that will undoubtedly follow, I will be sure and search for engine knock.)

No check engine light still - even though I still don't understand what the constant flashing during D-check mode was trying to tell me.

I now notice two modes that my car chooses from at random now, except when it hits 6+psi (EVERY time):
1) Everything's Awwww Rite! - car behaves normally, turbo spools smoothly and sounds awesome under partial throttle, can spool up to 5-6 psi and then enters other mode. WOT always results in entering the other mode, as well:

2) Crap - boost gauge jumps erratically in time with the engine's stumbling / jumping when at or under atmospheric pressure sometimes - car lurches about embarrassingly. Boost NEVER rises above 9psi (MBC was set to 15 psi) - boost in this mode does not feel like boost, although gauge reads steadyish boost under heavy load / WOT. I have just a hair above zero acceleration in this mode. It may be worth noting that it still can hit 9psi as before, and it is now under control of the stock boost control solenoid.

It will go away or not the next time I let go of the accelerator and press it again. So, to add to the motion of the stumbling engine, I have to keep hitting the accelerator to try and get into Mode #1 again.



So, I realize now that I probably haven't changed my spark plugs & wires for a good 40,000 km or more. Also, there were major oil leaks that eventually gave the wires a good coating (I think they are starting to swell). Then there was this whole situation where plugs 1 and 4 would keep getting flooded with fuel while the '94 ECU refused to provide pulse width to injectors 2 and 3... do you think my plugs and wires could be in bad enough shape that perhaps 6psi at this point is more than enough to blow out the spark and cause "reliable" misfires? (oh, what a term!)

Either way -- they're due. I'll drop by the store on the way home tomorrow and grab what I need and see what I can find out.

I've also gone ahead and ordered a fresh crank pulley from Subaru and am going to go ahead and pay my mechanic the $25 it will cost in his labor to replace it for me so I don't have to deal with that one.

(edited for accuracy on which plugs and which injectors were doing what)

[robertpaige]

Knock sounds like ... like someone just threw quarters into the throttle, right? I haven't heard that. (I've actually never heard that... perhaps I should look up a Youtube video of engine detonation. And after all the cool videos that will undoubtedly follow, I will be sure and search for engine knock.)

Knock sounds like, well, knock.. It has a lower pitched tap, or tick sometimes. Definition of Detonation: Unburned end gas, under increasing pressure and heat (from the normal progressive burning process and hot combustion chamber metals) spontaneously combusts, ignited solely by the intense heat and pressure. The remaining fuel in the end gas simply lacks sufficient octane rating to withstand this combination of heat and pressure. Detonation causes a very high, very sharp pressure spike in the combustion chamber but it is of a very short duration. It sounds like a higher pitches pinging.

No check engine light still - even though I still don't understand what the constant flashing during D-check mode was trying to tell me.

Good. Constant flash means you're OK and no codes are detected/being detected in d-check.

Crap - boost gauge jumps erratically in time with the engine's stumbling / jumping when at or under atmospheric pressure sometimes - car lurches about embarrassingly.

While your engine is stumbling your boost gauge WILL be all over the place.


So, I realize now that I probably haven't changed my spark plugs & wires for a good 40,000 km or more. Also, there were major oil leaks that eventually gave the wires a good coating (I think they are starting to swell). Then there was this whole situation where plugs 2 and 3 would keep getting flooded with fuel while the '94 ECU refused to provide pulse width to those injectors... do you think my plugs and wires could be in bad enough shape that perhaps 6psi at this point is more than enough to blow out the spark and cause "reliable" misfires? (oh, what a term!)

Honestly I have a feeling that new plugs and wires are gonna do a whole lot for the health and performance of your car. The plugs may just not be functioning well enough under boost to burn all your fuel.

I've also gone ahead and ordered a fresh crank pulley from Subaru and am going to go ahead and pay my mechanic the $25 it will cost in his labor to replace it for me so I don't have to deal with that one.

Good!! Also the crank pulley is not a fun job to do, especially by yourself. $25 is cheap labor.

[free5ty1e]

Thanks Robert!

It's good to feel like I'm starting to understand what might be happening with my poor car. I'm def. stopping on the way home at Autozone and grabbing some NGK V-powers and wires. I should have everything I need to get in there and change 'em out, methinks.

We can just get in there in our cars without removing the intake manifold, correct? Will I need a universal joint for my ratchet wrench to angle in to the spark plug holes or should I be able to sneak everything in and out of there without removing anything? (of course, I understand if I have to move my intercooler piping, lol -- I mean will I have to move anything from the stock setup to access the spark plugs, other than the wires?)

Then, dielectric grease on the part of the plugs that go into the boots to protect them, and thread anti-seize on the plug threads.

Gapping them to .039 - .040. The Subaru mechanic, who's used to working on WRXs and STIs, recommended I gap them at .022 or so. I mentioned the stock spec is around .040 and he balked. But I've read through our awesome spark plug FAQ

I also like Vikash's policy of always putting back in fresh spark plugs whenever I take them out. I forget where I read that but one of his posts somewhere went into how it's best since the crush washers they come with are only good for one application, really. I think I'll get some spare plugs to have lying around for just that purpose... NGK V-powers are inexpensive enough to not hurt my wallet with this policy.

[free5ty1e]

....REALLY, AutoZone? You don't normally stock the NGK V-Power copper spark plugs I need? It's a special order item? Wait, you can't even get them through special order?!

When the crap did the BKR6E-11 become a rare part?!?

I tried to find the Baxter auto parts that Google said was nearby but I just don't know this area well enough yet. When Google Navigation directs me to a residential cul-de-sac and tells me I've arrived at Baxter Auto Parts, I think it's time to go home and start over.

I'm gonna order online and just wait for delivery and have to deal with a(t least a) couple more days of this stumblastic lameness. I still might need an igniter or coil pack or ECU but ... I definitely need spark plugs and wires anyway so I'm startin' here.

[beatersubi]

NAPA stocks NGK, usually. You might need a u-joint for the driver side plugs.

[robertpaige]

I mean will I have to move anything from the stock setup to access the spark plugs, other than the wires?

Nope, with some extensions and a sparkplug socket you have decent access to everything. Driver side head will be a little more crammed than the pass. side. Our SOHC heads aren't bad compared to DOHC, there is a lot less space and it's a lot more of a pain in the ass, lol.

Then, dielectric grease on the part of the plugs that go into the boots to protect them, and thread anti-seize on the plug threads.

you got it.

Gapping them to .039 - .040. The Subaru mechanic, who's used to working on WRXs and STIs, recommended I gap them at .022 or so.

I dunno what he's talking about, but out owners manual states plug gap to be between 0.039" and 0.043. Here's a little faq for you about plug gapping;

Bigger gap =
Positive side: larger spark, better ignition.
negative side: your spark can get "blown out" by the air passing by.
Smaller gap =
negative: not enough spark, bad ignition, stuttering
positive: your spark is hard to blow out.

[free5ty1e]

LMAO -- tried to start it up to go to work this morning, and the battery is reading <10v. That ain't crankin' shit. It's barely powering dash lights.

It's a good thing I bought this jump starter; hopefully my alternator is still good, and I'm headin down to the local NAPA. 3.5 miles away, not bad. Thanks for the tip on Napa carrying NGK's -- they have 14 in stock woot!

So, gonna go (hopefully) jump start me car, drive to NAPA, get a new battery, get spark plugs, and get a U-joint for my ratchet and another extension just in case I need it. I have the grease and the antiseize and the wires already, so I should be good to go once I get back. Gonna charge the jump starter in the car on an inverter as I drive toward NAPA just in case.

I know this was an el cheapo Wal Mart battery and those don't usually hold up anyway, but this one lasted a long time and I made sure to always keep it charged so it didn't sulfate. I'm wondering if the suspected ignition issues may have accelerated this battery's demise.

Either way, she's gettin' a good fresh battery and good fresh spark plugs and wires today. I sure hope it makes a noticeable difference in the right direction.

[free5ty1e]

Oops. .not the battery. It held a charge just fine... I hear a rapid clicking and the dash lights flash in time with this. Then, the voltmeter in my car reads 10v and the dash lights are on and dim. Then, all off with the key still in and on.

Then, a couple seconds later.... all on full power lookin good and....all off. Shit.

Im going to try and find the problem but this is not encouraging.

Sent from my SAMSUNG-SGH-I337 using Tapatalk 4 Beta

[free5ty1e]

It started powering up again. I haven't found a wire that causes the problem when I wiggle it yet but I am searching. My remote starter had been acting up (I mainly used it for keyless entry) but I thought my remote failed and I lost my spare.

Last night when I got home, just before I turned the car off, It seemed to experience a power drop or something. The head unit and my gps and exhaust temp gauge and the lights in the dash all reset.

Im gonna go ahead and just remove my old remote starter...ive been using the key anyway. Maybe itll reveal some bad wire.

Sent from my SAMSUNG-SGH-I337 using Tapatalk 4 Beta

[free5ty1e]

Pulled codes 14 15 16 17... all 4 injectors. I wonder if thats from the power incident yesterday or from the power cycling and issues this morning? I think im gonna extend my green diag connectors to match my black ones so I can try the other diag mode later. For now.. still looking for wiring issues.

Sent from my SAMSUNG-SGH-I337 using Tapatalk 4 Beta

[free5ty1e]

I found one! Yikes... a wire coming out of the back of my turbo timer and the two on my amp remote turn on switch are all burnt crispy. ...insulation all melted off...

Yeah. Im tracing those and fixing that. Wow.

Sent from my SAMSUNG-SGH-I337 using Tapatalk 4 Beta

[Legacy777]

Yeah....you've got a short. Definitely double check ALL your wiring that has been added and make sure it's done right.....that very well could be your problem. Once that is fixed, clear the ECU memory by pulling fuse 14 EGI/TCU for about 30 minutes.

[free5ty1e]

I fixed it. Traced the issue to some forgotten wiring from an old half finished project. I had some electroluminescent lighting in my car with a controller I programmed and built, it was pretty cool, it coukd pulse colors to the beat of the music and such. Anyway it stopped working at one point and I had just forgotten about it until I traced the burnt wire over to it. I also had a burnt line going to my bass amp remote turn on (had it on a switch instead of the head unit but thats gonna change) and unfortunately they all tapped power from my poor turbo timers wires. Pretty close to the unit too...so I had to cut it out if there and I only managed to save an inch of some wires on it.

Oh well... the car has good power again and I still get bad stumbling and misfiring on boost starting about 5psi and I am going to try thr spark plugs n wires tomorrow after work. Whew!

Sent from my SAMSUNG-SGH-I337 using Tapatalk 4 Beta

[robertpaige]

Excited to hear results.