Emmm.... Now I'm Clueless. [Solved]

[LedSled]

I might have multiple problems going on here so, I'll just start from the top, sorry for a long post.

So, after a top-end rebuild to my 91' EJ22 (No Turbo) last winter, I added a K&N filter and removed the stock airbox. I noticed occasionally my idle sinking to like 500RPM and the car sounding like it would die. I would adjust the throttle cable, and all would be good, until the next time it either rose or sank... I figured it might be the intake sensor struggling since the K&N filter didn't utilize the airbox in the fender, maybe having reduced airflow.

I decided to vent the hood. I bought a fiberglass Duct and cut the hood right above the filter. All seemed fine, idle changing stopped, we were good. After a while though, I noticed periodically, when I drove the car, in first gear, if I romped on the gas pedal, the car choked, the RPMs dropped and sounded like it was going to die (it never did though, but wouldn't go more than 5 miles an hour unless I let off the pedal and slowly re-engaged). I changed both the fuel pump and fuel filter. Behavior continued.

I observed what I could, and felt this mostly happened when it rained, or after it rained. I thought perhaps water was dripping inside the hole cut in the hood onto air filter and flooding the mass air-flow sensor and choking the engine. I drilled holes in the duct so water would drip down instead of dumping on the air filter. Behavior continued.

I changed the air filter, and installed an "Electronic Supercharger", hoping the motion of this new fan blade would dry any moisture that would get into the intake tube. Behavior continued, now happening at all times.

I read about how IAVC's control the idle (and the RPMs did drop when I hit the pedal), and seeing a tube from the intake going to said IAVC, I thought maybe water had dripped into the module and destroyed it. I thought maybe it had been going bad for a while, the engine never sounded right unless it idled about 1200rpm, which seems relatively high. I figured this must be it, and I went to the junkyard and yanked a new IAVC and installed it with fresh gasket. Behavior continues, except now when I start the car, and it's in park, it idles at nearly 2500, and idles above 1500 in drive with my foot on the brake pedal. At this point i've stopped driving the car because I'm worried it will damage something.

I attached a timing light and checked out the ignition timing. My Chilton's said it needed to be 8 degrees
ATDC (I believe... I can't find my manual right now). When I adjust the IAVC it only adjust from about 10-15 degrees ATDC, I can't get it to retard to 8. The only thing I can think of is perhaps the timing sprocket was put on in an advanced position, which would be why I can't retard it far enough.

However, I'm not even sure this would solve the problem. Anyone got any ideas of where to go from here? I need to stop this "choking/stalling" and now, get the car to idle regularly.


Bit of basic information about the stalling/choking issue. It would happen when I very first started the car, and only in first gear, once I got the car up to about 20, the engine and car both performed as normal. At first, it happened once in a while, maybe once every couple of weeks. And it only stalled/choked maybe the first 3-4 times I was taking off from a stop. If I drove the car more than about 20 minutes that day, it performed as normal. It just perpetually got worse, although at the time, it was raining frequently. I have since replaced the hood with a non-vented one from the junkyard, but the behavior continues.

I haven't really heard of anything like this before. It's all very confusing. I don't want my Subie to die! Please help!

[rallyak]

Try putting your stock air box back in and see if that changes anything, my n/a didn't like me changing the way it breaths. My idle was high and the cel would come on 24 & 49. I ended up putting the box back and changed my TB with a spare ( possible bad sensor) runs great now.

[kimokalihi]

You can't adjust the timing. Its computer controlled. Have you or anyone messed with the idle screw on the throttle body? You should never adjust it, just for future reference. I would replace your coolant temp sensor and take your timing covers off and line up the crank pulley to see if the cams and crank lined up on their marks and check for vacuum leaks.

[Flatfourced]

I assume your CEL isnt on.

When were your plugs changed last? nice ngk v-powers?

Clean your maf. Check for vac leaks. Sea foam.

Basics are often overlooked.

As kimo said, check the timing belt. Was a new tensioner included in the top end rebuild?

[Legacy777]

As others have mentioned, check the basics first.

A couple things to note. Most people think the small tick below 1,000 rpm is 500 rpm. It's not, it's 750 rpm. So unless you had an external tachometer hooked up or a scan tool to read the rpm the ECU was seeing, if you thought the small tick below 1,000 rpm was 500, you could have been trying to adjust something that you didn't need to.

The k&n pod filters have a good tendancy to kill MAF sensors over time due to more turbulant air as well as oil droplets. In addition, the more turbulant air causes the meter to not read accurately. The bogging at initial throttle is some what typical with the pod filters. I too would recommend putting the stock airbox back on. You can remove the snorkel in the fender to help improve air flow and keep the intake growl.

[LedSled]

The CEL has been on since I bought the car, its always been on. I've only plugged in the plugs and checked for codes once, before the top-end rebuild. Never bothered to do it again. I will attach the stock box and pull another MAF while i'm at the junk-yard.

Plugs are brand new. Iridium Pulse Plugs.

"The k&n pod filters have a good tendancy to kill MAF sensors over time due to more turbulant air as well as oil droplets. In addition, the more turbulant air causes the meter to not read accurately."

If this is indeed the case, it would make sense as to why it seemed to "get worse" or do it constantly when I put an electric supercharger (basically a fan) next to it.

How can I check for Vacuum leaks?

Also, I thought I couldn't adjust the timing either... but then I noticed the screws on top of the IAVC module, and the marks on the timing belt cover behind the harmonic balancer. I connected a timing light to it, and twisted the top of the IAVC... it adjusts something...

Thanks for the help fellas! And yes, I'll check the CEL codes too.

[Legacy777]

Yeah, let us know what codes come up. Also try the active diagnostic check with the green connectors.

You can check for vacuum leaks by using starting fluid and spraying it around the hoses or area you think there's a leak. Engine rpm should increase if there's a leak.

Timing is a function of engine rpm. So when engine rpm changes, the timing will change. This is why the manual gives a base timing reading at a specfiic RPM. If you cranked up the rpm, sure the timing is going to change.

[LedSled]

Well, I connected the stock airbox back along with a MAF I pulled from the junk-yard. The idle was incredible low (Under 500, at the first bar above 0) and died if I didn't rev it. The intake also was backfiring.

I ran the codes.. and I got a lot of them. In order:

Code 42: Idle switch Abnormal voltage input entered from idle switch
Code 22: Knock sensor Abnormal voltage produced in knock sensor monitor circuit
Code 24: Air control valve Air control valve inoperative (Abnormal signal produced in monitor circuit)
Code 35: Canister purge solenoid valve Solenoid valve inoperative
Code 12: Starter switch Abnormal signal emitted from starter switch
Code 15: Injector #2 Fuel injector inoperative (Abnormal signal emitted from monitor circuit)

The light has been on since I bought the car, and actually, the reason I even installed a cold air intake was an intake back-fire broke the MAF apart from the airbox. So, something here might been an underlying issue all along. Is there any one-thing that could cause the bulk of these problems?

[Legacy777]

You need to clear the codes, and then run the active diagnostic check to see what the ECU is currently seeing. I doubt all those things are bad. The active check tells you what is the current issue. You can fix that, recheck and see if anything else is there. Typically you start with the lowest code first if there are multiple codes with the active check.

[LedSled]

Well, I plugged in the cables to clear the codes, and it immediately started flashing a code. It was Code 35, Canister Purge Valve Solenoid.... I replaced it. I then got code 22, Knock sensor... which was all cracked, so I also replaced. Now it will clear... I'll run the active diagnostic tomorrow. As of now though, after replacing the Knock sensor, the car actually runs somewhat, I was able to drive it around the block and such. While the idle is a bit rough, the acceleration is actually okay. I'll let you know what else happens in case anyone else has a similar issue.

[LedSled]

Oh... and it keeps backfiring in the intake box.

[Legacy777]

Backfiring may indicate a timing issue. You may want to pull the timing cover off and verify the timing marks line up properly.

[LedSled]

Pulled everything apart, re-did the timing, although I'm pretty sure it was already correct. The engine continued to back-fire in the intake box... I kept getting the codes oxygen sensor, or idle switch, or air control valve... I rewired the oxygen sensor (which was a new, "universal" bosch one), that made the oxygen code go off, and I slowly tinkered with the air control valve, and the TPS (which before today, I didn't know existed...). Tinkering with them both together, I was able to get the engine to start running smooth, with no back-fire. The first time I drove it, acceleration was kind of slow, and at 3000 RPM it would switch gears and lurch forward. Felt like the wheels had really low torque until the gear shift. I "lowered" the TPS, and it didn't lurch, but it just seemed... like it was gutless. The jerking stopped when it switched gears, but... it just seemed overly tame. I moved it half way between the the first point, and the second, and it's running good. Acceleration isn't as hard as it once was (Back when I had the super-charger and cold-air intakes), but satisfactory, it could climb hills, ran good, the only problem was the idle was pretty high (3000 in park, 1500 at stops). I adjusted the stopper screw on top (I know you're not supposed to, but I already did it countless times before I found that out.) The idle lowered to a rate I deemed satisfactory (Under 2000 in park, 1000, 900 at stops). So far so good. I can't really get out and see if everything's totally smoothed over because I busted coolant temp sensor plug while adjusting the the TPS.... I don't want to drive overly far in fear the disconnected plug won't tell the fans to kick on and over-heat my engine. So far however... GOOD.

I wish I could say there was one part that was "the problem", and if I had to, the Knock Sensor was the biggest (as when changed the car went from basically inoperable, to drive able, but shitty), but other problems included Misaligned TPS (most likely from moving stopper screw...). I'm just happy, it appears my Subie will live to ride another day.

[Legacy777]

This is how the TPS should be adjusted.

http://www.main.experiencetherave.com/s ... sting3.jpg

If you get a CEL after adjusting it with the idle switch, then the TPS may be bad or the throttle stop screws have been tweaked with too far out of calibration.


FYI, if the plug is busted on the coolant temp sensor, the fans should be running all the time.

[LedSled]

Alright... well I finally got it all fixed. The car seemed to be running okay. As I said I had installed a new Oxygen sensor. Upon doing checks of the engine, I realized, some how, the wires in the new sensor had touched, and fried. I fixed the wiring and the car began back-firing again. Puzzled, I unplugged the o2 sensor, and the back-firing stopped, plugged it back in, and it started again. I thought maybe the the Catalytic converter was plugged or something, and when the o2 sensor was plugged in, it detected it.

I gave up and took it to a shop. The mechanic ran a heat test on the CAT and told me it seemed good. He told me to try installing an Oxygen sensor, to which I replied I just had installed one. He asked me if it was Universal, or Direct Fit. I told him Universal, he told me to get a Direct Fit, and really only much older cars are compatible with the Universal ones. I thought he was full of shit, but never the less, I went and bought a direct fit.

Eureka! It worked. No more back firing.

I finally figured out the core of my dilemma though. As I said initially, my car seemed like it was choking/stalling when it rained.

Well, I was right. Here's how it worked. You see, when I adjusted the stopper screw at the top of the throttle bottle (a big no-no), it effects the TPS. Adjusting the TPS can effect when a car shifts, as in, you push the TPS all the way to the "top", the car will shift at 3500 RPM, you have it all the way at the "bottom", the car will shift at 2000 RPM-ish... Yes, I've gone through all the ranges, as I didn't have a way to check for conductivity.

Here's what happened though. When an engine is cold, such as a rainy day, and turned on, the RPM's are higher until the engine reaches a correct operating temperature, at which point it drops down. This is pretty much vindictive of all cars in a cold start. You can notice this when you perform a cold start and notice your RPM's around 2000 or so. My TPS had been modified via me moving the stopper screw to try and change the engines idle. The car shifted at low 2000's, and downshifted at 1700-1800. The problem? When the engine was cold, it idled at 2000, it never reached the 17-1800 range where it would have downshifted. Basically, I'd take off from a cold start, make it to 2nd, or 3rd gear, but the car wouldn't downshift. The sluggishness/stalling I felt was because after the first stop-sign of the day, the car hadn't down-shifted. I was trying to take off in 3rd gear, which is why my car had poor acceleration, and seemed to "Operate normally once it got to around 20 mph". This is why my car seemed to operate "Normally after 20 minutes of driving or so" Because my engine had reached a normal operating temperature, and a lower-idle rate, that was within the TPS's shifting range, finally allowing it to downshift.

Twas' a complicated night-mare. Just a compounding of trying to modify things to correct other things, without having full knowledge of what I was adjusting.

Don't touch your stopper screw! Bad TPS setting can lead to shitty acceleration in cold Whether! You can change your idle by adjusting the swivel on the IAVC or throttle cable!

Thank you to everyone who helped though. I hope this post can help someone else out.

[SiCkSTi04]

Good to hear it all worked out

[Legacy777]

Yup, glad to hear you got it sorted out....and just another example of why not to adjust your throttle stop screws

You said you didn't have a way to check for conductivity.....so you didn't have a meter or anything to run the procedure I posted? Even now, since things have moved around, I would highly recommend performing the TPS adjustment procedure to at least verify it is where it should be.