My car is mad at me! Random stalling, rough idle

[SoCal LS-L]

Originally Posted by macadamianut

Yes, I have tried pulling injector wires at idle, and did not find one that made a "no change" condition. So, I ruled out the fact that one injector was bad. Maybe still partially clogged/dirty but not bad.

Tonight I tried a couple things:
-Adjusting position of TPS and noticed that I would get a fast idle unless I shut down the car completely after the change. Basically I found no change.
-Running car with open airbox=no change
-Sprayed starting fluid on all known vacuum areas including manifold gaskets=no change in idle speed.
-Disconnected Battery and will let it sit overnight to clear ECU.
-Took video of tachometer while hesitation was occuring. Where to post? Youtube?

I see that there are lines and a plastic valve to the emissions canister. Could there be a potential problem with that area? On the bottom of the canister, there is a hose that returns into the fender well under the airbox with a nipple that does not seat well, (often found dangling.)

Still wonder what would happen if I disconnected the knock sensors or 02 sensors one at a time (like the fuel injector test.)

Somewhere in the back of my head (a very scary place) I keep hearing this little voice saying "Put the ECU, and everything else back to stock." Thing is, the only mods I could see effecting such a condition would be the ECU tune and the Stebro muffler.

I feel like I cannot rule out a clogged cat either, as it rattles, but with only 83,000 miles I'd find that hard to believe, plus the engine runs smoothly and builds power as the rpms climb.

Wow, this is crazy!

The connection from the canister to the fender is just a vent. I went back thru all the posts and it really seems like your MAF connection could be bad or corroded. Have you inspected the harness closely? You can probe each wire with a pin and measure continuity between the wire and the corresponding terminal, while moving the wire around to see if it loses continuity.

Also have you checked the coolant temp sensor? I dont remember you mentioning it.

[sowise]

Ok just a couple of suggestions that you may or may not decide to do, up to you.

First, you have been going to the dealer which is definitely not cheap, you keep having to perform checks and diagnostics on sensors that I would recommend you just replace anyway due to the age of the car and since most of them are easy to do, and may have saved you money in the long run anyway.

I would replace all cam and crank sensors, both knock sensors (knock sensors can cause your car to run like crap), EGR valve, and pieces of the Emissions i.e BPT valve, and also do the check on your iris check valve (suck/blow) to see if that is working. Your Iris valve may not be working correctly at the low RPM's causing issues, maybe worked better after they did that gaskets because they cleaned it up, replace any old vacuum lines that is cheap some of the sensors add up but if you do it systematically over time you won't have to worry much about them. Also like SoCal said check your coolant temp sensor on the driver's side rear engine area of the coolant cross over tube. Good luck and keep us informed of changes or corrections.

[macadamianut]

Darn,

Reset the ECU overnight. Still acting up. Still NO CODES.


I have noticed something that may help narrow this down - The problem only seems to occur at three instances:

-Idle
-Very small throttle opening
-When the throttle is abruptly closed

To refresh, in neutral, if I push the rpm above 4000 = no issue. If I mash the throttle in gear while driving starting at any rpm = no problem.

So this tells me that the issue lies in the "idle" and "just off idle" area. How about a thorough cleaning of the throttle body and checking that both valves are opening the same?

Here's to hoping that any of this helps identify the culprit.

Thanks.

[TN_fwdsvx]

Mac;

I have had your exact same problem since October 2009. I have been following your thread but had nothing constructive to advise. I did read several messages advising "Replace the MAF."

Finally bit the bullet and had Jackie at Anapolis Subaru FedEx a new MAF sensor to me. Installed it last Saturday and the SVX has been running GREAT since. I also ran a tie-tie around the connector and MAF sensor to ensure the connector does not loosen.

Good luck with yours.

Tom

[SoCal LS-L]

This might be a good candidate for Seafoam or throttle body cleaner, it almost sounds like your bypass air valve could be gummed up and sticking.

[XT6Wagon]

Have you cleaned both battery cables or atleast mesured the voltage drop across them when running?

Your problem sounds exactly like mine, and it was bad connections on the battery cable. It would spike a 1/2 volt or more drop across the negative cable which of course horks ALL the sensors so the ECU has no clue whats going on. Can't go by pure resistance as its a momentary spike for whatever reason. Cleaned the terminals and the connection to the body and the stalling went away.

[macadamianut]

Will clean battery connections and as many grounds as I can find. I found an interesting webpage from a guy who svx-ed his vanagon.

http://www.primafoundry.com/cars/SVX...ct%20Diary.htm

He also has some great pictures spread around good for finding the location of things.

He stated that he also had problems with the iris check valve being cracked and thus non-functional. I picked up a generic one at autozone today for $5.00. Will try to install tomorrow weather permitting and inspect the original. Does anyone know the correct direction of flow? Toward the front or toward the valve body? I suspect there is a chance it was re-installed backwards.

Thanks.

[SoCal LS-L]

Originally Posted by macadamianut

He stated that he also had problems with the iris check valve being cracked and thus non-functional. I picked up a generic one at autozone today for $5.00. Will try to install tomorrow weather permitting and inspect the original. Does anyone know the correct direction of flow? Toward the front or toward the valve body? I suspect there is a chance it was re-installed backwards.

Thanks.

Its easy to tell given the nature of the valve itself. The side you are able to suck air from connects to the intake manifold, the direction of flow always points to the intake.

[macadamianut]

Well,

Spent some time freezing my toes off in the driveway today. Focusing on the pass side of engine.

-Replaced iris check valve with the autozone part, (OEM part was working but took a lot of pressure to move the diaphragm.) Also replaced the vac lines associated = no change.

-Pulled all pass side coils and plugs. All plugs looked good and identical. All coils showed no obvious signs of damage or failure. All resistance measurements of coils were the same. = no change.

-Replaced vac line from EGR to manifold. = no change

-Ran engine at warm idle and measured voltage at battery terminals. 13.87v and would drop below 13.80v when engine would stumble. I am using a fairly inexpensive craftsman mulitmeter so I don't know how qiuck it responds to changes.

-I can swear that I am hearing a vacuum leak under manifold just behind the PS reservoir. Also a clicking chatter occurs just off-idle (again from under the manifold) and goes away when car falls to slow idle. Similar the the tick of fuel injectors but much louder and does not change frequency with rpm.

At this point I am thinking that I still have some sort of vacuum leak in some component under the manifold or as Socal suggested, a by-pass air clog, etc. Sounds like the manifold needs to come off again?

Will still try cleaning throttle body and checking snorkel carefully for any leaks.

On a side note: I found it interesting that the resin-fill on the tops of the coils are different colors. The rear and middle coil tops are off-white and semi-transparent while the front coil was dark brown and completely opaque. The part ID stickers on them are different colors too. Two were white text and one yellow text. Huh??

Thanks so much for all the helpful advice and suggestions. I really think I am getting closer to the root of the problem.

[macadamianut]

Took two videos this morning with my ipod, sorry if they aren't high q. One of the stumbling idle with suspected vacuum leak noise, and one of this strange clicking chatter that happens just of idle in which you can also hear the potential vacuum issue. Both seem like they could be coming from under the intake manifold and neither seem proper.

http://www.youtube.com/watch?v=iVbaKN5pM10

http://www.youtube.com/watch?v=aCgUY8ZyZ7Y

Hope this helps.

Thanks again.

[SoCal LS-L]

That air suction sound is normal, mine does it and every other EG33 I have ever heard does that off idle and stops shortly after idle return.

[macadamianut]

Well, I decided to suck it up and roll up my sleeves last night.

I pulled the snorkel and upon careful inspection, the spigot for the driver side breather is starting to crack away from the snorkel assembly. It is a small crack were they are plastic welded together. If I push on down on the spigot, I can see the crack open up. The pass side is cracked as well but not as bad. These are not big cracks by any means, but they are right in front of the mouth of the throttle body. Will seal up and try later.

Also decided to "go for it."

-I now have the throttle body and by-pass air valve off and soaking in seafoam deep creep. Will replace both gaskets upon reassm.
-DVOM test on injectors = good.
-DVOM test on knock sensors = good.
-Set TPS to 0.502 volts - heard a strange noise (like a solenoid freaking out)from some other component when adjusting position.
-DVOM test on coolant temp sensor was not in cold range spec. but then it was 28 degrees in the garage last night. Might take it into the house and try again. Will test warn too if I manage to get the car running ever again.

Still planning to remove intake manifold and replace vac lines, test and or clean all underside components. Will replace gaskets again.

Thinking of pulling crankcase breather valves and cleaning along with the basically brand new PCV

[SoCal LS-L]

If you are talking about the "spigot" that runs along the bottom front of the intake tube, it shouldnt matter if its splitting away from the intake tube, its a standalone tube thats merely attached to the intake, it shouldnt be allowing any intake leakage.... IF its the one im thinking of.

As far as the coolant temp sensor range, mine is also always out of spec for the temperatue, constantly reading too high of a resistance (therefore too high of a temp to the ECU). Is yours doing the same? Is it OEM or aftermarket (like Bosch).? Do you have an IR thermometer to record temp along with your resistance readings?

I ordered a new CTS from jamie at chaplins subaru, ill let you know if the readings are any different than my Bosch crappy one. I know a bad CTS plays all sorts of havok with the ECU.