The Dreaded Misfire Saga Continues

[hankosolder2]

I just noticed something interesting this weekend, although I don't know how relevant it is. We traveled from Gunnison to Durango for a triathlon carrying two bikes on top, and running the A/C. That route takes us over two smaller passes and three big ones. The CEL came on after just a short time on the highway. I checked it and turned it off a couple of times. Same old thing #4 and #3 misfire - no indication of anything wrong except the CEL.

 

I noticed something that occurred over and over. We would cruise up a pass, that meant anywhere from 15 minutes, to 1/2 hour running along mostly in third at 3500 to 4000 rpm. At the top, as soon as I let off the gas to start down the other side, the CEL would begin to flash, and continue for as much as a minute, until I gave it some gas at a flat or slight uphill. Then it would just go back to a solid light. I've read that flashing CEL is supposed to indicate something major, but again, there is never any noise, missing, surging - nothin' but the light.

 

Does this mean anything in-particular? I would assume there would be a big difference in vacuum between uphill and downhill, but I don't know much about this.

I understood that the blinking CEL was to warn about any condition which can damage the catalytic converter- like a misfire or excessive richness for an extended period.

 

Here's what I'm thinking... coasting after accell while in gear is a high vaccum condition (as you noted.) If you have worn valve guides/seals, oil could be sucked into the combustion chamber, causing a temporary miss. (but you have a fresh CCR engine, right?) Oil could also be drawn in through the PCV system. (Did you reuse your old engine's PCV valve?)

High vacuum could also accentuate a problem with a leaking injector- did you ever swap them? (The difference between the fuel pressure and the manifold vaccuum determines the injectors flow...hence the need for a vacuum sensistive fuel pressure regulator.) I know that you consider it improbable that two injectors could fail simultaneously (true) but consider that they may have been exposed to the same contaminants in the fuel. Oh, also you could have a trace vacuum leak at the O-rings sealing the injectors to the manifold.

 

I have another crazy idea. I think your car has a small filter mounted in the vacuum line to the MAP sensor. When it gets restricted, it delays the MAP sensor's reaction to changes in vacuum, which can throw an EGR code (Computer verifies EGR flow by monitoring the change in manifold vacuum when the EGR valve opens...or so I hear.) Perhaps a slow reponse from the map sensor could also briefly effect the mixture during an accell/decell transition. Tolerance differences in the injectors, design of the manifold, ignition system etc. could make cylinders 3&4 more prone to misfire if the mixture is off slightly. (I'm clutching at straws here...it's just a theory.)

 

Good luck,

 

Nathan

[subsince77]

I understood that the blinking CEL was to warn about any condition which can damage the catalytic converter- like a misfire or excessive richness for an extended period.

 

Here's what I'm thinking... coasting after accell while in gear is a high vaccum condition (as you noted.) If you have worn valve guides/seals, oil could be sucked into the combustion chamber, causing a temporary miss. (but you have a fresh CCR engine, right?) Oil could also be drawn in through the PCV system. (Did you reuse your old engine's PCV valve?)

High vacuum could also accentuate a problem with a leaking injector- did you ever swap them? (The difference between the fuel pressure and the manifold vaccuum determines the injectors flow...hence the need for a vacuum sensistive fuel pressure regulator.) I know that you consider it improbable that two injectors could fail simultaneously (true) but consider that they may have been exposed to the same contaminants in the fuel. Oh, also you could have a trace vacuum leak at the O-rings sealing the injectors to the manifold.

 

I have another crazy idea. I think your car has a small filter mounted in the vacuum line to the MAP sensor. When it gets restricted, it delays the MAP sensor's reaction to changes in vacuum, which can throw an EGR code (Computer verifies EGR flow by monitoring the change in manifold vacuum when the EGR valve opens...or so I hear.) Perhaps a slow reponse from the map sensor could also briefly effect the mixture during an accell/decell transition. Tolerance differences in the injectors, design of the manifold, ignition system etc. could make cylinders 3&4 more prone to misfire if the mixture is off slightly. (I'm clutching at straws here...it's just a theory.)

 

Good luck,

 

Nathan

Thanks, I think I am about down to straws all the way around. It is a new PCV valve, although I did put it in right before the engine change.

 

No I have not switched injectors yet, and that is next on the list, just haven't had time. i did go through and check the wireing as far as I could - found not breaks or worn spots. I rewrapped #'s 3 and 4 and added extra insulation to those wires anyway.

 

Am i correct in thinking that all I have to do to pull an injector is unplug it and remove the two screws? Do they just pull out at that point? Any sneaky springs or clips going to fly off and get lost?

[OB99W]

[...]

Am i correct in thinking that all I have to do to pull an injector is unplug it and remove the two screws? Do they just pull out at that point? Any sneaky springs or clips going to fly off and get lost?

There's a bit more to it, and doing it improperly can damage things. Here's what End Wrench says on the topic: http://endwrench.com/images/pdfs/ProperInsideEWFall05.pdf

 

By the way, have you ever had a vacuum gauge on this engine? It can sometimes reveal subtle problems that are otherwise hard to pin down.

[McDave]

By the way, have you ever had a vacuum gauge on this engine? It can sometimes reveal subtle problems that are otherwise hard to pin down.

I suggested that in post #5 but my post count isn't high enough to be taken seriously yet. :-p

 

Actually anytime a car comes in with a misfire the vacuum gauge is one of the first tools I grab, often before the scan tool if it's missing at the time. You have to make sure the engine is mechanically sound before you start looking elsewhere, and a vg can tell you that quickly. If the problem is intermittent like the op's, the gauge is on a long hose and stays hooked up while I take a test drive, along with the scanner (obd II cars). I usually slip the gauge under the windshield wiper so I can watch it without taking my eyes off the road.

[FSRBIKER]

Just out of curiosity where are you guys taking your vacuum readings from?

[subsince77]

There's a bit more to it, and doing it improperly can damage things. Here's what End Wrench says on the topic: http://endwrench.com/images/pdfs/ProperInsideEWFall05.pdf

 

By the way, have you ever had a vacuum gauge on this engine? It can sometimes reveal subtle problems that are otherwise hard to pin down.

 

Thanks for the link. I'll read it before doing anything. And no, I haven't. I don't have any of that equipment. I'm very much a low end mechanic. I will need to take it in for that, and the place I generally take it may or may not know much about vacuum readings on Subarus. I'll ask, he'll tell me, he's a friend and not a bad wrench, just not a Subbie guy,

 

 

I suggested that in post #5 but my post count isn't high enough to be taken seriously yet. :-p

 

Not True!! I am just starting with things I can do myself. I don't have a real good Subaru shop around here. When things get beyond me I either settle for what we have locally, or take it over to Denver which is a major trip.

[subsince77]

Just out of curiosity where are you guys taking your vacuum readings from?

 

Yeah - what he asked.

[McDave]

Not True!! I am just starting with things I can do myself. I don't have a real good Subaru shop around here. When things get beyond me I either settle for what we have locally, or take it over to Denver which is a major trip.

Ha! I was just teasing. Granted swapping injectors is not so simple, but a test vacuum gauge is and cheap too.

 

You tee it in at any Manifold vacuum port in the intake manifold. Vacuum ports above, or upstream of the throttle blade are Ported, not Manifold vacuum and only get vacuum above idle. You want vacuum at idle (and cruising speed).

 

Smaller ports like those that go to the AT vacuum modulator, canister purge, or MAP sensor (on some cars) are easier to find a tee that fits the gauge hose too. PCV and brake booster vacuum ports can also be used, but the tee will be a mismatched size and harder to deal with.

 

There are big universal tees that have all the sizes stepped from large to small on one tee, you just trim the smaller ends off to fit the different size hoses you're dealing with. It wouldn't hurt to get an assortment of tees while at the parts store.

[subsince77]

Thanks. The injector job doesn't look bad at all, and the vacuum gauge is probably something I should buy and learn to use. Assuming I can find the right port to Tee into, I would still need to know what I am looking for. Just a reading that is low at a certain RPM or something? I assume I'm looking for a leak here. I'll see what my Haynes tells me when I get home, but if there are some simple instructions on what I would be looking for on the gauge, I would need that - told you, very low end mechanic.

[McDave]

Thanks. The injector job doesn't look bad at all, and the vacuum gauge is probably something I should buy and learn to use. Assuming I can find the right port to Tee into, I would still need to know what I am looking for. Just a reading that is low at a certain RPM or something? I assume I'm looking for a leak here. I'll see what my Haynes tells me when I get home, but if there are some simple instructions on what I would be looking for on the gauge, I would need that - told you, very low end mechanic.

Basically what you want is a steady needle that only moves as you move the throttle. What you're looking for in diagnosing your particular problem is if the needle bounces when it is missing.

 

Actually you really don't want to see the bounce as it means you have a mechanical problem with your valvetrain. If the needle stays steady when it is missing, then the problem is electrical or ignition related instead.

 

Here's a real good site showing how the gauge reacts in different scenarios. (requires flash)

http://www.secondchancegarage.com/public/186.cfm

[subsince77]

Thanks again. I will start working on all of this as I get time. #1 I got the new igniter today, so I will install that first - easy and no reason not to.

 

#2 - I will swap the injectors and see if the misfire moves.

 

#3 - I will buy a vacuum gauge and figure out a way to hook it up so I can drive with it. I'll need some free time to get that one done.

 

I really can't tell you how much I appreciate the help. At least if I end up having to take it in someplace I will have eliminated a lot of possibilities by then.

[McDave]

You're welcome! Sounds like a good plan. Let us know how it goes.

[pdeltoid]

This site helped me out understanding how vacuum gages work.

 

http://www.secondchancegarage.com/public/186.cfm

 

I picked one up cheap at harbor freight.

 

Best of luck.

 

Mark

[OB99W]

I suggested that in post #5 but my post count isn't high enough to be taken seriously yet. :-p

Sorry, although I haven't personally found that a "high enough" post count ensures respect or that you won't be ignored -- but maybe I just haven't reached the right level yet.

 

 

Actually anytime a car comes in with a misfire the vacuum gauge is one of the first tools I grab, often before the scan tool if it's missing at the time.[...]

I've had a vacuum gauge a lot longer than a scan tool, and although it cost much less than the scanner, it's easily as valuable a diagnostic instrument for certain problems.

 

 

This site helped me out understanding how vacuum gages work.

 

http://www.secondchancegarage.com/public/186.cfm [...]

McDave, you have my sympathy.

[McDave]

McDave, you have my sympathy.

That's ok. Invisibility has it's advantages too.

[subsince77]

New igniter. Another CEL. Injector swap is next on the list. Maybe tomorrow night.

[subsince77]

I swaped the injectors this afternoon, but ran into some problems. Actually, the injector swap itself went just fine thanks to the help I received here and that end wrench article. However, even though I had loosened the gas cap, unplugged the fuel pump, and run the car dry, I still had gas flowing through the rails while I was moving the injectors around. The car was sitting level and it only had about 1/3 of a tank of gas in it. I'm not sure why I was still getting gas.

 

Anyway, I finished the job and put everything back together, but when I went to start the car the engine just jammed immediately. It would not turn over at all. I sounded like I had shoved a bolt into a cylinder or something. I had never experienced anything like that, and had no idea what was going on. I called a mechanic buddy who immediately suggested that if any fuel had run through while I was doing the injectors the engine might be hydro-locked. I had never heard of it, but I pulled all the plugs and sure enough a fair amount of gas came out of #1. I turned it over just for a second to get any more gas out. Then put the plugs back in. I thought the oil smelled like gas so before I did anything else I changed the oil and filter.

 

After that, the car started and ran fine. Now I just need to wait for the next CEL to see where the misfire is. Two questions:

 

Is there anything else I need to do after this hydro-lock deal? I couldn't think of anything else that would have been affected.

 

And, if I do have to change out another injector, what can I do to keep the gas from running into the engine? The only other thing I could think of is to park the car with the front uphill.

[McDave]

Ouch! Yeah, once you saw the gas was going to siphon you should have pinched off a rubber hose with a hose clamp or even vice grips.

 

Since it runs fine and you didn't mention anything about noise, I guess you got lucky and didn't bend a rod.

[OB99W]

[...]

Anyway, I finished the job and put everything back together, but when I went to start the car the engine just jammed immediately. It would not turn over at all. I sounded like I had shoved a bolt into a cylinder or something. I had never experienced anything like that, and had no idea what was going on. I called a mechanic buddy who immediately suggested that if any fuel had run through while I was doing the injectors the engine might be hydro-locked. I had never heard of it, but I pulled all the plugs and sure enough a fair amount of gas came out of #1. I turned it over just for a second to get any more gas out. Then put the plugs back in. I thought the oil smelled like gas so before I did anything else I changed the oil and filter.

 

After that, the car started and ran fine.[...]

Consider yourself lucky. See http://www.ultimatesubaru.org/forum/showthread.php?t=89121 to find out why.

 

I assume the gas didn't seem to be under pressure (it shouldn't have been).

[subsince77]

See, this is the kind of ignorance (my own) that keeps me from tackling new jobs on my engine. I know I don't know much, but I don't know WHAT I don't know, if you know what I mean.

 

No the fuel was not under pressure, just trickling along. I didn't really think about where it was going, I was just trying to work quickly to get the injectors swapped and stop the flow. I was unaware of any of this hydro-lock stuff, but now it seems obvious and makes perfect sense.

 

The engine barely moved, a partial turn, then a solid stop. I did give it a couple more quick tries, but it was clear that something was very wrong, so I just stopped everything until I had the plugs out. After replacing the plugs and changing the oil, it started a little hard, and ran poorly for just a minute, like a flooded engine. But then it ran fine, sounded good, idles smooth. After reading all this, I will see how it seems today, and really listen for anything unusual, but I think it's OK. I spent quite a bit of time under the hood listening and checking to see if anything was leaking, and I didn't notice any bad sounds.

[OB99W]

[...]After replacing the plugs and changing the oil, it started a little hard, and ran poorly for just a minute, like a flooded engine. But then it ran fine, sounded good, idles smooth. After reading all this, I will see how it seems today, and really listen for anything unusual, but I think it's OK. I spent quite a bit of time under the hood listening and checking to see if anything was leaking, and I didn't notice any bad sounds.

It sounds like you knew when to stop, got good advice from your buddy, resolved the problem, and did the right thing by changing the oil and filter. Since the engine seems to be running/sounding okay, it's reasonably likely that no damage was done. It makes sense to use the car, see how things go, and then let us know what (if anything) changed after the injector swap.

[Olnick]

I know I don't know much, but I don't know WHAT I don't know, if you know what I mean.

 

Wow! You're just like me!

[frag]

See, this is the kind of ignorance (my own) that keeps me from tackling new jobs on my engine. I know I don't know much, but I don't know WHAT I don't know, if you know what I mean.

 

According to Socrates, the real problem is when you don't know THAT you don't know. Philosophy and the art of Subaru maintenance...

[subsince77]

According to Socrates, the real problem is when you don't know THAT you don't know. Philosophy and the art of Subaru maintenance...

I mostly just really liked writing that sentence.

 

I didn't use the car today because we were going to a trailhead that required a bit more clearance. I did start it though, and it seems fine as far as I can tell, so on we go.

[frag]

I mostly just really liked writing that sentence.

 

 

And you should. Nothing wrong with it. I meant mine only as a complement.

Have a good day!