Subaru being a Beeeotch

[denverpilot]

The poor old Subie had started missing a little at idle and an occasional stumble on acceleration. I figured it was time to start a piecemeal tune up, buy parts, do a stage at a time... so a new set of NGK plugs and wires was tossed into the Amazon Prime order list for this week.

Symptoms: It would stumble at idle irregularly and do it more under any load. (Auto trans in gear, A/C on.)

Today driving over to the airport (it's the airport beater car ... all my flight stuff stays in it...) it stumbled a little more than usual at stop lights but not severe... and after I pre-flighted the airplane at the hangar and started it up to go back over to my CFIs office, it finally tossed a Check Engine light. Will pull the codes when I get home but I assume it'll be a misfire or knock sensor. (It's not knocking but the Subie knock sensors are sensitive.)

So... hopefully the plugs have shown up today. The wires showed up yesterday. Would be nice to pull them and see what they look like. Only 40,000 miles on them since the timing belt and head gasket stuff were done on the usual Subie schedule of around 90,000 miles. And I'm not sure the shop replaced the wires, so these could be the originals. Currently 130,000 total.

Since it's me, and cars, this is almost guaranteed to be an O2 sensor, probably the front one. Because you know. It has to be the absolutely most annoying thing to fail that will fail on my stuff. That's just Murphey doing his thing.

But here's the fun part: Since the CEL came on, no misfires. That's good and bad. Whatever isn't happy, the computer has decided to ignore it. That means it ain't the plugs and wires. Ha. So I bet I find an O2 sensor code in there too...

Guess I'll be buying an O2 sensor socket. I don't think I've had the "joy" of pulling one of those yet. And I don't feel like grinding a slot in one of my other sockets. Might as well add a new tool to the collection if it's going to cost me skinned knuckles and swearing.

 

[FORANE]

Get a h6 version subie. they have timing chains with no recommended replacement interval plus fewer head gasket failures.
The o2 sensor should set a code if bad. that code should be specific to which one is bad also.
Might be a cam position sensor.

 

[455 Bravo Uniform]

Check all your vac hoses for cracks, leaks.

 

[PaulS]

Get the codes, sounds like a sensor may be buying the farm, or you waited a little too long for your tune up.

 

[PaulS]

O2 wrench is cheap. Doesn't sound like an O2 sensor to me though, maybe air flow sensor? Let us know what you find.

 

[jsstevens]

Well what do you expect from O2 sensors up where there's so little O2?

I hope it's simple. I'd definitely buy the socket. It's worth it even for 1.

John

 

[denverpilot]

Well what do you expect from O2 sensors up where there's so little O2?

Ha. Usually stuff runs horribly rich up here if the O2 sensors or MAF sensors have failed. Which kinda matches the symptoms, of only being unhappy at idle. I suspect it's running super rich.

Just for fun, after leaving the airport with the CEL on, it's idling okay now but definitely stumbling during acceleration.

Haven't made it home yet.

 

[denverpilot]

Interesting data. Need to look up some stuff.

Code was a generic Cylinder 3 Misfire.

MAP looks sane at about 4 when idle and higher when reviving.

MAF (lb/s) sits at 0.01 at idle but never goes above 0.4 on no load rev to 5000 RPM and held there. That looks suspicious to me.

O2BIS2(V) hovers around 0.8 at idle and 0.9 revved.

O2BIS1(V) 1.5 idle and 1.1 revved.

I'll have to go look up those O2 sensor numbers and see if they're sane.

 

[Lachlan]

I thought my Grand Cherokee was fast in Colorado... since we moved to northern NY state and are at 1075' asl I've learned exactly how noticeable the hp increase hits you in the seat of your pants.

Also, I can get a fully fueled Cessna 150 off a grass field with 2 adults onboard in less distance here than it took to get a solo 180hp 172 to ping the ASI at KAPA!

Hope you get the Subi fixed. Cars suck.

 

[NoHeat]

Check all your vac hoses for cracks, leaks.

That was my thought, too.

 

[FORANE]

Coil pack? Corrosion?

 

[PaulS]

Interesting data. Need to look up some stuff.

Code was a generic Cylinder 3 Misfire.

MAP looks sane at about 4 when idle and higher when reviving.

MAF (lb/s) sits at 0.01 at idle but never goes above 0.4 on no load rev to 5000 RPM and held there. That looks suspicious to me.

O2BIS2(V) hovers around 0.8 at idle and 0.9 revved.

O2BIS1(V) 1.5 idle and 1.1 revved.

I'll have to go look up those O2 sensor numbers and see if they're sane.

Are the O2 sensors switching? I think the plugs and wires may be your issue, but those O2 sensors should be switching, may be time to change those too.

 

[Hank S]

Interesting data. Need to look up some stuff.

Code was a generic Cylinder 3 Misfire.

MAP looks sane at about 4 when idle and higher when reviving.

MAF (lb/s) sits at 0.01 at idle but never goes above 0.4 on no load rev to 5000 RPM and held there. That looks suspicious to me.

O2BIS2(V) hovers around 0.8 at idle and 0.9 revved.

O2BIS1(V) 1.5 idle and 1.1 revved.

I'll have to go look up those O2 sensor numbers and see if they're sane.

I've been having the same issues in my Mazda-built Ford Ranger, about the same mileage. It was very bad Monday evening after being parked in front of the house instead of in the driveway when a bad storm them blew through in the afternoon. Auto Zone checked the code, Cylinder 3 misfire during first 1000 revs. Their solution is a new head . . . But it's running beautifully now, better than the last several months. No plans for now other than an oil change.

 

[denverpilot]

Check all your vac hoses for cracks, leaks.

That was my thought, too.

It's a possibility. There's not much stuff on the '00 that's vacuum driven that would cause a misfire though.

Coil pack? Corrosion?

That's next. Plugs and wires first since this thing is designed for copper plugs. I realized as I typed that earlier that 40,000 miles is a bit long for copper tips. Don't know what's in there but platinum and iridium and what not aren't recommended in the 2.5L with the separate coils and wires. Mucho good in the coil on plug stuff but not what this old beast likes.

I bet I find eroded plug tips. Ha.

Are the O2 sensors switching? I think the plugs and wires may be your issue, but those O2 sensors should be switching, may be time to change those too.

Not sure what you mean by "switching"? Please advise. Ha.

Dumb me. Was in a hurry on Amazon the other day and the package arrived. There's all the stuff I ordered and ... one ... plug. LOL. Dumbass. So much for Prime being helpful tonight. Went to town and got a set of plugs. Ha.

More later. Have to get back and find my %^**%# 10mm socket to remove all the frogging plastic bottles and windshield washer bottle to get at the plug holes...

20 minute drive home first. LOL. Oh well. It's a nice night out. And I don't need the car tomorrow if I get in there and don't get done, but we'll see.

 

[denverpilot]

Oh. Forgot to mention. The coil packs are swappable. One per two cylinders. So if this doesn't help I'll swap and see if the problem follows a pack.

 

[GlennAB1]

Water in gas tank.

 

[denverpilot]

Water in gas tank.

LOL. With 10% corn juice in all our gas around here, not much problem with that.

 

[denverpilot]

Damn. Looks like the valve cover gasket is gone. Here's the offending "misfire" plug, #3. It doesn't look awful but it's definitely time for a change.

But the bad news was when I pulled #1. It's just covered in oil. Grrrr.

Boot also has some oil on the outside of #1 so the valve gasket cover replacement 40,000 ago has failed. I'm amazed it isn't #1 that's misfiring.

Here's 1 and 3 side by side. Grrrr.

 

[James331]

Brother, I hate to suggest it, but you should do a compression test.

2.5 missing like this, that's got a head gasket issue written on it.

How they went from the awsome 2.2 to that 2.5 is beyond me, especially for Subaru, but yeah.

 

[denverpilot]

2 and 4 looked much better. Still need replacing but no massive oil leak around the boots.

So that did it. Plugs were shot. Cleaning up that friggin number 1 hole to get the new plug in there is entertaining.

And whatever Japanese dickhole who designed those washer fluid bottle connections, both electrical and the hoses, should be shot on sight. Getting that bottle back in was harder than the entire rest of the job.

Here's 4...

And here's 2...

And yeah. Some idiot put platinums in the thing. They don't belong in this engine.

Proper NGK coppers/regulars are in, factory gap was amazingly, dead on, and it's idling right and needs a test drive to finish clearing the CEL after clearing the code.

It's idling about 100 RPM higher, no stumbles at idle or when goosing it. Gottta go find Karen's truck keys to get it out of the way so I can get the Subaru out of the garage for a quick test drive boogie.

Plug wires were $46, plugs were $14. Best $60 tune up ever. I also treated myself to a magnetic spark plug socket because I really hate dropping the things, even the old bad ones. But especially never the good ones! Ha. That was $12. Overpaid for that one but it also had a wobble head. Nice!

That plus the angled socket wrench from Champion that I've had forever for doing plugs, makes life so much simpler in tight spaces!

This is one very happy sounding 17 year old four banger.

 

[denverpilot]

Brother, I hate to suggest it, but you should do a compression test.

2.5 missing like this, that's got a head gasket issue written on it.

How they went from the awsome 2.2 to that 2.5 is beyond me, especially for Subaru, but yeah.

I found evidence that the stupid always leaking valve cover gasket is toast and probably the plug well seal, but had the head gasket done just after 90,000 along with the stupid timing belt. I know those go on these.

It's literally a recurring maintenance item on Subaru 2.5L motors. Annoying but if you're in there to do the timing belt, just do the head gasket too. No point in waiting. It'll blow around the same number of miles. Valve adjustment since you're in there too. No tick tick tick like the unmaintained ones.

Looks like she's got another 40,000 in her for sure. Just have to deal with that friggin oil leak next.

It's purring away here now. Got her up to op temp and waited for the electric fan to start cycling. Still happy.

Sounds better than it has in three years actually. Whoever put those platinum plugs in should be slapped. Not made for this old ignition system and they just start eroding immediately in this old remote coil stuff.

 

[denverpilot]

Whoa. Power. Just the little two mile jaunt to the gas station and wow... that's a big difference.

Going to run into town on the hills and go deposit a check at the drive up ATM just since it gives me an excuse to go get the computer to come out of "I just got reset and need to recalibrate" mode.

I don't think she's ever chooched this well!

 

[denverpilot]

Well 40 mile round trip, computer all back to normal, and holy crap the thing has more power than it ever has.

Almost embarrassing that I let the plugs get that bad. Had no idea. It's always been a dog on the hills out here and I just thought that was normal for the wimpy four banger.

Take that back. Not almost embarrassing. Just embarrassing. Ha. Oh well. Mucho better now.

 

[arnoha]

... chooched...

Is this an actual slang term? I heard some Canadian guy on YouTube using it. Meaning is clear enough, but I'd never heard it before. He invents lots of words, so assumed this was also one of them...

 

[Clark1961]

Good to hear you figured it out. Sorta interesting about the plugs. My 20 year old explorer uses plats...came from the factory with them. Similar style coil pack ignition system. The plats easily go 100k.

The crappy thing about the explorer is the 2 piece plastic intake manifold and the intake manifold gaskets that only last 18 years or so...the fasteners that screw into the plastic are very easy to strip and the repair is to replace the part. Techs aren't used to the plastic parts so they strip them.

 

[denverpilot]

Is this an actual slang term? I heard some Canadian guy on YouTube using it. Meaning is clear enough, but I'd never heard it before. He invents lots of words, so assumed this was also one of them...

Nah, I'm just a fan of said Canadian guy. Said Canadian guy is funny and he does stupid crap in his shop, which is a dying art in the land of Metrosexual males.

Engage Safety Squints!!! Corn-tact!!!

 

[denverpilot]

Good to hear you figured it out. Sorta interesting about the plugs. My 20 year old explorer uses plats...came from the factory with them. Similar style coil pack ignition system. The plats easily go 100k.

The crappy thing about the explorer is the 2 piece plastic intake manifold and the intake manifold gaskets that only last 18 years or so...the fasteners that screw into the plastic are very easy to strip and the repair is to replace the part. Techs aren't used to the plastic parts so they strip them.

God Ford drives me nuts. Plastic intake manifold!? LOL.

I swore off Fords a long time ago, but then Karen goes and falls madly in love with the Bling Truck (Lincoln LT) and has to have it, and against my better judgement we buy another Ford... one with a motor in it that has successful class-action lawsuits against it... LOL.

Oh well. They all do stupid stuff really. Just annoying when you're under the hood going, "There's no way any sane engineer thought THIS was a good idea..."

Now that I know the stupid valve cover gasket at least one plug well seal is leaking like a sieve, I guess I have to go price the set of gaskets and seals and then take all that plastic crap out again to get at the valve covers and do THAT job here sooner or later.

It's move all the plastic junk and ten bolts, so I'd survive it. Just a waste of four hours to "fix" a leak that'll just happen again.

Or I can be lazy and take it to the Subie shop over on Iliff. Those guys are great but I hate paying them for something that stupid easy to fix. I usually "save" those trips for when it's something that would take ten minutes on a lift and will take me ten days in my garage without one. LOL.

I really want a barn, a lift, and a welder. Then I could get going on some real havoc out here in the boonies. That's what you need for real fun, along with some much bigger breaker bars and literally breaking suspension components... LOL.

Let's just get this tie rod end off here... first we will start with a little heat... (Nothing ever goes well when the torch comes out to heat something so you can try again with a bigger breaker bar... well, until you've bought all new parts by the end of THAT kind of project. Hahaha.)

I did get a chuckle that so few do any work on vehicles anymore that some guy posted a YouTube video of how not to die when working on struts and strut towers...

"See these nuts here? These are the ones you should loosen for this job. If you loosen that one there, the strut launches itself under its own internal pressure through the hood of your car and bad things happen."

LOL. Cracked me up watching that one.

 

[Bill]

Get a h6 version subie.

I just took delivery of a new '17 Outback 3.6R Limited, it's really nice.

God Ford drives me nuts. Plastic intake manifold!? LOL.

I swore off Fords a long time ago, but then Karen goes and falls madly in love with the Bling Truck (Lincoln LT) and has to have it

I suppose it's just what you end up getting. My 2001 F-150 Supercrew (original 2v 5.4) is the best car I've ever owned, by an order of magnitude. In 17 years, I spend $1600 on unexpected repairs, and most of them were easy fixes like replacing a bad COP. That truck rarely broke, never left me stranded, and stood up to my driving style with no issues. I sold it to my brother last week, and he loves it. There is still not a thing wrong with it, I'd drive it cross country with no qualms. I just finally got the new car itch after 17 years.

 

[PaulS]

It's a possibility. There's not much stuff on the '00 that's vacuum driven that would cause a misfire though.



That's next. Plugs and wires first since this thing is designed for copper plugs. I realized as I typed that earlier that 40,000 miles is a bit long for copper tips. Don't know what's in there but platinum and iridium and what not aren't recommended in the 2.5L with the separate coils and wires. Mucho good in the coil on plug stuff but not what this old beast likes.

I bet I find eroded plug tips. Ha.



Not sure what you mean by "switching"? Please advise. Ha.

Dumb me. Was in a hurry on Amazon the other day and the package arrived. There's all the stuff I ordered and ... one ... plug. LOL. Dumbass. So much for Prime being helpful tonight. Went to town and got a set of plugs. Ha.

More later. Have to get back and find my %^**%# 10mm socket to remove all the frogging plastic bottles and windshield washer bottle to get at the plug holes...

20 minute drive home first. LOL. Oh well. It's a nice night out. And I don't need the car tomorrow if I get in there and don't get done, but we'll see.

So it sounds like you found your problem with plugs, 40k miles is a lot for regular plugs, I change them at 30k when I own a car without 100k plugs, which I don't now.


O2 sensors - The O2 sensors should be switching. The 1st one, the one before the converter, should go from high to low about every second or so. The one after the converter, should be pretty steady with occasional rises and dips. If they are stuck at one value then most likely the sensor is bad. If that is happening then I would replace the sensor in front of the converter first.

The sensors switching is driven by the computer. For the catalyst to work it needs to be recharged with oxygen, so the mixture goes lean. Once it is recharged with oxygen it needs fuel to keep the convertor hot enough to keep the reaction going so it goes rich, this happens over and over again.

Sounds like your sensor(s) are stuck, which is a sign they have failed. The car should run even better if you replace the failed sensor, if it has failed.

Read here for more info, this is one of many articles on sensors, google is your friend. Also, a stuck sensor won't necessarily turn on the CEL.

http://www.certifiedmastertech.com/2010/10/11/testing-oxygen-sensors/

 

[jsstevens]

There's so much "how to" stuff on youtube anymore it's great. But you have to know enough to watch and realize "This is good." or "This will be entertaining, but I'm not doing it."

Those plugs were really worn! One thing that's amazing to me about all these computer controlled engines is how bad basic stuff like that can get and they still run acceptably well. My first car was a 1969 Renault R-10. Conventional points and coil ignition. It would go from 35mpg to 25mpg in one tank of gas when the points went. The basic stuff (tune up stuff) was easy to do on it and I enjoyed it. (I tuned cars for pocket money when I was in HS & college. It was easy and fun for a car guy and I was badly undercutting the shops-sorry!) But those cars let you know right away when things got out of whack. Now the computer just keeps compensating until it really bad.

John

 

[denverpilot]

So it sounds like you found your problem with plugs, 40k miles is a lot for regular plugs, I change them at 30k when I own a car without 100k plugs, which I don't now.


O2 sensors - The O2 sensors should be switching. The 1st one, the one before the converter, should go from high to low about every second or so. The one after the converter, should be pretty steady with occasional rises and dips. If they are stuck at one value then most likely the sensor is bad. If that is happening then I would replace the sensor in front of the converter first.

The sensors switching is driven by the computer. For the catalyst to work it needs to be recharged with oxygen, so the mixture goes lean. Once it is recharged with oxygen it needs fuel to keep the convertor hot enough to keep the reaction going so it goes rich, this happens over and over again.

Sounds like your sensor(s) are stuck, which is a sign they have failed. The car should run even better if you replace the failed sensor, if it has failed.

Read here for more info, this is one of many articles on sensors, google is your friend. Also, a stuck sensor won't necessarily turn on the CEL.

http://www.certifiedmastertech.com/2010/10/11/testing-oxygen-sensors/

Didn't realize there's a cycle there. But saw it happening on the live data from the OBDII last night. Interesting. The front O2 sensor sees it as a very low number and then a higher number and it oscillates. Neat. Didn't know that's how they work the Cat. Know it works similarly with regen on the newer diesels, similar process just a lot slower.

 

[asicer]

God Ford drives me nuts. Plastic intake manifold!? LOL.

I thought all new cars have plastic intake manifolds these days. Which ones still use metal?

 

[arnoha]

Didn't realize there's a cycle there. But saw it happening on the live data from the OBDII last night. Interesting. The front O2 sensor sees it as a very low number and then a higher number and it oscillates. Neat. Didn't know that's how they work the Cat. Know it works similarly with regen on the newer diesels, similar process just a lot slower.

This is also why the computer goes into open loop mode at WOT. Want full power? Well, that means that you can't be switching between rich and lean: only one optimal place. Without a wide-band sensor, there's no real fine-tuning, just tells you rich or lean. (And wide-band sensors are way too expensive for consumer vehicles.) So, it goes back to the internal tables based off the upstream sensors, but ignores the feedback loop of the O2 sensor.

 

[denverpilot]

This is also why the computer goes into open loop mode at WOT. Want full power? Well, that means that you can't be switching between rich and lean: only one optimal place. Without a wide-band sensor, there's no real fine-tuning, just tells you rich or lean. (And wide-band sensors are way too expensive for consumer vehicles.) So, it goes back to the internal tables based off the upstream sensors, but ignores the feedback loop of the O2 sensor.

Ahhhhhh! Coolness. I've watched it go into and out of CL and OL and that info really closes the loop (pun intended) on what they're doing that for!

 

[denverpilot]

I thought all new cars have plastic intake manifolds these days. Which ones still use metal?

Hell if I know. I drive old stuff! Not that metal doesn't crack...

Karen's VW diesel was a nightmare on manifold replacements. Both intake and exhaust cracked. Metal.

Out of all the cars we've owned, that one was by far the biggest POS.

 

[arnoha]

I thought all new cars have plastic intake manifolds these days. Which ones still use metal?

My Subie is metal, and it's a bit newer than @denverpilot 's. It's also an STI, so I don't know if the more pedestrian models have switched to plastic. I'll bet not, though, due to the flat engine design. It's a guess, but, with the intake on the left and right being so far apart, I doubt plastic would have the material strength, especially with engine heat added. In particular, the flat engine rocking couple vibration mode would stress the plastic right in the middle where it's connected by fat, flexible rubber hose to the rest of the intake that's attached to the body instead of the engine. To make matters worse on the turbo engines, the intercooler is top mounted right there as well.

On inline and V engines, there are no long, skinny, and unsupported sections. Everything is bunched up together so it has bulk strength. Not true for a flat 4 with two long, skinny pipes going each direction.

EDIT: also, I love my Subie. 115,000 miles on that high-power, high-stress, high-revving engine and very little has broken. The TGVs failed early, so I had them deleted. The oil pickup cracked, a known weak spot. Replaced with the stronger Moroso part and no more problems.

Unfortunately, I think my rear main seal is now leaking. Bleh. It only leaks under load, so no oil puddle. But it's getting on the exhaust, so the car smells like the devil's ass and is somewhat noxious to drive. So I'm driving my truck while I decide if I'm going to fix it or sell it.

 

[arnoha]

Well, damn. I'm wrong.



EDIT: and holy crap, they moved the oil filter to the top of the engine! (1) Hella convenient! (2) OMG, the mess.

Also, I note they now use corrugated plastic, not rubber, for the join. Wonder if that reduces the stress.

 

[denverpilot]

Well, damn. I'm wrong.

View attachment 54075

EDIT: and holy crap, they moved the oil filter to the top of the engine! (1) Hella convenient! (2) OMG, the mess.

Also, I note they now use corrugated plastic, not rubber, for the join. Wonder if that reduces the stress.

The oil filter location looks incredibly stupid. LOL.

 

[mondtster]

This is also why the computer goes into open loop mode at WOT. Want full power? Well, that means that you can't be switching between rich and lean: only one optimal place. Without a wide-band sensor, there's no real fine-tuning, just tells you rich or lean. (And wide-band sensors are way too expensive for consumer vehicles.) So, it goes back to the internal tables based off the upstream sensors, but ignores the feedback loop of the O2 sensor.

There are plenty of production vehicles with wide band o2 sensors these days. Where do you think the ones that all the aftermarket sensor kits come from?

Running off look up tables with no feedback from an o2 sensor in open loop mode can be plenty accurate, no wbo2 needed. The engineers know how much air is going to be moving through the engine at different load points and can inject an appropriate amount of fuel without using a sensor to adjust. When in open loop under heavy load the fueling goes rich for the same reasons we run aircraft engines rich under heavy loads.

My Subie is metal, and it's a bit newer than @denverpilot 's. It's also an STI, so I don't know if the more pedestrian models have switched to plastic. I'll bet not, though, due to the flat engine design. It's a guess, but, with the intake on the left and right being so far apart, I doubt plastic would have the material strength, especially with engine heat added. In particular, the flat engine rocking couple vibration mode would stress the plastic right in the middle where it's connected by fat, flexible rubber hose to the rest of the intake that's attached to the body instead of the engine. To make matters worse on the turbo engines, the intercooler is top mounted right there as well.

Unfortunately, I think my rear main seal is now leaking. Bleh. It only leaks under load, so no oil puddle. But it's getting on the exhaust, so the car smells like the devil's ass and is somewhat noxious to drive. So I'm driving my truck while I decide if I'm going to fix it or sell it.

Subaru has used plastic intake manifolds, at least back as far as 2005 with the Legacy GT. The '08+ WRX has a plastic intake as well. I haven't looked at the F series engines for a while now but I think they have plastic intakes too.

For the oil leak, I'd start by looking at the turbo oil drain line or maybe a right side valve cover. Those both sit right above or near the exhaust.

 

[asicer]

The oil filter location looks incredibly stupid. LOL.

Just like plastic intake manifolds a few years ago, top mounted oil filters are also starting to become fashionable.