'99 Legacy Outback engine flutter... HELP!

[darrenbrigman]

Hello everyone! I'm Darren from south GA, and recently purchased a '99 legacy outback with the twin cam 2.5 liter engine/auto trans. The engine 'flutters' for lack of a better term. It almost feels like a cylinder misfire, but it usually isn't that bad. The car has 180,000 miles on it, but appears to have been taken care of fairly well. The CEL was on when I bought the car, and codes showed misfire on 3 of the 4 cylinders. I changed the plugs, reset the light, and it came back on within 100 miles. Same misfire codes. Then put new plug wires on, and reset. I put about 400 miles on it, and the CEL stayed out. The flutter was still there though. (although it did run much better after the plugs and wires.) I put a new fuel filter in, and replaced all 4 injectors. Car runs great above 2000 rpm, but below that, I still have the flutter. The exhaust seemed to be restricted, when comparing the output from the tailpipe with another '99 that a friend has. I pulled the cats off and... well, yeah, I removed a portion of the catalyst brick. Was going to see if this fixed the problem, and if it did, was going to order new cats. Anyway, the car has much better throttle response now, but still has the flutter. Strangely enough, when I reset the CEL light, it seems like the car runs perfectly with no flutter for a few miles. After 'fixing' the cats, I'm getting the CEL again, but this time it is showing a code for catalyst inefficiency. I'm sure this is due to the cats being open. I'm showing .08 volts on the front O2 sensor, and .075 on the rear. I certainly don't mind spending the money to get this thing running right, but I hate to just throw money at it until it works. My wife is a little ticked at me for selling her volvo V70 to purchase this one... guess I should have asked first! I do have a mechanic buddy who says that the problem is a bad inner CV joint, but after putting it in neutral and bringing the engine up to about 1200 rpms, the flutter could be felt. Then he said it has to be a motor mount. I think he's just guessing at this point. Anyway, if anybody has any suggestions, please let me know!

[svxpert]

<<I do have a mechanic buddy who says that the problem is a bad inner CV joint, but after putting it in neutral and bringing the engine up to about 1200 rpms, the flutter could be felt. Then he said it has to be a motor mount. I think he's just guessing at this point.>>

 

yeah, dont think he's a car mechanic. did you check the coil pack? no need to keep replacing plugs like you did.

[nipper]

Pay no attention to the Cat codes, you have no cat, so ignore it, as you can not do anything about it.

 

Have you done the water test? Spray a mist of water on the coils in the evening and look for sparklies? Thast where i would start.

 

Did you use subaru wires?

 

What kind of plugs did you use?

[darrenbrigman]

I haven't done the water test yet. I will give that a try tonight. Plugs are NGK, and the wires... well, no, the wires are not factory subaru wires. (was trying to save a few bucks.) If the coil is acting up, would it not trigger the CEL for cylinder misfire? As you can tell, I'm not much of a mechanic when it comes to these things.

[doc526]

Subaru wires are the best for the replacement. Most people have issues with using anything else. It is possible that coil packs on their way out can cause a missfire.

[bheinen74]

go buy the Subaru plug wires. You did not save money by buying the cheapies, that is probably the whole issue, you can search that and see.

 

Subaru uses a "wasted spark" design ignition. Hardly any other manufacturer does so. the plug wires are very critical that they are designed for this. Cheapy brands do not even know what wasted spark is, and they just make the plugs like they would for other cars. does not work.

[OB99W]

[...] I'm showing .08 volts on the front O2 sensor, and .075 on the rear. [...]

If those readings are accurate, you might want to investigate that area a bit -- that could indicate an extremely lean mixture, or bad sensors.

 

See http://www.aa1car.com/library/o2sensor.htm

[nipper]

If those readings are accurate, you might want to investigate that area a bit -- that could indicate an extremely lean mixture, or bad sensors.

 

See http://www.aa1car.com/library/o2sensor.htm

 

 

Is hard to take O2 reding seriously at this point as A- he has a miss B- he has no cats,

 

meowless he is.

 

nipper

[darrenbrigman]

Yep, I'm not overly concerned about the cat codes now. As soon as I get the skipping fixed, I'm going to put new cats on the car. I don't want to spend the money on new cats (they are quite pricey!) only to find out that I have a problem that will damage the cats. I have ordered a set of subaru wires, and as soon as they come in, I'll put them on. I'm also going to try swapping coils with my friend's car to see if that makes a difference. Thanks for all the suggestions! It's nice to see that there are people out there who actually understand subies. It seems that I'm the only one in this part of the state that actually likes a subaru!

[GeneralDisorder]

You can fix the cat code by installing a spark plug non-fouler between the rear O2 sensor and the pipe. This will keep the code from going off and you won't have to replace the cat.

 

But that has nothing to do with the "flutter" you are experienceing. I also would sugest OEM plug wires or NGK brand wires. Factory is excelent and I've used the NGK's for many projects without problem as have other member's here. Just put a brand new set on an EJ22 I swapped into a Brat and they are working great.

 

Other than that - if you could post a video of what your are experiencing it might help us.

 

GD

[OB99W]

Is hard to take O2 reding seriously at this point as A- he has a miss B- he has no cats

Having no cat would result in the downstream sensor reading about the same as the upstream one, which seems to be the case. It doesn't explain the low reading.

 

A misfire severe enough to set P03xx codes would cause there to be more unburned oxygen in the exhaust, which could cause the sensor to read low. However, the OP didn't indicate that the misfire codes recurred after he did some work. If there is now no misfire, or it's occurring but not sufficiently to set codes, then it isn't likely to drive the O2 sensor extremely low (lean) -- but something obviously is causing the low reading.

 

The OP indicated that the problem disappears for a while after the ECU is reset, then returns. It's hard to believe that ignition problems are being fixed, even temporarily, by having the ECU go to default settings. A more likely possibility is that the fuel trims are initially more correct after a reset, and then are modified to values that the engine doesn't like as much as time/miles go by. I've also seen an overly sensitive knock sensor unnecessarily retarding timing, causing symptoms similar to the "flutter" the OP described.

 

It would be interesting to see some scan data beyond trouble codes. STFT and LTFT numbers and ignition timing would be a good start.

[nipper]

You can fix the cat code by installing a spark plug non-fouler between the rear O2 sensor and the pipe. This will keep the code from going off and you won't have to replace the cat.

 

But that has nothing to do with the "flutter" you are experienceing. I also would sugest OEM plug wires or NGK brand wires. Factory is excelent and I've used the NGK's for many projects without problem as have other member's here. Just put a brand new set on an EJ22 I swapped into a Brat and they are working great.

 

Other than that - if you could post a video of what your are experiencing it might help us.

 

GD

 

If there is no cat no matter what trick he uses it isnt going to work.

[GeneralDisorder]

If you pull the 2nd O2 out of the exhaust stream with a non-foulder (or two), it will read low enough to not throw the code. Lots of folks with catless exhaust systems have used that trick. Done right, it will work without any cats.

 

GD

[darrenbrigman]

Well, I think I have good news! I put a new coil on, and put an anti-foul adapter between the rear 02 and the cat, cleared codes, and let the engine warm up. I'm idling at 712rpm, maf at 3.89 to 3.93, map KPA at 30, coolant temp at 188, 15 degrees ignition advance, STF trm1 5.1 to 5.7, fuel system 1 closed, 02s11 .72 to .74 volts, 02s12 at .785 volts, and best of all, no engine flutter! I honestly have no idea what most of the info above means, but that is what the scanner showed. It also shows catalyst normal! I'm going to drive it a bit today and see if the CEL comes on again. Looks like it is time to order new cats. Thanks for all the help, you guys are the best!!

[gibsonsk]

If you want to know about the flutter, quite a few cars use the wasted spark system, and what happens is the coil starts to break down and instead of producing 90k volts they produce only 50k-70k and you do not get a clean burn. Due to the fact that the cylender needing the spark has preasure and fuel mixed in, raising the resistance between contact points on sparkplug. Mix this with the fact that power takes the 'path of least resistance', well the plug you need to fire dosen't, the other one does.

[OB99W]

[...]Due to the fact that the cylender needing the spark has preasure and fuel mixed in, raising the resistance between contact points on sparkplug. Mix this with the fact that power takes the 'path of least resistance', well the plug you need to fire dosen't, the other one does.

Welcome to the forum. I hope you won't mind a small addition to your first post here.

 

The spark plugs and wires of the paired cylinders are effectively in series. To work correctly, when ignition is triggered the spark must jump both the the plug in the cylinder that's supposed to fire and the other one in the pair. The paired cylinder that's not on a power stroke is just ending its exhaust stroke, and indeed presents little resistance to the spark jumping its plug -- that spark is "wasted", but uses only a few kv (typically less than 5 kv, and can be as low as about 2 kv) in the process.