legacycentral bbs

I moved the new Legacy back to the garage this evening and was messing around in the drive way and got the car to bog and sound like it was missing. During that time I also heard a snapping. I opened the hood and saw an arc between the body of the coil pack and one of the plug wires.

I put a little bit of dielectric grease on the plug wires and that seemed to eliminate the arcing, but the engine didn't rev well from low rpm and still kind of stumbled. I'm assuming the coil pack is toast and needs to be replaced, but just looking for confirmation before I throw more money at this car...

Thanks

With the level of corrosion you've encountered on that car, there's no reason to assume it hasn't reached into the engine bay also.

Perhaps see if you can borrow someone's extra coilpack, maybe via a user on LGT.com.

You definitely need to befriend someone at the parts counter...

Yeah, the coil pack may well be junk (they should never arc from the body), but the problem may be the wire also. Where on the plug wire was the arc landing?

The stumbling you're experiencing may be because the ECU has learned weird fuel trims from misfiring all the time. Maybe try resetting the ECU? Not that that'll resolve the whole problem, but it might make it run better for the time being.

What kind of coil pack is it? If it's the 4-male-tower one with built-in ignitor (4-pin harness connector), I have a good spare off a 99 Impreza with about 90,000 miles on it.

Well the bottom part of the coil pack is rusty. I checked the secondary coil resistances and they were pretty much on spec, 10.9 kohms. Spec was 11.2 +/- 15%. But I'm not sure that really tells me anything.

The arcing was happening right at the boot of the spark plug wire. I've had the battery disconnected for several hours, so I know the ECU has been reset.

It is a female coil pack with the built-in-ignitor. It's off to the side of the engine. I don't have a good picture of it....this is the best I've got right now.

http://main.experiencetherave.com/subar ... G_6111.JPG

I've got new plugs & wires, and I think I'll start with that first. When I get to the engine work, I'll probably take the coil pack out and check it a little more closely.

Wow, new Subarus are so weird. They put the coil pack on the back of the intake manifold?

Yeah, you might as well put the new plugs and wires in. I suppose if a wire had a tiny crack in it and there was moisture around the boot, that could cause that kind of arcing even if the coil pack is okay.

Yeah I don't know.....I'm not a huge fan of this motor layout.

However, the things I HATE about this car are the throttle response, throttle pedal feel, brake pedal feel, clutch pedal feel, and clutch pedal travel. It's a very short clutch pedal that grabs almost at the bottom of the pedal.

I haven't driven it much, so haven't really played with anything to see if I can change those things.

I wonder if you could swap clutch hydraulics around. If you like how the GC 2.5RS clutch felt, maybe it would be possible to swap those master and slave cylinders in.

Throttle response seems like it would have a lot to do with ECU tuning. I wonder if they dull the throttle response for emissions reasons. Like if they back off on the acceleration enrichment or something.

Yeah I might play with the clutch hydraulics. I haven't really regularly driven a hydraulic push clutch setup. My Impreza is cable.

Since it's e-throttle, yes the throttle response is going to be ECU tuning. I'm not sure if Open ECU will have a rom dump for this car....will have to check it out.

My big issue is the throttle pedal is stiff. I HATE stiff throttle pedals. It makes it difficult to modulate the clutch and so on. The weird part is I've driven the STi and it doesn't have stiff throttle, so it may be as simple as swapping out the springs for the throttle. I need to buy a force gauge to measure the force required to depress the pedal compared to my 90 Legacy and Impreza, and then try to find replacement springs that produce a similar throttle pedal stiffness.

Ohhh... Drive by wire. Nuts, man. So the "throttle pedal" is really just a potentiometer and a spring, like in a video game. And your foot isn't really connected to the engine in any meaningful way; you press the control to ask the computer to do something, it decides what it thinks you "really meant," and then adjusts the butterfly accordingly.

Yeah, I'm sure you can just swap the spring out.

Yup, that's exactly what it is.

Well I think I may have found the problem....or found a symptom...I'm not really sure yet. I did the plugs & wires today and two of the four plugs had a VERY large gap. The electrode tip was almost worn down. Cylinder 4 and I don't recall whether it was 1 or 3 that had the other one. I'm pretty sure it was 1, because I pulled cylinder 3's first and don't recall being shocked by it. If it was 3 & 4, I'd say there may be a coil problem. However it being diagonal cylinders, I'm leaning towards it not being the coil pack.

I'll have to watch things and see how it runs. When I get the pics uploaded I'll post them.

The pics of the plugs are towards the end of these pictures.

http://main.experiencetherave.com/subar ... cywagon/10

Huh, the coil pack doesn't actually look like it's in bad shape, superficially..

I'm gonna make a guess here: The two plugs with worn center electrodes are from cylinder 4 and cylinder 2, and the wear is not in and of itself indicative of any coil pack problem, although it may have contributed to the arcing problem.

Take a close look at the other two plugs' ground electrodes. I can't really tell from the pictures you posted, but I bet you'll find a little divot worn into the part of the ground electrode that's closest to the center electrode.

This is actually the normal pattern of wear on waste spark engines like ours. When the coil fires, say, the rear two cylinders, it creates a huge eletrical potential between the #3 and #4 terminals of the coil. During the spark, the #3 coil terminal is thousands of volts higher than the #4 terminal. This huge potential makes electrons flow in the circuit: They go out the coil into the #4 wire, into the #4 plug center electrode, hop across the gap into the #4 ground electrode (creating a spark in the process), pass through the engine to the #3 ground electrode, hop across the #3 spark gap into the #3 center electrode (creating the other spark), go through the center electrode into the #3 wire, and back into the coil.

This happens whether #4 is the wasted spark or #3 is. It's asymmetrical. The wear happens on whichever electrode the electrons hit after jumping the gap.

Or maybe I have the polarities backwards, and the wear happens on whichever electrode the electrons fly off of. Whatever.

In either case, this is normal (although extreme) wear, creating a huge gap. And an extremely large spark plug gap may be so hard to jump that the spark finds another path, like the path you saw it take.

Thanks for the feedback Vikash.

I recall doing cylinder #3 first and it was not the severely worn electrode, so it would have been cylinder #1 and cylinder #4. So the fact they are diagonal cylinders is kind of odd. Until I realized how much a pain in the arse it is the replace the plugs I was going to do a compression test....but now I probably won't.

I'll keep an eye on things, but like you said, I'll chock the wear up to "normal" wear pattern.