2003 GC misfire

My mom's 2003 Grand Caravan Sport with 3.3L FFV (Vin 3) has a P0306 misfire. The code doesn't appear as often as the rough idle would suggest. In datastream mode the scan tool does register misfire counts on cylinder #6 but no misfire counts on any other cylinders. Dad changed out the plugs and plug wires searching for the problem, but all that did was take it from a P0300 random misfire code to the current P0306. We've double and triple checked the plug wire connections. I connected an inductive probe on the #6 plug wire and monitored on the oscilloscope. Randomly there is a goofy pulse that has a decaying fall time rather than the more typical quick fall time. We checked cylinders #2 and 4 as well and these two do not exhibit the same goofy pulse. We then swapped plug wires and the goofy pulse remained. Next we swapped plugs, no change to the goofy pulse. I made a new (8mm) plug wire, the goofy pulse remained. The plugs are OEM Champion double platinums.

I also installed a t-harness in series with the fuel injector for injector #6 and monitored it on the scope. The injector pulse looks fine, no indication of a problem with the injector and the injector resistance is within spec at 14 ohms. So, since the injector pulse is fine, I think the CMP sensor is okay.

We pulled the coil and checked the resistance of the wires from the PCM to the coil and they were all 0.1 ohms, so they seem to be clean and solid.

The PCV valve seems fine (rattles, only flows one way) so we didn't replace it.

Next we used SeaFoam to do a cylinder decarb. Don't yet know if when that gets done burning off it will make a difference.

We also replaced the coil. The goofy pulse seems to be gone, and the scan tool wasn't showing any misfire counts on any cylinders yet, but the engine still seems to idle just as rough. I don't yet know if things got better for Dad on his drive home after burning all the SeaFoam out, but I'm doubting it will.

Any other suggestions? I apologize for the really long post, I just didn't want to be accused of leaving out details.

-Rod

Scope plot showing what I presume is the misfire versus injector pulse.
http://i538.photobucket.com/albums/ff342/raschmidt/Automotive/th_Cyl6_plug_wire.jpg (http://s538.photobucket.com/albums/ff342/raschmidt/Automotive/?action=view&current=Cyl6_plug_wire.jpg)

One of the better posts on this subject I've seen in a while. Nothing to appologise for.

http://www.obd-codes.com/p0300
http://www.obd-codes.com/p0306

Is it just at idle or do you notice it across the board? Any indications of a leaking injector like hard starting or raw fuel smell? I'm thinking a scanner will detect a normal pulse from the PCM but not a leaking injector. Yes/No? Concider a compression check if you haven't yet.

It is only at idle that the roughness is noticeable. The van starts fine and the scan tool suggests that the exhaust is lean by giving high O2 sensor readings on bank 2 when idling rough. The fuel trim numbers, both short and long term, suggest the PCM is not trying to compensate either way. However, Dad did say the fuel economy is down.

We did talk about a compression test, but haven't done so yet. Would we need to pull all 6 plugs to do so?

-Rod

Try moving the plug and/or wire to another cylinder and see if the problem follows. I got a bad plug from Auto Zone once. It took me forever to figure it out. It can happed that either of the new wires or plugs are bad.

Try moving the plug and/or wire to another cylinder and see if the problem follows. I got a bad plug from Auto Zone once. It took me forever to figure it out. It can happed that either of the new wires or plugs are bad.

Yep, tried that (last few sentences or the first paragraph of OP), even made up a new 8mm wire and tried one of the old plugs.

-Rod

Yep, tried that (last few sentences or the first paragraph of OP), even made up a new 8mm wire and tried one of the old plugs.

-Rod

Missed that, sorry.

Missed that, sorry.

No problem, I appreciate you taking the time to offer suggestions.

-Rod

You could try verifying the misfire the old fashioned way- pull the plug wire (or injector plug) at the suspect cylinder, little change would confirm the miss. I assume the plug looked similar to all the rest when you pulled it?

I believe high O2 sensor voltage means low oxygen content in the exhaust (rich).

You're correct, the high O2 reading would indicate rich.... I think I need to make myself a cheat sheet, my head constantly tries to convince me that since the sensor is sensing oxygen, a higher voltage would indicate a larger content of oxygen in the exhaust stream. I guess I need to remember that it's backward of what I expect.

We haven't tried disconnecting the plug wire or the injector plug and running the van. We did install a spark test wire in series with the spark plug and didn't see an erratic spark, but could try pulling the wire off and connecting to a spark tester. I don't want to just leave the wire open circuit since I've heard that can be hard on the coil packs. Fortunately injector #6 is the only one that is relatively easy to get to the connector, so we could try disconnecting the injector as well. The misfire counts seem to be erratic enough that if we force a constant misfire, I'm not sure how much it will tell us.

When I talked with Dad Tuesday he said Mom hasn't mentioned anything about her van either way, but she's been driving it. Before the new plugs and wires she was afraid to even drive it. Also, Dad said it has a lot more zip now after the SeaFoam decarb.

-Rod

How is fuel pressure? Have you thought of a sticking injector? Have you tried a different coil? You might have a bad primary circuit in the coil itself which only manifests itself (now) occasionally at idle.

I'm glad you mentioned fuel pressure. That was one of the first things we planned to check. Alldata showed that the test port should be on the fuel injector rail right near injector #6. It wasn't there. We looked as far under the intake manifold from both direction as we could with a bright light, but not evidence of a test port. We followed the rail to where it switched to a flexible line under the intake air tubing, then as far as we could from under the hood. Nothing. We didn't climb under the van to look for a test port though. And, yes, we have experience measuring fuel pressure and know what test ports typically look like. This van has the plastic upper intake manifold and a single FIR runs underneath the center of the manifold. This isn't like the previous generation where there were two FIR and they were out in the open.

As mentioned in my original (long) post, the coil that's on it is new.

Your comment about "only at idle" resonated with me this morning. In the Ford forums that I frequent, if someone had posted that they had a rough idle and plugs and wires didn't have an impact, one of the first things I'd suggest is that they remove the EGR valve and clean the valve and passages. The EGR valve should be closed at idle, but if it's unable to close due to carbon build up on the valve, it provides what is essentially a vacuum leak. Based on our cylinder decarb procedure with the SeaFoam, and the amount of gray exhaust it created, there's a lot of carbon in the engine....

-Rod

I talked with Dad last night, he said that Mom thinks the van runs fine. She's not afraid to drive it anymore and the CEL has not come on again. Dad hasn't driven it recently to see if the funny idle is still there or not following the new coil and cylinder decarb. For now I guess we'll claim the money spent on the new coil was money well spent.

-Rod

Well, the misfire is back and it brought the CEL with it. Code is still P0306. Looks like we'll be pulling the manifold to determine if this really does have a fuel rail pressure port and to inspect the injector wiring and probably swap the #6 injector with a different injector. I'm not sure how soon we'll get to this as my garage is pretty full right now with other projects.

-Rod

So I was over at my parent's house tonight with my new scan tool. I checked with Dad and the van is still misfiring. It's particularly noticeable when they are backing up the driveway, but Mom's not afraid it will leave her stranded.

Anyway, using the scan tool I turned off the injector to cylinder #6, which is still reporting a misfire, and the idle speed dropped even further and was noticeably rougher, so that suggests to me that the fuel injector is still working properly. I think our next step will be to check compression and make sure that is fine.

Quick summary, since this problem started, the following have been replaced:


Spark plugs
Plug wires
Coil pack
Air filter

An oscilloscope shows that the fuel injector is getting the correct pulse, but the spark for cylinder #6 is not.

-Rod

I'm sorry if I break topic here, but was is seafoam?? And how to you decarbon an engine???

SeaFoam (http://www.seafoamsales.com/how-to-use-sea-foam-motor-treatment.html) is a magic elixir in a can that actually seems to do what it says. It has several uses, the only one I've used it for is to decarbon engines. The instructions on the can suggest using 1/3 can for the decarbon process. I typically use 1/2 can. Essentially you park the car in a well ventilated area that isn't near traffic. Get the engine up to operating temperature (approx. 2 minutes at 2000 rpm). Then shut off the engine and disconnect a large vacuum line that will let you feed in the SeaFoam liquid slowly and that will feed both cylinder banks in the case of a v-style engine. Start the engine and start metering in the SeaFoam, being careful not to let the engine stall. After you've poured in 1/2 can, shut off the engine and walk away from it for 5 minutes. Then start the engine and watch the gray exhaust billow out of the tail pipe. Here's why you don't want to be near traffic - 1) it will obstruct the view of drivers due to the thick cloud of carbon-rich exhaust, and 2) folks will call the fire department thinking something is on fire. You'll probably need to leave the vehicle to idle for 10 minutes or so before the exhaust thins out, then you'll take it for a drive around the block and think you blew a head gasket from all the exhaust that will continue to billow from the tail pipe.

-Rod

OK, so the P0306 code came on AFTER you swapped the coil, plugs and wires. Before it was a P300, random misfire, right? Have you tried putting the old coil back and see what happens? You've already tried moving the wires and plugs to a different cylinder to see if the miss follows and it didn't so its not the plugs or wires. I would try putting the old coil back. The new coil could be bad.

how bout a durango with the 4. 7l its an 06...where would I put the seafoam??

OK, so the P0306 code came on AFTER you swapped the coil, plugs and wires. Before it was a P300, random misfire, right?

Nope, it's had the P0306 since the new plugs and wires were installed. Prior to the new plugs and wires it was a P0300 but the misfire counts were showing up on the scan tool individual cylinder misfire counts (post #1). The coil was changed out after confirming a goofy spark to cylinder number 6, and said goofy spark pulse is still present along with the P0306 code.

In addition to a compression test I want to check the voltage to the coil pack primary as well as the trigger pulse to see if the coil is getting a poor trigger. If so, then I'll be back probing the CKP signal with an oscilloscope to see if the CKP is not a nice sharp square wave for cylinder #6 trigger.

-Rod

Last week Dad changed out the fuel injector on cylinder number 6 with a new one. The misfire is much less noticeable currently, but after clearing the code it returned (P0306) a few drives later.

Dad's going to check compression at least on the front three cylinders, then perform a cylinder decarb and check the compression again. Maybe, just maybe, the weak fuel injector allowed carbon to form on the valves and is contributing to the continued misfire. Since the new injector seemed to make an improvement, maybe the decarb will finally resolve the issue. Fingers crossed.

If not, I guess we're down to CKP sensor, CMP sensor, or PCM.

-Rod

Dad checked compression on the front 3 cylinders. Cylinders 2 and 4 both had 150 psi compression, but regardless of how many times he let the engine spin over, he was only getting 80 psi on cylinder #6. He pulled the valve cover and it appears that all the pushrods are traveling the same amount. He put a bit of oil down the plug hole and that didn't make any improvement so it probably isn't a stuck ring.

He did try Seafoam again and will see if that makes an improvement. Maybe there's enough carbon to limit the amount of air charge the cylinder can draw in.

-Rod

Well, we made some progress this morning. Dad had the upper intake off along with the fuel rail, etc. I ran my scope down the exhaust and intake ports, but nothing was obvious. We connected a cylinder leakdown tester and on cylinder #6 we could clearly hear and feel the air coming from the exhaust port. We rotated the engine a bit to confirm we were at TDC and still, air from the exhaust. The leakdown tester dropped to about 60%. To confirm we were using it correctly, we switched to cylinder #4 and this time the tester only dropped to about 9%. So off came the head. Nothing noticeable with the head, head gasket, or pushrods so we pulled the intake and exhaust valves. The intake valve looked pretty good, but the exhaust valve definitely has issues. My celphone didn't do a very good job of capturing the issues with the valve, but it was only seating on probably 60-70% of the surface. The valve is burned down in the area where the the head shows a lack of sealing.

Dad's going to spend some of the early afternoon finding a machine shop without a huge queue, new head and lower intake gaskets, a new exhaust valve, and new head bolts.

There really wasn't any play in the valve guides so we're not too concerned with those, plus it wasn't using/burning any oil. At least we are highly confident we finally found this cylinder #6 misfire! Based on the lack of availability of the parts in stock locally, this must not be a real common issue. Of the 3 O'Reillys, 4 AutoZones, and 3 Advance Auto stores in town, only 1 of the Advance stores has the gaskets, bolts, and valve in stock.

Poor picture of the valve:
http://i538.photobucket.com/albums/ff342/raschmidt/Automotive/CIMG0227.jpg

Underside of the head, cylinder #6 valves removed.
http://i538.photobucket.com/albums/ff342/raschmidt/Automotive/CIMG0225.jpg

-Rod

Dad was able to get a new valve from the dealer for less than from a parts store, plus the dealer had it in stock and the parts houses would have needed to order it. He checked with an acquaintance who still has a shop and the guy told him where the valve machine was and to have at it. Dad gave him a $20 tip for letting him use the grinder, so for $37 plus new gaskets, head bolts, and coolant, the van is again running and running well. The engine idles smooth and the CEL has not come back on. The compression on cylinder #6 is still slightly lower than 2 and 4, but within 15%. I think we finally have this issue resolved, and I will no longer suggest to people they can assume compression isn't the problem when they say the misfire is intermittent.

One other note: Once Dad had the top half the engine off he carefully inspected the fuel rail. There is no pressure port on the fuel rail like the service manual suggests there should be.

-Rod

Glad you got it working and thanks for posting the results. Someone will definitely use the information someday.