Subaru WRX Forum banner
1 - 10 of 10 Posts

·
Registered
Joined
·
7 Posts
Discussion Starter · #1 · (Edited)
Foreword: I don't have a WRX but NASIOC has to approve my thread and UltimateSubaru has to approve my account, which I'm still waiting on, so I was hoping to possibly get some help here. I also thought it would be better to post here instead of the "General Maintenance, Troubleshooting" subforum since I'm using RomRaider and have a very specific question. Also, I'm a noob and I recently started to learn into how to do repairs on my own. I'm using the FSM, RomRaider, and Google to find my answers. But, being a noob, I'm ignorant of certain concepts so I'm not sure what to do at my current point. Hopefully this ain't against the rules too much.

Car: 2003 Impreza Outback Sport (USDM)

Problem: I've been getting a P0304 (4th Cylinder Misfire) with no other codes. This happens at idle after warming up the car (driving it around for a few minutes). This initially happened after I put the pedal to the metal and started hitting high RPMs (3k+), after which I stopped at a light and the CEL came on. It seems the problem goes away with high RPMs, and comes back with low RPMs. I have some jackstands in the back that rattle due to the misfire (in the same way the car bounces a bit when idle), and at 2.5k RPM I got a sort of machine gun rattle. Increasing the RPMs stopped it. I tested this multiple times on one of my trips.

Things I have done:
  • Replaced spark plugs
  • Swapped 2nd/4th cylinder sparkplug/coil cables. (nothing changed, 4th kept misfiring)
  • Checked ignition coil with multimeter, it was fine (though regardless, afaik, a problem with the coil would have both 2nd and 4th cylinders misfiring, or all 4)
  • Check freeze frame for CEL (the fuel trims were normal as was the intake manifold pressure, but I think it wasn't reliable as realtime data on TorquePro showed short-term fuel trims at 10-12% when idle)
  • Logged a 20 minute trip with RomRaider

Things I have not done:
  • Use an endoscope camera to check the sparkplug housing/area (in case of cracks)
  • Preform a leakdown or compression test
  • Check the new sparkplug in the 4th cylinder after the misfire (as according to this guide in the "IGNITION MISFIRES AND COMPRESSION MISFIRES" section).

Suspicion: I believe this to be a small vacuum leak looking at the data I logged using RomRaider (I started logging after 20 minutes driving). I noticed that the long-term fuel trim (A/F Learning) was hitting a bit above 10% sometimes, as well as -10% (while driving). However, using TorquePro (w/ Bafx Scanner), I was getting 10-12% short-term fuel trim realtime (while idle and P0304 was currently going off). So I'm thinking it's too lean because of a vacuum leak somewhere.

Current Plan: Preform a "smoke test" (something I saw on YouTube) for the vacuum leak. Though how helpful it would be I have no clue.

Can anyone offer any advice as to what my problem can be, or what other tests I can do to find out?


Here is some of the data I've gathered, hopefully it's helpful.

Freeze Frame from the CEL
Code:
Freeze frame report generated by Torque for Android
=========================================================

Vehicle VIN: Not present
Vehicle Manufacturer: Unknown
Vehicle Calibration ID: D6XR2U0CE43F

Freeze frame information:
------------------
Fuel Status = 0 byte
Engine Load = 3.137 %
Engine Coolant Temperature = 181.4 °F
Fuel Trim Bank 1 Short Term = -0.781 %
Fuel Trim Bank 1 Long Term = 4.688 %
Intake Manifold Pressure = 34 kpa
Engine RPM = 712.5 rpm
Speed (OBD) = 0 mph
RomRaider CSV Log File (20 minute drive)
1.57 MB file on MEGA

Some Graphs Based on RomRaider Log File (x axis is seconds)
*A/F Correction is short-term fuel trim (I'm not sure of the relevance of the 16-bit/32-bit cpu corrections).
*A/F learning is long-term fuel trim




 

·
Super Moderator
Joined
·
11,560 Posts
Check your oil make sure it isn't milky, check your coolant. A head gasket leak in cyl 4 could cause your issue. Did the plugs have odd tints to them?

Compression and leak down would be my next test. Usually easy and can be done with loaned equipment.

It's been a long time since i messed with the torqued app, I don't recall it being super time accurate or particularly detailed due to a low polling rate.
 

·
Registered
Joined
·
7 Posts
Discussion Starter · #3 · (Edited)
Check your oil make sure it isn't milky, check your coolant. A head gasket leak in cyl 4 could cause your issue. Did the plugs have odd tints to them?

Compression and leak down would be my next test. Usually easy and can be done with loaned equipment.

It's been a long time since i messed with the torqued app, I don't recall it being super time accurate or particularly detailed due to a low polling rate.
The Torque App is detailed to the point that it shows the values OBDII gives it, which is all I needed really. The polling rate is the normal default baud (4800). The slowness comes from the bluetooth, but there are usb versions.

Regardless, I used the SSM protocol with RomRaider (usb into my laptop), and the misfires can be seen on the A/F Correction #1, where it hits nearly 40% a bunch of times. Those are the lean mixtures that are happening (which cause the misfire if I understand it correctly). I'm not sure of the causes for a lean mixture though other than a vacuum leak (as before, I'm a noob).

Note that my CSV log is in milliseconds, and the baud rate was at 10.4k, so if detail and accuracy is required, then it would be in the graphs/log file.

I'll check the oil, coolant, and the 4th cylinder plug, and report back.
 

·
Registered
Joined
·
7 Posts
Discussion Starter · #4 ·
The coolant is low, but still there. Looks normal. Oil isn't milky and is normal.

Here is the 2nd (left) and 4th (right) cylinder plugs. Both were put in at the same time. The 4th one looks virtually brand new. I'm not sure what to make of this.

I can't tell what a sparkplug would look like when it's an ignition misfire vs a mixture misfire. It looks in almost too good condition for it not to be a ignition misfire. I'd figure if the spark happens, then there would be at least some wear and tear. Then again, it's the fuel mixture going off that makes it black.

329189

329190
 

·
Super Moderator
Joined
·
11,560 Posts
Check that cylinder for fuel. Even if it was rich or lean I'd expect more discoloration.
 

·
Registered
Joined
·
7 Posts
Discussion Starter · #6 ·
Check that cylinder for fuel. Even if it was rich or lean I'd expect more discoloration.
I'm not sure what that entails. I'm going to preform a compression test though as soon as I figure out how to do that via SSM or OBD for the single cylinder.
 

·
Super Moderator
Joined
·
11,560 Posts
I'm not sure what that entails. I'm going to preform a compression test though as soon as I figure out how to do that via SSM or OBD for the single cylinder.
You can do current draw or buy/rent a tester. It just screws into the sparkplug hole. Then pull the fuel pump relay and the coil then crank it.
 

·
Registered
Joined
·
7 Posts
Discussion Starter · #8 · (Edited)
You can do current draw or buy/rent a tester. It just screws into the sparkplug hole. Then pull the fuel pump relay and the coil then crank it.
I'm going to do a compression test first then via laptop. Then try what you're saying.
 

·
Registered
Joined
·
7 Posts
Discussion Starter · #10 ·
Fuel compression?

That'll probably do the job for cylinder compression
Yea. For the leakdown I'll have to go to a mechanic, or buy my own kit. But I haven't found any kit online below $30 that reviews say isn't garbage.
 
1 - 10 of 10 Posts
Top