02 Honda Odyssey P1399 Random Misfire - What test have I missed?

Hi Folks,

I have in our shop a 2002 Honda Odyssey with 148K miles on it. Came in due to a P1399 Misfire and 1 dead miss. I have done a bunch of diag on it and I'm getting stumped. Usually I find the problem by now.

Scope Equipment
Old Snap on MODIS w/ 4Ch Scope (Shop Scope) No amp probe
Waekon AutoWave 1ch Scope (Personal Scope)

History
Previously 1 month ago, 3 Coils and 6 Spark Plugs were replaced at our shop before I was employed there due to random misfires. The vehicle is now back with a random misfire once again. Sometimes it will give specific 4 specific cylinders but it seems the cylinders it gives, changes from time to time after erasing the codes. I don't know if the engine run without misfires after the previous work was done.

Symptoms
1. Very Light Misfire at idle and under load. Creating a very very slight shutter to the point you can't even really feel it.
2. Stalls at times during consecutive snap throttle tests while engine falls to idle. MAP Sensor reading steadily climbs as engine starts to stall. Haven't looked to much at the IACV yet

DIAG Performed
1. Basic Hands on check first, found one coil unplugged. - FIXED Dead Miss
2. Performed Coil Stress Test (Spark Gap Test) - All Coils had hot bright loud blue spark at 25KV Gap
3. Replaced 1 Coil Found to fail 2nd day into Diag creating another dead miss. 4 Coils Replaced
3. Scoped Control Signal to Coils - Consistent with no drop outs 3 Wire coils so I only get a pulsed square wave. I did have a funky square wave on one coil but that seems to have disappeared after moving that coil and retesting that specific control circuit and the new control circuit the coil is now on. The "turn off" event was mostly straight down except at the very bottom where there were lots of small oscillations which was consistent the entire time I was scoping that circuit. It's no longer there.
4. Scoped Fuel Injector Control Side - Consistent with no drop outs. I should probably upload a pic of the waveform as its hard to see any pintle hump in the waveform in any of the injectors.
5. Checked EGR Operation and Position Voltage Value. Manually activated EGR and engine stalled immediately. While EGR is inactive, position voltage is around 1.1 volts.
6. Checked MAP Sensor voltage at idle - 0.8 to 0.9 volts. Fluctuates linearly with throttle response on a scope and stays consistent at idle
7. Checked Valve Clearance - Intake to .008 to .009 and Exhaust to .011 to .012
8. Checked B1S1 O2 Sensor values. Switching seemed normal.
9. Check ST & LT Fuel Trim - .98 to 1.03 (1.00 Seems to be 0 Correction) Sometimes it will hang around 1.03 but does the same now and then at .98
10. Check Spark Plug Gap - .40 to .43
11. Checked Compression - 240-248 PSI on all cylinders
12. Disconnected Purge Control Valve - No change in misfire Previously tested PCV for proper operation and leak test

Questions
1. Could EGR Still cause a misfire at idle and under load even though values seem good?
2. Can someone suggest a few other ideas?

I have not checked Timing or Fuel Pressure. I figure if it was fuel pressure, the engine would break up under heavy load or high rpm.

Thanks in advance!

Big Thanks to Paul Danner for all your videos and your website! You taught me a lot! I will be buying your book very soon.

Nice testing summary. Much appreciated!

I think the EGR could account for a miss during cruise, but not at idle. I can't remember which Honda V6 engines worked this way, but I do know some of them have cylinder specific EGR ports in the intake, while others just feed the EGR gasses into the common plenum. If the individual ports clog, then some cylinders get more EGR than others. You could try disconnecting the EGR valve, and see if it'll reset a misfire code.

At idle... That's tougher. You could try moving an injector? I see that you scoped them, but I'm thinking of a flow rate difference, or a poor spray pattern. Sucks, 'cause that requires pulling the upper intake again. :silly:

This is a bit of a stretch, but have you performed the ECM Reset since the repairs were done? Not just a code clear, the actual 'ECM Reset' under Functional Tests in the Modis. This resets the idle adaptives AND the misfire detection profile. It's not technically required after a tune up, but I do it after every misfire repair to guard against false misfire codes.

I am slightly suspecting fuel delivery at injectors so i may swap injectors around.

SMALL UPDATE
I removed the EGR valve and inspected the actual valve to see If it was closing and it was. I pressure tested the egr ports on the valve itself while the egr was closed using tire leak check fluid and i was getting some bubbles.. nothing i knew was causing an issue but i cleaned it anyways and got it so the valve seals now. Though that did not fix the problem.

I forgot to mention i have not done a ecu reset. Ill give that a try

SERVICE UPDATE

1. Performed Intake System Cleaning (Spray BG Cleaner into intake before throttle body.) Not big on this service, never seen it work. We don't have the adapters to hook up the fuel system with cleaner so the cleaner is dumped into the tank and drove around for awhile, again, probably won't do much. No Change Noted
One thing I did notice is the oil warning light was flickering at idle afterwards. I had the engine held at around 1300rpm for about 10-15 minutes. After engine idled, warning light would flicker but after shutting the engine off a few times the light stopped coming on. Just something I will note down. There is a slight tick sounds like from the engine.

2. Performed ECU/PCM Reset as Suggested. No Change Noted

Notes:
Engine has plenty of power but does have a slight hesitation when there is a misfire under load. ECU quickly picks up misfires under load and takes time to catch them at idle.

I ordered a Injector Balance Tester OTC 3398 but won't be in till next week. I have not moved injectors around yet. I wan't to do a balance test first before charging customer for intake gaskets and all the time to do that process.

Other Suggestions?

You could try a running compression test with the schrader valve core removed from your compression gauge. If you find any suspect cylinders, you can follow up with a leak down test. I have seen some of those 3.5l engines with a burnt #4 exhaust valve. Doesn't necessarily fit the random nature of the misfires though.

I have an ultasonic injector cleaner and flow tester to view spray paturns with if you want me to see if anything is up with the injectors?

Sent from my Moto G Play using Tapatalk

Thanks Tutti57. I am from New Hampshire are you in the North East?

Customer picked up vehicle till next week or when they can drop it off again. It runs good minus the minor misfire.

FUTURE DIAG DIRECTION
- So I think I will treat the idle misfire and the under load misfire as different problems. Since I don't know if both became a problem at the same time or if at separate times. So I'll check below;

1. EGR Passages for Clogs and Restrictions - Under Load Misfire
2. Fuel Injector Balance Test - Both Idle and Under Load Misfire.
3. Cylinder Leak Down Test - Both

QUESTIONABLE TESTS
Not sure if below could cause a misfire.

1. VTEC Operation - Under Load Misfire ???

graywave wrote: Thanks Tutti57. I am from New Hampshire are you in the North East?

Customer picked up vehicle till next week or when they can drop it off again. It runs good minus the minor misfire.

FUTURE DIAG DIRECTION
- So I think I will treat the idle misfire and the under load misfire as different problems. Since I don't know if both became a problem at the same time or if at separate times. So I'll check below;

1. EGR Passages for Clogs and Restrictions - Under Load Misfire
2. Fuel Injector Balance Test - Both Idle and Under Load Misfire.
3. Cylinder Leak Down Test - Both

QUESTIONABLE TESTS
Not sure if below could cause a misfire.

1. VTEC Operation - Under Load Misfire ???

I am in Rochester NY.

Sent from my Moto G Play using Tapatalk

Have you tried clearing the CKP pattern and performing a relearn procedure?. If not, it is worth a try.
I have chased a ghost in the past only to find the CKP pattern was not complete.
Good luck sir!

11. Checked Compression - 240-248 PSI on all cylinders


This is an awful lot of compression and if the reading was true you probably are dealing with a completly carbonized engine.......
Maybe you want to try another compression gauge just to verify

juergen.scholl wrote: 11. Checked Compression - 240-248 PSI on all cylinders


This is an awful lot of compression and if the reading was true you probably are dealing with a completly carbonized engine.......
Maybe you want to try another compression gauge just to verify

Juergen's right, that IS a lot of compression. :blink: I've seen timing issues cause abnormally high compression, but it usually also causes the engine to barely run. The MAP reading at idle is right on, too. Still, maybe a timing check is in order? That'd certainly explain the random misfire counts, the rough idle, and misfire while driving.

I've never known the VTEC system to cause a miss, but you could certainly disconnect the solenoid and recheck.

i had a 2007 pilot 3.5 with a misfire at idle. not bad but could feel it slightly. turned out the valves needed adjusted

The only output control options I have with our Snap On Solus is ECM/PCM Reset which allows it to relearn all the values. I did perform that but nothing changed.

I will Recheck the compression with the valve out of the gauge. I usually allow 3 compression strokes on the gauge to get my final reading. I don't remember what it hit on the first stroke. Next time I get the van in I'll test it again. I never did look for the compression spec of the motor.

Compression Spec I found on Identifix is 135psi. i am guessing that is minimum spec

Yes, this is minimum spec.

Although an overadvanced intake timing will rise the compression considerably I am not convinced that this is happening in your case, Probably you would have cam/crank/sync codes as well as different map readings and so on....

If you truly got that high of compression you really might be dealing with a carbonized engine. Might wanna take a glance with a borescope.....
.

I would want to see a relative compression test. Then depending on what is found, a pressure transducer in the cylinders.

Once the vehicle gets dropped off again, possibly this week, I"ll do another compression test with the schrader valve removed. I don't have a pressure transducer on hand. If I had one I would scope, crank,cam, fuel injector & single cylinder pressure transducer and get a cranking and running waveform of engine timing and compression/valve operation. Scoping crank/cam/injector is still on my mind.

Unfortunately this shop also doesn't have a amp probe to conduct a relative compression test. I'm pretty sure a relative test will be fine considering those compression numbers I got were consistent even though the numbers are high, each cylinder I allowed 3 compression cycles to hit before releasing the key.

I'll give everyone an update with my findings once I start diagnosing the vehicle again.

This week I will try to see if I can save scope snap shots to my laptop from the MODIS scanner. Ive used a Verus Pro for awhile previously so I feel like I am using a dinosaur and need to learn to grab waveform snap shots off it. Seems like I need to hook it up to my laptop via USB to grab em. It also doesn't seem to have as good of a sample rate and screen resolution as the verus.