My wife has been driving a 2002 LX four-cylinder with manual trans for at least 10 years. Since November 2012 we've been getting intermittent check engine light of various codes, usually misfire codes with no change in car performance. The only thing we've noticed is an intermittent miss at idle.

The first time it happened was November 2012 with 119k miles and the code was P0301. The light was going off and on. The Honda shop adjusted the valves and replaced the distributor cap and rotor and "serviced" the fuel injection system. They suggested I replace the number one cylinder fuel injector. We chose to wait. About five months later the light came on and the code was P0301 and P1456 (Evap system leak). We took it to a new mechanic who replaced the gas cap and switched the #1 cylinder injector with the #4 cylinder injector. Compression in the #1 cyl was 145..Since that time the light goes on and off multiple times per year and more recently multiple times per month. The codes most often are P0301, P0303 and P0300. Sometimes I see P0302 and P0304, all misfire codes. I have also seen P1399 and quite recently P1607.

Repair work performed: New plugs and wires, new distributor cap and rotor, adjust valves, fuel injector service, new gas cap, clean EGR valve, adjust valves again, switched fuel injector #1 with fuel injector #4, replace plug wires again, ignition rotor again, plugs regapped, added Lucas octane booster. I also removed and inspected an electrical relay from under the dash, looking for broken solder connections.

Like I said there has not been a change in performance that is connected with the check engine light. The only thing we have noticed is an intermittent miss at idle which does not correlate with the CEL being on or off. Other than a loose connection somewhere I am not sure what else to do at this point. The original Honda repair shop is more than happy to replace anything and everything that could cause misfire codes. That is not what I'm looking for. Any suggestions?

 

Your description sounds like you have a misfire that might be getting worse.

Your PCM (vehicle computer) counts the number of misfires that you feel at idle. Once you get over a set number of counted misfires in a certain period of time, the check engine light will turn on.

Do you remember having any work done on this car not long before you noticed the misfire problem? It may be hard to remember back to 2012.

Do you notice the misfire more when you first start your car and it is at idle, or do you notice the misfire all the time?

I'd start simple first. When it gets dark outside, start the engine, and use a spray bottle with water and lightly spray over the spark plug wires. Try to look for an arc (see video below, just use the mist setting on an old Windex bottle).

The next step would to get some live data. Do you have a scanner that can read vehicle data? Misfire counts, fuel trim, codes, etc.?

Here are the other codes that aren't P03xx.
P1607 is a code for a PCM circuit failure. This could be your problem, or this could be the result of your problem.
P1399 is a temporary code for a misfire in one of the cylinders.

 

Thank you for your quick reply.

I do not remember having any work done shortly before the misfire codes started showing up. It looks like we have kept most of the receipts so I will go through them and see if I can find anything around November 2012.

We only notice the misfire (rough idle) when the car is warmed up and idling without any fans running or air conditioner. The misfire is so subtle that we rarely feel it inside of the car. When we do it is felt as a rough idle while sitting at a stoplight with no accessories running. However, it is unpredictable as to when that will happen. If we open the door we can hear an intermittent pop coming from the exhaust pipe. It is not very loud so things need to be very quiet to hear it or be standing right next to the tailpipe. If the fan for the radiator kicks on the intermittent pop goes away. The same thing with the air conditioner. Frequency is about 5 to 8 times in a 10 second time period. It is erratic and can give two or three quick pops in a row followed by a three second pause before popping again. I should point out that the tailpipe pop seems to always be there if you listen for it. However, the rough idle can only be felt on occasion. I tried listening under the hood for the sound of an arc but heard nothing and have not had a chance to use the spray bottle. I will try that tonight.

I want to emphasize that the only time we ever feel anything abnormal is the above described OCCASIONAL ROUGH IDLE ONLY which I assumed is a misfire. The rough idle is so subtle that the mechanics have always said they couldn't duplicate what we are experiencing. I also want to point out that this phenomenon is no worse than it was three or four years ago.

I have a code reader that also will read live data. There are at least 15 to 20 different items listed under live data. Unfortunately they are abbreviated so I am not sure what to look for beyond the obvious such as RPM. By the way today at idle the reading was 700 RPM which increased to around 770 RPM when the radiator fan or air conditioner came on. As I mentioned above, I could no longer hear a pop at the tailpipe. There is slight variation in the idle but seems to stay within a 11-12 RPM range. The tachometer on the dash seems pretty steady.

The CEL came on after driving a few miles today. I did the above readings and checked the computer codes without clearing them. Stored codes were the misfire codes again P0300, P0301, and P0303. The pending code was P1399. I did not clear the codes because I wanted to try the live data reading after checking with you.

So what should I be looking at in the live data with possible abbreviations? Thanks so much for your help with this problem.

 

List the abbreviations that you reader has for live data, and we may be able to help steer you in the right direction. You can try searching for your reader online to get a manual with better descriptions.

Fuel trim and a misfire count on the cylinders would be the first places to start.

 

My tester is the OBDMATE OM123. The printed manual is no help and neither is the company website.
Live Data:
DTC_CNT
FUEL SYS1
FUEL SYS2
LOAD_PCT
ETC
SHRTFT1
LONGFT1
MAP
RPM
VSS
SPARKADV
IAT
TP
O2S B1S12--B2S
O2B1S1
SHRTFTB1S1
O2B1S2
SHRTFTB1S2
OBD
So I found a video online which indicated that the SHRTFT and LONGFT are for fuel trim. What about misfire count? I also tried the spray bottle technique shown in the video. The result was no arcing or sparking produced anywhere.

 

look at shrtft1 and longft1 for the fuel trim at idle when the engine is at operating temperature. Let us know what you see for a value.

Also see how fuel trim changes if you increase the rpm from ~750 at idle to 1000 or 1500.

SHRTF1B1S2 and SHRTFTB1S1 also look like fuel trim for short term. Not sure how they are different than shrtft1.

Try DTC_CNT as well. That might be a misfire counter, or your scanner doesn't have that ability.

 

First I will report the LONGFT1: It started out at -4.7 and went down to -7.0 after it ran for a while.

As for SHRTFT1 at full operating temperature, at idle it mostly fluctuated from -4 to -19. First it was -4 to -9. After the engine ran a little longer it was staying below -10 (-10 to -19). It then rose a little but stayed in the negative numbers. At 1200rpm the range was -5 to +3.1.

By the way SHRTFTB1S1 and S2 seem to be sensor one and sensor 2. They stay pretty close to each other and read very similar to SHRTFT1. So does that mean there are two sensors for one bank (engine)?

The DTC_CNT is the number of stored codes. It started out with three and became zero after I cleared the codes. By the way the P1399 returned almost immediately under pending codes.

Went ahead and checked the O2 sensors. They both read in the range from 0.035 to 0.83.
Thanks again for your help.