Hiya, so I've go this Honda Accord LX, 1998, V6, 222K miles. She's my baby, and she's sick

The following items are relatively new but are all up for speculation:
plugs (NGK, just re-gapped them a bit ago, all looked equally burnt)
wires
cap
rotor
EGR valve (cleaned out)
IAC Valve w/ gasket
MAP sensor
fuel injectors (refurbs from ebay)
rear O2 sensor

note: I've done the EGR port clean out and tube extension.
When it was running fine it was getting 23 mpg highway, now it gets 15 max, and it's getting worse.

Symptoms:
1. takes a few seconds longer to start (possibly unrelated, not really an issue to me as is)
2. seems to bog down during normal driving conditions. Have to push the accelerator down further and further until it finally revs higher and launches the car forward.
3. Using cruise control on the highway, set to 65 mph, accelerates to ~67 mph, bogs down until it coasts down to around ~64 mph, then accelerates again. She's basically surging but I may not be too understanding of the definition of it.


Codes: P0171: System Too Lean (Bank 1) P0170: Fuel Trim Malfunction (Bank 1)

I was able to catch a P0130 randomly (Sensor Circuit Malfunction (Bank I Sensor 1)) but it quickly disappeared.

I initially suspected a vacuum leak but have sprayed all gaskets in and around the intake with carb cleaner, no changes in REVs. I have not done a smoke test.

I pulled and plotted some data. First, highway driving. (It may be better viewed if downloaded and opened in excel) The first tab is all the data. The second is driving on cruise control around 63 mph and having the car hold the speed great. Right before this on the graph you can see i was coasting in neutral, just down a big hill for about a mile (trying to save some gas). The third tab is the same cruise control situation but the car is surging trying to keep the set speed. Check out how everything is just going up and down, what the heck.

https://www.dropbox.com/s/su8n5tg6ug...20PM.xlsx?dl=0


At this point I'm thinking it's an electrical issue, either the O2 sensors, the main ignition switch wiring harness, the vtec solenoid, or the ECM itself.. others have suggested the thermostat, vacuum leaks, and the cat being clogged.


If the vtec solenoid was infact stuck, it'd lead to bad timing. When it bogs down at low speeds under load, the sound coming from the tail pipe is rough, possibly from the timing issue. It sounds thin and like coughing, like a slower "chh, chh, chh, chh", like it's barely chugging along, then i press the accelerator a bit more and it launches me forward with a nice vrooomm.

The intake boot had a crack in it but i taped it up for now. The engine runs a tiny bit better but I have a strong feeling it's not a vacuum leak issue based on how erratic it is. Soon after I taped it up the surging came back. In fact, when the car is cold it'll run fine for about a mile then start acting up. Each time I reset my codes or the ECM totally, the problem goes away for about 5 miles, then slowly comes back more and more until it throws the lean code.


I'd like to fix this an I am a mechanical engineer, it's just driving me crazy.. If you need more data pulled at a particular time or while I'm doing a certain action let me know!

The second and third data is some acceleration data from stand stills. You can see how the RPMs and speed will drop while the throttle position stays the same, then once i press the accelerator down more, the RPMs and speed quickly pick up.

https://www.dropbox.com/s/n1lz79kvoz...ccel.xlsx?dl=0
https://www.dropbox.com/s/36l9cmul66...ccel.xlsx?dl=0

This fourth document is all the work that's been done to the car since my family has owned it (minus some oil changes and inspections).

https://www.dropbox.com/s/hewgsyy9wxl1914/work.pdf?dl=0


I hope you guys can help me figure out what's going on! it's quite a doozie I know.

Thanks,

-Andrew

 

I don't see any data relating to testing fuel pressure actual, not surmising through fuel trim data, nor any maintenance on changing the fuel filter.

 

1. I have no way to test and record fuel pressure. especially while on the go. The ECU in this car has no direct sensors for fuel pressure.

2. This car does not have a fuel filter. There is a nylon mesh attached to the fuel pump in the tank but I removed it a month ago for inspection and it was clean.

 

Correct.
You will need to attach a mechanical pressure gauge and verify the fuel pressure(actual) is normal and does not drop under load.
If there isn't a test port you will need to attach via the feed line coming from the tank.
FPR will be on the return side.
This will also allow you to test the FPR, which if blown, would not maintain correct pressure at the rails.

You need to collect data yourself outside of what the PCM can provide, even then the PCM can be fooled and give false data. It's a SD system there is quite a bit left on the table.

Actually, there is. There is the sock as you mentioned, but attached to the fuel pump assembly is the fuel filter. Not readily replaceable metal cannister like older Accords but there is still a filter.
Rock Auto sells the part by itself...


Or you can purchase the main body of the fuel pump assembly less the pump/sock filter/sender unit/etc #1 in the below pic...

 

Switched out the Fuel Pressure regulator.. That didn't do it.
Patched a small leak in my exhaust, but i dont think that was a real issue as it was right before one of the mufflers at the end of the system. ended up switching the O2 sensor right in front of the main CAT and then reset the codes. drove great to work today, hopefully the issue has been fixed!

 

Fuel trim data is all thats needed. it can tell you everything from the valve timng being off to a weak or failing fuel pump.



This isnt the 70s....

 

I cant view the data. Post it again in a more user friendly format.

 

OBD2 ECUs are of the rationality type. Fooling a PCM would be extremely hard to do with out setting a code.



The ability to interpret the data is whats needed.



I like you MIKE,youre always helping people out here and Im not trying to bust your chops, but you gotta learn to trust the PCM first,then perform intrusive testing second.

 

So it threw a P0135 code.... Before I switched the sensor out I was getting about 200 miles per tank. Now I suspect ill get ~300 a tank because it's running so smoothly, even with the P0135 code thrown, and i've gone 60 miles and the fuel gauge has barely budged. Now unless I nicked a wire in the harness of the sensor i'm not sure why this would happen. But the thing runs great. Despite the P0135 code and the lack of the two original codes... I contacted the vendor of the sensor to see if i can return it as I suspect it may be faulty.. they said I can return it. I'm gonna test the resistance of the old O2 sensor to see if it was really bad in the first palce. Tomorrow I pln to test the resistance of the installed sensor (new), but it's downpouring today. Yes, I purchased the correct O2 sensor, I just hope they sent me the correct one.

 

what data would you need to look at? besides fuel trim... rpm? speed? absolute throttle position? timing advance? let me know so i can cut down the amount of data on the graph so it's more easy to read.

 

I've heard of testing the resistance as such:
"You can check resistance of O2 sensor (should be 5-15 ohms) on pins 3-4 (lower pins looking at latch on top of connector). If resistance is good, and fault repeats after a reset, then wiring, connector, or PCM."

thoughts?

 

The resistance for the heater is 10-40 ohms.

If it is lower than 10 ohms, it may have damaged the ECM. (if it is higher it will simply need to be replaced)

A bad heater can fry the ECM,so prepare for that.

 

Have you checked the ECT sensor? Ensuring that the engine is actually going into the closed loop mode? Also, can you watch the voltage on the O2 sensor(s)? Just thinking out loud.

 

Pins 3&4 have a resistance of 13.3 ohms. Pins 1&2 appear to be an open circuit, infinite resistance... But this is on the old sensor I took out. It was down pouring today so I didn't get under my car to test the one I installed. Will do that this weekend. It the resistance is different at all I'll switch them out. Plan to return the new sensor as it is

 

Figured it out. Upstream O2 sensor!

 

Was that the one you just replaced?

 

I originally switched it out with an eBay one I bought as a Bosch 13007 but they shipped the wrong one. I returned it and got a legit 13007 from rock auto. Talk about a pain in the ***

 

Fuel pressure ok? Valve adjust?

 

no idea on the fuel pressure... but it runs fine now. I think what was happening was the Denso upstream sensor was worn out, not enough to trigger the P0135 code, but enough to make the CPU think the car was running lean. I'm not sure how it thought that only bank one was running lean to begin with but it started to make sense to me with the temporary fuel trim malfunction code.

My theory is that the damaged upstream sensor made the CPU think the engine was running lean, it made it think there was a vacuum leak or more air than is supposed to was somehow getting into the cylinders. to counteract this, it tried to dump as much fuel in as possible. but no matter how much fuel it dumped in (the fuel trim would out), the upstream sensor still suggested that it was running lean. because of this, the CPU thought that the fuel trim was broken because it simply couldnt dump enough fuel in, when in reality it was dumping too much fuel in, bogging down the engine and killing my gas mileage. when i switched the sensor, the feedback was accurate and the CPU didnt mistakenly think the car was running lean. Since then, it appears ill be getting the best gas mileage i've ever gotten, almost 350 miles to a tank!

Just to check, I took out the CPU to inspect for shorts and burns in the circuitry, nothing was found. But it reset the comp which was nice. So when i started the car again after installing the new legit 13007 Bosch upstream sensor, the light wasn't on and hasn't come back since.

The valves were not adjsuted. Not looking to spend any more cash on her for a while.