My '97 Civic is 150k miles on it has started to act up. It surges under light throttle when cold and eventually will kick on the check engine light with a Code 45 (too lean/too rich). No other code are thrown. The fuel milage has gone from 35 to the low 20's but I am not sure if it is because it is running to rich causing the code (the surging seems like too lean to me) or because I can't keep the check engine light off and it is going into a default mode of operation that has sucky fuel mileage.

It has a new fuel filter, and plugs. The injectors all have the same impedance and the air filter is clean. I just adjusted the valves. The timing belt was changed at 90k.

I am not sure how to approach this. Anyone have any ideas.

Jim

 

Making repairs that eliminate the CEL should solve your problem.

The pictures show how to troubleshoot code 45 (P0171 or P0172). Take your car to AutoZone to determine whether your code is P0171 or P0172. This is a free service.

 

Hi,

Thanks, the trouble trees confirmed my thoughts. My main finger pointing was the O2 sensor or the fuel pump. With the mileage it could be either. I sort of ruled out the fuel pump or regulator as if I floor it is stops surging and accelerates normal. Also when changing the plugs I saw no signs of plug fouling so I am leaning toward too lean based on the way it runs and the shape of the plugs. (I worked as a GM mechanic for a decade before changing careers.... wow, a decade ago).

I have read discussions about what Civics do or don't have EGR valves. Where are these typically located?

Also it says "Sensor 1", that implies there is a "Sensor 2". Are there two sensors on the Civic?

Thanks,
Jim

 

Your Civic indeed has two O2 sensors. I sent you a private message (PM) with information that should answer your sensor and EGR questions.

 

Yes, it has two VERY EXPENSIVE sensors and VERY EXPENSIVE fuel pumps. Guessing will be very costly on this car!

Jim

 

Ok I think I found it. I hook my O-scope up to various sensors inputs at the computer. Everything looks great except:

O2 Sensor 1 - Runs between 0.4V and 0.6V
O2 Sensor 2 - Runs between 0.4V and 0.6V... for a few seconds. Then it runs up to 1.2V and sticks. Occasionally it will dip down to 0.8V. When I shut the key off the O2S1 drops to a few tens of millivolts, and O2S2 will stay at above 1.0V for a long time.

What do you think. Need O2 Sensor 2, seems sick?

Jim

 

Yeah, the way that the secondary sensor readings bounce around doesn't make much sense, so I think it's a safe assumption that the sensor is bad.

 

Update:

Valuable lesson. The O2 sensors must be read with a differential amp in the computer. Don't use the case of the computer as the ground, use the the returns from the O2 sensors. Once I did that the wierdness went away.

Over the last few day I have had a DVM on the O2 sensors and trying to correlate the light with the readings.

I leave my home in the morning and the temp is in the high 20's F. I drive a 2 lane country road at 60mph with the cruise control for 30 miles then a freeway at 75 miles for 20 miles then 5 miles at 40mph or so with a couple stop signs and red lights to work.

First few minutes:

Startup
Main O2 = 400mV
Secondary O2 = 400mV

When the coolant starts to reach normal it must transfer into closed loop and:

Cruising (cruise control)
Main O2 = 700mV
Secondary O2 = 600mV

Downhill closing the throttle
Main O2 = 70mV
Secondary O2 = 70mV

Wide open throttle
Main O2 = 625mV
Secondary O2 = 525mV

Full warmed up in neutral air off.
Main O2 = 800mV
Secondary O2 = 700mV

The car sits at work the day. Repeat the drive in the reverse direction temp in the mid 50's. One stop light before warmed up then set the cruise at 45 and before the next redlight the engine light comes on just about the time the temp comes up to normal.

I don't see anything funny in the O2 sensors during this.

I stop and clear the code and it is fine the rest of the way home. The next day and it will repeat the above. The light comes on at same point in the trip home.

Went through a tank of gas keeping the code cleared. It went from the low 20mpg to the mid 20's, still well below the 35 it use to get. Something is still wrong.

I stopped by AutoZone on the way from a meeting today (the light came on after sitting all day after the morning drive). It was code 45 - P0171. The tool said "System lean on Bank 1". the manual just says fuel system too lean. Nothing about "Bank 1". It is a D16Y7 so it has no EGR. It is a 97 LX sedan. It is the only code.

It does not skip, studder or stall (other then the surge with light throttle when cold). It idles at 500RM without skipping a beat.

It looks like time to test fuel pressure and regulator.

Can anyone suggest where I can get a pressure guage and fitting adapter that will work on the Civic outside of the Honda Special tools?

Also is there anyway to test injectors on a Honda?

Anyother thoughts on what it may be?

Jim

 

Replace your primary O2 sensor (First sensor), your problem will go away.

Happens a lot, primary O2 sensor goes into closed loop mode and the CEL will come on.

Secondary O2 sensor doesn't do much of anything honestly, it's a simple emissions tool whereas the primary is what gives you the great gas mileage and good driveabillity of your car.

 

Thanks for sharing your interesting data. Were the O2 sensor readings above taken only on the drive to work, which didn't throw the CEL. Did you happen to take readings on the way home when the CEL turned ON. When this fault occurs, do you think that the system improperly remained in an open loop, consistent with a problem with the primary O2 sensor?