nate_o

Hey folks, sorry in advance for the long post,
I recently purchased a 1999 Civic HX 5-spd manual. The code P1165 came on. I took it to a mechanic. They told me there was just a loose wire, they reattached the wire and the code turned off. They said the car was otherwise in great shape. The code turned on again after about 50 miles of driving. I bought a OBD-II code reader, and it's picking up the same code. I've reset it a few times, but it keeps kicking back on.

When driving, the car will intermittently feel like it's struggling a little bit, for a few seconds, almost like you're going up a short steep hill. When starting, and in neutral, it will idle fine for about 30 seconds, then jump around a little bit between 700-1200rpm. When in gear the idle doesn't skip around.

I've found a few aftermarket mods on the car, including a K&N oil filter and an aftermarket exhaust. I don't know enough about Civics to tell if there are any other mods.

I have a Honda Service Manual, and that tells me I should replace the primary heated O2 sensor.

Before I do that, is it possible there are some mods I don't know about that are making the engine run rich, and fouling up the O2 sensor? It is possible the K&N air filter or the aftermarket exhaust is causing strange readings on the O2 sensor? I don't want to replace the sensor just to find out some mod is the root of the problem, or some mod is causing the O2 sensor to fail prematurely.

Former User

P1165 almost always indicates a bad primary O2 sensor.

fleabag

Clean/repair ALL electrical grounds. If you have dirty/weak grounds, especially the ground for the o2 sensor, it may give a code to the ecu. Consult the helms service manual, find all the electrical grounds, then with a multimeter, use the OHMs function and measure the ohms from the ground point being checked to the battery negative cable. If the ohms are higher than .5, they need to be cleaned and or repaired. Clean all the metal connections so they're shiny, but do not scrape paint off the body but instead clean the hole for the bolt which the ground screws into with compressed air and or a brush. (Only scrub unpainted parts) One ground you will obviously overlook if you consult the helms electrical grounds diagram is the o2 sensor grounds. With an ohm meter, put one probe on the body of the o2 sensor (where you put the wrench on it) and another probe on the battery negative. If the ohms are higher than 0.1ohm, then either the hole the o2 sensor screws in needs to be cleaned or some other electrical ground in the engine bay needs to be cleaned/repaired which is why I mention before to clean/fix all electrical grounds.

If your electrical grounds test out ok and you don't have any other electrical problems yet still receive the check engine code for bad o2 sensor, then replace the sensor.

I know you may not want to start cleaning your electrical grounds, but even on my car which had a very clean engine bay, absolutely no rust (California car) yet cleaning the electrical grounds helped my o2 sensor perform much better than before.

nate_o

@ RonJ - I was pretty sure that was the problem, but for the HX I need a wideband O2 sensor, and those aren't cheap. Before I spent $200+ I wanted to be reasonably sure that was the problem.

@Fleabag - Scrubbing isn't fun, but it's free, and it won't hurt, so it's a good place to start.

Thinking back to when I had the mechanic take a look at it, he said there was something else that was a performance part, and I vaguely remember it was something to do with the throttle body, but I went over the paperwork again and he didn't write it down. If I took some pictures of the engine, would anyone be able to tell me if anything looks non-stock?

Lugzilla

Yeah, throw some pics up.

Former User

Basically, only a bad primary O2 sensor will throw P1165:



I don't know what the mechanic did to "fix" the problem or how much he charged you, but my guess would be that he only reset the ECU. It then took a bit of driving to throw the code again. Unfortunately, I think you are on a wild goose chase that eventually will bring you back to the primary O2 sensor. In the meanwhile, the diagram says you should also check for a poor connection at the primary O2 sensor plug and the ECU plug.

Kista20

How many miles are on the car? They are notorious for going bad sometime past 100k miles, it could also be a cracked manifold allowing for the o2 sensor to throw off the readings sent back to the computer.

Former User

P1165 indicates an electrical problem.

nate_o

@ Kista - It could just be old and worn out. No idea if it's the original O2 sensor, but the car has 160k miles on it. The manifold is aftermarket and looks pretty new, so I don't think there's a crack in the manifold.

@ Ron - That looks like the same manual I have. Could you describe how I should check for poor connections? I did a visual inspection on the O2 sensor conneciton, and they looked ok, not obviously oxidized or loose or anything. I haven't checked at the ECU yet. I thought about checking for shorts from the sensor to the ECU but my multimeter probe cables aren't long enough.

@ Fleabag- I checked resistance on the O2 sensor and I was getting readings in the 0.2-0.4 range, depending on where exactly I put the probe tips. I'll try scrubbing the hole in the manifold and see if that helps.

nate_o

Here are a few pictures. As I was looking around at the engine, I noticed the timing belt was a few degrees off of zero. Is that ok? Any other comments on anything?

My mechanical experience has been a pro bicycle mechanic for years, and I restored an 80's carbureted Yamaha motorcycle. This is the first car I've owned in a long time, so I have some mechanical aptitude, but I'm out of my depth with all these sensors and EFI and stuff.

Kista20

ah yes you are right... from the pictures it shows they aren't using the oem manifold so I would start there first because the wires would've had to of been extended to reach it's new location.

Kista20

[QUOTE=Kista20;45918398]ah yes you are right... from the pictures it shows they aren't using the oem manifold so would've been nice to have known that up front.

nate_o

Kista - In the OP I said I had an aftermarket exhaust. I don't know that much about cars, so I thought the exhaust manifold and all the rest was all one piece. Sorry for the confusion.

Former User

As Kista mentioned, you should trace the primary O2 sensor wires to see whether they have been cut and spliced.

Ohm test all O2 sensors wires that run between the O2 sensor and ECU. Also test each wire for a short (continuity to body ground).

fleabag

What the **** did you do? You need to explicitly say exactly what you did...

nate_o

@ Fleabag - I got different readings depending on where on the O2 sensor I placed my probe. I held one probe point very still on the negative battery terminal, and the other probe point very still on one of the hex faces (the flat part the wrench goes on) of the oxygen sensor, I got a reading of 0.2. When I moved the probe from one hex face of the oxygen sensor to a different hex face on the sensor, I got a reading of 0.3. When I tried a third hex face on the O2 sensor, I got 0.4.

Former User

Your test was completely uninformative because the O2 sensor grounds at the ECU, not locally.

exb18c1

Ron is right.The best thing would be to scope the signal out,or at least to monitor the ocsilations in visual mode on the scanner.May be the sensor is tired and sluggish.

fleabag

What are you talking about? All O2 sensors use the exhaust manifold as an electrical ground. Period. If you're thinking that because o2 sensors today have more conductors at their plugs than in times past that they no longer use the exhaust manifold to complete the electrical circuit to send its signal to the ecu, you would be wrong. Have you seen a single wire o2 sensor before? Do you really think that sensor would work with just a single electrical conductor? Of course not, you run one wire to the o2 sensor and then use the body of the o2 sensor which is screwed into the exhaust manifold as the ground.

Former User

Please post evidence from a service manual circuit diagram showing that any 96-00 Civic O2 sensor grounds on the exhaust manifold. You won't respond because you are wrong. Stop posting BS in threads. Period.

fleabag

Have you EVER tested an o2 sensor off the car before?



Btw, how would the service manual have a circuit diagram of the o2 sensor using the exhaust manifold as a ground? That doesn't make any sense at all! That would be the equivalent of a circuit diagram for a bolt that screws into the body of a car. There is no diagram, you screw bolt into car and you're done. What would there be to illustrate? You screw the o2 sensor into the exhaust manifold and that is it, how could there possibly be a diagram showing the relationship between the two? I don't think you understand what you are saying. A service manual is not an exhaustive list of all the possible things that could possibly go wrong in a car, but merely a guide to help you test and repair various things on the car. There is absolutely NOWHERE in the service manual illustrating the replacement of the TPS sensor nor the cleaning of the throttle body, IACV, etc. yet those are all things strongly advocated on here because we're not of the mindset of replacing parts but repairing them first. To the point about the o2 sensor, if you don't have a rusted out exhaust manifold and all of your electrical grounds in your engine bay are in perfect condition, testing of the o2 sensor ground is pretty much moot... however since people usually do not have impeccable electrical grounds and DO have rusted out manifolds and or bolts, my ohms test is useful because it helps illustrate the relationship of the o2 sensor's body and the car's electrical ground.

Why so serious?

Former User

You must immediately stop posting in this thread until you look up a detailed circuit diagram for any 96-00 Civic O2 sensor. You will find that NONE grounds on the exhaust manifold. If you find otherwise, post the evidence in this thread. Otherwise, stop posting in this thread and also stop posting bad information on this forum. PERIOD. LAST CHANCE.

fleabag

I have gone over the electrical wiring diagram on the helms manual several times and I'm well versed in it just so that you know. Just because the circuit is not shown in the electrical wiring diagram, it doesn't mean it is not there. The circuit would never be listed in the helms manual because it is internal to the o2 sensor and the o2 sensor uses its exterior as a ground, hence the reason for it not being shown. The circuit diagram in the helms manual is for the wires from the ecu to the o2 sensor, that is it. How about you keep this conversation civil so that everyone, including yourself can learn from it. If you insist, I'll post a copy from the Electrical Wiring Diagram book.

Former User

You are wrong as usual. The service manual shows when any component uses a local ground, and none of the 96-00 Civic O2 sensors uses a local ground. Don't make another BS post in this thread or in this forum unless you are prepared to back up what you say. You need to separate what you think from facts. Your BS tech posts will no longer be tolerated.

nate_o

I didn't find any obvious splices between the HO2S and the 7-pin connector. Between the 7-pin connector and ECM the wires are covered up with some black sheath, so it's hard to tell if it was spliced between the connector and ECM.

I checked the pins going from the connector to the ECM and the from the connector to the HO2S.

Here are my results so far:
H is heater, S is sensor, U is unused
I've included color codes after the letter
/ - no connection
<> connection

Going to the HO2S:
H1 (blue) <> H2 (yel) / S1 (x) / S2 (x)
U / S3 (gry) / S4 (wht) / S5 (blk)

3.3ohm at connection, no connections to ground

Going to the ECM:
U / S3 (yel) / S4 (blk) / S5 (red)
H1 (blk/wht) / H2 (blk) <> S1 (green) / S2 (wht)

0.2ohm at connection between H2 and S1. H1 and S1 both connect to ground @ 0.2ohm.

I've been looking at the wiring schematic in the manual, and it looks like the Blk > ground connection is ok. I'm not sure if the green cable going to the ECM should be a ground though.