1998 Ex 2dr Rough Idle at Low RPMs

[JSpiers]

1998 Honda Civic Ex two door. Car has 185,000 miles, has been very reliable car. Have done basic maintenance over the years. CEL has been on for last few years but hasn't seemed too impact function. Yesterday, car began to idle very rough. At higher speeds was hard to notice but didn't seem to have as much power. Figured there was a bad plug or wire. Remove one of the wires as car was running to see if problem got better or worse. Saw white spark, but must have pulled it out too far as the car died. Tried to crank back up, no dice. Tried to read CEL codes but couldn't without cranking the car. Pulled a wire and tried to crank to watch spark and didn't see any spark. Pulled distributor, tested ignition coil and ICM. Ohms were low on ignition coil, parts store said ICM showed bad at low RPMS but good at high RPMs, but to replace. Bought new distributor, installed and car cranked back up. Still have rough idle problem though at low RPMs. Sounds pretty good when I idle it up. Checked CEL codes and am getting 41. Reset CEL and get code 41, Primary Oxygen Sensor heater. Should I replace it? Would it cause rough idle at low RPMs? Thanks in advance for the advice.

3/17/12 Update

Pulled the IACV this morning and thoroughly cleaned it. Seems to run a little smoother, but still rough idle. On initial start up with motor cold, idle is very smooth (2000 RPMs). Idle is still smooth at 1500rpms, but once it drops below 1500 (between 1500 and 1000) it begins to idle rough. Typically I only notice the rough idle as the idle is in its down stroke. Once it begins to idle back up the idle goes back smooth. As the car warms the idle typically lands in the 600-1200 rpm range (average 700/800).

Note, while driving engine runs rough with increased throttle too. Major loss of acceleration.

CEL shows code 41, 71, 72, and 23. Today is the first day I've gotten anything other than 41.

Primary oxygen sensor heater
Random misfire cylinder 1
Random misfire cylinder 2
Knock sensor

 

[JSpiers]

Bump. Anybody have any advice for my situation?

Here is my current plan of attack and reasoning:

1. Since I wasn't getting any random misfire or knock sensor CEL codes before the distributor swap, I'm going to check timing to ensure timing is right. I read in my Haynes manual that the distributor can't be installed 180 degrees off. Based on my research for some basic checks, the wire post on the distributor associated with cylinder 1 is the bottom left hand post (viewing from the passenger side of the engine compartment). Moving clockwise around the distributor, it should be 1, 3, 4, 2. Looking at the engine from the bumper, these should correlate at the valve cover with 4, 3, 2, 1 (viewing left to right).

I've never checked the timing, so here are the steps I was planning on taking. Please correct me if I'm wrong.
A. Hook up red & black clips to battery. Clamp pick up lead to number one cylinder wire
B. Warm up engine, jump out CEL with wire
C. Line up pointer with white mark? on fly wheel by twisting distributor housing after loosening 3 bolts.
D. Lock down bolts
E. Recheck timing

2. Tap number one cylinder plug hole and replace all spark plugs. I pulled 3 of the 4 plugs and they looked fine. Had issues on the number one cylinder plug. Used penetrating oil and let it soak over night. Finally loosened up. Appears to have a some galling on first 2 threads. Appears to be some corrosion on the base of the threaded portion of the plug. Could this be due to the issue with the threads?Click for Pictures. Have a tap that I will slowly run in and out of the hole, wipe off any small pieces of metal. Repeat until run tap through entire hole. Will use a lot of oil to keep the shavings on the threads of the tap.

Is it possible to remove the spark plug tube easily before I clean up the threads?

 

[RonJ]

CEL shows code 41, 71, 72, and 23. Today is the first day I've gotten anything other than 41.

Primary oxygen sensor heater
Random misfire cylinder 1
Random misfire cylinder 2
Knock sensor

First off, never ignore CEL codes. You'll pay in the end.

If you reset the ECU, do all 4 codes repeat?

When was the last time you replaced the plug wires and plugs?

Have you ever replaced the fuel filter?

What's the resistance of the primary O2 sensor heater element? Is 7.5A dash fuse 15 blown?

 

[Ek_Rider]

didi you check and make sure everything was top dead center your timing cuz i know that if you try and crank your car to much you can jump timing i know cuz i did before and car had a real rough idol

 

[JSpiers]

First off, never ignore CEL codes. You'll pay in the end.

If you reset the ECU, do all 4 codes repeat?

When was the last time you replaced the plug wires and plugs?

Have you ever replaced the fuel filter?

What's the resistance of the primary O2 sensor heater element? Is 7.5A dash fuse 15 blown?

Ron, thanks for the additional checks.

Plugs and wires replaced probably 4 years ago.

Replaced fuel filter probably 6 years ago.

Is the 02 sensor heater element the same as the 02 sensor sticking out of the exhaust manifold? What is the easiest way to check it if so? I think there are 3 wires coming out of it with a shroud around them.

I checked all the fuses in the dash and under the hood with the continuity tester on my ohm meter (didn't pull, but used probes on face of fuse). They all checked out.

However, there is one fuse that is hard wired to the battery that I checked and it was blown, replaced with a new one and it blew again upon cranking. Appears to be factory, but makes me think of a hot that could've been ran to the trunk for an amp. Previous owner traded in with 45K miles when the wife bought from the dealer. Runs directly from battery through firewall, but has in line fuse right after connection to battery. Has plastic corrugated shroud around it. Looks very factory.

Will double check fuses, reset CEL and recapture codes, chase threads on number cylinder, adjust timing and report back.

 

[JSpiers]

didi you check and make sure everything was top dead center your timing cuz i know that if you try and crank your car to much you can jump timing i know cuz i did before and car had a real rough idol

EK Rider, do you mean I should physically check TDC or just ensure timing is set to TDC via light? Forgive me if I don't make any sense. Much of this is very new to me.

 

[JSpiers]

Update

Chased the spark plug threads and replaced the fouled plug. Adjusted timing (it was off quite a bit). Pulled fuse to reset CEL, warmed car up, and got new CEL codes. Here is what I got:

Code 71 Random misfire cylinder 1
Code 72 Random misfire cylinder 2
Code 73 Random misfire cylinder 3
Code 22 VTEC Pressure Sensor (could have misread code before or this time, before I read 23 Knock Sensor)
Code 41 Primary Oxygen Sensor Heater

Observations: When I started the engine cold, I really couldnt tell that there was anything wrong with the engine. Even after the engine initially warmed up I could only hear a faint miss when I listened to the muffler. As the car stayed at operating temperature for longer, it began to get worse. The frequency of the misses or roughness got progressively worse. Still major loss of power when trying to accelerate.

What next? Will verify CEL code 22 first.

 

[RonJ]

1) Figure out whether the code is 22 or 23.

2) Focus first on solving misfire codes 71-73. Remove the coil and ICM from the distributor. Resistance test the coil and have an auto parts store test the ICM.

3) Code 41: Unplug the primary O2 sensor (closest to header) and then measure the resistance across the two black wires in the plug on the sensor side.

 

[epic_civic]

Ronj, wasn't there an issue where the wiring under the throttle body could throw the code for 41 and blow your fuse #15?

 

[RonJ]

Ronj, wasn't there an issue where the wiring under the throttle body could throw the code for 41 and blow your fuse #15?

Yes, click the Fuse 15 TSB link in my signature. Is your fuse 15 blown?

 

[JSpiers]

1) Figure out whether the code is 22 or 23.

2) Focus first on solving misfire codes 71-73. Remove the coil and ICM from the distributor. Resistance test the coil and have an auto parts store test the ICM.

3) Code 41: Unplug the primary O2 sensor (closest to header) and then measure the resistance across the two black wires in the plug on the sensor side.

Here are the results:

1) It was code 23 - knock sensor. I reset and continued to get 23, until the last check when it disappeared.
2) Pulled the coil and ICM from the brand new distributor I put in last week. The coil resistances were from A to B = 0.5-0.6 ohms, from A to Secondary = 15.8 kilo ohms. Had the ICM tested and it checked out good.
3) Unplugged the 02 sensor and tested resistance across the 2 posts in the connector that correspond to the black wires on the 02 sensor side. Couldn't get a reading. Tried between other wires and couldn't get any readings either. This was when the car was off, should I have done this while it was running?

In addition, I checked the plug wires for resistance. I got the following:
Cylinder 1 = 8.5 kilo ohms
Cylinder 2 = 7.8 kilo ohms
Cylinder 3 = 6.5 kilo ohms
Cylinder 4 = 6.0 kilo ohms

Also checked fuse 15 under the dash again. It was good.

Observations:

The longer the car runs the worse it runs rough. While not in gear it seems to run smoother at higher RPM's than lower. When in gear, it runs terrible at high and low rpms. Major loss of power while in gear.

I initially pulled the codes when the car was cold today. Only got 41. As it heated up, I pulled them again, got 41 and 23. When it got to full temp and let it run for a bit, began to run even rougher, put it in gear etc. I got 41, 71, 72, and 73. Code 23 disappeared the last time I pulled them.

While sitting in the car, noticed different smell. Exhaust doesn't smell normal, maybe richer?

Last observation. Hope this is my problem, not a result of another issue. The last time I pulled the codes, the idle was so rough the car died. This is the first time it has died on its own while running. I cranked it again, it barely started, still very rough idle. Decided to turn it off, noticed smoke from under the hood. I do have some oil leaks so I wasn't surprised given the car isn't being driven daily noq. Did some snooping around saw that smoke was mainly coming up from around the firewall. Got under the car, notice that the catalytic converter was glowing red. Could this have been my problem all along? Or is this a result of the car running rough? Might explain why it runs rougher the longer the car is on.

Ideas?

 

[RonJ]

The engine misfires result in unburnt fuel in the exhaust that ends up burning in the cat. That's the reason the cat was glowing red. The substrate inside the cat is likely melted = cat is toast.

Is the new distributor OEM or aftermarket?

No resistance across the two black wires of the primary O2 sensor indicates that it's bad and needs to be replaced. The non-functioning primary O2 sensor is also contributing to the engine running rich. In the meantime, test whether unplugging the primary O2 sensor eliminates the misfiring and misfire codes. It probably won't help, but it's worth trying.

The knock sensor (blue in picture below) is attached to the rear of the block. It has one Red/Blu wire. Check the wire and sensor for damage.

 

[RonJ]

The coil resistances were from A to B = 0.5-0.6 ohms

The TEC coil A-B reading should be 0.63-0.77 Ohms. Is your A-B measurement corrected by subtracting the internal resistance of your multimeter?

 

[RonJ]

Major loss of power while in gear.

This is likely due to a combination of engine misfiring and a clogged cat.

 

[JSpiers]

The engine misfires result in unburnt fuel in the exhaust that ends up burning in the cat. That's the reason the cat was glowing red. The substrate inside the cat is likely melted = cat is toast.

Is the new distributor OEM or aftermarket?

No resistance across the two black wires of the primary O2 sensor indicates that it's bad and needs to be replaced. The non-functioning primary O2 sensor is also contributing to the engine running rich. In the meantime, test whether unplugging the primary O2 sensor eliminates the misfiring and misfire codes. It probably won't help, but it's worth trying.

The knock sensor (blue in picture below) is attached to the rear of the block. It has one Red/Blu wire. Check the wire and sensor for damage.

Aftermarket distributor. Couldn't locate an OEM quickly enough. Original was TEC...coil resistance was low on it and ICM failed test. How do I determine the internal resistance of my multimeter? What is the coil resistance supposed to be for an aftermarket distributor?

I will go ahead and test unplugging the 02 sensor to see if that helps and then replace it. The resistance test was completed with the car not running. Was that correct?

Will inspect knock sensor for any obvious issues and report back.

Will the cat need to be replaced before I do any further testing to accurately troubleshoot? Or, will the bad cat just impact emissions testing?

 

[JSpiers]

The TEC coil A-B reading should be 0.63-0.77 Ohms. Is your A-B measurement corrected by subtracting the internal resistance of your multimeter?

No, would get 0.6 about half the time, then it would drift to a 0.5 and back again. Mine doesnt appear to measure in the hundredths.

Thanks for the advice.

 

[RonJ]

Aftermarket distributor. Couldn't locate an OEM quickly enough. Original was TEC...coil resistance was low on it and ICM failed test. How do I determine the internal resistance of my multimeter? What is the coil resistance supposed to be for an aftermarket distributor?

Our Civics don't do well with aftermarket distributors.

Internal resistance is the resistance reading when the two meter probes are touched together. Subtract this reading from the A-B coil reading.

I will go ahead and test unplugging the 02 sensor to see if that helps and then replace it. The resistance test was completed with the car not running. Was that correct?

The O2 sensor resistance measurement is normally done with the engine off, but could also be done with the engine running and would yield the same result.

Will the cat need to be replaced before I do any further testing to accurately troubleshoot? Or, will the bad cat just impact emissions testing?

You first need to solve the misfire problem. Otherwise, you would quickly kill the new cat. If the cat substrate has melted and is preventing exhaust flow, then you need to replace or gut the cat immediately after fixing the misfire and CEL code problems. Once the O2 sensor is fixed but before replacing the cat, then code P0420 will likely be thrown, indicating that the cat is bad.

 

[JSpiers]

RonJ,

I unplugged the OEM primary 02 sensor and cranked it. It sounded no different. Seems as though it is running worse and worse. It will hard crank at all now. When it does, feels like it is running on one cylinder, very rough. Basically just sputters very very slowly.

I went ahead and replaced the 02 sensor to see if that would help. Didn't help. I ohmmed the new 02 sensor and got 7.6 I think. After this, I disconnected the front side of the cat and tried to start. No change from before. Rechecked the wires for spark and saw spark on all wires (with fuel injectors unplugged). Sparks seemed whitish yellow to me. Then went ahead and pulled spark plugs out to inspect again. All but number 2 were wet with gas. So here is what has been checked or replaced:

1)Replaced Spark Plugs
2)Checked Wires (ohmmed)
3)Replaced distributor (with aftermarket) Checked new coil, ICM, crankshaft position sensor, TDC sensor, and cylinder position sensor
4)Replaced Primary 02 sensor (Bosch)
5)IACV (cleaned)
6)Checked crankshaft fluctuation sensor
7)Cat unhooked on manifold side

Ideas?

 

[JSpiers]

Update

Brought a buddy over and rechecked everything I've touched. Everything appeared to be right. New observations:

1) Inspected knock sensor. It appeared to be plugged in before I lightly pulled on the clip. Sensor came out. Tried running without knock sensor, car responded same. Tried to reinsert the knock sensor (get the clip to hold sensor in and was unsuccessful). Inspected clip and found the brown piece that inserts into the block was actually on the clip. So either I broke the housing off or it was already broken when I pulled on the sensor. I didn't pull very hard, use pliers, etc. so I think it was already damaged.

2) Advanced and retarded timing within normal range of distributor housing slots to try to get engine to run and had no success. Removed distributor bolts to retard and advance timing even more and got motor to turn over and run with it extremely retarded. Was tough to keep it going though. Tried to extremely advance timing and it ran like a champ. Found a sweet spot probably 15 degrees out of time and the car will crank and run like normal (extremely smooth, couldn't hear misfire). Once running, adjusted time back closer to the normal operating window within the distributor housing slots and car first began to misfire and run rough and then would eventually die as I got closer to the 12 degrees. If i put it at 12 degrees the car mimics the symptoms I have been describing in this thread. Reset CEL, cranked engine with timing advances, car fired right up ran smooth, brought to temperature, no CEL codes.

Why would I need to advance timing so much to get car to run right? I checked mechanical timing to make sure I hadn't jumped a tooth and everything was fine. Took coil and ICM out of new distributor put in old distributor and car still required advanced timing to run properly.

 

[JSpiers]

Update - Just wanted to let everyone know what I finally found and fixed.
I found an intermittent spark plug wire, that was causing all of my rough idle issues. Even though I ohmed out each wire, and they were will within the tolerance described in my service manual, at least one was defective. I had previously replaced them in the past 5 years. After putting a new set it, the car runs perfect.