You likely have a TEC coil.

Dial your multimeter to read Ohms on the lowest scale (e.g. 20 or 200). Touch the two meter probes together to measure the internal resistance of your meter. Write down the reading. Measure the primary winding resistance and then subtract from this reading the internal meter resistance. This will give you the true primary winding resistance.

Now set the meter to a scale close to the expected secondary winding resistance, such as 20K. Now measure the secondary winding resistance. No need to subtract the internal meter resistance because it is insignificant on this higher scale.

 

I do indeed have a TEC coil. I don't have a multimeter and my dad is out of town. (He has the high end multimeter for work...) I was unable to test the coil (so far). Maybe once all of my moving expenses are out of the way, I'll buy myself a meter.

I got brave and took the distributor apart for the first time ever. There was minor corrosion on the contact points so I sanded them very lightly with fine sandpaper. There was similar corrosion on the rotor, so I also very lightly sanded it. Got the ICM out and got a ride to Autozone. He tested it 10 times. It tested good 7 out of 10. There were 7 individual tests in one test run. It intermittently failed on "threshold" 3 times.

Smoking gun?

Ugh, I really don't have $100 for a new ICM, so it looks like I'm going to be running it as-is for a little. It did start right up when I reassembled everything at least.

Does the intermittently bad ICM theoretically explain most or all the symptoms? Flashing CEL at idle, extremely low idle, occasional stall?

 

You can buy a digital multimeter at Harbor Freight Tools for $4, or less if there's a sale.

If the ICM is bad, as it seems to be, the coil is likely also on the way out. Also considering the bad shape of the cap and rotor, start saving $ for a new OEM distributor, which generally has a lifespan of 150K miles.

 

Harbor freight is known for cheap tools. Dunno if I trust them for accurate readings on electrical equipment. But I'll look into it.

The car started fine yesterday after reassembly but threw a CEL as I left for work today. Gotta check the code...

Then I have to sell my body and worldly possessions to get the cash for a new distributor. When the engine goes in Saturday, it'll literally be all new.

 

I think it's about $150 for the distributor, including coil and ICM (but not cap and rotor). So when it comes to spending 3 figures it seems worth it to do that instead of just the ICM, and lessen the possibility of the coil being bad as well.

I spent about 3 days trying to get a moped to run. The coil passed the ohmmeter test with flying colors. Finally just said screw it and swapped the coil out; she fired right up. It was a bad coil. I don't know why they still print the ohmmeter test-- it's not reliable since coils can fail due to shorted turns and/or internal arcing. This does not affect the DC resistance of the coil but it will remove it's ability to generate voltage high enough to spark.

 

I can't speak about % reliability of the coil test, but many members that I have helped on several different forums had bad coils when the test indicated so. I am an advocate of the test. Sometimes the test is done incorrectly because people don't know to subtract the internal resistance of the multimeter from the primary winding reading.

 

I'm looking through the service manual for more information. It says that there should be 4 wires connected to the ICM. I only have 3. It's missing the blue wire on the right side, the single post on the side. I took that picture just after I took the cap off. There isn't even a blue wire to connect to it. Not sure if that's significant or not.

 

The blue wire is for the tachometer. 99-00 Civics don't get the tach signal from the ICM. They get it from the ECU. So it's fine that the blue wire is missing.

 

I checked the codes on my lunch break. There were two: Random misfire again, and misfire on cylinder 4. Not sure why it's misfiring, unless that's a symptom of the ICM being intermittently bad. (Am going to pick up a new ICM today, had to shop around for the cheapest price.)

I replaced the plug wires with Autozone ones, Duralast or whatever, 6 months ago or so. I suppose it's possible that the wire has prematurely failed. All of my problems seem to be on cylinder 4. It has the worst rings, the worst valve seals, the dirtiest spark plug, and now the misfire.

Has anyone mentioned that you two, RonJ and mk378, are exceptional members (or, I dare say, pillars) of the online Honda community, and that your troubleshooting skills and knowledge are invaluable?

 

I finally bought a digital multimeter. I'm sure it'll come in handy.

Checked the coil. Set it to 200 Ohms, checked internal resistance, checked primary winding resistance. Minus internal resistance, 0.71 Ohms. Switched to 20K Ohms. Checked secondary winding resistance. 17.7 K Ohms. It's within spec. Coil is good. I did let the engine cool for a few hours, but the coil was still a bit warm when I tested.

ICM has gotta be the culprit. Not sure why it's only misfiring on cylinder 4 though.