I have a 2003 Accord LX 2.4L 4 cyl manual 137800 miles. The check engine light shows code P0141, meaning secondary HO2S (sensor 2) heater circuit malfunction. The light glows for a few days at a stretch and then goes out for a few days and comes back on again, and so on. I used the Honda official shop manual to troubleshoot in the following order.
1. Check #18 fuse in the under-dash fuse box. Ok.
2. Resistance between sensor connector terminals (female) 3 and 4 should be between 5 and 6.4 ohms. If not, replace sensor. My reading is 8.2 ohms. While there is a deviation, it is close to the range. I wouldn't have a question if it were open circuit or short circuit. So I kept going down the troubleshooting list.
3. Check for continuity between body ground and terminals (female) 3 and 4 individually. No continuity, which is normal.
4. With ignition on, measure voltage between harness side terminals (male) 3 and 4. No voltage present.
5. Measure voltage between male terminal 4 and body ground. Battery voltage present.
6.Turn ignition off, jump the SCS line with the HDS (Honda Diagnostic System tool), disconnect ECM/PCM connector E, turn ignition on and measure voltage between A5 and E21. If it is 0.1 volt or less, repair open in the wire between E21 and the oxygen sensor.

I couldn't do step 6, as I don't have the HDS tool. Also, the manual doesn't say what an SCS is. At this point, what should I do? I could replace the oxygen sensor since step 2 above didn't completely succeed. If someone can shed some light on how to accomplish step 6, I can proceed with this step, which would be the last troubleshooting step. I still keep thinking about the fact that the check engine light is intermittent and maybe it is a loose connection somewhere. Suggestions are welcome.

 

Back in the OBD-1 days there was a connector with 2 wires that you jumper with a paperclip or a piece of wire. That's the SCS connector. Those 2 wires are now in the OBD-2 connector where you would plug in a scan tool. If I only knew which pins … (?)
Don't jumper randomly, because you can make some connections that can damage the PCM.

Step-2 - a couple ohms over might just be your meter's uncertainty. Put the 2 leads directly together, and it might read 1 or 2 ohms already, so subtract that value from your measurement of the sensor heater.

Step-3 - continuity to ground - I wonder about the same test on those same 2 pins on the car's side of that connector in addition to the sensor side.

Since the CEL goes off & on, maybe there's a flaky connection or a wire rubbed bare that occasionally touches ground. When you check for continuity to ground, move the wire around in case that makes it short.

 

Jim, thanks for your response. So far I haven't figured out which terminals to jump for the OBD-2 connector type. I agree that I should not jump random terminals. I hope someone on this forum can shed some light on this.
Step 2. You're correct, the meter didn't read 0 when the leads were shorted together, it read 1 ohm. This would mean that the resistance I measured would be 8.2 - 1 = 7.2 ohms.
Step 3. I did measure the resistance for the male terminals on the harness side. Terminal 3 showed infinite resistance and terminal 4 showed 3.6 - 1 = 2.6 ohms.

I wasn't able to move the wire around when checking for continuity because the wire is underneath the car and I was performing continuity measurements from the back seat, meaning the two were not simultaneously accessible. I did get under the car to look at the sensor cable. It is tucked in the right places, in good condition, with no signs of fraying or a bare wire being exposed.

 

Maybe the heater resistance is just borderline where sometimes it triggers the error?

It looks like the point of step 6 is to see if the wire harness from pin E-21 at the computer out to the O2 sensor is open. You could check that with your meter, if you don't turn on the car at all, disconnect the E connector from the computer, and check continuity in the wire from E21 to the sensor. At the sensor's plug, it'll be one of the wires that connects to the heater.

If that's OK, maybe a new sensor is the thing to do. Normally I do troubleshooting before throwing parts at it, but the O2 sensor is almost a tune-up part, It might not cost much money, and you've tried to do the troubleshooting with the only result being the borderline resistance of the heater.

After all, the whole point of that troubleshooting is to eliminate the possibility of wiring problems in the car, before you assume it's a bad sensor.

 

Look at my thread from last week. The car I worked on was a 1999 Accord.
https://www.hondaaccordforum.com/for...-heater-67120/

Maybe disconnect the connector and measure the resistance of the 2 heater wires from the sensor (not the wires going to the ECM). If it is OPEN, then replace.
If I remember right, there are 2 sensor wires and 2 heater wires. On mine, the 2 wires I measured were both black.

 

You can try a test light on the harness side connector to the O2 sensor. Just make sure you connect the bulb to each wire that is feeding the heater circuit. A 25 watt bulb would draw about the same amount of current as a good heater circuit inside the O2 sensor. Turn the key on to see if the bulb stays on to verify the wiring upstream of the O2 sensor is working properly.

 

Jim, rockhoundrob and PAhonda, thanks for your response.

It suddenly occurred to me why my diagnostics haven't turned up anything concrete (except for the heater resistance which is borderline). The check engine light hasn't been on for the last several days, so one would legitimately expect all tests to be normal. I'll repeat the diagnostics if and when the light comes on again.

 

Guys, you'll never believe this ...

As expected the CEL came on again a couple of days ago. I checked the resistance between terminals 3 and 4 (black heater wires). It was around 200 ohms and stayed the same with repeated checks. Today I checked it again. It was slightly under 300 ohms. As I watched the multimeter, the resistance reading kept creeping up slowly by a few ohms at a time and it crossed 400 ohms. This is weird behavior for a resistive element. At that point I ordered online a replacement oxygen sensor (Denso 234-4797) and a 22 mm offset socket, the special tool required for removal. Later I took the car out for an errand, the CEL was still on. However, on the return home, the CEL went off! I measured the resistance again. It held steady at 8.2 ohms. I don't know what is causing this behavior and what to make of it. It'll probably continue to run fine for a few days and then the CEL is likely to come on again. Regardless, I feel that the oxygen sensor is flaky and should be replaced. Your comments are welcome.

 

You measured the flaky behavior of the O2 sensor. Normally, the problem someone faces is they cannot catch it "in the act".

 

JimBlake, good point, normally it is hard to catch a defective component in the act. The new sensor has arrived. I measured the resistance right away, it is 5.6 ohms, which is well within specs. I'll attempt to install it in the next few days.