Take the 2-minute tour ×
Motor Vehicle Maintenance & Repair Stack Exchange is a question and answer site for mechanics and DIY enthusiast owners of cars, trucks, and motorcycles. It's 100% free, no registration required.

After months of normal driving with no problems, my 2003 Honda Civic EX check engine light came on during a long trip. Got it scanned and it's P0420: Catalyst efficiency below threshold bank 1. They said it could be a bad catalytic converter or oxygen sensor.

I borrowed a scanner and tried to measure the oxygen sensor readings over time. Doesn't seem to have good time resolution:

enter image description here

This is voltage vs "time", though I don't know what units "time" is in, or if this is useful in any way. The code did not occur during this graph, anyway. Apparently the sensors are supposed to switch on and off like this?

The lambda sensor's output voltage should not remain constant, however. It should flip-flop back and forth from rich to lean as the PCM increases and then decreases the amount of petrol injected into the intake manifold. Every time the voltage reverses itself and goes from high to low or vice versa, it's called a "cross count." A good O2 sensor on an injection system should fluctuate from rich to lean about 1 per second. If the number of cross counts is lower than this, it tells you the O2 sensor is getting sluggish and needs to be replaced. http://www.fordscorpio.co.uk/obd2scan2.htm

Though another site says:

Once a fully functional catalyst is up to working temperature and is “lit”, we should see a nice “flat line” signal from the downstream sensor, which shows that the exhaust gases have been successfully converted. http://www.picoauto.com/tutorials/cat-efficiency.html

An example of a single frame:

ABSLT TPS %     25.4
CALC LOAD %     74.9
MAP KPA           75
COOLANT °F       194
IAT °F           109
ST FTRM1 %       0.8
LT FTRM1 %       3.9
FUEL SYS 2       N/A
O2S11 V        0.875
ST FTRM11 %      3.1
O2S12 V        0.780
ST FTRM12 %     -1.0
OBD2 STAT         CA

Frame    21
Time    74.3

I cleared the code and it didn't come back during normal driving, but then during the long trip in the opposite direction, the engine light came on again. Both times the light came on were shortly after filling the tank and getting back on the highway. (I measured 38 mpg during this trip, which doesn't sound like anything is seriously wrong with efficiency.) I assume it's the same code this time but haven't read it yet.

It passed emissions inspection 02/19/2011.

I have oscilloscopes and volt meters if that would help figure out the cause.

Update: I replaced the upstream oxygen sensor and the light still comes on. It comes on after filling the tank with gas, and goes off after filling the tank with gas. So something about the gas being slightly different from one station to the next?

share|improve this question
Yep, most likely a bad O2 sensor and/or bad cat. –  Brian Knoblauch Sep 8 '11 at 12:44
@BrianKnoblauch: Yeah, but which? An O2 sensor is $$, while a cat is $$$. –  endolith Sep 8 '11 at 15:57
If the O2 sensor is really bad, you'd replace the rear one on bank 1. The rear/secondary sensor is the one that determines catalyst efficiency. Bank 0 vs 1 is manufacturer/model specific. You'd want to consult the manual. If the cat is bad, you'd want to find out why (oh, and it should be under warranty, I believe they're required to last 10 years nowdays). A mixture issue can lead to catalytic converter failure, and if you are forced to buy one, it'd be pretty unpleasant to just ruin that one and have to replace it again! –  Brian Knoblauch Sep 8 '11 at 16:22
Narrowband O2 sensors (which is what you have) switch between rich and lean indications exactly as your graph shows, so they are at least partially working... You'll need to find out which sensor is which but if S12 is the secondary sensor, this graph looks a little weird to me (assuming it uses the high voltage as a 'rich' indicator, which is what I'm used to seeing) as the second sensor seems to be staying 'rich' a little longer than I'd expect. Nevertheless it's hard to tell from the graph if it's the cat or sensor. I'd get the sensor checked out first, though... –  Timo Geusch Sep 9 '11 at 14:36
hey what cat did you replace?? the one in front of the motor? or the one in the back of the motor? i have a 2003 accord v6 ex model mine has two cats with two oxygen sensors comming right off the engine –  user2528 Dec 20 '12 at 7:41
show 1 more comment

3 Answers

up vote 5 down vote accepted

It was indeed the catalytic converter (the EX does not have a pre-cat). After replacing it, the downstream sensor graph spends most of its time above 0.5 V, as it should if the converter is working.


I started getting this error code again, a little over a year after replacing the cat.

I also finally found a guide on how to read these graphs, from a reliable source: Acura ServiceNews: Diagnosing Catalytic Converter DTCs

You should see these voltage readings from the oxygen sensors on a healthy converter:

  • Cruising speed, steady throttle: About 0.6 to 0.8 volts at HO₂S Sensor 2 (the heated oxygen sensor downstream of the catalytic converter).
  • During deceleration or fuel cut: 0 volts at HO₂S Sensor 1 (the heated oxygen sensor upstream of the catalytic converter), and 0 volts at the HO₂S Sensor 2. (The 0 voltage is due to no combustion and lots of oxygen in the exhaust.)
  • Throttle opening after deceleration: A gradual rise to 0.8 volts at HO₂S Sensor 2.
share|improve this answer
add comment

A common problem in civics of that year. It's not the main catalyst that goes bad, it's the pre-cat (the one hanging right off the exhaust manifold). This one is for cold start and once your car is warmed up, the lions share of the work is done by your main catalyst.

Take it in to the dealer to have your precat checked. Many cars have a 10 year warranty on emissions equipment.

share|improve this answer
how do you know that's the problem? –  endolith Sep 21 '11 at 19:32
For the record, my car is the EX, with the D17A2 engine, which only has one catalytic converter in the exhaust pipe under the vehicle, so that's not the issue. The D17A6 has two catalysts (one in the exhaust manifold and one in the exhaust pipe). The D17A1 has only the one in the exhaust manifold. –  endolith Feb 8 '12 at 2:29
add comment

I got the same code with 2000 tundra. I found you need to compare HO₂S Sensor 2 to engine load. If engine load is low enough, the sensor voltage should be 0 for a period of time as well. When decelerating, engine load is low, rpm is low and throttle position probably is close. Not a technician, just leaning.

share|improve this answer
add comment

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.