91 accord code 43

I have a 91 accord with 205k miles on it. Recently, it has been running
kind rough during acceleration and after a little while, the check
engine light will come on and then the car runs just fine. It is
throwing a code 43. I have used some fuel injection system treatment in
the gas, and had a tune up with new plugs, wires, pcv valve, air
filter, and fuel filter (all honda parts).

I have been searching old threads in this group, but haven't been able
to find any where the car was running bad until the check engine light
comes on, and then runs good. When I am accelerating, it just doesn't
seem to want to go, and it also has a very rough idle (until the light
comes on) and sometimes it is so bad that I have to keep my foot on the
gas a little to keep the car from going dead.

Anybody know, or have any idea, what could be wrong? I was thinking it
might be an O2 sensor, but I would like to know more before I shell out
$300+ for a new one.

Thanks!

That the symptoms go away after the CEL and code come up is
a little odd. I wonder if the ECU cuts out the oxygen sensor
signal at this point, so the car seems to run better after
the code comes up.

Otherwise, given the symptoms (including Code 43), I would
strongly consider replacing the oxygen sensor, especially if
it's the original. IIRC correctly, some folks have helped
quickly confirm a failing O2 sensor (on older cars like
yours without allegedly more sophisticated onboard
diagnostics) by disconnecting its electrical lead and
observing how the car runs.

Which engine does your Accord have? At
https://www.automedicsupply.com/ , for the F22A1 engine, you
can get an OEM O2 sensor for about $90. For the other 91
Accord engines, about $130. I have used this online site and
service is excellent. Haven't found better prices for OEM
sensors.

http://www.pelicanparts.com/BMW/tech...gen_sensor.htm
has some good discussion.

New OEM distributor cap and rotor, and timing check, are a
good idea, too.

> wrote
>I have a 91 accord with 205k miles on it. Recently, it has
>been running
> kind rough during acceleration and after a little while,
> the check
> engine light will come on and then the car runs just fine.
> It is
> throwing a code 43. I have used some fuel injection system
> treatment in
> the gas, and had a tune up with new plugs, wires, pcv
> valve, air
> filter, and fuel filter (all honda parts).
>
> I have been searching old threads in this group, but
> haven't been able
> to find any where the car was running bad until the check
> engine light
> comes on, and then runs good. When I am accelerating, it
> just doesn't
> seem to want to go, and it also has a very rough idle
> (until the light
> comes on) and sometimes it is so bad that I have to keep
> my foot on the
> gas a little to keep the car from going dead.
>
> Anybody know, or have any idea, what could be wrong? I was
> thinking it
> might be an O2 sensor, but I would like to know more
> before I shell out
> $300+ for a new one.

Elle wrote:

I would
> strongly consider replacing the oxygen sensor, especially if
> it's the original. IIRC correctly, some folks have helped
> quickly confirm a failing O2 sensor (on older cars like
> yours without allegedly more sophisticated onboard
> diagnostics) by disconnecting its electrical lead and
> observing how the car runs.

------------------------------

Crappy signal versus NO signal will make a big difference. It's not a
very valid test because the computer KNOWS when it's sensor has been
unplugged. I'm toying around with some of this stuff right now on a '00
CR-V right now..

'Curly'

"'Curly Q. Links'" > wrote
> Elle wrote:
>
> I would
>> strongly consider replacing the oxygen sensor, especially
>> if
>> it's the original. IIRC correctly, some folks have helped
>> quickly confirm a failing O2 sensor (on older cars like
>> yours without allegedly more sophisticated onboard
>> diagnostics) by disconnecting its electrical lead and
>> observing how the car runs.
>
> ------------------------------
>
> Crappy signal versus NO signal will make a big difference.
> It's not a
> very valid test because the computer KNOWS when it's
> sensor has been
> unplugged. I'm toying around with some of this stuff right
> now on a '00
> CR-V right now..

I believe you. Just thought I'd seen someone say they had a
rough running car and when they unplugged the oxygen sensor,
things settled down, assisting with a decision to replace
the sensor. Presumably this was at idle, 'cause for the
reasons you gave, and knowing the function of the sensor,
I'd expect the computer to be all messed up without the O2
sensor input signal.

But I don't have any substantial documentation for this
little test. The net on the other hand has plenty of
resources that describe more sophisticated generic and
specific (to one's particular car model) tests for an O2
sensor.

"'Curly Q. Links'" > wrote in
:

> Elle wrote:
>
> I would
>> strongly consider replacing the oxygen sensor, especially if
>> it's the original. IIRC correctly, some folks have helped
>> quickly confirm a failing O2 sensor (on older cars like
>> yours without allegedly more sophisticated onboard
>> diagnostics) by disconnecting its electrical lead and
>> observing how the car runs.
>
> ------------------------------
>
> Crappy signal versus NO signal will make a big difference. It's not a
> very valid test because the computer KNOWS when it's sensor has been
> unplugged.



If the sensor is unavailable for *any* reason, the system will go into open
loop, so unplugging it will be a good initial test.

As far as I know, the '91 system is in open loop under the following
conditions:
* Idle (at any time)
* Below full operating temperature
* Full throttle



--
TeGGeR®

The Unofficial Honda/Acura FAQ
www.tegger.com/hondafaq/