Revisit 90 Accord EX O2 sensor

I still have a Code 41 on my F22A4.

Since heater resistance is spec 10-20 ohms, the NGK that came out read
19 ohms, and my current Bosch read 5 ohms, I spliced a 14 ohm 1W
resistor into the Bosch. The heater circuit read 19 ohms after the
splice. Cleared the ECU via fuse. Radio is reset, so I know I pulled
the right fuse. Code 41 comes back.

I snagged a new Denso (OEM) O2 sensor off Ebay. Plugged it in (didn't
bother to install it). Cleared the ECU via fuse. Code 41 comes back.

There is 12 volts at the O2 harness.

I'm baffled. I guess all I could really do at this point is an ECU
swap?

 

eh? how is it going to give correct signal if it's not installed???

 

You might also consider checking the wiring harness. Check the
continuity of the signal wires at the ECU harness by jumpering the
sensor signal wires at the sensor connector. You want to use high
impedance ohm meter to measure the voltage swings while the engine is
running at the ECU harness or the sensor may be damaged.

I wouldn't temper with the heater BTW. If you select the correct
oxygen sensor, particularly the excellent Bosch Planar type, then it's
a matter of Plug-and-Play. I've used many of these planar types,
including the "universal" type where you have to splice the old
connector back on, and Bosch sensors are excellent.

I had to use one of the many types of Bosch "universal" sensors just
to find a planar for some cars. Note, use only those "universals"
approved in the catalog for your car. I even fixed a P0420 (catalytic
converter efficiency below threshold) problem dumping the Denso
sensors for Bosch planars. Yeah, it turned out to be crappy oxygen
sensors from Denso. No more Denso crap for me. NGK should be OK, I do
like their Iridium -IX / -Laser plugs. Great stuff.

Best-in-class: Bosch Planar type oxygen sensor
http://www.boschautoparts.com/Products/OxygenSensors/PlanarO2.htm

 

wow, spot the shill! you get paid for that?

here's the thing tho buddy, most other posters on this [_honda_] group
disagree with you and think bosch sensors are crap. based on bad
personal experience, that includes myself.

want more bad bosch? their spark plugs. inherited a set of platinums
with my crx - lousy. only 2k miles old. slow starting, poor pickup,
poor top end. replaced with standard ngk's, and now she purrs like a
pussycat.

bottom line, bosch suck.

 

Your mileage obviously vary, which is fine. We talk about the good,
bad and ugly of cars.

I have the exact opposite, good experience with Bosch planar sensors
than you. I use exclusively NGK Iridiums, but will lean towards trying
Bosch x2 and x4 next time because of positive experiences from some
owners and myself with the sensors. Although I must say I had just
assumed the Bosch plugs are bad, like most people in this group.

There are talks of Japanese cars like NGKs and not Bosch; German cars
like Bosch not NGKs, etc. But when a Honda part (Denso sensor, Honda
DOT3 brake fluid) didn't work so well in a Honda, then I'd be open to
try other options. And I happy to find that Bosch makes excellent
oxygen sensors for a variety of cars.

 

Note that the problem I'm trying to eliminate is a Code 41 (Heater
circuit malfunction), not a Code 1 (Incorrect oxygen content).

 

Yeah, quick search showed it as sensor fault.

But some cars use current draw as a test. You have to find the range
it accepts for your year. So if current draw is above or below, the
code is set. But if current draw is correct, then it may indeed be the
ECU. But codes are not always accurate, so you may have to test
coolant temp sensor too.

 

but does it not need to be installed to have all elements earth correctly?

 

I actually did strap it to the exhaust for that reason (maybe a Code
41 is thrown for other reasons?) But as far as I can tell, if the
heater circuit grounds anywhere besides the return wire, it's a fault
in the sensor.

Re: John's post, does anyone know where I could find those current
draw specs? I wouldn't have the first clue where to look.

 

Why coolant temp sensor?

the O2 heater is the fuel injectors.

 

Low temp (or faulty sensor) puts the system in open-loop mode. It's
recommended to check this with sensor codes. Some cars expect current
draw to be 0.3-2.5 amps. Outside then it throws heater circuit code.
Don't know about that year of yours.

A quick look shows you have to check connection to ground at the
sensor connector, and also making sure the proper wire has a path to
ground and the other wire doesn't (wiring test at sensor, and between
the sensor and ECU). So just strapping the sensor may not provide the
"ground" it wanted. Best to refer to the proper manual for that year
and model.

If wiring checked out it's recommended you swap in a known good ECU
and see if the code disappears.

 

BTW, what other changes were made besides taking out the old sensor
that gave you the 41? So the Bosch was still in the pipe? Maybe try
removing the resistor you spliced in? (particularly *if* ECU tests
current draw)

The resistor was 1W. But if the heater operates at 12v and if some can
draw 2.5 amps, then you'll have 30W on your hands, am I right?

rockauto.com shows:
Bosch 15710 w/ OEM plug (looks like a planar type btw, not sure)
Bosch 15733 Planar type universal type
F22A4 uses: 36531-PT6-A02

 

All I have is the dash code, which is a 41.

Okay. that would equate to a range of 4.8 ohms to 40 ohms on the
heater circuit.

Something I didn't think to check is the resistance of the ECU side of
the circuit... hmm.

None of the wires in the plug show continuity with the threaded part
of the sensor.

I don't get it. If anything it doesn't provide the heat it needs to
enter closed loop. Whether the sensor is threaded into the pipe or
strapped to it should not matter at all, right?

Anyone have one to offer?

 

None aside from replacing the sensor and trying the resistor splice.

It threw the code both with and without the resistor.

Yeah, I screwed up. Even at 19 ohms, that's 7.5W. I'd have to find a
power resistor rather than the electronics resistor I used.

I don't care about the Bosch that much except as an exercise. I've
moved onto the OEM sensor. Why the OEM sensor wouldn't work is a
complete mystery to me and what suggests I'm dealing with a bad ECU.

 

You checked:
So you check the resistance of the sensor side (plug lead #3/4),
checked no-continuity either of #3/4 to GND on the sensor side, and
battery voltage from #4 on ECU side.

I didn't read that you checked:
Just want to make sure the #3 wire going back to the ECU side isn't
grounded to chassis with the ECU disconnected; but #3 is grounded with
ECU connected. (This is the heater return from #4 for a later model,
F23, so YMMV).

If this checks out you should test the ECU. But you'll need to find
the correct manual and procedure for the F22. And another ECU, maybe:
www.pickapart.com

Regarding install: I was just wondering if the heat from the exhaust
is needed in addition to the heater is all to go closed-loop. May not
matter. Anyone?

Was the Bosch you used one of the numbers I listed? Just for my
reference only. Thanks.

 

the probability of ecu failure is very small. check the feed circuitry
- unplug the ecu and check continuity for the whole cable route from the
ecu plug to the sensor - then get a sensor from a junkyard and see how
you go with that for testing. iirc, there was a prelude guy here a
while back with a similar problem that disappeared on fitting the oem
sensors - and he'd tried all the alternatives.