The gauge in my road runner is acting pretty funky, some pretty serious lean conditions. But it isn't consistent and isn't exactly reproducible.

The 02 sensor doesn't have that much time on it, but this is also not a normal vehicle!

I've seen both things on the net, from ways to clean them to they can't be cleaned.

Any thoughts?

high zinc oil and leaded gas are enemies of O2 sensors.
my friend had a chassis dyno with a heated O2 sensor. he dynoed a car with water injection. it destroyed the O2 sensor on the first pull.

i have heard you can take a propane torch and heat up the sensor part and that sometimes helps.

An additive called Techron does it for my Honda Accord and clears the check engine light. You can get it at auto parts place or its in Chevron gas.

I know nothing about O2 sensors in antique cars though.

Shouldn't need cleaning, but you should test it. I have done this many times. As long as you can get close to 1 volt, and it reacts quickly, it is good. It sometimes takes a little experimenting to get the heat just right to make the O2 react.

www.youtube.com/watch?v=NKsS6AELPuM

The afx one I use asks to be calibrated in open air periodiclly.
says to take it out and set it in open air and let it heat up for like 30 min and the you can put it back in.
but I guess it depends on your system.

Wow, great answers/opinions everyone.


Yes, you can clean an O2 sensor with oven cleaner Mick. I've done it a couple of times and it works.

Fill a small glass with oven cleaner and soak from the threads down in it for an hour or two. Blow it out with compressed air and good as new. I do it twice to make sure everything gets taken off the element.

If you still get wild readings after cleaning it then the sensor has to be replaced.

A propane torch burns at 3500 degrees and normal car exhaust burns at about 4 to 500 degrees. I guess you can imagine what will happen to the element in an O2 sensor when you touch it with a propane torch. Yep, probably wreck a perfectly good $150.00 sensor.

oven cleaner soak. i'll try that sometime with a couple of sensors [4 wire] i have here [used] on the shelf.
when my 99 minivan set the cat O2 code, i took it out and it looked dirty. [had 200k plus miles on it]
i had a couple of fairly low mile O2 sensors from a ford ??? that had the same wire colors in the same pattern on the sensor, but with the ford connector on the end.
i transferred the minivan plug to the ford sensor, installed it, and code gone. it seemed to be working as it should through the scan tool in real time frame, and 20k something miles later, still no codes while my "bus" is running like a top.
so are these sensors basically the same, but with different connectors ?

Thanks. Will give that a shot.

Makes sense about the propane torch, somehow that didn't seem right...

Don't know when I'll get to it, but will try to report back.

A propane torch is a valid test to know if the O2 is reacting correctly or not. Actually since the element is shielded, it can be difficult to get it hot enough to react. If the O2 is good, as soon as it comes into it's operating range, it reacts nearly instantly. And looses it just as fast so it is not necessary to hold the heat on for any length of time. I guess it is possible to overheat it, but is not as easy to do as you might think. And as an FYI, I have tested dozens of O2s over the years like that and they either reacted okay or they didn't. Even multiple tests rendered the same outcome. No deterioration in performance.

An O2 needs to typically exceed 600* in order to operate and their heaters often hit 1,200*. If propane scares some of you then use a candle, it only goes to about 1,000*.

Dismiss these tests if you like, but how else can you know know if the O2 actually functions correctly? Whether they are clean or not.

I'd recommend the torch, since using a candle may deposit even more carbon on the sensor...

Good point.

By your experiment, I think they are all the same Moparx. They all go from 0.1 to 5 volts, and I think Bosch or NTK are likely the supplier to the OEMs.

Physically, the only difference I see is the size of the holes on the outer shield. They all have to relay 14.7 air fuel ratio which is 0.5 volts I believe. Every engine has to hit that number, so I can see why that Ford sensor works for you. Glad you went ahead and did this, I'll know for next time I'm at Pick n Pull.

Keep in mind that there are two types of Oxygen Sensor,
narrow band 0.1 to 1.0 volts
and
Wide Band 0.1 to 5 volts.

https://en.wikipedia.org/wiki/Oxygen_sensor

If you have a small gasoline engine powered device like a lawn mower or electric generator
you can make a test device by making a pipe section with a threaded side hole for the Oxygen sensor.
Slip this pipe over the muffler outlet and let the exhaust flow through past the Oxygen sensor.
With the engine running normally the Oxygen sensor should read slightly lean, engage the Choke on the engine and it should read rich.

You can actually use this same test device to check the correct running of your natural gas furnace.

Most Oxygen sensors will go bad if you get metal deposits on their CeO2 ceramic,
such as Zinc from burning lubricating oil in the cylinder,
or Lead from running Tetra-Ethyl leaded gasoline, such as Aviation or Race gasoline.

We used to inventory more higher mileage vehicles than we do now which meant that we had to deal with more O2 sensors. I always kept a number of O2s on the shelf that I had pulled from the pick-a-part yards. All we really had to deal with were the 4 wire narrow bands. There were basically 2 brands; one with the round end on the shield and one with a flat end on the shield. Often they would interchange and work, no problem. But occasionally for what ever reason, a vehicle would not like anything but the style it came with. So, I would usually replace an O2 with the same style that came out of it. Wasn't usually necessary, but since I had a handful to choose from, it just ensured that I wouldn't need to revisit it.

And being used, I always tested them before installing them. They needed to swing through their voltage range very quickly, and needed to hit at least .9v. Preferably, a full 1v. The connectors were usually different from one car to another so it would need to be changed. And although the sensor itself was usually the same, downstream O2s usually had much longer leads.

I also used the narrow O2s in a tail pipe fixture that I had made with a meter to set the cruise on some of my buddy's carbed stuff. Probably like the fixture that was mentioned in an earlier post. It was good to get the cruise in the mid 14s. Every Holley that I have checked has been way rich on the cruise, often 4 jets sizes or more.

Because of our change in inventory, I don't see much O2 trouble codes anymore. Now it's TPMS. So, I carry a scan tool and pull them at the yards so that I can keep them handy. It's always something.

Wouldn't a random misfire cause this, too?

Yes, it could. But a random misfire usually sets it's own code, too. It is important to figure out if there is one code and one code only. Or if there are multiple codes and which is the primary offender. That often takes clearing the code(s) and seeing what comes back and how soon and in what order.

If you wanted to clean the CeO2 ceramic of an oxygen sensor,
should you use an oil, an alcohol, an acid or a strong base like lye?

There may be a clue that a strong base like oven cleaner is safe to use from this quote from Wikipedia:

sample quote

One small but illustrative use is its use in the walls of self-cleaning ovens as a hydrocarbon oxidation catalyst during the high-temperature cleaning process. Another small scale but famous example is its role in oxidation of natural gas in gas mantles.[20]

end quote

If the walls of self-cleaning ovens are coated with CeO2
where the oven manufacturers should know that strongly basic oven cleaner will be sprayed onto those walls
then one might guess that oven cleaner would not do serious damage to it.