Can an O2 sensor be tested with a voltmeter? If so, do you just test the terminal to ground with the lead connected, disconnected or hook it up in series with the lead to the sensor?

TG

The oxygen sensor cannot be tested with ANYTHING.........PERIOD!

Phil

------------------
87 FIERO GT 2.8 5spd
0-60 in 6.8 seconds!

After fighting a "code" for months on my first Fiero , on my second fiero I just replaced it before I even drove it.
I feel it is a tune up part. Just like plugs. For the price,25-30 bucks. replace it and you don't have to worry about it.
wheelie

Taking y'alls advice, I didn't bother trying to test the sensor. But, I did find a mechanic that is a MPFI expert. He had one of the Snap-on scan tools that cost about 3 grand, and he knew how to use it.

After hookup, it kept showing rich, lean, rich, etc. He was suspicious of the O2 sensor and ran a thorough check of the system. He checked all the sensors, then fuel pressure and then checked for vacuum leaks. He told me how to service the system and what to look for. Then did a test drive.
For all he did, he charged me $35 and I felt like it was a bargain.

Turns out the guy has a bunch of Fieros in a field that he keeps for parts and restoration! There are some good mechanix out there, you just have to find em.

I'll gonna chunk the sensor and consider it a major tune up part.
Thanks
TG

You can test an O2 using propane but in the end it's not a foolproof test. Outside of a dead (pulls the O2 bias down to 100mV all of the time) a slow switching O2 is the first thing that happens when an O2 is going bad. The output lags the AFR change. It throws off the AFR in closed loop.

Considering how cheap the O2's are, it's just easier to throw in a new one.

TK

O2 sensors CAN be tested. If you have a graphing DMM or a DSO.

A correctly working sensor generates a voltage between 0 and 1v. When driven rich or lean it should respond in under 100ms. It should also read over 800mv and under 200mv when forced rich or lean. If you don't have a graphing meter or a DSO you won't be able to check response rate. If you do have the needed equipment you can test the sensor in car by backprobing the sensor connector, monitor the sensor as it warms up (hold throttle about 2000 rpm for a couple of minutes). If you drive the system lean (make a big vacuum leak) the voltage should drop like a rock to under 150mv in less than 100ms. When you drive it rich (plug the leak or use carb cleaner in intake) voltage should rise to over 850mv also in under 100ms. Things to watch out for... some contaminated sensors will get slow or flatline at varius voltages after warming up. Also if you see a negative voltage the sample cell if cracked.

Make sure the exhaust isn't cracked or leaking near it... that will cause false lean (under 450mv) readings. The O2 sensor bung is also a good spot to do backpressure testing, you can buy backpressure gauge sets that screw right in place of the sensor.

Hope these few tips help someone out.

Phil... why would you say they can't be tested? They can even be benchtested with using propane to heat them and drive them lean and rich... just clamp in a vise, ground your graphing multimeter or scope to the vise, red lead to sensor lead. Heat tip with propane torch... should take less than 30 seconds to start generating a voltage which should go over 850mv and stay over 800mv for 2 minutes. Then flick the torch away and watch reading drop to under 150mv in under 100ms. When you flick back over tip should go over 850mv in under 100ms. You can often see certain types of contamination moving around on the tip of the sensor when doing this. It has the benefit of burning the carbon off the tip. I rarely do this sort of testing anymore except as a demonstration.

Hope it helps.

One reason I asked if they could be tested is to know how and if it could. It seems that some parts are getting hard to get and if I could test them, I won't have to spend 2-3 days getting a part I may not need.
This may sound sort of picky, but I had a heck of a time tracking down a throttle position sensor, then getting it ordered only to find it was not an exact fit (I returned it).
That and it is fun for me to dig into this stuff and learn something.

GM is getting to be less helpful and good parts may be hard to find. If I learn how to test parts from junkers or whatever, then they can be used by me or someone on the forum without a big waste of time.

Thanks
TG

Bosch oxygen sensors for Fieros are under $20 at Advance and Autozone.

One reason I said that they can't be tested is because the service manual SAYS that it cannot be tested with....ANYTHING! I am speaking out of the vehicle of course.

Phil

------------------
87 FIERO GT 2.8 5spd
0-60 in 6.8 seconds!

You can test the switching across stoich but it's hard to test the speed. That's why it's better to just replace it.

If you have a scanner, you can check for the cross counts.

It probably is tough/impossible to do a reliable test outside the vehicle. There seem to be several factors for it to work right. If you had baseline data to go on (fuel/air ratio under certain conditions) ,you might be able to get a better-than-rough-idea with it installed.
At less than $20, it is not that big a deal. But when you are in sub-suburbia like me, it is easier to hang a meter on something and test it than waste a drive to town and blow $20.
Or order something and have it not come in or be wrong.

BTW, how much info does the Win-ALDL give?
Since my wife got a laptop, this might be worth looking at. It is a 1100 mhz machine and runs Win XP.

TG

I can't really think of anything useful that the aldl needs that my 3k Snapon does. WINaldl is perfect if you're not in the business.

Phil

------------------
87 FIERO GT 2.8 5spd
0-60 in 6.8 seconds!

$20 I wish mine is $85 before tax what a rip off GM is.

------------------
Stock 85 GT 4 speed 2.8

I wish I could justify the Snap-on tool!
I was impressed.

TG

like GTdude said, WinALDL is just as useful as the snap-on scan tool, if not more so

A few points here....

Win ALDL is nice IF you have a motor it supports. Otherwise the least expensive scan tools I know of are the AutoXray GM ODB1 scanners. I'm told you can get referbished ones cheap. (Call AutoXray) New ones run between about $150 and $200(usd) so shop around.

The method of testing an O2 sensor with a torch is iffy at best. It is very easy to trash a good sensor doing that test. It often heats and cools the sensor very fast, much faster than normal, which can lead to imediate or near future failures caused by stress cracks.

If you think the sensor is bad always check both the wire to the sensor and ALL the grounds bolted to the engine/transmission. O2 sensor issues are very often caused by a cruddy wire someplace. (With one wire O2 sensors, one of those grounds on the block is the O2 sensor return line.)

If the O2 sensor hasn't been changed in a couple years, replace it. They usualy have a rated life of about 30,000 miles. I use only Bosch or the Carmaker's units. ACDelco/GM on all GM, MotorCraft for Ford, Etc...

While you can see what it is doing with a Fuel/Air Meter, DVM, or ECM scanner, there is NO way to be sure those numbers accurately reflect what is really going on.

The voltage may change within range but the sensor could be a hair slow responding. That results in the ECM responding slowly. All it take is a sensor that is a tiny bit slow to mess things up.

Plus... ECM data streams are laggged. How much depends on the ECM at hand. By the time the ODB1 ECMs report the data to the scanner there have likely already been several changes. BTW... The Scanner should show it changing rich/lean. If it doesn't change then the Sensor or wiring s shot. Keep in mind tho that the ODB1 stream only updates a few times a second. It's not quite real time. The rich/lean flag may not change with every update because it may have changed multiple times between updates.

Again, if it's more than a couple years old, or you simply don't know it's age, replace it.

------------------
RIP Crew of Columbia. 01-Feb-03

All these big nice tools....I just used my voltmeter, hooked it up in series with the sensor.
and it read .1v at idle and went up to .9v at full speed...That's just fine..

quote Originally posted by theogre: The method of testing an O2 sensor with a torch is iffy at best. It is very easy to trash a good sensor doing that test. It often heats and cools the sensor very fast, much faster than normal, which can lead to imediate or near future failures caused by stress cracks.

I much prefer to do response tests in car driving the sensor rich/lean using propane enrichment tool and a big vacuum leak.

If your going to bother taking the sensor out to bench test, it should probably be for educational purposes, in a shop environment it makes more sense to just spin a new one in and move on.

And stop using the Oxy/Acetylene torch for bench testing Ogre ;-) Sure it assures that the sensor is bad... just like tap testing the ECM with your 3lb deadblow ensures your replacing a known bad ECM.

Just kidding, thanks for all the great info you give here.

omg.......i loved that analogy..........lol
how true how true

Phil

------------------
87 FIERO GT 2.8 5spd
0-60 in 6.8 seconds!

I usually don't bother with any testing at all. If it's more than a couple years old, or no-one knows how old it is, it gets a new one. This is cheap insurance on older cars. (It's a bad idea on new cars... more in a second.)

Trust me... even a lighter can heat the O2 sensor too fast and make it croak. The comment timming is what's funnier when we sort of "fixed" a device this week at work... It went to scrap. They got a newer used unit in it's place. So they are happy and I have one less 10 year old heap to keep working.

Enrichment and leaning aren't bad ways to test.

Again, you can only test them so much. I don't know of any test that will detect slightly lazy O2 sensor. These look ok on tests but respond just a hair tooo slow. It can be enough to be trouble.

If you have a newer car with multiple sensors enrichment and leaning is about the only way you can test them. Blanket replacement on these cars is very expensive. Some of these newer, usually heated, O2 sensors are $50-$100 each. some might be more.

Many new cars have 4 or more O2 sensors. Fortunatly, the newer cars all have OBD2 now. OBD2 has a faster data stream and more brains. It usually not very hard to find the batty sensor with these systems.

As always, if the computer says a sensor is bad, don't assume it's the sensor. You've still got to check all the bloody wiring to them. (Most new cars use heated O2 sensors. A common problem is low heater voltage caused by a dirty/loose wire.)

I wonder how Bosch and whoever tests them at manufacture. If it is like most plants, they probably run a go/no-go test of a representative sample. Since I started this thread, I think I ought to track down Bosch's Web page and see if I can get something of value.

I can't believe modern engines use up to four sensors! It seems too complicated.
What happens in several years or so when a $40-$50k vehicle has wiring harness problems?
Or a 1 yr old car? I am not against technology but some things just don't seem practical.

How fast are the ECM's in a Fiero? I assume they are OBD1. 2-4 MHZ? What processor? 8 bit Texas Instrument? I bet Von Neuman (sp?) laughs everytime a car starts.