Wideband O2, why a heating element?

[honclfibr]

I noticed that most of the wideband O2 sensors have heating elements. I assume that this is to keep the temperature around the sensor stable in order to keep the sensor linear. My question is, why is it not possible to use a temperature sensor and compensate for the nonlinearity of the O2 sensor in the controller?

I've had experience designing accelerometers for industrial applications (I'm an EE) and this is what we found, we originally focused on keeping the accelerometer at a constant temperature but have found that it is much more reliable to compensate for the ambient temperature change.

 

[WombatTSi]

Its simply easier and less complicated to have the heater built in rather than run a whole extra sensor (not to mention the precious ECU space required for such calculations) and the difference would be negligable. BTW, most 02 sensors are heated, not just widebands (all 4 wire ones are)

 

[Defiant]

Like everything else in cars, this way was 3¢ cheaper.

 

[NinjaTech]

I may be completely wrong on this but I have heard that the turbo DSM's have pyromiters built into the O2 sensors, not for use in compensating for temperature diffrence though. I was just wondering if this was true or not so that I dont give any false info in the future. I kind of dont think they do simpley becouse there are not enough wires on the O2 sensor for this function, but just wanted to clarify.

Thanks,
Pat

 

[steve]

I have heard that the turbo DSM's have pyrometers built into the O2 sensors.

Not as far as I know.

Steve

 

[GTM]

Not as far as I know.

Steve

Actually they are the same as best I know. Pyrometer was the origin and became the standard nomenclature for science, marine, aircraft, commercial trucks all had pyrometers. It wasn't until the automotive industry started mass producing them with less accuracy that they adopted EGT to describe it's function. The pyrometer is considered a percise scientific instrument and it's calibration becomes important whereas the EGT rarely has any such stipulation.
.........

The problem is the thermister, or thermocouple sensor and employing a thyrister for switching.

It's impossible to make a sensor which has a linear curve, thus at low temps they do not reproduce accurately if their normal range is say 600+. If you could keep the temp below 900F you could use a thermister and have the accuracy needed. But we cannot use these because we are seeing 1.500-2,000F temps so the sensor needs to be heated but monitor how much current is being consumed, plots what the sensor is reading and compensates until the sensor gets up to it's operating temp and the heater is not needed. It's been 40 years since I had to think about these things so hopefully I got most of it right. Actually I had a medical research project where we were on the cutting edge of measuring temps of less than 1/10000 of a degree in joules. Hehehe no electronic calculators ...

There will be a test, the right number of answers will be subtracted bla bal bla.

Cheers,
GTM

 

[steve]

Actually they are the same as best I know.

What's the same? I'm confused as to what your refering to.

The comment was "I have heard that the turbo DSM's have pyrometers built into the O2 sensors" and I replied "not that I know of". Although it may be possible to read a temp from the heater in the O2 sensor based on it's resistance as it heats up, it's not a pyrometer. I don't think your would be able to get any repeatable temp data from the O2 side of the sensor with any sort of processing likely to be found in a car.

I can assure you that the O2 sensor as used provides no temp information to the ECU and there is nothing else in the car from the factory that could be considered a pyrometer from the point of measuring exhaust temp.

Steve

 

[GTM]

What's the same? I'm confused as to what your refering to.

The comment was "I have heard that the turbo DSM's have pyrometers built into the O2 sensors" and I replied "not that I know of". Although it may be possible to read a temp from the heater in the O2 sensor based on it's resistance as it heats up, it's not a pyrometer. I don't think your would be able to get any repeatable temp data from the O2 side of the sensor with any sort of processing likely to be found in a car.

I can assure you that the O2 sensor as used provides no temp information to the ECU and there is nothing else in the car from the factory that could be considered a pyrometer from the point of measuring exhaust temp.

Steve

Yes you are right I didn't complete explaining.

A pyrometer is a complete unit which consists of a gauge and a sensor. A thermometer isn't a thermometer if it doesn't have liquid inside. Off the top several of your electrical meters are just an ohm meter measuring resistance like the pyrometer. So while you don't have a gauge on the O2 sensor it's change in resistance is being plotted in the ECU.

Were you to mount an ohmmeter on your dash, hook up the O2 sensor you would have a what was errantly described as a pyrometer but in fact become an O2 meter. In the case of the O2 it's _reacting _to_ O2_, plotting the resistance change so not a pyrometer, nor is the heater though it is measuring the current used to heat the O2 probe inside the unit but you can't call it an ammeter though you could measure the current flow. There are 1,2,3, and 4 wire O2 sensors if I'm not mistaken and I think you can use a 3 wire in a 2 but not the reverse and can only use a 1 wire in the one wire sensor, not sure about the 4 wire but think one is a ground. Check schematics. As I recall it has to do with the special alloy being changed to improve to it's reversable reactance to O2. I don't remember the alloys used but the reason for the heater has to do with their reaction at lower temps is less accurate.

When you test these circuits/devices you use the ohm meter and the volt meter to compare with factory specs. Does that help or have I just made it worse.

Cheers,
GTM

 

[steve]

We are really off topic so I beg forgiveness.

Yes you are right I didn't complete explaining.

A pyrometer is a complete unit which consists of a gauge and a sensor. A thermometer isn't a thermometer if it doesn't have liquid inside. Off the top several of your electrical meters are just an ohm meter measuring resistance like the pyrometer. So while you don't have a gauge on the O2 sensor it's change in resistance is being plotted in the ECU.

Using your analogy a thermocouple or thermistor is the liquid inside a thermometer. There are various types but they either act be generating a voltage when heated or by changing resistance.

The Nernst Cell in a Lambda (O2) sensor doesn't measure heat, Once the temp of the cell goes above 625°F the cell becomes electrically conductive and transfers of oxygen ions from the reference gas inside the cell to or from the outside exhaust gass generates a voltage. This voltage reflects the difference in the oxygen content between the exhaust outside the sensor and the air inside. The reaction is sensitive to changed in temp but that's secondary and really noise to the function of the sensor. The narrowband sensor used in most cars also has the property that it's output voltage changes rapidly around the stoichiometric point.

The ECU doesn't monitor the change in resistance, it digitizes the sensors output voltage and starts decreasing the amount of fuel injected while watching the voltage for the swing around the stoichiometric point. The actual code in the ECU just looks for the sensor output to go below one voltage before it starts increasing the fuel and watching for the voltage to go above another value. Each itteration of the software loop it lowers or raises the fuel until it see the switch. This is repeated whenever the ECU is in closed loop to maintain an average of 14.7:1 AFR.

Were you to mount an ohmmeter on your dash, hook up the O2 sensor you would have a what was errantly described as a pyrometer but in fact become an O2 meter. In the case of the O2 it's _reacting _to_ O2_, plotting the resistance change so not a pyrometer, nor is the heater though it is measuring the current used to heat the O2 probe inside the unit but you can't call it an ammeter though you could measure the current flow. There are 1,2,3, and 4 wire O2 sensors if I'm not mistaken and I think you can use a 3 wire in a 2 but not the reverse and can only use a 1 wire in the one wire sensor, not sure about the 4 wire but think one is a ground. Check schematics. As I recall it has to do with the special alloy being changed to improve to it's reversable reactance to O2. I don't remember the alloys used but the reason for the heater has to do with their reaction at lower temps is less accurate.

The ECU doesn't do anything with the heater in the sensor but power it. On the turbo cars the heater isn't even directly connected to the ECU. It's powered off the MPI relay just like the MAF, CAS, Solenoids, ISC, and injectors. So the heater might be usable as a thermistor as a side effect, it's not used that way.

The 1, 2, 3, and 4 wire sensors have to do with, 1 how the ground reference is generated and, 2 heater or not.

The alloy used is zirconium-oxide or titania-oxide ceramic. The heater is used to bring the cell up to it's reaction temp quicker than it would without it.

When you test these circuits/devices you use the ohm meter and the volt meter to compare with factory specs. Does that help or have I just made it worse.

Cheers,
GTM

I think I understand why you brought this up but I disagree that the O2 sensor could be considered part of a pyrometer. It's simply wouldn't be very good at it and it's not used that way.

Steve

 

[GTM]

We are really off topic so I beg forgiveness.


The Nernst Cell in a Lambda (O2) sensor doesn't measure heat, Once the temp of the cell goes above 625°F the cell becomes electrically conductive and transfers of oxygen ions from the reference gas inside the cell to or from the outside exhaust gass generates a voltage. This voltage reflects the difference in the oxygen content between the exhaust outside the sensor and the air inside. The reaction is sensitive to changed in temp but that's secondary and really noise to the function of the sensor. The narrowband sensor used in most cars also has the property that it's output voltage changes rapidly around the stoichiometric point.


The alloy used is zirconium-oxide or titania-oxide ceramic. The heater is used to bring the cell up to it's reaction temp quicker than it would without it.

I think I understand why you brought this up but I disagree that the O2 sensor could be considered part of a pyrometer. It's simply wouldn't be very good at it and it's not used that way.
Steve

I fell into the Pyrometer/EGT because someone previously else stated one of those urban legends.

I had looked this up some years back but forgot most of what I had read. However I don't remember any discussion of the "stoichiometric point" but if you say it is a factor you are doing better than me. The info I'd found was a discussion between 2, O2 sensor makers and their lab tests results. At the time I wasn't aware there was a heater so thought it might be a thermistor because they were using platinum wire on one side of the circuit which is used in thermistor design but then quickly realized this was not relative.

Cheers,
GTM