Know your Errors - TRMS pro's and con's

[Phyllomedusa]

Hey fellow,
after I am currently crawling under a car again and would like to give it a check up (after all the rust is gone) I was thinking about the Lambda-Sensor and a check-up.
So what is it....just an oxygen sensor that tells the car's ECU how much oxygen is left in the exhaust gas and therefore helps the ECU to control the amount of injected fuel for the best fuel/air mixture.

The signal is an alternating signal that ranges somewhere between 0,2 and 0,8V (warm and healthy engine). So you could see it as an ±0,3V AC Signal with a 0,5V DC offset. If there would be a problem with the engine: let's say the fuel filter is aged and there is not enough fuel delivered, it means that there would be more oxygen left over compared to a good burn (as there will be less oxygen needed to burn the fuel). This mixture is called lean and has some problems: less engine power and even worse... higher burning temperatures that aren't healthy for your motor....
Best is to check this with a scope. In fact it is fairly interesting to watch when the engine is warming up and then the control circuit with the lambda sensor kicks in to watch over the mixture.
So somebody suggested to watch with your multimeter and talked about looking for 0.2 and 0.8V as Min/Max or that the signal somehow reaches these levels from time to time (if they don't it could be that the sensor is aged....).

As you may have seen the topic's name you may already guess my thought: well, if is is really an AC+DC Signal, our modern TRMS multimeters will consider it the same and don't tell you the Min/Max, but the effective voltage which is much lower. So in this case you would actually consider your oxygen sensor to be broken, but he isn't. Your meter is just telling you a lower voltage.

As there is actually a lot of Sensors in car and a lot of them are working in a way like that you need to consider this. My U1253B has a Peak-Hold feature that brings you a lot nearer to that if I compare the signal on the scope.
Well, it is described as 1ms Peak hold...so I suppose the max. frequency to accurately detect it is 1kHz.

....so why not build id a feature to measure and show you the actual maximum of the amplitude....?!?!.... TRMS is nice to have but if that's all you can get into problems.

[Kiriakos GR]

I did read your story three times and I am still confused about accepting those as facts, which will lead in to a logical conclusion about the TRMS.

I am speculating, that what you care mostly is to verify that the sensor works correctly (it is close at their calibration standards).
a) The expected output in mV, is just an indication that the sensor is alive.
b) The pulse frequency is a key factor, and must be known so to tell  if the multimeter is capable to measure volts in the specific frequency range.
c) The Min/Max Peak  it is the faster mode to detect Peak voltages ( and so no one lies here of what this function does).

I am speculating again that you do not have any guidance from the manufacturer of this car, and you are walking blindly.   

If I am suspecting that I am dealing with pulses, the first thing that I would measure is the frequency.

One good oscilloscope which haves one significant amount of memory in it, it can record one number of samples, which they are a larger window of data compared to multimeters.

But If you do not aware of what you are looking for (from the start), I do not think that even this it would be a great help.

[Phyllomedusa]

As I said: A scope is the best way, but not everybody has one. Also high memory isn't necessary as the control of the injection isn't that slow.

To the sensor.....it's no magic, most of the oxygen sensors work on the same principle. They compare the ambient oxygen with the exhaust oxygen and a difference there produces a voltage. The oscillation just comes from the ECU and injection systems that tries to mix it the best way.
Here is the problem: The sensor will be getting lower in amplitude over the time and therefore the ECU thinks that there is less fuel in the mixture (it's lean) and shove in some extra-fuel. That increases the fuel consumption (you'll notice that) and actually isn't healthy for the exhaust system and the catalysator.

As I said the Sensor is just measuring what the ECU control and it is an AC Signal. I doubt that the multimeter is smart enough to know that you want to measure the amplitude/Positive peak.
As far as I remember the TRMS measures the effective voltage (230V on the mains doesn't mean the Max. Voltage is at 230V, but 230*SQRT(2)).

So the sensor won't be bad, but there is a high possibility that you think it's dead when you use your TRMS meter as it shows you the effective voltage.
For example:
If I use a sine of 600mVpp the meter gives me a reading of roughly 211mVrms. If you don't remember/or know that this is the effective Voltage and not the actual Maximum of the Signal.
And regarding the Sensor output....you don't need any advice from the manufacturer to read the voltage . If it just is cycling for about 200mV or stays high/low with a warm engine...... put in a new one

Maybe check this one for the principle of the sensor. Perhaps you'll understand then what I suppose.

[Phyllomedusa]

As I said: A scope is the best way, but not everybody has one. Also high memory isn't necessary as the control of the injection isn't that slow.

To the sensor.....it's no magic, most of the oxygen sensors work on the same principle. They compare the ambient oxygen with the exhaust oxygen and a difference there produces a voltage. The oscillation just comes from the ECU and injection systems that tries to mix it the best way.
Here is the problem: The sensor will be getting lower in amplitude over the time and therefore the ECU thinks that there is less fuel in the mixture (it's lean) and shove in some extra-fuel. That increases the fuel consumption (you'll notice that) and actually isn't healthy for the exhaust system and the catalysator.

As I said the Sensor is just measuring what the ECU control and it is an AC Signal. I doubt that the multimeter is smart enough to know that you want to measure the amplitude/Positive peak.
As far as I remember the TRMS measures the effective voltage (230V on the mains doesn't mean the Max. Voltage is at 230V, but 230*SQRT(2)).

So the sensor won't be bad, but there is a high possibility that you think it's dead when you use your TRMS meter as it shows you the effective voltage.
For example:
If I use a sine of 600mVpp the meter gives me a reading of roughly 211mVrms. If you don't remember/or know that this is the effective Voltage and not the actual Maximum of the Signal.
And regarding the Sensor output....you don't need any advice from the manufacturer to read the voltage . If it just is cycling for about 200mV or stays high/low with a warm engine...... put in a new one

Maybe check this one for the principle of the sensor. Perhaps you'll understand then what I suppose.

So in this case you may be able to see it as a kind of pulsing DC and check it that way. But what if the frequency would be about 40Hz or so. Then I am pretty sure you'll get the DC component only. An example for a similar signal but with higher frequency could be wheel speed sensor. It can get dirty or the gap between the sensor and the wheel could be too wide (really happens). Then the voltage of the sensor itself.
I agree with you that the peak feature is meant for voltage spikes and so on.....but well, it seems to work in this case too. And I could really appreciate the Max Peak of a signal. Could be interesting for educational purposes as well, as I think a lot of folks out there think that there is only a Max of 230V on the mains ....

[Kiriakos GR]

The above video speaks clearly that what they are interest at, is the Duty Cycle.
And this is something that the DMM can do.

The scope that this dude has looks interesting, and I made a quick search about it and  found the below video.






About True-RMS in automotive repairs, what I know from experience is that they use average measuring multimeter, but is not my field of expertise, and I do not wish to get more to it.   

[Phyllomedusa]

Yes, the Verus is nice. It's a Windows Slate Laptop with integrated scope and OBD interface..... very nice, but quite expensive, too

About the Duty Cycle: That was about the injectors. They control the injection time through the duty cycle. So that's tze response to the oxygen measurement.

I gave it a try and set up my function generator as Sinus with 600mVpp + 500mV DC at 1Hz. That could be more or less an idealised signal of the control loop (some have more a square waveform).
Took my DMM's at DC and set it up at Max/Min. You can take that route and get your +0.2/+0.8 after a while.
I hope I can weld the rest of the body tomorrow and then give it a try when the car is finally back on the ground.

So in this case it works in that way, but as I said.... on higher frequencies you just get the DC component and Effective AC voltage.
About the meters.... Most of them I see dangling around with the Fluke 87V or 187 or Fluke that advertises the 233 as automotive and both of them are TRMS....so I am not too sure about that.


...I hope you won't ever need to get your hands dirty because of your motorcycle

[Kiriakos GR]

...I hope you won't ever need to get your hands dirty because of your motorcycle

My father was one automotive electrician which died by a health issue at the age of 45.
I was 16 years old at 1985, he was planing for me to be the next who will succeed him as master electrician.
I got allot training by him since the age of 10 as helper, actually I was the washing machine for starters and alternators 

One reason that I loved electronics was the minimum dirt factor. 
Later own that I did my practice as industrial electrician for three years, I had to deal with the triple amount of dirt, than be an automotive electrician.

About my motorcycle, yes those Honda it is hard to make your hands dirty.
 

[liviu2004]

Sorry to bring an old topic up, but herewith my contribution and experience.

When I wanted to measure at the O2 sensor the voltage output, I set the DMM on DC autorange, run the engine and if the engine was ok, I could see the voltage fluctuating slowly between 0.1 and 0.9V. For older cars, a cycle of 1s is acceptable.

Keeping in mind, until the O2 had warmed up, the ECU will output a bias voltage, for circuit integrity checking. This bias voltage is somewhere at 0.4 to 0.5V on the cars I know. Once O2 starts working, it will ride indeed on this bias voltage.

So in this example, I am not directly seeing a problem with TRMS on multimeters.

[Kiriakos GR]

So in this example, I am not directly seeing a problem with TRMS on multimeters.

Unfortunately the topic starter Phyllomedusa, he is not among us any more.
Life style change or other, and I got forced deactivating his account.

Within the years, I am now better aware that TRMS will assist at measuring accuracy when the AC signal this is no pure sinusoidal.
Does it mater if  low voltage actual value will be within 30% of error ? = Non TRMS meter.   

In conclusion when you are up to measuring for a first time a specific ACV, it best using an Oscilloscope to probe it.
If the ACV frequency along voltage value, both are within the offered specifications of specific multimeter, then you may use just the multimeter as quick check.

Handheld DMM they have small internal memory so to measure accurately slow signals at low voltage.
Another detail this is the DMM measuring range, an accurate measurement starts approximately at 6% and over.

To accurately measure a single volt  the DMM should be at 6V range with 10V range as highest.
Bench-top DMM they have higher number of ranges, and they can do much more, especially at low volts and down to mV.     

[Kiriakos GR]

The 8846A this has and a single volt range, plus enormous digits resolution.
With it you may compare two identical sensors behaviour, and with increased certainty to detect the one working out of specifications.