Differential probe & measurements

[lewisP]

When measuring a power amplifiers output using a differential probe it will display on the oscilloscope the difference of the positive and negative cycle leaving just the cross-over distortion and DC component. I'm guessing this makes it easier to bias power amplifiers. What else can you use a differential probe for?

 

[KlausST]

Hi,

Maybe there is some misunderstanding in "differential" measurement.
With a differential probe you measure the difference voltage between two points. You can´t measure the difference between "cycles".

Imagine a BTL power audio amplifier in a car. Each output uses is a simple push-pull output stage and the speaker is connected between two outputs.

The output of each stage is biased to about VCC/2. If you use a measurement related to GND you will always see the Vcc/2 in your scope reading.
But if you use a differential probe and connect the two wires to the two output stages. Then you will see the same voltage on your scope as the speaker sees.


Klaus

 

[lewisP]

With a differential probe you measure the difference voltage between two points. You can´t measure the difference between "cycles".

What circuits would you want to measure the difference voltage?

But if you use a differential probe and connect the two wires to the two output stages. Then you will see the same voltage on your scope as the speaker sees.

True if they are matched and no difference between the two power amplifiers. If there is cross over distortion between the two cycles you can measure the difference of voltage and just measure the cross over distortions peak to peak voltage. Positive cycle goes to ch1 and negative ch2, add mode and invert ch2 leaving the difference between the channels. This will only display the crossover distortion.

If you have any other examples of differential measurements or using the add/subtract mode on the oscilloscope to measure the difference.

 

[ads-ee]

Differential scope probes are regularly used in the digital realm to look at differential signaling standards (LVDS etc), besides being used to look at differential amplifier outputs. Those differential scope probes will look at the difference between two signals and reject any common mode on both signals, so you can look at a noisy LVDS signal and see a "clean" waveform (which is what the receiver would see).

 

[Audioguru]

A pure sinewave has extremely low distortion (any kind of distortion) because it has no harmonics, it just has the fundamental frequency.
You can see distortion (even harmonics, odd harmonics, clipping or crossover) if the fundamental frequency is reduced or removed by a notch filter.
Differential probes have nothing to do with it and are not used.

 

[lewisP]

What circuits are you measuring LVDS signals?

A/B amplifiers or push-pull amplifiers produce cross over distortion. The Cross-over distortion is caused by the transistors cutting off as the input drops. The differential measurement will display on the oscilloscope the cross-over region only.

 

[KlausST]

Hi,

What circuits would you want to measure the difference voltage?

For example the output voltage of a bride mode amplifier, but you didn´t understand. It is a voltage measurement, and no distortion measurement, and you can use it even if both amplifier stages are not matched.

--> and you (alomst) gave an answer to your questions: LVDS signals for example, or RS422/RS485. Quadrature encoded sine signals. Professional differential analog audio lines. BLDC motor control signals.
I have built an Xray sensor with virtual ground. There are a lot of applications.

The differential measurement will display on the oscilloscope the cross-over region only.

I don´t know how you want to do this....
If you mean to measure the difference between input and output, then you usually have the problem caused by the amplifiers gain and phase shift.

If there is cross over distortion between the two cycles you can measure the difference of voltage and just measure the cross over distortions peak to peak voltage. Positive cycle goes to ch1 and negative ch2, add mode and invert ch2 leaving the difference between the channels. This will only display the crossover distortion.

What do you mean with "cycles"?
Do you mean the two power stages? Ther is no "positive cycle to´ch1" and no "negative cycle to ch2". One stag is inverting, the other is not.
Often bridge mode amplifiers use a single power supply. this means both amplifiers produce positive voltage only.

Your description is more like displaying the "sum" (insted of "difference") of both inputs.
The result is by far not the distortion only. It includes DC offset, gain error, phase error .... Maybe on first sight it looks like distortion only, but it´s surely not.

I agree with audioguru.

Klaus

 

[Audioguru]

A properly designed class-AB audio amplifier produces no crossover distortion. A poorly designed audio amplifier uses a trimpot to adjust the idle current in the output transistors so that the crossover distortion is eliminated. A class-B amplifier always produces crossover distortion.

Why do you say "The Cross-over distortion is caused by the transistors cutting off as the input drops" which is not true? It shows that you do not know anything about crossover distortion so please read about it in Google.

Use the two differential probes on the 'scope to show the difference between the input (a pure sinewave) of an audio amplifier and its output that might have crossover distortion. You must adjust the levels and maybe offset them so that they are the same then invert one channel of the 'scope and add it to the other channel. All types of distortion will be seen.