Adding Oscope Channels to test for phase shift

[PrescottDan]

When using the ADD mode on the Oscope to ADD 2 channels to test for phase shift of parallel circuits or identical stages or circuits

Inphase test:
Both input signals have the same input frequency and voltage. Both outputs should have no phase shift between the two output channels

Out of phase test:
One input is in-phase and the second input is 180 degrees out of phase. Both outputs should have a straight horizontal line on the Oscope, if you anything other their is a phase difference between the 2 outputs of the circuits or stages

What else can a tech use the ADD mode on the Oscope for? for what kind of circuits or applications

 

[betwixt]

Probably the most common use is with one channel inverted so the probes work differentially. You see the voltage from one probe to the other instead of from each probe to ground.

For phase testing, the better method is to use X/Y mode, where one input moves the dot up/down and the other moves it left/right. The linearity of the signal is seen as the straightness of the line and the phase difference as it's angle.

Brian.

 

[PrescottDan]

Probably the most common use is with one channel inverted so the probes work differentially. You see the voltage from one probe to the other instead of from each probe to ground.

What kind of circuits or applications or signals would you want to use the Oscope like this? to measure what?

 

[D.A.(Tony)Stewart]

• What kind of circuits or applications or signals would you want to use the Oscope like this? to measure what?

• A-B or Ch1&Ch2invert allows you to make Differential measurements on differential signals like RS485 to examine common mode noise.
• It also permits easy current measurements on high side series 75mV current shunts.
• More frequently in high speed transients the ground inductance causes ringing and using two probes in differential mode with short ground leads improves the signal capture accuracy, when the ground has local noise transients as well or is isolated from ground.
  
  Many applications use this measurement method.

 

[PrescottDan]

Many applications use this measurement method.

can you please list some more? so I can learn from

•A-B or Ch1&Ch2invert allows you to make Differential measurements on differential signals like RS485 to examine common mode noise.

So the common mode noise voltage should be close to zero or in micro volts or millivolts between the differential signals?

•It also permits easy current measurements on high side series 75mV current shunts.

Do I need to use current probes on my oscope for this?

What is a current shunt?

I'm not sure how to do this

 

[betwixt]

No, no, no and more no.

Please read what we wrote. Differential VOLTAGES are MEASURED, it has nothing to do with current or the quirks of RS485. You can calculate the current if necessary using Ohms law but oscilloscopes only measure voltage.

You can use differential measurement when there is no safe or suitable ground to measure against. Each probe has an input pin and a ground clip, in some circumstances the ground clip cannot be connected to the circuit. It may be that the circuits own ground has a high voltage on it or it might be that connecting the clip will make the probe influence the operation of the circuit. In differential mode you do not (normally) connect the ground, instead you measure from input pin to input pin and use the 'add & invert' function so the oscilloscope does the math to show the voltage across the two inputs.

Brian.

 

[PrescottDan]

instead you measure from input pin to input pin and use the 'add & invert' function so the oscilloscope does the math to show the voltage across the two inputs.

Yes I understand, but can you give some examples of when I would use this? for what kind of circuits or applications

To measure the differential voltage across two inputs of what kind of stage or circuit or application?

 

[betwixt]

Example - be careful if you try this, dangerous voltages may be present!

You want to check the voltage across a component on the AC side of an SMPS. You cannot connect the oscilloscope ground because it would short out the incoming power to the wall socket earth pin. You place one probe at each end of the component, (using a voltage dividing network if appropriate to keep it within safe measurement range). You do not connect the probe ground clips, in fact they should be removed or taped back to prevent risk of them touching anything. Adding and inverting the probe signals will show you the voltage across the component.

Another example -
You want to measure the voltage across two data lines. Normally when using a scope you would connect the probe ground clip to one line and the probe pin to the other. However, connecting the ground will disrupt the signal on that wire. Measuring with two probes and no ground avoids the problem.


Brian.

 

[PrescottDan]

You want to measure the voltage across two data lines. Normally when using a scope you would connect the probe ground clip to one line and the probe pin to the other. However, connecting the ground will disrupt the signal on that wire. Measuring with two probes and no ground avoids the problem.

Why would the ground disrupt the signal?

What kind of data lines are you talking about?

Thanks for the help and examples

- - - Updated - - -

I found other examples

Example#1:

Channel1 is the GOOD working reference board
Channel2 is the board under test

You ADD Or subtract both channels to measure the "difference" of the 2 waveforms to view and measure the "comparsion"

Example#2

Channel1 is an AC waveform , set Oscope to AC coupling
Channel2 is a DC voltage, set Oscope to DC coupling

Set Oscope to ADD

Now the Oscope will display the AC waveform on channel#1 and the DC offset on channel#2

This is how to measure the DC voltage difference from ground reference how much the AC voltage is from ground to the DC offset voltage the AC voltage is riding on

Channel#2 is your ground or DC offset voltage
Channel#1 is an AC or Logic waveform

When you ADD them together you can measure the difference between the DC voltage or DC offset voltage to the ground reference and also the AC waveform or Logic signal waveform difference from the DC offset voltage to the ground reference

 

[D.A.(Tony)Stewart]

You should never ground a differential data driver with a scope ground. surely you know by now what a differential signal looks like.

 

[PrescottDan]

You should never ground a differential data driver with a scope ground.

I guess a differential data driver has a different ground potential to the Oscope ground?

 

[betwixt]

Differential data has no ground, both wires carry signal. If you ground one of them the signal on that wire is lost and that would almost certainly prevent devices receiving the signals working properly.

Referring to your examples:

#1 is only valid if both boards are working in exact synchronism. If they are not, there would probably be differences between the voltages anyway.

#2 I'm not sure what you mean. When you measure differentially then invert one channel and add it to the other, the result is a single voltage. If you only want to look at the AC part of a waveform the usual method is to use AC coupling at the scope input.

Brian.

 

[PrescottDan]

#1 is only valid if both boards are working in exact synchronism. If they are not, there would probably be differences between the voltages anyway.

Why would there be a difference in voltages? I would think they would be different in time because they aren't synched together

How would I take two circuit boards and synch them together to do a "time comparison" between a good working board compared to a board under test?

I'm not sure how to synch a good circuit board with another board under test so i'm able to the timing signals of both boards and testing the comparisons between the inputs vs the outputs

- - - Updated - - -

#2 I'm not sure what you mean. When you measure differentially then invert one channel and add it to the other, the result is a single voltage. If you only want to look at the AC part of a waveform the usual method is to use AC coupling at the scope input.

If you have waveforms that have a DC offset voltage and the AC waveform is riding on top of a DC offset voltage , you separate the AC waveform and the DC offset voltage on 2 separate O scope channels

How can you put only the AC waveform on Oscope Channel#1 and the DC offset on Oscope channel#2?

ADD both O-scope channels and then you invert channel#2 it will cancel out the AC waveform on channel#2 with channel#1 and leave u just with the DC offset voltage on channel#2

- - - Updated - - -

When you measure the voltage difference from the zero reference to the DC offset voltage or a DC voltage, is this an amplitude measurement?

What other kinds of signals, waveforms, Logic TTL/cmos or DC offsets voltage would you want to do Comparison measurements?

What kind of circuits or applications would you want to compare or do a comparison to multiple waveforms, , Logic TTL/cmos or DC offsets voltage?

When would a tech want to compare or do a comparison to multiple Logic TTL/CMOS waveforms?

example. to make sure the Logic signals are in synch for what kind of circuits or applications?

Logic signals are mostly synch in XXX?

- - - Updated - - -

Example, a tech would want to "compare" or do a "comparison measurement" the in phase and out of phase of waveforms for what kind of stages, circuits or applications?

- - - Updated - - -

I'm not sure when a tech would want to "measure the voltage difference" of 2 inputs or 2 outputs

The only thing i can think of is when you want to compare a good board to a board under test to see of the inputs and outputs are the same voltages or there is a voltage difference

I have measure the voltage difference when measure the voltage that is across a component like a resistor

But I haven't measured the voltage difference of an output or an input of a circuit, I'm not sure when a tech would want to do this

 

[betwixt]

Consider this scenario then:

You have two boards, each has an oscillator on it, they run at the same frequency, they have the same amplitude and wave shape.

Q. if you place one probe at each oscillator output, invert one of the channels and add it to the other, what do you see? Think carefully before answering.

How would I take two circuit boards and synch them together to do a "time comparison" between a good working board compared to a board under test?

The point I was making is that in virtually all circumstances you can't do it. Your two boards would have to be identical in operation right down to the minutest detail and be doing exactly the same operation simultaneously for there to be no difference between them.

Brian.

 

[D.A.(Tony)Stewart]

Right... Thats what logic analyzers are for....

 

[betwixt]

If it's a logic signal!

Brian.

 

[PrescottDan]

You have two boards, each has an oscillator on it, they run at the same frequency, they have the same amplitude and wave shape.

Q. if you place one probe at each oscillator output, invert one of the channels and add it to the other, what do you see? Think carefully before answering.

There will be 2 waveforms with a phase shift/time difference between them because both boards aren't synced

How would you sync them?

Q. if you place one probe at each oscillator output, invert one of the channels and add it to the other, what do you see? Think carefully before answering.

If both boards are both the same with no component tolerance differences, there should be a straight horizontal line , but in the real world it will never be a straight line, you will measure the difference