-3
\$\begingroup\$

I'm having problems displaying two channels simultaneously on an oscilloscope when both probes are connected to measure voltages "across" the components.

To demonstrate the problem two resistors are connected in series and fed by a function generator. So the aim is to display the voltages "across" both of these resistors.

Okay so I first connect only one probe to one of the resistor's terminals as in the photo below(this photo is a bit blurry sorry):

enter image description here

And in this case what I measure in oscilloscope's CH1 is in the below photo: enter image description here

So far so good. But now I hook up the other oscilloscope's probe to the other resistor as in the photo below: enter image description here

But in this case what I measure in oscilloscope's CH1 is totally different as in the photo below: enter image description here

My question is:

1-) Is my setup wrong? Is this a wrong way of displaying voltages across the components in an oscilloscope? (Is that because the probes have the same ground?) Or the oscilloscope is defect?

2-)How can I work around this issue? I was planning to see the phase difference in an AC circuit between a capacitor and a resistor for instance. For that I need voltages "across" their terminals not relative to circuit GND.

Note: I'm using a "Velleman HPG1 1MHz Pocket Function Generator" with 100Hz sine wave, and a Siglent 100MHz oscilloscope.

EDIT: Here is the basic schematics:

Each probe is connected in a way that their crocodiles and probe tips wired across resistor terminals as in my photos.

enter image description here

\$\endgroup\$
  • 2
    \$\begingroup\$ The first thing I saw when the page for this question loaded was a blurry picture that was also way bigger than it needed to be. Screw this, and -1 for the gross disrespect and sloppiness. \$\endgroup\$ – Olin Lathrop Dec 28 '15 at 11:55
5
\$\begingroup\$

Look at your picture here: -

enter image description here

Because the crocodile clip connections on most common o-scopes are grounded to each other at the o-scope you have effectively shorted out R1 (indicated by the red line).

\$\endgroup\$
3
\$\begingroup\$

You have your oscillscope probe grounds connected to two different nodes. With very few exceptions, all the grounds on every scope channel are shared. They need to be connected to the same potential. Right now you are effectively short circuiting across R2 and it is having no effect.

If you place Probe 1 where you have it, but move its GND to the circuit GND, you can then use the math function of your scope to get what you need. Plot Probe 2, and Probe 1 - Probe 2 and that should give you your points of interest from a purely voltage point of view (no phasing/etc).

\$\endgroup\$
  • \$\begingroup\$ Please use an isolation transformer on the DUT in this case. Using a cheater plug is a good way to get somebody electrocuted. \$\endgroup\$ – Turbo J Dec 28 '15 at 7:56
  • \$\begingroup\$ Updated answer to call out an isolation transformer in bold. \$\endgroup\$ – Krunal Desai Dec 28 '15 at 7:57
  • \$\begingroup\$ My scope is connected to a plug which doesn't have earth already. I don't think it is the problem. \$\endgroup\$ – user16307 Dec 28 '15 at 7:58
  • 1
    \$\begingroup\$ I hope that's intentional, but regardless of Earth ground connectivity, the return / ground of each channel is shared. If you connect the return of CH1 to a different potential than the return of CH2, that is not good and you risk instrument and/or circuit/DUT damage. \$\endgroup\$ – Krunal Desai Dec 28 '15 at 7:59
  • \$\begingroup\$ So my set up is wrong because the probes the same grounds? \$\endgroup\$ – user16307 Dec 28 '15 at 8:00
1
\$\begingroup\$

Here's the basic setup for a simple RC lowpass filter:

schematic

simulate this circuit – Schematic created using CircuitLab

Note: you can use the built-in circuitlab simulator to run Time Domain analysis (which simulates what you will see on oscilloscope) or Frequency Domain analysis (a Bode plot). Follow the link under the picture caption.

Both of the scope probes use the same ground reference. You don't need to use a differential probe to see the phase response of the filter.

Time Domain Response of RC filter

\$\endgroup\$
  • \$\begingroup\$ Thanks, just curious one question. Imagine you have a circuit with hundred components. And you wanna measure two component's voltages across their own terminals like I tried to do. What I understand from your words that if one uses two probes, those probes' grounds should be connected to each other which is also connected to the circuit's GND. Is that right? So my setup was wrong? \$\endgroup\$ – user16307 Dec 28 '15 at 10:20
  • \$\begingroup\$ Yes your setup was wrong. Note that many scopes can compute A-B (or "invert channel 2, A+B" on analog scopes) to show the voltage across the resistor in this example. \$\endgroup\$ – Brian Drummond Dec 28 '15 at 10:51
  • \$\begingroup\$ A-B only works in DC. In AC the voltages from R C L are not linearly added. They are phasors. So it seems it s not possible to see two voltages "across" any random components. \$\endgroup\$ – user16307 Dec 28 '15 at 11:58
  • 1
    \$\begingroup\$ Not true. The A-B mode on an oscilloscope will show the voltage between the 2 probes for the full bandwidth of the oscilloscope. This mode is only limited by the common-mode rejection capability of the oscilloscope which generally is good but not as good as a true differential measurement using a dedicated differential probe. \$\endgroup\$ – Barry Dec 28 '15 at 13:42
  • 1
    \$\begingroup\$ If your scope has two channels and an external trigger input, you may be able to display CH1-CH2 as one channel and display the external trigger as a second channel. Or, if there's only a momentary "view trigger" function, you can at least set up the external trigger to a known reference position (0 degrees at center graticule) and see the phase difference of CH1-CH2 compared to external trigger (filter input signal). \$\endgroup\$ – MarkU Dec 28 '15 at 22:40

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.