I'm having trouble getting a stable trigger on my Rigol DS-1202 oscilloscope for a complex waveform. The waveform is composed of a 25.5kHz carrier square wave modulated by a 200Hz square wave, both with a 50% duty cycle. Trying to measure the RMS voltage across a resistor with this signal flowing through it, I've connected channel 1 and 2 across the resistor.

Here are the specifics:

  • Carrier frequency: 25.5kHz square wave
  • Modulation frequency: 200Hz square wave
  • Duty cycle: 50% for both carrier and modulation Modulation type: Amplitude Modulation (AM)

I've tried various trigger settings, but the waveform keeps moving across the screen and I'm unable to get a stable display. I'm aiming to measure the RMS value of the difference between two signals (A - B).

Could anyone provide detailed steps or tips on how to set up the trigger correctly for this type of waveform? Any advice on ensuring an accurate RMS measurement once the waveform is stable would also be appreciated.

What I've Tried:

  • Setting the trigger source to both Channel 1 and Channel 2
  • Adjusting the trigger level and slope
  • Using different trigger modes (Auto, Normal, Single)
  • Changing the horizontal time base settings

Despite these attempts, I can't seem to stabilize the waveform for a reliable RMS measurement.

images of my oscilloscope setup

  • 1
    \$\begingroup\$ You haven't even mentioned how the modulating waveform modulates the carrier. Is it AM, FM, phase, something else? \$\endgroup\$
    – Justme
    Commented Jun 7 at 16:29
  • \$\begingroup\$ Added in the question that it is AM modulated \$\endgroup\$ Commented Jun 7 at 16:33
  • \$\begingroup\$ Are you aware of the trigger holdoff setting? \$\endgroup\$
    – Hearth
    Commented Jun 7 at 16:36
  • 1
    \$\begingroup\$ I was not aware of this, but just looked it up. So correct me if I am wrong, but I can set a holdoff time to between half and 1x period of the 200Hz signal, and the oscilloscope will wait that long before triggering? If this is done correctly, it should ignore the 25.5kHz inside the burst, then only trigger again when the next burst begins? \$\endgroup\$ Commented Jun 7 at 16:44

2 Answers 2


Couple of suggestions:

  1. Single shot.
    Do you really have to view the waveform in real-time? With the method of triggering you are now using, each screen update will be slghtly different to the previous, since the 200Hz may not not synchronised to 25.5kHz the carrier. This results in the problems you are seeing.

  2. Trigger from the 200Hz signal, if you have access to it.

  3. Adjust trigger hold-off time to get a better synchronisation with the 200Hz component.

But why do you need a stable image? On a good scope, the accuracy of RMS measurements should not require stable images, provided that:
(a) the sample rate is sufficient to capture the highest frequency of the signal, to the required accuracy. For a 25kHz square-wave carrier, I would suggest a sampling frequency of at least 10 times that, so 250kHz sampling rate.
(b) the measurement system uses the entire data record, not just what is shown on the screen.

  • \$\begingroup\$ Using edge triggering currently, and the waveform slides around the screen, since there are multiple edges in the 25.5kHz signal. This causes the RMS voltage to also jump around. Triggering a separate 200Hz signal would be a good solution, but my oscilloscope only has two channels which are being used to measure the voltage across the resistor. I will try increasing the trigger holdoff time so that it picks up on the burst edge rather than the carrier edge. Thanks! \$\endgroup\$ Commented Jun 8 at 5:18
  • \$\begingroup\$ Trying doing single shots, and compare the RMS readings for those. Try timebase with just 2 cycles of the 200Hz on the screen to start with, take 10 single-shots; for each shot record the RMS & average. Then change the timebase to fit more of the 200Hz cycles, and repeat. \$\endgroup\$ Commented Jun 8 at 5:24
  • \$\begingroup\$ but my oscilloscope only has two channels which are being used to measure the voltage across the resistor. I don't understand. From your question, it seemed that the AM modulated signal is the voltage across the resistor, so why do you need two channels? \$\endgroup\$ Commented Jun 8 at 5:25
  • \$\begingroup\$ Thanks Fabio, I'll try the single shot approach and see if it's a tight enough spread of readings. I am using one channel probing each side of the resistor, then the RMS of A-B will give me the average voltage drop across the resistor \$\endgroup\$ Commented Jun 9 at 22:10

From the posted DSO pictures I assumme amplitude modulation (AM):

Then you have to get the modulation signal out of it - here e.g. a schematic for a simple AM demodulator, that produces only a triangle signal, not a rectangle, but its output on the right gives a rising slope and a falling slope, when the modulating signal goes high or low (more or less exact time, expect jitter).


simulate this circuit – Schematic created using CircuitLab

Perhaps another diode is required? Amplifier of the signal?


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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