5
\$\begingroup\$

I just bought my first oscilloscope and have been tinkering with the measurement of different signals just to learn the ropes and familiarize myself with the scope.

One of the first things I noticed was the ambient 60Hz "noise" that is picked up when a probe is connected to the scope. It is obviously originating from the electrical lines in the building, and I can verify this by holding the probe in my hand and placing my other hand nearer or farther from an electrically powered object or cable - the amplitude of the noise increases or decreases respectively.

Then I probed the mains voltage to see how clean the incoming power is. Next, I viewed the aforementioned 'noise' signal on one channel and the mains power on the other channel.

What I saw was surprising and I don't understand the reason: The mains voltage shows a clean 60Hz sine wave. The 'noise' signal is a jagged sine wave, still obviously 60Hz but with a lot of static. However, the two signals are out of phase, by what appears to be approximately 90 degrees if my measurement is correct. I would have expected the phase of the mains power to match the phase of the noise. I was concerned my scope had some sort of delay between sampling the two channels, so I switched the inputs and still saw the exact same phase difference. Attached is a screen capture showing the two signals (red = mains, yellow = noise).

A little more research shows that transformers can cause phase shift between the input and output sides. (TI pdf) Is this what I'm observing? or is something else in play?

Screenshot from oscilloscope

If I'm missing a basic point, please help me understand it or at least guide me in the right direction.

\$\endgroup\$
  • 2
    \$\begingroup\$ I personally don't trust these scopes enough to probe mains with them and I suggest that you do not either. The AC trigger line is in phase with the AC signal coming into the PSU. Here are the screenshots for me: first,second The phase difference does seem to be 90 degrees to me. \$\endgroup\$ – AndrejaKo Dec 8 '13 at 20:27
  • 1
    \$\begingroup\$ I agree with @AndrejaKo, probing mains with an oscilloscope can be very tricky, even for the experienced EE. Better to use a transformer. \$\endgroup\$ – jippie Dec 8 '13 at 20:49
  • 1
    \$\begingroup\$ @RyanGriggs on my scope this mode is called 'X-DEFL' (deflection), sometimes it is called X-Y mode and probably other names are used too. Instead of using a time base for the X-deflection, you use channel B for X-deflection and channel A for the regular Y-deflection. It is very handy to check phase difference or frequency difference between the two channels. If the circle slowly changes shape to a line the frequencies differ. Check Lissajous on Wikipedia. This is a nice image: en.wikipedia.org/wiki/File:Circular_Lissajous.gif \$\endgroup\$ – jippie Dec 9 '13 at 7:16
  • 1
    \$\begingroup\$ @Ryan Griggs You got the background from jippie and I'll tell you how to get it in practice: Press the display button and after that the H3 button. Above H3, the XY mode setting should change from OFF to ON. Same procedure turns it off. \$\endgroup\$ – AndrejaKo Dec 9 '13 at 7:19
  • 1
    \$\begingroup\$ @Ryan Griggs About the trust side: A bit of both, since my scope had some measurement problem, but the main issue is safety. In theory the scope says it's CAT II rated for 400 V and that in theory might be OK for connecting it to power plug. Due to problems I already mentioned, I had to take it apart completely few times and after seeing internals I wouldn't trust it with mains power at all. Also I don't know what probes you're using, but if they're the one that come with the scope... Well they aren't the safest either. If you're going to probe mains, do it from behind a small transformer. \$\endgroup\$ – AndrejaKo Dec 9 '13 at 7:25
4
\$\begingroup\$

Looks like an almost perfect 90 degrees phase shift. The difference between:

  • directly coupled therefore no phase shift;
  • forming an RC phase shift (high pass) filter, coupled through the capacity between the mains wires and your body, and high (but not infinite!) impedance of your scope.
\$\endgroup\$
  • \$\begingroup\$ If the filter is RC, then it depends on C component. If I'm waving my probe tip next to a power cable, the phase shift should change, because the capacitance changes as well. I don't think that I see that effect right now. \$\endgroup\$ – AndrejaKo Dec 8 '13 at 21:05
  • 1
    \$\begingroup\$ @AndrejaKo If the capacitance changed from 10pF (270Mohms) to 1pF (2.7Gohms) and the scope input were 10Mohm, the phase angle would change from 87.9deg to 89.8deg. I'm+1 totally convinced with this explanation +1 \$\endgroup\$ – Andy aka Dec 8 '13 at 21:28
  • \$\begingroup\$ @jippie - thanks for this info! I was not aware that RC circuits caused phase shift. Time for some more reading! \$\endgroup\$ – Ryan Griggs Dec 9 '13 at 3:18
  • 1
    \$\begingroup\$ @Andyaka - that explains why the phase shift appears to be constant... Can you suggest other configurations of this test where the phase shift could be altered enough to observe it on the scope? \$\endgroup\$ – Ryan Griggs Dec 9 '13 at 3:20
  • 1
    \$\begingroup\$ @RyanGriggs In theory you should be able to move the scope probe closer to the mains conductor. By moving closer and closer the capacity between conductor and scope probe increases and the phase difference decreases. In practice however, with the size of the scope probe (= capacitor plate size) you need to be so extremely close to the conductor to have the capacity increased enough to see the effect on scope, that this is not viable. \$\endgroup\$ – jippie Dec 9 '13 at 8:33

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.