0
\$\begingroup\$

I discovered an "funny" thing today.

I had used a GW instek probe GTP-150B-4 150Mhz in 10x mode and documented the result.

Today I noticed in the testing department that what I've measured doesn't correspond entirely to what they see. The pic below is what I got with the GW-probe, where the red lines is what the peaks are with the probe they use, which is a Tektronic P3010 100MHz 10X.

enter image description here

I tested the exact same probe on my own scope which is the same model and I got the same result. Firstly I thought it might would have something to do with the input capacitance, but according to the datasheets they have the same nominal input capacitance.

Before I tested I also ran the probe against the scope's calibration signal.

Any more ideas what this may depend on?

Additional:

In the pic below the same settings is used, just different probes. As can be seen the high frequency spikes is a lot higher with the the tek-P3010 probe. I've yet not found an logical explanation.

With GW-probe 150 MHz 10x With GW-probe With Tek-P3010 100 MHz 10x With Tek-P3010

\$\endgroup\$
6
  • 1
    \$\begingroup\$ I have no idea what the difference now is between the measurement with one probe and another, maybe show pics of both and explain the difference you see? Also they have different bandwidths, that can already explain a lot \$\endgroup\$
    – PlasmaHH
    Apr 4, 2018 at 9:14
  • \$\begingroup\$ @PlasmaHH basically what I have explained, however I've added 2 additionally pics now. As can be seen the probe with highest bandwidth gives the lowest spikes. So I can't see that the bandwidth wouldn't be a logical reason, unless it would be the other way around. \$\endgroup\$
    – user103776
    Apr 4, 2018 at 10:07
  • 2
    \$\begingroup\$ Are they using a longer earth lead on it? What happens if they use an earth clip immediately next to the probe tip? Those spikes look like inductive pickup from a big earth loop. \$\endgroup\$
    – user_1818839
    Apr 4, 2018 at 10:27
  • \$\begingroup\$ The question was not the source of the spikes but Brian might have a point here. Are you positive that both probes are in good condition ? A probe with broken ground can do well while compensating on a signal that have that same ground already but fail in some insidious way on a bench with a weak ground path. Do you have a third probe? \$\endgroup\$
    – Dorian
    Apr 4, 2018 at 11:27
  • \$\begingroup\$ @BrianDrummond I will check that. \$\endgroup\$
    – user103776
    Apr 5, 2018 at 15:09

2 Answers 2

Reset to default
1
\$\begingroup\$

Are you testing in different rooms/environments, too? I'd also make sure that you are using the same grounding setup (lead length, etc.). It looks to me like a shielding/environmental noise issue.

\$\endgroup\$
2
  • 2
    \$\begingroup\$ This should be a comment on the original post. I don't think it qualifies as an answer. \$\endgroup\$
    – Transistor
    Apr 5, 2018 at 18:07
  • \$\begingroup\$ Sorry! I'm still fairly new to the SE system \$\endgroup\$
    – Daniel Bogdanoff - Keysight
    Apr 5, 2018 at 18:17
0
\$\begingroup\$

Looks like the oscilloscope input capacitance resonating with something in the circuit. Close to resonance a it makes a big difference. Have you tried to slightly under or overcompensate the probe ? Also as Brian pointed a broken probe without ground might be the source.

\$\endgroup\$
2
  • \$\begingroup\$ This has nothing to do with scope probe compensation. \$\endgroup\$
    – user_1818839
    Apr 4, 2018 at 10:34
  • \$\begingroup\$ Only the input capacitance is not the same. For Gtp-150 8.5 to 18.5 pF , for P3010 from10 to 15 pF , in a resonant circuit it can make a difference. \$\endgroup\$
    – Dorian
    Apr 4, 2018 at 11:13

Your Answer

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

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