2
\$\begingroup\$

As I was trying to validate our pcb design, I measured the impedance matching on the digital transmission lines, and got stuck with the oscilloscope not showing what I expected. We are using the PicoScope 4227, but I suspect that the strange measurement is not directly related to the specific oscilloscope but more to general on how to measure digital signals.

My setup is as follow: I measure the output-pin of an FPGA with both oscilloscope input A and B.

Input A (blue) is measured using the oscilloscope probe set to x10.

Input B (red) is measured using a 50-ohm coxial cable terminated at the scope input with 50-ohm.

As you can see, Input B seems correct, no reflection ect.

But Input A (the probe) has some overshoot, and ringing.

The probe should be calibrated and valid to 250MHz, so I don't get was is wrong. ?

I suspect that the small probe-clip adds some inductance, and therefore we see the ringing. But I really don't know.

oscilloscope screenshot

EDIT After reading your answers, I walked got into our lab storage and started looking, and found a probe GND clip like the one showed in the answer below.

When I use that clip, I get the output below. Which is really a huge improvement.

enter image description here

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

Yep, this is typical when your ground connection is not suitable for the frequencies you're working with. You may want to use a ground spring to reduce both the inductance-related peak and the noise:

enter image description here

\$\endgroup\$
  • 1
    \$\begingroup\$ In have tried moving the ground clip even further away from the signal measured with the probe and can see that the ringing gets worse (as expected). \$\endgroup\$ – JakobJ Apr 15 '16 at 8:58
  • 1
    \$\begingroup\$ You where completely right, the GND clip was the issue. In have edit my answer with new picture. \$\endgroup\$ – JakobJ Apr 15 '16 at 9:12
  • \$\begingroup\$ @JakobJ Hey, I'm glad you have found that GND clip and it helped! \$\endgroup\$ – Dmitry Grigoryev Apr 15 '16 at 12:49
2
\$\begingroup\$

enter image description here

I suspect that blue is a probe problem - maybe a slightly displaced ground connection forming a small inductive loop that resonates with the probe capacitance producing ringing when hit with a fast transient. Red shows some small signs of not being perfectly terminated (nearly gets there first time) but is fine for data.

\$\endgroup\$
  • \$\begingroup\$ Have you tried 'tuning', or compensating the probe? There should be a trimmer in the housing for that. \$\endgroup\$ – F. Bloggs Apr 15 '16 at 8:20
  • 1
    \$\begingroup\$ @F.Bloggs yes I always take care of this but maybe you should address this comment to the OP or make your own answer. \$\endgroup\$ – Andy aka Apr 15 '16 at 8:28
  • \$\begingroup\$ You where completely correct. \$\endgroup\$ – JakobJ Apr 15 '16 at 9:14
  • \$\begingroup\$ @Andy - sorry, yes, that was directed at the OP. I see I posted my comment in the wrong place! Oops. \$\endgroup\$ – F. Bloggs Apr 15 '16 at 12:57

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.