3
\$\begingroup\$

I understand about input impedance and capacitance... My question relates to the cable: What is the characteristic impedance of the cable connecting the probe to the scope? It must be terminated in a matched impedance to avoid reflections but the input impedance is typically 1 M ohm in parallel with a few pF - that can't be the characteristic impedance. So what gives?

I have noticed that the cable has a centre conductor that is very small in diameter - is that to create the correct Z0?

I've been asking this question of professional EEs for many years and no one knows...

\$\endgroup\$
  • \$\begingroup\$ Keep in mind that many probes are only a meter or so long, so transmission line effects are negligible for any frequency whose wavelength is more than 10x that value -- 10 m corresponds to 30 MHz. This is why passive 1x probes have very limited bandwidth specs. \$\endgroup\$ – Dave Tweed Jan 16 '15 at 12:22
  • \$\begingroup\$ OK, so a 10x probe, how does that not have dreadful reflections and stuff? \$\endgroup\$ – Hugh Stevenson Jan 17 '15 at 10:38
  • \$\begingroup\$ That's addressed in the material linked in the other answers. Also, Dave Jones addressed this topic in his EEVblog #453 - Mysteries of x1 Oscilloscope Probes Revealed. \$\endgroup\$ – Dave Tweed Jan 17 '15 at 12:30
8
\$\begingroup\$

Probe cable is lossy coax.

Achieving a matched condition with an oscilloscope probe is virtually impossible because the source impedance of the circuit under test is unknown and generally different from the scope's 1MΩ or 50Ω input impedance.

On top of that the input impedance of the oscilloscope has a reactive component as do most circuits under test, which makes it impossible to achieve impedance matching over the full BW of the scope (eg 100MHz). To dampen the catastrophic frequency response as a result of the reflection on the probe cable, the latter is made of lossy coax. If you measure the core resistance between the tip and the BNC connector with a multimeter you can observe this.

You may want to read "Tektronix ABC's of Probes" primer and "The Secret World of Probes" from Doug Ford. Both are excellent primers on a correct understanding of how probes work.

\$\endgroup\$
  • \$\begingroup\$ Sounds Plausible. I'll have a read of the reference texts that you give. The source impedance is not important for reflections on a transmission line since, if there are no reflections at the receiving end then, there are no waves travelling towards the source... :) \$\endgroup\$ – Hugh Stevenson Jan 17 '15 at 10:41
  • \$\begingroup\$ Fantastic! The mystery is resolved. Also found out who invented it: \$\endgroup\$ – Hugh Stevenson Jan 17 '15 at 10:57
3
\$\begingroup\$

It's probably 50 ohms. It's basically just coax cable.

One reason for choosing 50 ohms over any other impedance is that some testgear has switchable 50 ohm terminations on its inputs, to improve accuracy for HF measurements, and in that case a 1x probe could be terminated correctly.

\$\endgroup\$
  • \$\begingroup\$ Sorry, not possible. It is terminated in 1 Mohm so would suffer dreadful reflections and ringing on fast edges. \$\endgroup\$ – Hugh Stevenson Jan 17 '15 at 10:42
2
\$\begingroup\$

I hardly ever use a x1 probe. But grabbing one off the wire rack, I measured ~330 ohm of series resistance and ~100 pf parallel capacitance. (The 100 pF number is a bit squishy.) The 100 pf is a killer at high frequencies. And this probe has a 3dB point near 7 MHz. (about where 100 pf is 300 ohms) The lossy coax is important for operation as a x10 probe... I've read about the details, but to be honest I have forgotten them. Jim Williams had a nice discussion of probes... I think in this app note. And this is a classic... but not too much about x1 probes.

\$\endgroup\$
  • \$\begingroup\$ Just a quick comment to thank you for these links, George, and highlight them to others who might arrive here on search. I've found these extremely helpful in lifting the fog on understanding the challenges of probes and scope measurement. The second link is broken, but I found this one that works. If it breaks again, the title is Oscilloscope Probe Circuits by Joe Weber (Tektronix). I found it by searching on "tekprobecircuits pdf". \$\endgroup\$ – scanny Feb 24 '16 at 4:36

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.