1
\$\begingroup\$

My question is related to impedance matching when using a function generator and oscilloscope to identify a type of passive filter, i.e. low pass, high pass, etc.

If I output a function sweep from a function generator with an output impedance of 50 ohms, to the DUT (unknown filter), and the output of the DUT is connected to an oscilloscope with an input impedance of 50 ohms, do I need any additional impedance matching components?

The minimum and maximum frequencies of the sweep would be dictated by the minimum and maximum measurable frequencies of the oscilloscope. In addition BNC connections are being used to connect the equipment.

\$\endgroup\$
3
  • 1
    \$\begingroup\$ What frequencies are being used? If above 100 MHz or so, everything becomes more difficult. Also, can you use a high impedance oscilloscope input, instead of a 50 Ohm input? On the other hand, what you are describing is very similar to what a network analyzer does, and its inputs are generally all 50 ohm. \$\endgroup\$
    – Pigrew
    Commented Jul 30, 2016 at 17:52
  • \$\begingroup\$ Thank you for the reply. Would I be safe just adding a unity gain, non inverting op amp as a buffer stage between the function generator and the DUT, and between the DUT and the oscilloscope? \$\endgroup\$
    – Patrick
    Commented Jul 30, 2016 at 18:52
  • 1
    \$\begingroup\$ You can rent a network analyser for about $1,200 USD a month, good for 9KHZ to 6GHZ or so. Limit the frequency range to less than 50MHZ if possible, so a 50 ohm load is not mandatory. Use the same coax cable type for signal and measuring, so they do not add 'unknowns' to your testing. \$\endgroup\$
    – user105652
    Commented Jul 30, 2016 at 20:34

2 Answers 2

Reset to default
1
\$\begingroup\$

Filters are designed to work only with a given source impedance and a given load impedance. The most common cases are :

  • in audio frequencies : source impedance = 0 Ohms (<10 Ohms for example), load impedance = Infinity (> 10KOhms for example)

  • in radio frequencies (> 10Mhz for example) : 50 Ohms at both ends

In general, these impedances doesn't need to be very precise (say for radio frequencies : 10% error should not be a problem in most cases).

In your case, if you don't know for which impedances the filter has been designed, and furthermore, if you don't have an idea of the transfert function you should obtain, the problem can become arbitrarily difficult.

Hope your filter is somehow classical (50 Ohms, Tchebychev & Co)

\$\endgroup\$
0
\$\begingroup\$

Your instruments will be the most sensitive with matched input and output impedances, though measurement equipment generally can handle significant mismatches while still providing good results.

Impedance matching is preferred due to the fact that measurement equipment generally is the most accurate at high power levels, and that maximum power transfer happens when the source and load impedances are equal. Another difficulty is that at high frequencies, reflections of signals at impedance discontinuities become quite significant. These reflections are reduced when impedances are matched.

Another issue to contemplate is the number of measurements that you need to do. A 2-port network (one with an "input" and "output") is described with a set of four complex values. You will need to devise a measurement for the input impedance, the output impedance, the forward gain, and the reverse gain of your filter. This requires four separate measurements to be made. Each measurement should record a magnitude and phase shift.

\$\endgroup\$

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.