0
\$\begingroup\$

I'm working with a sensing material and want to better understand its response. I have the theoritical elements worked out, but now want to begin testing and have no idea what type of tools/instrumentation would be required for the task.

What I'm testing:

  • Two terminal device
  • A photon interacting with this device creates a change in current (10-100nA).
  • The signal generated from this interaction could last between 10-100ps (picoseconds).

My question is what is the instrumentation required to measure at such high speed (10-100ps) with such a small change in current (10-100nA)?

Thank you for your time. Cheers

\$\endgroup\$
  • 1
    \$\begingroup\$ A really, really well designed transimpedance amplifier connected to a really low capacitance photodiode. With a minimum bandwidth of 10-100GHz (you did not actually specify a bandwidth, just a time interval within which the things of interest happen) you probably have to buy rather than build. You will also need an equally good oscilloscope (completely out of the question for most people) or additional circuitry or equipment to read the amp. \$\endgroup\$ – DKNguyen Jun 27 '19 at 1:09
  • 2
    \$\begingroup\$ You might want to explain what the device actually is, what the current nominally is upon which the change you indicated occurs, the interval between events, if you only need to measure the peak itself or if an integral (or average) might be acceptable, required accuracy and precision, etc. \$\endgroup\$ – jonk Jun 27 '19 at 1:32
1
\$\begingroup\$

We'll assume a bandwidth of 10 GHz. And a system noise resistance of 62 ohms.

10Ghz requires a (single pole) timeconstant of 16 picoseconds. For each 16 picoseconds of settling time, you will improve the amplitude measurement by 9dB (1 neper).

We will assume your pulse flat-top has long enough duration to provide the amplitude resolution.

The total integrated noise of 10GHz bandwidth, using 62 ohms and its exactly 1nanoVolt RMS noise density, will be

Vnoise_total = sqrt(10GHz) * 1nanoVolt

Vnoise_total = 100,000 * 1 nanovolt = 1E+5 * 1E-9 = 1E-4 volts RMS

Vnoise_total = 100 microVolts RMS

Can you live with that?

| improve this answer | |
\$\endgroup\$

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.