Monday, June 2, 2014

Use an oscilloscope to collect optical spectral data

I'm using my new Tek oscilloscope to collect data from a DIY spectrometer.  The scope is in X-Y mode with infinite persistence.  The X axis is controlled by the rotational position of the diffraction grating in the spectrometer, and the Y axis is controlled by the photomultiplier tube, which is detecting the spectra.  In this video, I discuss the scope setup, and in future videos, I will talk more about the spectrometer, and how to use it.


8 comments:

  1. Nice - I've just recently bought a PMT module myself and was planning on making a spectrophotometer too. I'd really like to see how you arranged the optical side.

    Are you using any output circuitry on the PMT or is the anode connected directly to the scope? I briefly saw close to a 1kV out of my Hamamatsu module when I accidentally turned the lights on when it was at full gain (something that I believe can permanently damage the tube, though mine seems to have fortunately survived) - your output voltage seems much lower.

    ReplyDelete
  2. I'd be interested to see a bunch of different light sources, particularly with regards to the IR and UV portions of the spectrum. Also, have you tried calibrating it with a monochromatic light source such as a laser diode?

    ReplyDelete
  3. Damien, the PMT output (anode) acts very much like a current source, so an easy way to convert the output current into a voltage is to put a resistor between the anode and system ground. Since the cathode is at a negative potential (and emitting electrons), the anode will be pulled negative when said electrons impinge on it, thus the output voltage will be negative. I used a 100K resistor, so that 100uA of output current will equal 10V. The scope has a high input impedance of at least 1M, so it can read this 100K source directly. If you have nothing connected the anode (eg near infinite resistance), even tiny nanoamp currents could make high voltages appear there. You must have an output load.

    ReplyDelete
  4. Why did you use a PMT instead of a solid state detector? Do PMT's have flatter frequency response?

    ReplyDelete
  5. Jeff, I believe PMTs are still the most sensitive optical detectors of all modern tech. They are very low-noise, and often do not require an external amplifier at all, thus eliminating another source of noise. Unfortunately, PMTs do not have a flat response -- they are sometimes made to be very responsive to a particular phosphor.

    ReplyDelete
  6. Hi Ben,

    First of all, thank you for sharing your work. I enjoy watching your videos. I have a question regarding PMT and Oscilloscope.

    Currently, I'm trying to build a scanning laser confocal microscope. I have a Hamamatsu PMT that is connected to a tektronix TDS714L Oscilloscope. I used Pyvisa to query the oscilloscope as my microscope stages move from point to point, and used python to get the average voltage at every point, forming an array of voltages before I convert them into a heat map for the resulting image. Right now this point to point collection is very slow when I try to go finer in resolution.

    I have seen your scanning electron microscope images with an oscilloscope, and I'm not sure how I can collect the voltage readings against the position of my image. I'm thinking of collecting voltages every line instead of every point, so the scan takes faster.

    Could you point me in the right direction please? Thank you :)

    ReplyDelete
    Replies
    1. Tze Chong, it's true that capturing a whole oscilloscope trace will be faster than capturing the voltage at each point. I haven't used Pyvisa, but I would suggest looking for a method to download the entire oscilloscope stored trace instead of just the current voltage. You'll need to move the microscope stage at a constant rate, and start the oscilloscope sampling at the start of the stage movement, thus the trace will contain all of the data for one scan line. After the acquisition, you could download the trace data to the computer, then repeat the process for the next scan line. Your stage controls must allow for very constant, consistent movement for this work.

      Delete
    2. Thank you so much for your advice!

      Delete