0
$\begingroup$

I am in need of suggestion regarding a frequency modulation experiment. I am splitting a 80 MHz repitition rate laser using a 50 : 50 beam splitter. Both arms are frequency shifted using two AOM's driven at 40 MHz and 40.66 MHz. After that they are recombined using another beam splitter. I need to confirm the resulting frequency that is 660 kHz. However, when I use a oscilloscope, the only signal it displayes is the laser rep rate that is 80 MHz. Is there a way to just see the modulated optical frequency ?

Thank you

$\endgroup$
1
$\begingroup$

Side bands may exist, but I think that's not what you are looking for.

You mention an 80 MHz repetition rate in your laser. This implies that it is a pulsed laser. You do need to make sure the pulses overlap at the output port of your interferometer. That means the two beam paths should have the same length. You also need to make sure the beams have the same polarization.

Your pulse length will be on the order of nanoseconds, most likely. In that case it is not possible for the 660 kHz beat frequency to show up within individual pulses. Instead, the beat frequency can only be apparent as a modulation of the envelope of the pulse amplitudes. One cycle of the beat frequency will correspond to about 121 pulses, so you need your detection system to be able to show a large number of laser pulses in order for the pulse amplitude envelope to be apparent.

$\endgroup$
  • $\begingroup$ Thanks Mr. McGrew. The polarizations are same as the beam splitter is nonpolarizing. For the current wavelength I am working on the band width is 0.8 nm and pulse width (fwhm) is 2.3 ps. I have tried to keep the distance same but I dont know if they are exactly the same. Is there a way to ensure that ? $\endgroup$ – user188062 May 24 '18 at 0:54
  • $\begingroup$ Also, if I understood you correctly this can be detected using a oscilloscope, right ? or do I need a spectrum analyzer ? $\endgroup$ – user188062 May 24 '18 at 0:55
  • $\begingroup$ You don't need an optical spectrum analyzer, because the signal you are looking for will not be at an optical frequency. An oscilloscope with a photodetector that can respond faster than 660 kHz is all you should need. $\endgroup$ – S. McGrew May 24 '18 at 3:27
  • $\begingroup$ Tou will need to capture a single sweep of 2 or 3 microseconds to see the beat signal clearly. Otherwise, you'll need to trigger in phase with the beat frequency, e.g., directly from the AOM driving circuits. $\endgroup$ – S. McGrew May 24 '18 at 3:33
  • $\begingroup$ Depending on how fast your oscilloscope and detector are, you might be able to use the oscilloscope to help adjust the path lengths. Or, you might just run both AOMs at the same frequency and adjust the interferometer until you get high contrast fringes, visible to the eye. That ensures that the path lengths are the same. Then get the alignment perfect so that only one broad fringe lands on your detector. If you have any mirrors that direct the beams out of plane, you might be rotating the polarization. Best to test using a polarizing filter. $\endgroup$ – S. McGrew May 24 '18 at 3:52
0
$\begingroup$

Have you tried using a spectrum analyzer rather than an oscilloscope and then looking for side bands?

$\endgroup$
  • $\begingroup$ Hello Gilbert, could you please elaborate a little on side bands? Also, Is it necessary for the two arms of laser beams in this particular set-up to be temporally synchronized, (as it is a pulsed laser), for the 660kHz frequency to be detectable ? $\endgroup$ – user188062 May 24 '18 at 0:00
0
$\begingroup$

Gentlemen,

I used both of the suggestions and I finally got it. Being a pulsed laser it was slightly tricky than the CW laser.

  1. Practically everything after the AOM's was placed on a translation stage for better temporal optimization.

  2. Instead of 907 nm which was invisible we started with 720 nm which was visible and help us hitting same spot inside the splitter. Once this was achieved, the wavelength was changed to 907 in steps of 40 nm with caliberation at each step.

  3. We have a SR865A LOCK IN with partial spectrum analysis capability, we used to first get the 660kHz signal, and maximized the signal using the translation stages (temporal overlap)

  4. We were able to observe in oscilloscope by rescaling the time axis to 1 mmicrosecond.

Thank you to both of you for the suggestions. enter image description here

The attached image shows when the modulation was observed for 460kHz

$\endgroup$

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.