Results 1 to 6 of 6

Thread: Mode Locking / Q-Switch Questions

  1. #1
    Join Date
    Jun 2017
    Location
    Seattle, WA
    Posts
    47

    Default Mode Locking / Q-Switch Questions

    A year ago I darkened Photolexicon's doorway with a lot of questions about building a frequency doubled DPSS laser. I built it, but the frequency doubling never performed well. "I will fix it when I learn to do the math to understand what's wrong." I learned to do the math. Built a nice library in Matlab that does a pretty good job of predicting both IR and frequency doubled outputs. I knew the intensity in the KTP was too low but I need a predictive model so I didn't just randomly change the cavity.

    I got my answer and my answer is I'm probably not going to be able to get the KTP intensity high enough just running CW. I'm now looking at either mode-locking or q-switching (as was pointed out on this site last year but I wanted to learn the math).

    I'm considering either a saturable absorber or an AOM. I have both and have been experimenting. And now I have questions. So many questions.

    Mode Locking
    My understanding of mode locking is that you're aligning the different cavity axial modes so they're all in phase. This way you get a regular pulse train due to constructive interference. Average power is about the same but the pulses are very intense.

    To do mode locking with a saturable absorber you choose an absorber that "encourages" phase alignment by becoming more transparent when the phases do align. This feels like it needs some very careful calculation, and it also seems like the pump power needs to stay pretty well locked to one value. If you turned up the pump more doesn't this just bleach the absorber and you lose mode locking?

    I bought an AOM of eBay that is designed for mode locking and the way I understand it they've set it up so the operating frequency causes then AOM to resonate and this is what is creating the modulation -- IE it's not driven by an external modulation like a Q-switch. This means that the AOM frequency has to match the resonant frequency of the cavity or you won't get mode locking. That's tricky in my case. My AOM requires a cavity that's about 2.2m long. I don't have that much bench space, and I imagine you get CW all day until the cavity length is precisely tuned. How well will that stay with temperature changes? Seems hard to maintain without some element guiding the frequency of the AOM.

    Q-Switching
    For Q-switching you need an absorber that has enough loss to prevent lasing. Cr:YAG has a florescence lifetime of 3ms so this essentially becomes a capacitor and continues charging until it reaches enough to bleach. Stronger pumping means the pulses happen more often. But with lasing prevented there is very little flux in the cavity. Does the absorber need to be right next to the laser crystal for this to work?

    AOM Q-switching is all up to the AOM. The AOM I have I can trigger it manually and see that it diffracts the beam. But when I put it in cavity the diffracted beam still lases, just weaker. Is this a case of needing an AOM that has more diffraction? My AOM is only 2W and has a diffraction efficiency of 50% (this is the AOM I have). Or is this a case where I haven't configured it right?

    Test Setup
    I built this test rig to try things out. I can swap either the absorber or AOM into the upper area of the cavity. The lower area is currently configured for extracavity SHG -- this way I have a pretty strong signal that either mode locking or Q switching is working since CW will yield really poor harmonic generation:

    Click image for larger version. 

Name:	Setup.jpg 
Views:	3 
Size:	724.0 KB 
ID:	59972

    With a .8 initial transmission saturable absorber and a focusing lens to get the intracavity beam down to 60µm I can get the laser to lase with just a little more pump power then when the absorber isn't there. IR output measures the same on my power meter but the 532 output is pretty different.

    With no saturable absorber the green output is almost unnoticeable in this photo:

    Click image for larger version. 

Name:	WithoutAbs.jpg 
Views:	4 
Size:	757.0 KB 
ID:	59973

    With the saturable absorber I measure 4mw:
    Click image for larger version. 

Name:	WithAbs.jpg 
Views:	4 
Size:	726.8 KB 
ID:	59974

    I seem to be getting either mode locking or Q switching but I don't know which. What tools do I need to detect this? I have a pyroelectric joule meter I also bought off eBay hooked to my oscilloscope. It does output a signal when the laser is on but it seems DC. My cavity resonance is about 265Mhz which is just under my scope's 300Mhz bandwidth limit so I would expect to see something. Are these not fast enough?

    I also tried .5 and .2 T0 Cr:YAG crystals but could get no output. My assumption here is that lower T0 causes more build-up so a larger Q-switch pulse, but you wouldn't want this for mode locking. I've turned the pump power up pretty high and neither of these would lase. The crux here is I don't know how to calculate the fleuence on the absorber when the cavity is not lasing.

    Also, any good papers or books that would help with computing the best initial absorber transmission for my setup and help me model this? I have books by Siegman, Koechner and Silfvast and what looks like an excellent paper here that I haven't totally gone through yet. I want to be able to calculate the best T0 and the intracavity pulse intensity so I can figure out the best waist into the KTP.

    Sorry for the insanely long post and so many Q's. Complicated subject and I don't know what I'm doing so I don't know what to ask.

    Thanks for any info! -- Brian

  2. #2
    mixedgas's Avatar
    mixedgas is offline Creaky Old Award Winning Bastard Technologist
    Infinitus Excellentia Ion Laser Dominatus
    Join Date
    May 2007
    Location
    A lab with some dripping water on the floor.
    Posts
    9,890

    Default

    Usually I use a DET-10A from Thorlabs for watching Q-switching / Modelocking , with a minimum of a 100 Mhz scope, usually faster is preferred.

    The core of the DET-10A is a FDS010 (Thorlabs) or similar photodiode with less then 1 mm^2 area reverse biased with 5 VDC , and a 0.1 uF cap to return pulse current to ground. The Det-10A schematic is out there. They mount the PD on a piece of microstrip, but if you build it dead bug with very small loop area, it can work. The DET-10A is designed to work into a 50 Ohm scope input so buy a 50 Ohm terminator if you dont have one.

    Usually we're watching the flashlamp puise shape, but since your diode pumped, that is not an issue.

    You need some form of attenuation for the PD, I use cheap thin alumina disks from Ebay and combinations of ND4 and ND10 filters...

    DISKS from Seller:
    advancer53 who usually has something useful.

    https://www.ebay.com/itm/303950558680?hash=item46c4dd75d8:g:MwcAAOSwKPNTyPl Y&amdata=enc%3AAQAHAAAAoNOhhPqwzDIxWD1iJPQW732CKRF ofDH%2FyfXV13UgmVlqFvG7POUq5hRo7rXf13dCyZ%2F6sM%2F AMcJSqHZfEpTGFBBMsfadK0%2BfrQ6assNXoDLJV5DwmEYksqX xFS%2FoKgLvZ03wyoryEx8QjjpW8ZNZVTf3U8wlJOn2TmfZJnp 7LPBA3%2FR9GMWlqgycHRc%2FOkuwQ5RFko4iP1J%2BUhdO1M7 4KQU%3D%7Ctkp%3ABk9SR6iTifrRYQ

    Steve
    Qui habet Christos, habet Vitam!
    I should have rented the space under my name for advertising.
    When I still could have...

  3. #3
    Join Date
    Nov 2008
    Location
    Cleveland Ohio
    Posts
    2,599

    Default

    Mode locking does as the name implies locks the modes together so they are all in phase. Q switching changes the Q of the cavity so it only lases after the power in the cavity reaches a certain point. Up to that point the population just builds. Once a threshold is reached the inversion falls down rapidly in a giant short high energy pulse. When the pulse ends the Q drops back out of the lasing regime and the build up starts again.

    WIKI has a nice explanation of each
    https://en.wikipedia.org/wiki/Mode_locking
    https://en.wikipedia.org/wiki/Q-switching

    The there is cavity dumping which is even something else.

    I think you might actually be seeking this:
    https://eng.libretexts.org/Bookshelv...d_Mode_Locking

    This is more what i think of for a solid state laser for mode locking due to the Kerr Effect. There are many ways to do this.
    https://eng.libretexts.org/Bookshelv...e_Locking_(KLM)

    I am NOT an expert on this so this is just my opinion and what I've learned. I working to convert a Ti:Saph to modelocking that is now CW. I want to use laser diodes to pump it so I wont see enough action to make a KErr mode locking setup. Hence a sesam. Looking to use a SESAM but man are they expensive and the power range is limited.
    https://en.wikipedia.org/wiki/Semico...bsorber_mirror

    LAST:KICK BUTT FOR MAKING IT WORK!

  4. #4
    Join Date
    Jun 2017
    Location
    Seattle, WA
    Posts
    47

    Default

    Quote Originally Posted by mixedgas View Post
    Usually I use a DET-10A from Thorlabs for watching Q-switching / Modelocking , with a minimum of a 100 Mhz scope, usually faster is preferred.

    The core of the DET-10A is a FDS010 (Thorlabs) or similar photodiode with less then 1 mm^2 area reverse biased with 5 VDC , and a 0.1 uF cap to return pulse current to ground. The Det-10A schematic is out there. They mount the PD on a piece of microstrip, but if you build it dead bug with very small loop area, it can work. The DET-10A is designed to work into a 50 Ohm scope input so buy a 50 Ohm terminator if you dont have one.

    Usually we're watching the flashlamp puise shape, but since your diode pumped, that is not an issue.

    You need some form of attenuation for the PD, I use cheap thin alumina disks from Ebay and combinations of ND4 and ND10 filters...

    DISKS from Seller:
    advancer53 who usually has something useful.

    https://www.ebay.com/itm/303950558680?hash=item46c4dd75d8:g:MwcAAOSwKPNTyPl Y&amdata=enc%3AAQAHAAAAoNOhhPqwzDIxWD1iJPQW732CKRF ofDH%2FyfXV13UgmVlqFvG7POUq5hRo7rXf13dCyZ%2F6sM%2F AMcJSqHZfEpTGFBBMsfadK0%2BfrQ6assNXoDLJV5DwmEYksqX xFS%2FoKgLvZ03wyoryEx8QjjpW8ZNZVTf3U8wlJOn2TmfZJnp 7LPBA3%2FR9GMWlqgycHRc%2FOkuwQ5RFko4iP1J%2BUhdO1M7 4KQU%3D%7Ctkp%3ABk9SR6iTifrRYQ

    Steve
    Thanks, Steve -- DET10A (now DET10A2) on its way and thanks for the advancer53 tip -- those do look great and also purchased.

  5. #5
    Join Date
    Jun 2017
    Location
    Seattle, WA
    Posts
    47

    Default

    Quote Originally Posted by kecked View Post
    Mode locking does as the name implies locks the modes together so they are all in phase. Q switching changes the Q of the cavity so it only lases after the power in the cavity reaches a certain point. Up to that point the population just builds. Once a threshold is reached the inversion falls down rapidly in a giant short high energy pulse. When the pulse ends the Q drops back out of the lasing regime and the build up starts again.

    WIKI has a nice explanation of each
    https://en.wikipedia.org/wiki/Mode_locking
    https://en.wikipedia.org/wiki/Q-switching

    The there is cavity dumping which is even something else.

    I think you might actually be seeking this:
    https://eng.libretexts.org/Bookshelv...d_Mode_Locking

    This is more what i think of for a solid state laser for mode locking due to the Kerr Effect. There are many ways to do this.
    https://eng.libretexts.org/Bookshelv...e_Locking_(KLM)

    I am NOT an expert on this so this is just my opinion and what I've learned. I working to convert a Ti:Saph to modelocking that is now CW. I want to use laser diodes to pump it so I wont see enough action to make a KErr mode locking setup. Hence a sesam. Looking to use a SESAM but man are they expensive and the power range is limited.
    https://en.wikipedia.org/wiki/Semico...bsorber_mirror

    LAST:KICK BUTT FOR MAKING IT WORK!
    Thanks for the links. Need to read that paper more carefully -- lot of dense math in there. I've also looked a bit at sesams. Haven't found anyone who sells them yet and they sound like they still need somewhat bespoke configuration, which for my hobby of 1 unit puts it pretty out of reach. I've also read up a bit on KLM. Seems like a really good way for me to destroy all my optics trying to get it to work and seems to require a pretty dialed in pump power, but really clever idea (and no excited state losses like a saturable absorber).

  6. #6
    Join Date
    Jun 2017
    Location
    Seattle, WA
    Posts
    47

    Default Sensor Arrived

    The photodiode and ND filters arrived and I managed to get a scope image of what this is doing. It does look like it is both Q-switching and mode locking. The mode locking modulation depth is pretty poor and the Q-switch repetition rate is between 4-15kHz depending on power input, which seems really low. And I can only run it for a few seconds before it heats up thermal lensing messes up the waist in the saturable absorber and it stops lasing entirely (this is a test setup and I don't have enough extra hardware to provide decent cooling for anything but the pump diode), But hey, it's a promising first step.

    Click image for larger version. 

Name:	Pulsetrain.jpg 
Views:	1 
Size:	755.3 KB 
ID:	60002
    Click image for larger version. 

Name:	Pulse.jpg 
Views:	1 
Size:	738.7 KB 
ID:	60003

    Now I need to debug a lot of math. The mode locking pulses are about 330MHz and this lines up pretty well with my cavity length. But the Q-switch repetition rate is way off (I measured 15kHz....I calculated 2.2Mhz) as is the pulse width (80ns measured, 1120ns calculated). I'm pretty sure my calculations for power output and intracavity intensity are off too and since the whole point is to get the SHG conversion better I definitely need good numbers there.

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •