Welcome to Laser Pointer Forums - discuss green laser pointers, blue laser pointers, and all types of lasers

LPF Donation via Stripe | LPF Donation - Other Methods

Links below open in new window

ArcticMyst Security by Avery

Spatial filtering

Joined
Oct 30, 2014
Messages
4
Points
0
Hi,

I have a 500mW 405 laser diode with a G2 lens. I need to spatially filter the beam to produce a beautiful gaussian profile. I am trying to produce a clean set of interference fringes through a Mach-Zender interferometer.

I have a optics table and a 12.5 micron pinhole with nice positioning capability. I can produce a nice gaussian central portion but I always have a lot of rings around the central portion. These rings cause problems later in my setup. I was told the spatial filter isn't matched correctly if you still have rings.

Has anyone had success spatial filtering with a similar setup?

Thanks.
 





Joined
Oct 30, 2014
Messages
4
Points
0
Thanks for the reply.

Yep. Multimode diode lasers are tough.

I seem to have a nice spatial filter working. However, I am trying to get some high contrast fringes through a Mach-Zender interferometer. I get fringes but low contrast.

I think it is the multiple longitudinal modes killing the contrast, but I am not sure.

A SLM 405nm diode laser is $1000s. Out of my budget.
 
Joined
Feb 5, 2008
Messages
6,252
Points
83
That is out of my league, I'm afraid.

Somebody else can chime in on this, I hope. Sorry :/
 

LSRFAQ

0
Joined
May 8, 2009
Messages
1,155
Points
83
Multiple off axis transverse modes, a horrible mode structure, both transverse mode and longitudinal modes. As well as a very broad wavelength spread, are hurting you. Trying to get good fringes with a modehopping laser that may have a 2-3 nm wide spectrum is very tough.

Can you get your hands on a HENE? Very narrow.... Maybe ~10 Ghz total line-width on a good day with a short cavity.

Or a Tem00 Long cavity DPSS green? ~1.1 nm FHWM.

Anything other then a direct diode laser at that point.

If I was STUCK with 406 or 445 blue for experimental reasons with a interferometer... Well, I'd try to find another way..

If there is no escape, try adding some Littmann-Metcalf style grating feedback to your blue diode. Do not feed 100% of the diode's energy back into itself, it will blow the faucet off the diode. But you can narrow the spectral width by grazing incidence grating feedback and adjusting current. That can be further improved by controlling the diode temperature. That will not help you with the beam profile problems. It will however sharpen the fringes greatly. Any technique you use to clean up a multimode will cost you dearly in output power.

In fact, I'd tune the laser with the grating. Then I'd up collimate, kill the nasty edges of the beam with a circular mask, then try to spatial filter. Or use anamorphic prism correction to circularize the beam, mask it and spatial filter.

By doing the above steps, I'm basically building a clone of Coherent's "Cube" series of lasers. That is a lot of work to perfect. 405 Cubes without the extra wavelength stabilization show up on Ebay for reasonable costs.

Your basically trying to make a silk purse out of a sow's ear, when using a multimode blue/uv diode.

Books you might want to find to read up on theory:

The Late Tony Siegmans' book, a bit tough to read:

Lasers: Anthony E. Siegman: 9780935702118: Amazon.com: Books

William Silfvast's book, wonderfully easy:

Laser Fundamentals: William T. Silfvast: 9780521541053: Amazon.com: Books

This gentleman is the master of the Homemade ECDL, a good read:

Laser-Projects Page

Laser-Projects Page

Before I went to all that trouble, I'd do some math:

There is always the chance that you need to change pinhole size or get a much better focusing lens. A few rings are pretty normal with even the best spatial filter. If I had followed the pinhole sizing requirements on the CVI Melles Griot or Newport App notes, and I still had lots of rings, I'd start narrowing the wavelength and "pre-cleaning" the beam.

I wish you patience and good luck. :)
Steve
 
Last edited:
Joined
Oct 30, 2014
Messages
4
Points
0
Thanks Steve,

Very thorough response and very much appreciated.

At this point, I am a little stuck. The 404-405nm diode is being used mostly because the module is small and high power >500mW compared to the "cleaner" modules. Also, we have built our own laser driver so we can pulse the beam synched to our CCD camera. Anyway, all for not if the laser isn't appropriate. The high power is needed because the beam illuminates faint fluorescent particles at the output of the interferometer. More power is more signal.

I think doing some kind of external cavity stuff is more than I am up for at this point.

I am using a G2 405 lens focused through a 12.5um Edmund pin hole. I can get decent looking central pool at this point but have some rings around it. I can trim those rings with a second aperature. The result is now a nice circular pool of light. Very clean. Still, the fringes are poor contrast although very stable, no shifting or changes over several minutes. I have been on some holography forums that suggest up to several millimeter coherence length from a 405 diode. This is more than enough. My path lengths are matched to 100 microns probably. So, it might be that my fringe contrast needs are greater than my diode can deliver.

Your final comment. Patience. Yep, I have used a lot so far. A week of work just in alignment. Hope I have some patience in reserve.

Thanks again.
 

LSRFAQ

0
Joined
May 8, 2009
Messages
1,155
Points
83
It sounds like your beam profile is on target.

I spent 8.5 years of my life using lasers, leds, metal halide, cermax, xenon strobes , and ultrafast air flashes to film Electrospun Nanofibers with high speed cameras. I've filmed objects smaller then 1/2 my illuminating wavelength, without microscopes... I have some tricks for using slow CCD cameras, as we only had one high speed one. This includes using external modulators.

Talk to me... I may be able to direct you to an expert.

What is your pulse duration? Driving the diode with pulses opens a new dimension. However some of that is fixable...

Things like DC pre-bias, preheating the diode, injecting RF through a bias Tee, grating feedback, attenuators to stop cable ringing, can help..

Even Fresnel Reflection from a slightly tilted microscope slide or cover glass can act as a etalon or frequency selective element.

Avoid undesired optical feedback like the plague. Remember, the black spaces from one end of the interferometer are travling back and becoming white spaces at the laser end. This is where a waveplate and polarizer get very handy to form a isolator.

Is your MZ polarization sensitive?

What is your pulse time and PRF? Can you gate the camera?

Steve
 
Last edited:
Joined
Oct 30, 2014
Messages
4
Points
0
LSRFAQ,

Thanks for the reply. Sorry for the delay. I have been buried in the lab. Trying to improve things. Several of your points were very helpful. I seem to have improved things quite a bit. Here is what helped:

Properly sized pinhole - about 10um seems best with these diode lasers and lenses I have. I melted several stainless pinholes searching for the right match. National Aperture high power pinholes have been working well.

Properly spaced pinhole - I had been using the pin hole very close to the laser output. The idea being that closer focus would produce smallest possible point size and would get the cleanest, highest power through. However, too close caused back reflections to destabilize the diode. This produced a fuzzy set of interference fringes through the MZ interferometer.

Pathlength matching - Uggg, this was a bear. The MZ interferometer contrast is sensitive to path length if your longitudinal coherence is very short. This is the case with the multimode diode lasers. I had to set up my beamsplitters on micrometers and search for the position of highest contrast. It turns out, pathlengths were off by about 100microns. This was enough to change the contrast from barely passable to great.

Suitable laser power - At high power the diodes coherence went down a lot. Unfortunatley, I need a lot of power to illuminate my target. So there is a tradeoff here.

Patience - It takes a lot of mind numbing set up to get two clean beams to interfere in the MZ.
 
Joined
Sep 25, 2007
Messages
434
Points
28
lightweight you run the risk of catastrophic destruction by putting the optic up close to the diode like that, the back reflection is not good for a diodes health.
Last summer I lost a diode because of this I think.. I had it in a C6 host for probably 2 years and I suppose all the use of it made it weaker, as my finger passed by its front the diode died. I think it was my fingernail that did it.
 




Top