where to find AR-coated glass?

i'd like to put a window in front of my lasers and possible future projects, as protection from outside. with antireflective-coated glass the loss would be less than one percent. now.. where to harvest it from? the closest i came to was inside a projector, but thats not exactly inexpensive.. ;-)
sure, i can buy it from "real laser companies", but that would then be the most expensive part of the setup..
i would need a small piece only, three cm² being enough (when i split it into smaller pieces).
my only idea so far is to order a glasses-lens without any "strength", as flat glass, with AR coating. would be high quality, but quite thick, and still around 50$ i guess..

anyone any idea?

manuel

looking around, i found two sources:
camera-filters and flashlight-lenses!

http://www.flashlightlens.com/str/index.php?app=ccp0&ns=display&ref=apps_specs

i ordered some uv-block camerafilters from hama, lets see if they block 405 as well. they are ar-coated anyway.
i'll report back, as well on how easy they are to cut!

manuel

also http://www.edmundoptics.com/onlinecatalog/search/index.cfm type laser windows

hey, thanks for the link! i didnt know edmundoptics was actually affordable?
this here looks good:
http://www.edmundoptics.com/onlinecatalog/displayproduct.cfm?productID=1919
3x3 inch, 1mm thick, 18.50$.
its transmission:

a nit low for 405, higher gets quite nice though. and for that price..?
i'd get it, if i didnt order other stuff already and if i was in usa..

manuel

The optics from the BR Sleds have AR coated pcs. in them. rob

my camera-filters arrived today, "hama uv o-haze 49mm".
in short: they dont work as hoped.

laser input output
405 13mw 12mw
532 36mw 30mw
650 177mw 149mw

they look clear, but you already see they arent that well AR coated. all "lost" energy is reflected, nothing is absorbed (as far as i tested).

great, now i have three useless filters.. but with 3$ a pop, i *had* to try it!

back to planning..

manuel

Try the 'SURPLUS SHED' manuel, Your Pal, rob

I think there is a glass called amaran or armaran glass, which does not reflect light.. Not sure, look over at laserfreak.net. There is a topic somewhere...

ArRaY said:

I think there is a glass called amaran or armaran glass, which does not reflect light.. Not sure, look over at laserfreak.net. There is a topic somewhere...

Click to expand...

hmm, interesting.. i was already thinking about some glass with the same refractive index as air, which then would be "invisible" as having no reflection at all.. no luck finding that "amaran or armaran glass", neither google nor laserfreak.

anyway, i just ordered some UCL lenses from flashlightlenses.com. i will report back when they arrive!
and/or will check in a local photo shop for those more expensive camera-lenses..

manuel

holy sh*t!
i got those UCLs from lighthound. you dont really see them when looking through it! weird thing, like some foul magic trick! you only see the (nicely sandblasted) boarder, with nothing inside! you barely see the reflection of roomlamps when trying!

the reflection depends on the degree of incidence. below 45° it stays quite constant, and almost ZERO! yes, i see a dot being backreflected, but its so little, i wont bother to measure with my thermo-head. couldnt measure that anyway, too little power! its less than one percent, i am sure. and thats whats being backreflected. the important part for a laser-window is the reflection of the reflection, the "forward-reflection" or whatever you'd like to call it. which creates a second (and more) dot right next to the original dot. the second dot has to "pass" the reflection twice, so its strongly reduced in power. regular glass with 4% (per surface) and a 200mw pointer as an example (to why i was searching for this AR coated stuff):

200mw - 96% = 8mw first reflection, going back to the laser, from the surface at the outer side
8mw - 96% = 0.32mw second reflection, from the beam above , from the surface at the laser's side

so when you shine 200mw through regular glass, you get 192mw out, 7.68mw back into the laser, and 0.32mw right nect to the original dot.
no, that doesnt sound much, but is definitely annoying. remember, brightness is perceived logarithmic, so you have to turn the power up 10x to actually see it twice as bright. so the original 200mw, now 192mw beam, looks only 2.5 times brighter than the garbage-dot with 0.32mw!
(if these 10x are right. the direction is right, anyway)

well, you guessed what the second dot looks like with those UCL lenses: they look like not being there! seriously, even when pointing the first (strong) dot at a black surface , i couldnt locate any second dot!

I couldnt have hoped for more, i am seriously impressed! and for that price? damn, best investion for a good while!

lets see how easy it will be to cut/smash the lens into smaller pieces!

manuel

The only way to ensure no reflection at all is to obtain AR coated glass and mount it at a predetermined angle with respect to the incoming beam.

At a certain angle (the value of which depends on several factors including the wavelength of the light, the glass composition and the medium through which the beam is traveling) all light is transmitted and nothing is reflected. This angle is referred to as the Brewster angle. This site contains a calculator for figuring the Brewster angle for a specific setup: http://brewstersangle.info/

Truthfully, if you do the calculations and determine the correct Brewster angle for your setup, it doesn't matter if the optics are AR coated. A plain piece of glass mounted at the Brewster angle will not reflect anything and allow 100% of the beam to pass.

If you were to calculate the Brewster angle for the optics you just mentioned, you'd be able to eliminate reflections altogether.


EDIT: I neglected to state that a Brewster window's efficiency (whether or not it introduces losses due to reflections) is polarization dependent. Even with unpolarized lasers, using optics mounted at or near the Brewster angle will reduce beam noise and reflection losses dramatically, and eliminate any possible feedback issues..