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

Buy Site Supporter Role (remove some ads) | LPF Donations

Links below open in new window

FrozenGate by Avery

Mutil watt 445nm using beam splitters?

It's the enteral limitation in combining laser power, isn't it? You can go for twice the power using PBS cubes, but that option simply ends with only 2.

Knife edging can sometimes work out nicely, but i still consider it 'cheating' in most applications. Surely the beams can overlap at larger distances and give the impression that they are one beam, but when you break it down, its still just a bunch of lasers roughly aimed at the same piece of wall (or sky).
 





Benm said:
It's the enteral limitation in combining laser power, isn't it? You can go for twice the power using PBS cubes, but that option simply ends with only 2.

Knife edging can sometimes work out nicely, but i still consider it 'cheating' in most applications. Surely the beams can overlap at larger distances and give the impression that they are one beam, but when you break it down, its still just a bunch of lasers roughly aimed at the same piece of wall (or sky).
Click to expand...


well generally speaking yes a cube can only combine two beams.

yes all you are doing is overlapping beams but not roughly. if done properly it really is just one beam you cant see lots of single beams.

subject you knowing what your doing you can achieve lots of power with a great beam.

take arctos (above pic i posted), 48 red beams into a 5mm by 5mm final beam at about 1mR

thats impressive.
 
Very interested in doing knife edging properly. Particularly selecting lenses to reduce the size of the final beam.
Could you share more about it & perhaps links if you have some?

With 1W of 445 now available cheap red is such a weakness in RGB.
 
Last edited:
Ok, So here is my optimistic new project I'm going to begin next week when my Casio XJ-A140 arrives.

There are 3 rows of 8 diodes. I'll remove two rows. Cut the block with a band saw so that only the top row remains. Therefore I'll have one row of 8 diodes side by side. Then I'll make/buy eight adjustable mirror mounts and try the "knifes edge" process to get all eight beams close together..That should give me close to 8 watts blue :)

One little problem I can think of: the eight beams will be similar to a multi emitter c-mount laser. Do I need a lens similar to what one would use on these c-mounts to correct the fast/slow axis?

I'll keep you all posted.
 
Last edited:
daguin said:
Yep. However, they are "stacked" four high ;)
Ya want a mirror assembly? I can sell ya one . . . .

bb4-1.jpg

bb3.jpg


Peace,
dave
Click to expand...


Could you clarify when you say stacked four high?

Thank you
 
VectorField said:
Very interested in doing knife edging properly. Particularly selecting lenses to reduce the size of the final beam.
Could you share more about it & perhaps links if you have some?

With 1W of 445 now available cheap red is such a weakness in RGB.
Click to expand...


red isnt a weakness, not if you build your own ;)

if youve got the money ive got the red.


Ok, So here is my optimistic new project I'm going to begin next week when my Casio XJ-A140 arrives.

There are 3 rows of 8 diodes. I'll remove two rows. Cut the block with a band saw so that only the top row remains. Therefore I'll have one row of 8 diodes side by side. Then I'll make/buy eight adjustable mirror mounts and try the "knifes edge" process to get all eight beams close together..That should give me close to 8 watts blue

One little problem I can think of: the eight beams will be similar to a multi emitter c-mount laser. Do I need a lens similar to what one would use on these c-mounts to correct the fast/slow axis?

I'll keep you all posted.
Click to expand...

that wont work.

if you knife edge 8 beams in a row together your final beam will be a long line.

what you need to do is have the input beam into the first telescopic lens as square as possible. that way your final output beam will be square and not a long line.

does that make sense?

you need to arrange your knife edge beams into a square ish beam not a long line.

example -

you need to measure the size of the beam at apeture. say the beam measure 5mm by 3mm.

rotate the diode so the 3mm is left to right and the 5mm is up and down.

with that information you would then knife edge 4 beams across measuring 12mm (3mm x4) and stack 3 beams on top of each other measuring 15mm (5mm x 3)

so your input beam into the first telescopic lens would be 12mm by 15mm which is square ish and then your final output beam would be a usable shape.
 
Hi Andy,

Thanks for the info.
So I couldn't knife edge 8 diodes in a row. I'd have to stack some as you said? I really didnt think about rotating the diodes. Good advice
 
Or figure out how to make the diodes transparent, then just put one behind the other. :P

Actually, I've been thinking about this for a while, and one possibility has occurred to me, though it would be tricky. Theoretically you could merge many (tens) of beams directly on top of one another.

The concept is to use temperature control to shift the wavelength of the diodes, get the beams close with knife-edge combining, then use a prism compress the beams the rest of the way together, taking advantage of the differences in wavelength.

Suppose you have a series of diodes. You would mount them in a line, then create a carefully controlled temperature gradient from coldest to hottest along that line. Due to how laser diodes react to temperature, the hot end would have longer wavelengths while the cold end would have shorter. Combine the output knife-edge style and you have a line shaped beam that is bluer on one side and redder on the other (though the range would probably only be about 10nm). Now use a couple of prisms with fancy coatings (specific for your wavelength to minimize loss), and you should be able to get the beams practically co-linear. Your output would be more broad-spectrum than a typical laser, not that that matters.

You would do two of these at opposite polarizations (reusing the temperature gradient), then PBS them together.

Ok, that's crazy. It might be fun to try though.
 
my guess is that theoretically you could have lasers with the polarization changing in 20 degree increments,
and combining them repeatedly with cubes.allowing for 68 or 9but you would need to reflect one and pass
a large range of polarization angles,but everything works in theory. that would give like 10 watts,
without driving the diodes too high.if you had said combining lens, it would be simple, but it would
be easier to simply make a much bigger diode
 
Last edited:
lasersgopewpew said:
my guess is that theoretically you could have lasers with the polarization changing in 20 degree increments,
and combining them repeatedly with cubes.allowing for 68 or 9but you would need to reflect one and pass
a large range of polarization angles,but everything works in theory. that would give like 10 watts,
without driving the diodes too high.if you had said combining lens, it would be simple, but it would
be easier to simply make a much bigger diode
Click to expand...

Unfortunately polarizing beam splitters don't work that way, even in theory. With 20 degree increments, you would only gain 45% of the second beam (or lose 55% of the first beam). The next PBS would be a similar scenario. In the end the most you would be able to approach is 200% of any single beam (which is the same you would get with two beams with 90 degree offset and one cube).
 
The polarization things doesn't work with more than 2 beams, there are only 2 orthogonal polarisations and the rest is a mix of both.

The wavelength shift trick could work in theory, but in practice you won't get far. You could select 2 diodes with wavelength far apart and temperature tune them to get them futher apart, a separation of 5nm would be enough to combine them with less than 10% loss. It would require a custom coated beamsplitter (unless they're already made for this wavelength) but it's possible.
The prism trick probably doesn't work, prism have a low dispersion, so the setup would be huge even to combine 2 wavelengths. Efficient gratings would be worth an experiment. It would still make a huge setup, but you might be able to combine more than 2 beams.
In the end the wavelength trick is probably not worth the trouble.
 
You must log in or register to reply here.


Back
Top