Beam combiner tests (new battery-driver holder)

[laserluke]

Here are some shots of the tests of a pbs cube and mirror.
Two diodes in series powered by my ims pcb driver (here the thread) in a new high power 4x18650 battery holder (here the thread) with ultralow power losses.

 

[T_Warne]

How small a head will you be able to get the lasers and optics into? I don't know if you are going to keep the diodes oriented as they are now, but I like the + look of the dot.

 

[laserluke]

Into how small a head will you be able to get the lasers and optics? I don't know if you are going to keep the diodes oriented as they are now, but I like the + look of the dot.

The optical head diameter is 5cm. It protrudes of about 4.5cm to the top of the battery-driver holder.
The shape of the beam dot and diodes orientation is mandatory because of the light polarization. With an anamorphic prism pair after each diode you can obtain a round dot, but with some power loss and increased dimensions and costs.

 

[tsteele93]

I am loving this - I think taking two 445's and making a + dot is an awesome idea.

 

[joeyss]

I think it's easier to get it to stay together that way since you're not trying to clump them over one another or right next to each other.

 

[Cyparagon]

...since you're not trying to clump them over one another...

But... that's exactly what's happening here.:can:

 

[joeyss]

Only it's rotated so it dosen't look so obvious. Instead it looks like just another muiltmode diode.

 

[laserluke]

I am loving this - I think taking two 445's and making a + dot is an awesome idea.

This pbs cube is very efficient, the output power is close to twice the power of a single diode.
The driver can easily power up a couple of new 9mm diodes. The output power could be more than 6W... a very powerful + dot :yh:

 

[Benm]

What would be the benefit of the + shaped output?

If you get two - shaped outputs superimposed you can still use prism pair optics to get a somewhat decent beam, but that option is clearly out ofter combining the beams this way.

Combining two - shaped beams with a pbs would require a quarter wave delay in one of the source beams to rotate the polarization by 90 degrees, but i stilll think that would be the better way to go - unless you wanted to project + signs for some reason.

 

[laserluke]

What would be the benefit of the + shaped output?

If you get two - shaped outputs superimposed you can still use prism pair optics to get a somewhat decent beam, but that option is clearly out ofter combining the beams this way.

Combining two - shaped beams with a pbs would require a quarter wave delay in one of the source beams to rotate the polarization by 90 degrees, but i stilll think that would be the better way to go - unless you wanted to project + signs for some reason.

You are right, but to follow your suggestion you need the quarter wave delay and the prism pair and so power loss, more complex optical head etc.
If you want you can also use a prism pair after each diode to obtain the same result with components without narrow bandwith.
At the end the simply question is: do you want only optical power or a circular dot?
If you need power to burn or even to pump doped materials like fluoride glasses the dot shape it's not so important.
And moreover... at even 2W it's very difficult to clearly see the dot shape with naked eyes because of its terrible brightness...

 

[T_Warne]

I just like the cool + target reticule look.

 

[Cyparagon]

At least this time you are not using your usual language.

Welcome to the internet.

 

[ARG]

How much power is being lost to the optics?

 

[AutoItKing]

Nice! Looks like you could stuff this into a maglight... just drill two holes for diodes.

 

[laserluke]

How much power is being lost to the optics?

This setup is very imprecise because it's only a proof of concept. You need to block a diode and rotate the other until you find the minimum power at a direction and the maximum at that perpendicular. You must then repeat this procedure at inverse for the other diode. I have only a too approximative measure at this stage and i'm completing the optical head. I'll post the correct results.

 

[Cyparagon]

You need to block a diode and rotate the other until you find the minimum power at a direction and the maximum at that perpendicular.

That's the slow way of doing it.

Laser diodes are polarized along the plane of their PN junction. The beam profile is an image of the junction. Therefore, if you have one beam profile at 90° from the other beam profile, their polarizations will also be 90° from each other and will sum with the greatest efficiency. No power measurements required.

It's not a coincidence you got a + pattern.