Set-up for combining 4 LDs

[Light superglue]

Since today is the long weekend in Spain when I do not have to go to work, but Labour Day what means that one has to work and not just stay idle, I decided to finally build the set up for combining 4 LDs. There was a lot of talking on LPF how this could be easily done but no one yet has posted a "working" set up.

So here is mine. Sorry for not making it very ergonomic but for doing experiments it seems fine: 4 laser modules, 2 Broadband PBS cubes, 1 445-462nm Dichro mirror.

This Blue-Blue Dichro mirror as I posted before can also work for combining 405+450nm, so here I have put 2 BDR-209 and 2 PLT5 450B diode modules.

And now the "unexpected results" of the 1st experiment:
Combining 2 PLT5 works fine - the beams are easily combined by finding the right polarisation positions as you see and most of the light is passing through the Dichro (picture made at 26mA current).
But combining 2 BDR-209 does not seem to work with same type of PBS - by rotating modules I cannot find a position where PBS would be transparent for any of the two beams, it looks that both beams in any rotational position are mostly reflected by the PBS! (I am talking here about PBS not Dichro)
What does it mean? Can the reason be that the 405nm lasers from BDR-209 may be not V/H but circularly polarised and I would need a 1/4 and/or 3/4 waveplates to combine them?
I thought that PBS-combining of single mode LDs does not need use of any WP but maybe I was wrong???

In 2014 Jander6442 has made a Maglite with two PBS-combined BDR-209 and it produced 1335mW if I remember it right. From two LDs each capable of 800-1000mw according to DTR tests!!! Could he encounter a similar problem without knowing it?

 

[BobMc]

Looks like a very fun project. Hope you get it figured out. I'm not much help, but I'm hoping for the best. Love to see the finished product. Find your your idea very intriguing, best wishes. Like your pictures. :gj:

 

[diachi]

Good to see your setup!

They should be linearly polarized... Anyone else able to do some testing?

 

[Light superglue]

OK, I think I have found why these PBS are not working for BDR-209.

Checked other PBS which I have been purchasing meanwhile from different suppliers.
These 2 big ones were cheap from ali/ebay with description to have 405-450nm ARC. But having 405nm ARC only means....having 405nm ARC and nothing else. Apparently polarising mirrors themselves in these PBS were not made to let pass any polarisation of 405-415nm to good extent.

However, other PBS cubes (from Stanwax and Live lasersystems) work as expected - let pass one and reflect other P. Interesting that these two were not promoted as for combining 405nm, but 445 (Stanwax) and 440-670 (LL).

Also funny that PBS from Lasertack which also has 405-680nm ARC in descrition behaves like the two from Ali/ebay. But to say the truth its description mentions that it can be used for combining 445, 532, 650 etc. but says nothing about 405nm...

 

[diachi]

OK, I think I have found why these PBS are not working for BDR-209.

Checked other PBS which I have been purchasing meanwhile from different suppliers.
These 2 big ones were cheap from ali/ebay with description to have 405-450nm ARC. But having 405nm ARC only means....having 405nm ARC and nothing else. Apparently polarising mirrors themselves in these PBS were not made to let pass any polarisation of 405-415nm to good extent.

However, other PBS cubes (from Stanwax and Live lasersystems) work as expected - let pass one and reflect other P. Interesting that these two were not promoted as for combining 405nm, but 445 (Stanwax) and 440-670 (LL).

Also funny that PBS from Lasertack which also has 405-680nm ARC in descrition behaves like the two from Ali/ebay. But to say the truth its description mentions that it can be used for combining 445, 532, 650 etc. but says nothing about 405nm...

Figured it might be something like that, glad to hear you got it figured out. Thanks for updating us!

 

[Alaskan]

Thank you, if you post photo's of your setup working in a darkened room so I can see the beams I'd love to see that.

Chris

 

[Light superglue]

Hi Alaskan,

I would like to but holidays are over now and I have to be at work all days long.

I will try to return to the subject on the WE. Now I can get ~1.5W from 2 BDR-209 combined but the "new" problem is that collimators are shaky and I cannot get one spot at 5m, there are always 2 spots some millimeters apart on the wall. I need to get closer into the subject of thread fixing - teflon tape, external springs from Survival lasers...what else can you suggest?

 

[Light superglue]

Yesterday I have received the package with external springs from SL.

Unfortunately they are 10mm long and have 4 coils - this does not pass into the tiny space between Cu module body and focusing ring when G-2 has to be focused to infinity. But at the end after pushing hard I could get the 3 modules with G-2 to focus on the wall at 5m and for the 4th module I had to cut half of the spring off. In this way the collimators were not shaking any more and I could start the adjustment of the beams at 10% of current (60mA to both BDR-209 and 30mA to both PLT5 450B).
But I would not recommend to anyone using these springs with G-2 unless for burning when the barrel is more screwed out of the module - look at the picture how the coiling is pressed out making a lot of pressure on the M9x0.5 thread inside... I was just trying not to use teflon tape, maybe I was on the wrong way...

These single mode spots were so tiny that it was not possible to see if they coinside at 5m away or not. After 1 h of running to the wall and back to the set up best I could do is to position the 4 spots into 5mm square but no way of getting one perfect spot!
So I realized that I need a mirror which would reflect all beams to the wall next to me... this story will continue.

 

[Light superglue]

Hi Alaskan, Diachi and Bob (hope other guy are interested, too),

This night I finally managed to snap some pictures which you may like.
So this is the final confirmation that this Dichro can be of use to combine 405 and 445nm, not only 445+462.

So I bought a mirror, put it 2.5m away and did the adjustemnt of 4 beams looking at the reflected spot on the wall next to me (10% of max current). Would not say that the result is completely perfect (single mode spots are not as forgiving as big multi mode spots) but at least the spot is max 3mm in diameter now. Might be better with some spacial filtering...

The old LPM (with zero line -0.05W) shows 1.98W what means the beam has 2.03W in it. It is well visible in the dark, but unfortunately not at a sunny day (this is why you want to make DPSS, right?). The direct beam burns through white sheet of paper at 1m, not further. Sure it would if focused with a kind of BE, but here I have all lasers focused to infinity.
However it made a perfectly round hole in a blue post-it which was attached to wall 5m away and left a 2mm mark on the white painted wall behind!!!

 

[Exerd]

On a side note to your testing you could use a webcam connected to a laptop. Then you could remote into the laptop from your Android and view the webcam image on the screen. I don't know which method you'd find easier for your scenario but it's another thought.

I honestly was not aware that 4 beams could be combined--that way. I always thought that you would get 2 with the same polarization causing an interference reduction when setup like that, I.E. only 1 beam of two lasers. Of course this would occur only in the scenario of absolute stacked beams, or at least I had went on for some years believing as such. Light has 2 polarization states. How do we combine 4 lasers into 1 beam when there are only 2 states they can exist within?

Or am I overthinking this in that you are merely aiming 4 lasers at a distant point and it's not a truly combined beam?

To explain how I am viewing this problem I think of it like what follows:

Lasers A, B, C, and D.

In your setup I view it like you are adding together [A+B]L1+[C+D]L2.

A enters and passes, it is state 1.

B enters and reflects, it is state 2.

C enters and passes, it is state 1.

D enters and reflects, it is state 2.

Beams meet and now that I've just fully confused myself, L1 beam reflects one direction and L2 beam reflects the other--they are not combined.

Yes I really am curious now.

 

[diachi]

On a side note to your testing you could use a webcam connected to a laptop. Then you could remote into the laptop from your Android and view the webcam image on the screen. I don't know which method you'd find easier for your scenario but it's another thought.

I honestly was not aware that 4 beams could be combined--that way. I always thought that you would get 2 with the same polarization causing an interference reduction when setup like that, I.E. only 1 beam of two lasers. Of course this would occur only in the scenario of absolute stacked beams, or at least I had went on for some years believing as such. Light has 2 polarization states. How do we combine 4 lasers into 1 beam when there are only 2 states they can exist within?

Or am I overthinking this in that you are merely aiming 4 lasers at a distant point and it's not a truly combined beam?

To explain how I am viewing this problem I think of it like what follows:

Lasers A, B, C, and D.

In your setup I view it like you are adding together [A+B]L1+[C+D]L2.

A enters and passes, it is state 1.

B enters and reflects, it is state 2.

C enters and passes, it is state 1.

D enters and reflects, it is state 2.

Beams meet and now that I've just fully confused myself, L1 beam reflects one direction and L2 beam reflects the other--they are not combined.

Yes I really am curious now.

He's using two PBS cubes and a custom dichro. Some really narrow combine blue custom dichros came up in a GB on Photonlexicon a while back, good for combining ~445nm with ~465nm. So he's using a PBS to combine 2x ~445s and another to combine 2x ~465s, those beams are then combined with the dichro. There's no simpler way to superimpose 4x beams of a single colour directly on top of each other. There are other ways to do this, but they are more complex.

 

[Light superglue]

Hi Exerd,

What you write would happen if all 4 lasers were of same wavelength. Then yes, both combined beams would be reflected from the Dichto mirror into different directions. Same if I would try to combine two PBSed beams with a third PBS - the beams combined with 1st PBS in a row would be splitted again by the following PBS...

But here I have 2 pairs of lasers of "slightly" different WLs.
Thank you Diachi for explaining except here I combine 405 and 445nm beams because this Dichro turned out to be Multi Band (what was not expected at start).
So it is both good to combine 405+445 and 445+462nm.

 

[paul1598419]

Glad to see this thread as it confirms what I was trying to say in a different thread. To combine four lasers of the same wavelength is not possible using PBS cubes and dichros. And when you said that combining "slightly" different wavelengths of laser light, in your case was 40nm and 17nm.

 

[Exerd]

Thanks a lot guys. You made me realize, I should not read right before bed on some late nights. I surely sat there looking at Light Superglue's photo for some minutes like this: :thinking:

Starting off in the 405nm forum I had presumptions, and upon seeing "BDR-209" a few times, I jumped the gun before seeing "PLT5". I had gone on thinking this was 4*BDR209 diodes. As you can imagine, the idea that suddenly 4*405nm diodes could be combined and also that the way I viewed photon polarization had been wrong all this time... Well, you get the picture.

What's indeed still interesting in respect to this current discussion is the topic of wavelength regions for PBS setups. By this idea I start to wonder, how narrow of a band gap would be permissible (in theory) if one could obtain the diodes? If one already had the diodes, how narrow of a band gap will current-tech manufacturing/coating processes allow for dichro mirrors?

Those questions aren't entirely hovering around "Would it make sense to do?" or "Are the diodes readily available?" but rather around the opportunities that may be possible now or in the future with a bounty placed on the proper diodes. As an example, lets say there are '465nm' diodes and 2 will hold at 480nm while another 2 on hand hold around 474nm. 6nm is a rather small gap, but I simply wonder if such mirrors/cubes are obtainable.

 

[Benm]

The question would be 'at what cost'.

I suppose you could actually make a dichro that allows you to combine 470 and 480 light. The problem is the gigantic r&d costs to actually make one, as well as the precision required in manufacture.

That could boil down to a million dollars to develop the first one, and an additional cost of $10k for each additional one built. In theory that would work, but unless you have NASA budgets it would probably not be feasible for practical use.

 

[Light superglue]

Hi Exerd and Benm,

I do not know about 470-480nm but if you would like to combine 4x BDR-209 I think there is a way to do it on practice at a reasonable cost.

Cyparagon has measured some and found the WV to vary from 404 to 417nm. So if you try several and find a couple emitting at 407 or less and a couple emitting at 414+, then use 2 PBS cubes and this dichro beamsplitter with 412nm cutoff:

https://www.chroma.com/products/parts/t412lpxt

Not super cheap at 225usd (+vat/shipping I suppose), but big enough 24x36mm, so one with skills could cut it into 4-5 usable smaller dichros...

Of course the 1st couple will have to be kept cold...