The brightest single blue beam (4 diodes combined)

Yes I know that some hero members here have done videos with up to 200W blue beams long ago but those consisted of multiple knife-edged beams and I wanted to try producing a single beam where output of 4 LDs would fit perfectly on the same optical axis.

This year I have assembled 2 optical benches: one with 2 PBSed NUBM44 diodes at 4.5 and 5A and other with 2 PBSed NUBM07E diodes at 4.5A each.

Before disassembling them for other project I thought it would be fun to combine the beams with my new 445-465nm dichro mirror and look how bright it will be (I expected at least 18-19W beam).

Of course it was impossible to perfectly combine 2 separate benches by just positioning them correctly on the table, but at least in one dimention (horizontally) I have succeeded, because I needed both beams to enter 3x cylindrical beam expander at the same point to stay combined after expansion.
I could not reach exact vertical alighnment - so at 5 meters distance 2 round spots were visible one exactly under another, but on picture attached the beam looks great, does not it?

The optics in use here are: one prism mirror, 2 PBS cubes, 2 waveplates, one B-B dichro and 2 cylindrical lenses. One of 07E has original Ball-lens, 3 other modules have G-2 collimators.

The LPM I have capable to mesure this power is not calibrated, so you have to trust it with a coefficient 0.7-0.8 but I think 19W should be there even if as you can see a lot of light is reflected by many optical pieces in use all around the room.

By the way, can anyone (better with residence in Europe) help me in calibrating this LPM in 10-100W range by comparing to a calibrated one?

Mmmmm..Isnn't the alignment of multiple beams...well..ENOUGH TO SET YOUR HAIR ON FIRE !!.....BUT...well the result is wonderfull ~~ Thanx !! Keep up the sharing !!! It is somehow ...comforting to know that others...feel my pain !! HAHAHA Beam Out !

Awesome, please put that into a pointer!

I was thinking of that.

Searched at aliexpress for a host with biggest head and found this:

https://www.aliexpress.com/item/Alo..._2&btsid=08b4866d-6714-48c6-9e90-138f9a5c4063

The head diameter is 88mm! Could bear a baseplate with 4 12mm modules and optics installed, or? In a way like ballista was produced by CDBEAM and Minamoto...
But, to one of similar flashlights there was a comment from a german user (they are very demanding customers and good in mesuring), who said that the highest current he mesured in the purchased torch was only 1.5A. What means all seller´s specs are exagerated.

https://www.aliexpress.com/item/NEW...id=f795a9e6-7f85-41f3-b307-d379adbf0894&tpp=1


So step by step. First I want to try making a drawing how to set up all the parts on a 70-80 x 120-150mm baseplate (do not think it can be smaller - it will have to include some mirrors with mounts as well).
And Sanwu expander maybe as the final touch for beam shaping.

EDIT: fixed the links but there is no that comment now, maybe they erase critics from time to time...

Hmm... I can't get your links to work for me, expired so quickly? CDBEAM777 makes extraordinary stuff, he is working on a Cyan Cannon for me now. I used to have Ehgemus build me custom laser pointers for me with large 3 inch diameter tubes to mount big PCX lenses in for expansion, but he is making more money machining tops now, so got out of the laser host business. If the beams are on-axis, an expander is easy to make, the relatively high divergence of that particular diode really needs at least 8X, it has about a 12 mRad divergence using a 6mm lens.

I have fixed the links but that comment with critics seems to be removed by now.

If current is not high enough maybe from 2 hosts one twice longer can be made with higher voltage instead of current (host sticks on some pictures look modular consisting of 2 threaded halfs) - so a "super long maglite" with 8 batteries (>30V) could be made to drive 4 LDs in series using a 30 to 20V downstep driver???

I am thinking to use 6x cylindricals and 3x Sanwu after that like RC said was working by him for long range burning.
And 6x expanded beam will not pass into small 10x Jetlasers expander entrance hole anyway.

I saw the flashlights, those are impressive to me, just what the doctor ordered for what I like to do too, and 4 X 26650 (or 8 X) would be the way to go. You could use what I will call a pre-regulator, any kind of voltage regulator ahead of the constant current regulators for the diode to drop part of the extra voltage to get down to the range you need for the laser diode regulators. I have considered doing so for a project myself.

Wow, that's a heck of a lot of light all combined.

I have my doubts about making that portable though as it also produces a lot of heat. I presume you'll need to resort to fan cooling to get any usable run time on this?

Here are a couple examples of how small of an array you could build if you were willing to limit the device to 12mm modules. Everything is within a mm of scale.




I'm was planning on sandwiching the assembly/s between a pair/s of 2"x2"x 1/2" copper blocks and use copper shims to get everything in the right place.

Assembly would require the use of those little adjustable mount cubes to get things aligned before "gluing" the parts in place and removing the "cubes"

I guess for a host you could use an 6 cell maglite.

Aaron, I don't understand how those cubes will work like that, what does that blue rectangle represent? Are those PBS cubes?

Benm said:

I have my doubts about making that portable though as it also produces a lot of heat. I presume you'll need to resort to fan cooling to get any usable run time on this?

Click to expand...

It's relative to what is an acceptable weight for your laser pointer, for me 3 pounds of heat sink isn't so bad if you can get that much light out! Using a fan will allow the use of a smaller heat sink, so that's a good idea, save the weight if you can, but the heat sink should have fins to help exchange the heat more efficiently at that kind of power. I write this after building a portable multi-diode laser which weighed about 60 pounds all together, using a 12V auto radiator fan to keep it cool, so 2 or 3 pounds of heatsink ain't nothing to me now At least you can pick the thing up and hold it for a few minutes, the monster I built is almost too heavy to put in the back of my truck without trouble.

Adding a photo:

Here is a TK75 flashlight with a ~3" diameter heat sink (3.25"?) from a 100 watt 50 ohm RF "dummy load" attached to it. Sure, it is heavy, but it will do the job of cooling a lot of power, the challenge is whether their is enough room in the solid section which couples with the flashlight head to mount all of your laser diode assemblies and optics, that and having someone machine a slot to mount them in and bore a hole through the center for the output. You could slice off the finned section and add it to something else with all of that built into it, joining them together somehow but that reduces the heat transfer some between each side, although with that much surface area, not a problem, finding a machinist who can do a good job is the issue, and cost, of course.

Would be easier to make something ugly and hide the thing inside a large diameter tube, if you have a way to get the heat out. I had considered using a 3 inch diameter piece of solid aluminum round stock, slicing 75% or more of its length in half as a ledge to mount my laser diode and optics on, then put the ledge portion of it inside a tube. The mass of the aluminum itself will sink a lot of heat before it gets too hot, but better to have part of that rod finned to help some, where the tube isn't covering it. For that configuration, the fins would be on the back side instead of the front, and one piece as well as far more straight forward and cheaper to have machined. Not saving anything using a surplus dummy load as in the photo below, but it does look good.



The TK75 has a long battery tube because it has a total of four extensions on it, each holding 4X 18650 batteries in series-parallel to produce twice the voltage as well as twice the mAH capacity of a single battery. Due to the way each additional extension is wired in parallel with the tube in front of it, instead of series, with this flashlight you can keep adding battery extensions and the voltage remains at 8.2 VDC peak for fully charged cells, but the AH capacity goes up with each additional section. Using 3000 mAH 18650 batteries and four tubes, the total capacity is 24 amp hours at 8.2 VDC peak. Better have a lot of chargers, that's 16 batteries.

Here is something Ehgemus made for me for that same flashlight, it is made from a piece of 3 inch aluminum round stock with a flat 3/8" thick piece of aluminum flat bar stock machined to bolt into it. This did not require the mass of the heat sink I was writing about above for higher power, just to hold a knife edge assembly and optics. All of this hidden inside a thin walled aluminum tube.

Hi Benm,

Do you think this host will be good enough - it has fan for cooling installed!

https://www.amazon.com/DEWALT-DCBL7...&qid=1479152109&sr=8-1&keywords=dewalt+blower

Unfortunately this model is only cheap in US but not for us Europeans...
Alaskan, you sure would like its impressive size!

Hi Aaron,

Both of your layouts are bigger than maglite head (total sizes). And how do you want to cover all module holders (grey squares) with 2"x2" copper plates?
I think 70x120mm is absolute minimum for the (one or two) baseplates to touch module holders completely, that´s why host head has to be >80mm in diameter...
But I know premade Alu 15mm high "bricks" with 12mm holes which you could use for your assembly. I have tried them, these really have close to 12mm holes, not 12.5mm like most of crappy module heatsinks on ebay.

https://www.aliexpress.com/item/Fre.../1704546930.html?spm=2114.13010608.0.0.xZDVkf

I do like power, thus I need big heat sinks, just trying to help out, what I shared took me a long time to figure out, or find a good flashlight for such a project.

Alaskan said:

Aaron, I don't understand how those cubes will work like that, what does that blue rectangle represent? Are those PBS cubes?

Click to expand...

Alaskan,

Two 445s combined with PBS and two 465s combined with PBS, then the output of each of those combined by a 445/465 dichro. The dichro is the blue rectangle, PBS cubes are the cubes the line going corner to corner. I'm guessing the gray part on the cubes is supposed to represent a waveplate?

That design can be made smaller, or at least, not as wide. Throw in an additional turning mirror and you can have all 4 modules side by side. Alignment if the dichro will also be easier with my suggested layout. My paint skills aren't great, let me see if I can come up with something.

The just don't look like they will work in that configuration. I need to see ray diagrams to show what is going on with those, I understand using wave plates and how the cubes work, but things don't look lined up right at all.

Edit, I finally see what he is doing, duh. I don't know why I couldn't see it before.

Alaskan said:

The just don't look like they will work in that configuration. I need to see ray diagrams to show what is going on with those, I understand using wave plates and how the cubes work, but things don't look lined up right at all.

Edit, I finally see what he is doing, duh. I don't know why I couldn't see it before.

Click to expand...

Yeah, be easier with the beams drawn in.

Here's my crude drawing of an improved setup (IMO) for fitting in a portable host.



Black squares are laser modules, light blue is PBS cubes, dark blue is the dichro, gray for first surface mirrors, red is the beam path. Left out the waveplates but you get the idea.

Host would be the same length but nearly half as wide.

In theory Aaron´s configuration might work because all is there like in my setup (except cylindricals), but optomecanical part is not clear from the drawings.

And this is where most of issues with portable can be (baseplate may be deformed, glue debonded, PBS and mirrors displaced, screws loosened...). But let´s see, maybe he will succeed in assembling this at the end. Me also had to reposition all parts hundred of times before it started to work...

My way was trial and error, even if I pretend to know the theory a little, something always went wrong (such as "unexpected" need for rotational adjustment of cylindricals).