7W 450 nm handheld build

[Buffo]

So, yeah, I know it's irresponsible (some would say reckless) to build a hand-held laser with this sort of output power, but it was something I've always wanted to try.

Huge thanks to Rich (Lifetime17) for his assistance with the custom-machined host. He had to do a good bit of extra machining to accommodate the large 25 mm copper heat sink module that I selected for this build. If anyone is considering a build like this, speak with Rich first! His work is first-rate, and his prices are very reasonable. The host is ~ 25 mm long and holds two 18650 Li batteries in series.

I went with the NUMB44 (V2) diode for this project. I got it from Jordan at DTR laser. (Another great guy to deal with! Fast shipping, and never any problems.) I'll freely admit that I sort of cheated here in that I ordered an all-up module with the diode, driver, and G2 lens already installed and the current pre-set for 5 amps. I've worked with bare diodes before, but for this project I was in a bit of a rush and wanted everything to be plug-and-play when it arrived.

I selected the 25 mm copper heat sink for the diode, although if I had it to do over again I think I would have gone with the smaller 20 mm module. In my opinion both are overkill for a handheld, but the 25 mm unit is really big and it forced Rich to do some extra machining to make it fit in the host I selected. For sure I won't ever have any overheating issues though!

I'm very pleased with the results of the build. Of course, because there's no secondary correction optics, the fast axis of the beam rapidly spreads to a line after a few meters. But up close (less than 25 cm) the beam is pretty much lethal even before you try to focus it - anything remotely combustible will immediately begin to char at close range. This is one of those projects where at the end you sort of chuckle nervously to yourself as you wonder if maybe, just maybe, this wasn't such a good idea after all.

Now that I've gotten the itch out of my system, I don't think I'll build any more handhelds. Truthfully this thing is impractical as hell - both because of the extreme danger due to the output power and because at anything beyond close range the beam quality rapidly degrades due to the lack of fast-axis correction. But it's still fun to fire it up and marvel at the fact that it's making north of 7 watts and it's a handheld! Part of my brain is still scolding me for being so irresponsible, but another part is cheering about the memory of smoke rising from one of the fence boards when I first tested it.



Here's a video of the brief test that followed the picture above:

Adam

 

[Alaskan]

LOL, why such low power? PBS combine two! That's what I have to say about being reckless Poor fence, had no defense. Nice work on that baby, great looking host.

 

[Buffo]

Thanks for the compliments! I agree that the host looks fantastic. Can't say enough good things about Rich's work. (Lifetime17 here on the forum) He had quite a few designs that caught my eye, but I finally settled on this one.

It's got considerable heft to it, due to the large heatsink and the solid copper lens adjustment ring that he machined to match. But all that extra thermal mass means that it has no problem operating continuously for longer periods. I've never gone past 30 seconds or so (no real need to!), and it's never gotten more than slightly warm to the touch.

A dual-diode PBS arrangement in a handheld build would indeed be a challenge! Although I think that before I attempted something like that I would first try to implement some sort of secondary optical correction method for the fast axis. But that's a project for the distant future - if ever.

Adam

 

[Lifetime17]

Hi,
Glad the build came out for you, dont be ashamed about the high output in a handheld unit. They are always a treat when you fire it up and see something sizzle. Steak anyone ??? Vey impressive greta job and the match with that25mm sink from DTR is the bomb. Thank You for sharing this build, buddy.
Rich

 

[gazer101]

I'd love to see someone build a handheld chemical laser: https://en.wikipedia.org/wiki/Gas_dynamic_laser

 

[Snecho]

That looks sexy. Really nice buiid, and of course amazing host from Rich!

Which lens did you end up going with?

 

[Buffo]

I decided to go with the G2 lens, as it should give the best performance for close-focus applications. Since I wouldn't have any fast axis correction I knew the beam quality in the far field would always be poor, so I figured my best option would be to optimize the near-field performance.

But to be fair, at this power level you really don't need to worry too much about focusing - at least in the 5-10 inch range anyway.

Adam

 

[RedCowboy]

Wow that copper head looks nice, good job and happy/safe lasing.

 

[Shakenawake]

You think that's big? Lol. my numb44 build has a 2" copper heat sink. Even then overheating is still a possibility

 

[gazer101]

You think that's big? Lol. my numb44 build has a 2" copper heat sink. Even then overheating is still a possibility

At some point heatsinks stop being effective and you need to introduce heat pipes and radiators into the mix XD

 

[hakzaw1]

I am still 'giddy' over my 1W405 in a 'Sinner' solid copper host.
its all heat sink. Its a 'keeper' IMHO
TY Riccardo--

 

[Cyparagon]

At some point heatsinks stop being effective...

This is a ~12W heat load, my dude. That's nowhere near any practical limit. For comparison, A pentium 2 has 3 times that power dissipation and still used a heat sink with no fan or heat pipes.

 

[gazer101]

This is a ~12W heat load, my dude. That's nowhere near any practical limit. For comparison, A pentium 2 has 3 times that power dissipation and still used a heat sink with no fan or heat pipes.

True, but I guess what I'm trying to say is that at some point you'd get better form factor from a radiator system than from a large block of metal. Obviously that isn't possible within a handheld laser, but yeah

 

[Buffo]

Wow, a 2 inch diameter heat sink does seem rather excessive! Bet it makes for a heavy unit too!

I agree that there is a point where just adding more thermal mass to the heat sink ends up being counterproductive, but in this case the diode is never operating long enough to saturate the heat sink anyway.

And yeah, if I wanted to build another handheld that was capable of a longer duty cycle I would probably machine some fins into the circular heat sink rather than merely increasing it's diameter.

There is one advantage I can think of to having a larger heat sink though: a 2 inch unit might just give you enough space to cram a set of cylindrical lenses into the front end to correct the fast axis. That would yield a superior output beam! But it would also be a ton of work to design and machine the custom mounts, and aligning it would be a real challenge. Definitely not something I want to attempt...

Adam

 

[Snecho]

There is one advantage I can think of to having a larger heat sink though: a 2 inch unit might just give you enough space to cram a set of cylindrical lenses into the front end to correct the fast axis. That would yield a superior output beam! But it would also be a ton of work to design and machine the custom mounts, and aligning it would be a real challenge. Definitely not something I want to attempt...

So like a built-in beam expander?

 

[Buffo]

A beam-expander in one dimension only, yes. This would be positioned after the primary collimating lens.

To further clarify...

Most high-power multi-mode diodes have a profoundly rectangular beam where one axis (the "fast" axis) diverges at a much greater rate than the other. Interestingly when the beam is first collimated using a standard aspheric lens the fast axis is actually the narrow portion of the beam. But because the divergence is so much worse on the fast axis the beam width on that axis rapidly overtakes the beam width in the other direction, and within 10-20 cm the fast axis becomes the wider part of the beam.

To correct this problem you need to quickly expand the fast axis and then re-collimate it. (Expanding the fast axis will reduce it's divergence while leaving the other axis unchanged.) The goal is to bring the fast axis divergence down to match the other axis, but often this results in another rectangular beam profile, except that the wide portion of the beam is now the slow axis. So you still have a rectangular beam profile, but now it has been rotated 90 degrees and it should have more or less equal divergence on either axis.

Choosing the power of the secondary correction optics and the distance they are positioned from the primary collimating lens is always going to be a trade-off between having better divergence matching on either axis in the far field vs having a more square-looking beam profile in the near field.

Commonly this sort of fast axis correction is done with a pair of cylindrical lenses or a pair of anamorphic prisms. Nearly every commercial laser module that uses a multimode diode has one of these two solutions inside to correct the fast axis divergence. And of the two, the pair of cylindrical lenses is much more compact, so that's probably the best route to take if someone wanted to build this into a hand-held. But even so it would be tricky indeed to shoehorn this into the business end of a pointer, not to mention the difficulty in aligning it. Also it would be nearly impossible to adjust the primary collimating lens once you added the secondary correction optics so you would not be able to focus the beam to a pinpoint anymore. (Remember that the secondary correction optics must remain aligned with the fast axis, which means they can not be allowed to rotate with respect to the diode.)

But if someone did manage to build something like this, they would end up with a high powered hand-held that had the same beam quality as a commercial laser module, which would be quite the achievement.

Adam