UPD: My portable builds with NUBM31/36: Rocket, Radar and Boombox

[Light superglue]

It must be almost a year since I saw on LPF last story about making a big torch with 100W power. These things were started by Styro, then by Trinh and then Smackitup has posted pics of his portable working from 1200W driver. But since then, no new builds... until last week when Kcc posted a YT video of his gun with Osram MDP chip.

Many buiders must have run into same problems as me: crappy drivers which do not work, little space for batteries in any available torch body, no space for big enough heatsink etc...

Well, if there are no more pictures, I will start a thread about some new funny builds what I am working on from time to time, trying to combine them into a 100W torch by trial and error.

Recently I found that the head of a pretty big 21LED torch from ali (I had 3 of them) can accept the 250W driver from ebay. 250W is on the edge because the consumption of NUBM31 is already 250W but this driver was working fine with NUBM31, when I tried it many times before (40V, 6A output, 30 seconds on).

Then trying to find the highest voltage LiPo batteries of the smallest size I ran into these Gaoneng 6s 22.2V, 500mAh batteries which could fit into the tube of Fenix TK75 which I purchased earlier with 2 extensions in order to try making 24V battery holder from 8V original holders. But this was not easy, so I will try two of these these small LiPo instead. Here I arrange them in parallel to have 1Ah. If you do calcultions the 250W consumption could last up to 5 min with this capacity.

All assembled with NUBM31 on a big circular heatsink from ali with a fan behind, here is a pic. of the trial, it worked well.
But once I charged batteries up to 100% (25V) and finished the setup, once switch on there was no light but only smoke pushed by the fan from inside of the tube! Why does it always happen to my new builds???
So opened it again, and yes, the driver coocked! Funny that coocked part was the small transistor on the bottom and not 2 capacitors marked by red, which were heating during tests!

The story is to follow...

 

[Light superglue]

Coocked 2 drivers one by one: first was set at 6A, second at 5.5A, now installed the 3rd one set to 5A, what is 2.5A through each LD row.

And I decided that TK75 tube was too tight for the batteries with parallel adapter and thick wires which had to be pushed inside on their side. I replaced it by a wider transparent acrylate tube. Now one can see what is inside and it was also possible to make a hole for battery monitor.

Here is the size comparison with Maglite, where I previously installed 5 DTR modules with 1W violet multimode diodes.

5x GH04W10A2GC in a Maglite

Yes, me too always wanted to make a focused burner like this one: but regarding how much it would consume with all NUBM44 diodes I have first tought to build a smaller similar thing with lower consuming 405nm diodes. The original idea was to try with BDR-209 single modes because of the high...

laserpointerforums.com

One lens from Thor was glued to one more torch head, so the laser has now the separable focusing lens. Once focused the beam burns through 0.4mm thick stainless steel reflector (holes marked by red square).

 

[RedCowboy]

I like your heat sink and fan fit into a tube, very straight forward approach.

 

[WizardG]

So...you were planning to repaint the kitchen anyway?

 

[Mattronium]

Nice build!
Not to many builds using these due to price, and difficultly of design.
Definitely needs active cooling for any real runtime. What is the duty cycle on that? Can you get it to run continuously with the cooling method you've got or does it still get too hot? It looks like it should be able to run continuously.

 

[Light superglue]

This battery monitor has temperature input but I still have to replace the sensor by one which can be atached to the laser chip in order to monitor its temperature. Without this done I only tried laser for 30sec on, not longer. Maybe at the end I will have to replace this small fan by a faster one.
Meanwhile I finished another build in Falcon radar body which looks like the original build of Styro. This has smaller 80mm HS but much faster and stronger server fan inside, I will post some pictures tomorrow.

I cannot find the way of using active cooling even if this sure would be better for the duty cycle.
Once I mesured the TEC efficiency in an ice cream maker and it was transferring 30W while consuming 90W. So here we have to extract 150W of waste heat (if a proper driver was in this build set to 6A to make 3A through each LD), and a TEC capable of doing this would consume 450W (twice as laser itself), and the HS and fan should remove the total of 600W. Too big for a portable build I think...

 

[Light superglue]

This week finished another Steampunk build wich has same Al radar body as Styro used in his YT video.
Main difference from the Big torch is the use of 2 same drivers, each is set to deliver 3A at 42V, so the chances are high that they do not cook as fast as one set to 5-6A in the Big torch.

NUBM31 is placed on a 80 (or 82) mm circular HS and cooled by a super fast server fan (80mm, 24V, 6000rpm) which makes a noise of a jet engine when turned on. Hope it will cool well...
Both drivers did not want to work in parallel, so I had to use each one separately for 2 rows.

The battery capacity monitor on the back shows the voltage of LiPo and can also show the temperature of the chip once I receive thermal sensors ordered from ali. I also soldered a 30A fuse in the main circuit, so it can work as safety switch (taking it out or pressing it back in) in addition to a robust switch which is the trigger.

Once assembled it looks very tight full of wires inside but there is no bigger body I could find in ebay and ali up to now...

 

[Light superglue]

Here is the 3rd build which I will call Boombox. You sure know why - just take a look at this big black body from the Double Head XHP100 spotlight recently purchased on ali!

First, I found that a big Al HS taken from an ice ceam maker couple of years ago enters into this body really well. Then I attached two NUBM36 arrays to it.
Second, I tried to power it all with one BST900 driver which was fine for one array (250W input). But no, the real maximum I could get even using a 56V battery (L14) were 330W only.

After this I had to add the 2nd BST900 driver set parallel to the 1st one. Now the system could deliver 500W needed but 2 drivers worked not equally: you can see that one is giving 3.5A output while other - 2.5A. This makes 6A in total.
But at the end the spots on the wall are very close and the beams seem to be parallel enough, so luckily I did not need to use any steering mirrors here!

 

[Mattronium]

Nice, With 2 units you're sure cranking out the power.

 

[Light superglue]

It seems that many members who were interested in high power lasers yeas ago have disapeared from LPF and those who stay are more discussing politics than lasers today...

After making a Boombox with 200W blue output I have tried to upgrade it adding 2 more NUBM36 MDP and raise the total output to 400W!
The 800g Heatsink in it was for sure small for that (600W of heat to dissipate) but I suggested that beaming half a minute at full power should be possible.

The main issue was where to find the right driver and batteies which would fit into the body. This body was big for a spotlight but not super big like toutan 3D printed for his laser. Maybe two 6s LiPo will do the job and fit inside as well but first I did experiments with 56V Li batteries from a Leaf blower.
The "1500W" boost driver from ali/ebay which looks similar to what Shmuckitup used in his build can really give 1000W output - on the pic I test it with big resistors to give almost 12A at 83V. But it does not have easy to handle regulator, nor screen showing voltage/current/power, so I have tried to use a couple of DPX800S boost drivers. Their dimentions were just right to fit two into the Boombox body.

Do not trust that they can do 800W as given in specs! Already after a minute of outputting 500W (6A at 84V) the HS was very hot and the fan only turnes on when outputting power. At the end I purchased 4 of these drivers and in one of them the fan feed was not working, the fan of another one burned (making the fuse to pop) after less than one minute of outputting 500W!

After that, I made a separate 5V feed for the 2 fans,so they could be on all time the Boombox is turned on, to make sure that transistors are not overheated and on board fan feed would not fail again.

Here is the screen of one LCD when giving 6A at 80.5V for 2 of the chips. The both drivers work in parallel separately giving ca. 480W to the 2 chips each. All 4 spots are pretty close on the wall what means the beams well parallel.
30 sec on is maximum duty I tried - the thermal sensor on one of NUBM36 shows almost 50ªC after that. Then I wait HS fan to cool it all to rt and start again...

And it makes a lot of blue light!!!

 

[CurtisOliver]

400W of stable 450nm! Let's restate that. 0.4kW! That is insane. Excellent job! I can't believe you had this thing running at half a kilowatt for a short while.

And it makes a lot of blue light!!!

That was a British level of understatement there.

 

[farbe2]

Use something high end if you want a small driver that lasts.
hint hint ColorDrive 6: Link

 

[Light superglue]

Thank you Farbe, maybe later I will try it.
Back in 2016 I purchased blue-blue dichroics from them. Do you know if they still have those in stock?