The brightest single blue beam (4 diodes combined)

[Light superglue]

Yes, Alaskan, I know this pain, it is very difficult to find a decently priced host for more than 10W ouputting pointers.

One either has to machine himself (not my case, I have little tools) or just take biggest available torch and try to design devices to fit in.

But would not you agree that maybe an issue would be to switch from torches as hosts and pay attention to the currently growing choice of Li-ion hand tools like headlights, drills and blowers? Big heatsink is great but enough space for big HS and big fan may be even better...

 

[Alaskan]

... Big heatsink is great but enough space for big HS and big fan may be even better...

I really like my TK75 for use as a host, I believe this flashlight is better suited to the higher power builds which need a big heat sink than many others, but that particular flashlight is very expensive, so few will go that direction due to that.

I don't disagree with your statement adding a fan to a heat sink to get more heat dissipation is better than just using a larger mass of metal, using a fan causes any heatsink to be far more effective, reducing the size of the heat sink by half if it has fins is probably easy to get when using a fan. If the heat sink is designed for the fan, you can get as much as a 5:1 improvement, that is huge!

 

[Light superglue]

Hi Alaskan,

High price for a right host may not be the highest part of the total cost of the assembly I have in mind.
Only parts to be in use would add to 600-700usd not counting experimental setups needed to chose/test right LDs, tools to buy, parts to break/discard etc.
But the most disgusting thing is waiting a month for every small piece from aliexpress/ebay to arrive and find out it is not exactly what was on picture...the other side of cheap price.

There is a video in Youtube "Blue laser handle 10W" where in comments the guy (from Vietnam) writes that he sells kits for 150 or 200usd apparently ready to install 2 NUBM06 and a PBS cube. The handle on video looks to me similar to TK75, maybe he machines his hosts from those?

Are the cells in TK75 connected serially? I see the head has the right 88mm diameter, but how high current can flow in this body? Is it possible to combine 2 hosts into a one host with 8 18650 cells? I do not think it can accept 26650s like the flashlight I found in aliexpress but the handle will sure be much shorter with such cell packing.

 

[Alaskan]

The TK75 has optional battery extenders which cost over 30 dollars each, each extender has a battery holder to hold 4X 18650 batteries which are wired in series-parallel, for each extender, which doubles the voltage to 8.2V and you get the current rating of two batteries, instead of four, in each extender.

Adding extenders increases the capacity, not the voltage, as you would get from adding more batteries in series in a regular flashlight without this special series-parallel battery holder. If you buy 3000mA rated 18650 cells, you get 6000 mAH per extender, each time you add another one that much more. Again, the voltage never gets higher than 8.2 V peak no matter how many extender sections you add. You can look at some mods done to the battery packs which go in each extender to increase their current capacity, CPF has members posting mods to the battery packs there.

I saw over at CPF that someone modified the battery holders to be able to pull up to 25 amps at 8.2 VDC. Here is a review of the flashlight:

Review: Fenix TK75-2015: 4000 lumens, beamshots, compared (also)to TK75-2900 version

Only thing is, there isn't a on off button on the battery cap, only a small tiny low current momentary switch under the on-off rubber button which would need to be wired to the driver to turn it on, some drivers allow this using the TTL control. That switch might only be good for 100-300 mA or so, it would need to be wired to TTL control which is very low current.

Link to higher current mod, but it is in German, good photo's though: http://www.taschenlampen-forum.de/threads/fenix-tk75-3xxhp50-modding-beratung.44245/

 

[AaronT]

Now this is why I love sharing my ideas on this forum, you guys jump in and make them better.

In theory this arrangement would be much simpler to construct/align as 2 modules, one on one block of the heat sink sandwich and one half on the other, then align the halves to make the 28 watt or so module.

 

[diachi]

Now this is why I love sharing my ideas on this forum, you guys jump in and make them better.

In theory this arrangement would be much simpler to construct/align as 2 modules, one on one block of the heat sink sandwich and one half on the other, then align the halves to make the 28 watt or so module.

Your drawings are much better than mine! That certainly is a good chunk smaller, as expected. Maybe not quite half as wide, but close.

You could keep it around the same size and throw in an additional bounce mirror to help align the near field on the dichro (As shown in my other drawing). May be an extra step but it'd be way less frustrating getting it to work.

Just align both PBS cubes, stick them side by side and align the dichro, adjusting the near field with the third bounce mirror rather than moving modules and messing around with further PBS alignment.

Extra cost of course, but it'd be what? $20-40?

 

[AaronT]

Your drawings are much better than mine! That certainly is a good chunk smaller, as expected. Maybe not quite half as wide, but close.

You could keep it around the same size and throw in an additional bounce mirror to help align the near field on the dichro (As shown in my other drawing). May be an extra step but it'd be way less frustrating getting it to work.

Just align both PBS cubes, stick them side by side and align the dichro, adjusting the near field with the third bounce mirror rather than moving modules and messing around with further PBS alignment.

Extra cost of course, but it'd be what? $20-40?

I was thinking once each pair diodes is aligned I would mount the dicro to one and align, then I would temporarily mount both arrays to one of those multi axis mount cubes and use that to get everything lined up.

I would be aligning the lasers from start to finish in a calculated position in relation to the blocks/sinks I am gluing them to. At least to get "rough" ideas of position so the little multi axis adjustable cubes can fine tune everything.

The trick will be finding a thermally conductive adhesive that won't cause the components to shift as it cures while the components are temporarily held in place with the adjustment cubes.

The first step would still be getting my "theory of assembly" to work for just a pair of diodes.

I have a NUBM06 powered Phaser project to finish first. I will do at least 1 or 2 more single diode builds (HL63193MG/10440 microbuild, and a NUBM44/2x26650 lightsaber) then I will attempt to construct a 14 Watt Array.

 

[diachi]

I was thinking once each pair diodes is aligned I would mount the dicro to one and align, then I would temporarily mount both arrays to one of those multi axis mount cubes and use that to get everything lined up.

I would be aligning the lasers from start to finish in a calculated position in relation to the blocks/sinks I am gluing them to. At least to get "rough" ideas of position so the little multi axis adjustable cubes can fine tune everything.

The trick will be finding a thermally conductive adhesive that won't cause the components to shift as it cures while the components are temporarily held in place with the adjustment cubes.

The first step would still be getting my "theory of assembly" to work for just a pair of diodes.



I have a NUBM06 powered Phaser project to finish first. I will do at least 1 or 2 more single diode builds (HL63193MG/10440 microbuild, and a NUBM44/2x26650 lightsaber) then I will attempt to construct a 14 Watt Array.

Yep, that's the way to do it for sure - that's how the lab lasers (that don't have adjustable mounts) tend to go about it.

If you're just using the adhesive to adjust the optics it doesn't need to be thermally conductive, only if you're using it to attach the lasers themselves to the heatsink. UV cure epoxy is a great choice for attaching optics and can be cured quickly using a 405nm laser. Avoid super glue, it likes to leave a film on nearby optics and will make them useless.

Personally I'd use screws and threaded holes to attach the lasers where need be (with some regular thermal paste of course) and then use the UV cure epoxy to attach the optics once aligned.

Hope that makes sense and I'm on the right track?

 

[AaronT]

Yep, that's the way to do it for sure - that's how the lab lasers (that don't have adjustable mounts) tend to go about it.

If you're just using the adhesive to adjust the optics it doesn't need to be thermally conductive, only if you're using it to attach the lasers themselves to the heatsink. UV cure epoxy is a great choice for attaching optics and can be cured quickly using a 405nm laser. Avoid super glue, it likes to leave a film on nearby optics and will make them useless.

Personally I'd use screws and threaded holes to attach the lasers where need be (with some regular thermal paste of course) and then use the UV cure epoxy to attach the optics once aligned.

Hope that makes sense and I'm on the right track?

This

That you guys have shown me today has inspired the following.

One solid copper block drilled to take 9mm diodes with leads and designed to have aluminum cooling fins bonded if desired

After testing each diode (fixed focused) and locating the beam in exact orientation to the diode the diodes will be pressed at spec into the block.

Optical components would be glued directly (with gap) to the diode/block assembly.

Red indicates adhesive.

I'm thinking a 2x26650 host modified to take 8x10440s (Efest IMR 350mAh High Drain). 2x10440s each in its one tube providing power to the driver+external pot X 4

Cylinder with a shelf holding the diodes/optics threaded at the base and drilled inside to hold drivers, holes on outside so a "key" could be used to adjust the output of each diode 0-100%.

Obviously this is a short run time device but off the top of my head the smallest most powerful build using off the shelf components and one spec machined cylinder of copper.

Of course just because you could run it at 26 Watts @ 100% power doesn't mean you have to. But having the option in such a portable device...

This gives me mad scientist giggles.

 

[Benm]

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

Haha.. that would indeed solve any heat problem quite quickly, though the air it blows out also cools the motor so it will be a bit above ambient temperature.

A leaf blower as a casing would seem insane, and for diode lasers it probably is.

It could be a realistic option to consider for those wanting to constuct a portable CO2 laser though.

For those that consider scuba lights as hosts: Beware that some scuba lights are only capable of running at full power for any length of time when submerged. They will overheat on dry land quickly. Obviously if you use them as laser hosts you don't need to deal with the full power of the orginal LED light installed, but still be very aware of this.

Proper high power diving lights usually do not have large fins on them, but simply rely on water flow to remove the heat (which works extremely well). Brand name lights will often have a warning that you should not use them out of the water, and/or have some thermal overload protection to dim them if you do anyway.

Also LED's are comfortable with much higher temperatures than laser diodes, so if you get something like that 30 watt led light, expect the LED in that to run fairly hot (maybe 70 celcius die temperature). This does not damage the led much, but could be very uncomfortable for your blue diodes especially when already overdriven.