Advice on 26 Watt Array

[AaronT]

If I sandwitched this array (All components are to scale, 12x30mm diode modules used) between two 2in X 2in X 1/2in aluminum blocks 0.2 lbs/ea

Would that be enough sink for 2xNUBMb44 + 2xNUBM06E?

 

[RedCowboy]

Two 2 inch by 2 inch both 1/2 inch thick top and bottom correct? And this for 4 diodes such as 44 and 06 running 4.5 amps each?

If so it would get too hot in under 2 minutes, and as warm as you would ever want it to get in about 1 minute. Estimated.

That's about the amount of mass and surface area, well a bit less actually than I run 1 nubm44 in for 5 minutes.

4.5 amps at 4.7 volts in to each diode is 21.15 watts into each diode to get 7 watts out before the lens, so 56 to 60 watts of heat with lens waste and drive heat or a solid 55 watts of heat just in diode by product.

Imagine 11 x 5 watt lasers all pointed at your heatsink without any energy reflected, that sucker is going to get hot fast.

On top of your plates you need some thermally bonded fins and a small fan blowing across them.

Here's two 2 x 3 x 1/2 plates attached to a 1/4 inch plate with some small fins and just 1 nubm44 gets this hot in 5 minutes.
The reason is these diodes are much more efficient at factory spec of 3amps? but at 4.5 they lose efficiency.

These nubm44 and a nubm08 run for 4 to 6 minutes before they are getting good and warm, but that's 1 diode each, the small is a 2 inch wide round bar 2 inches long the large is 2.5 by 2 and that adds 2 minutes but it's a 4.0 amp nubm08 and the long is a 2x2 round bar with a 2.5 OD 1/4 wall sleeve 6 inches long and has a 6X pair, it's good for a solid 5 minutes but it gets hot after that except during winter I can run longer, the heat just keeps building up without forced air and surface area multipliers such as fins.

I had 3 NDB7875's in a 2.5 x 2 inch round bar all at 2.4 amps each and it was really hot in 2.5 minutes or so.
Multiple overdriven diodes are going to make a lot of by product heat, just mass alone won't give extended duty cycles, not reasonable mass, not for multiple diodes.

 

[AaronT]

1 sec bursts every 5 sec sustained.

1 sec burst every other sec for 2 min than 5 min cooldown.

1 min constant and 5 min cool.

1/2 lb diode unit, 1/2 lb worth of 18650s

I wonder how many you could knife edge before the beam would be more or less square...

I also plan to use BlackBuck drivers in their own sinks, maybe driving pairs of diodes. Would give adjustable output.

Possible application https://www.reddit.com/r/EmDrive/comments/42uzo7/opticallaser_emdrive_revealed/

Would also be building one of these.https://en.wikipedia.org/wiki/White%E2%80%93Juday_warp-field_interferometer

I think I can enter the warp drive race...

 

[RedCowboy]

Give it a shot, I will love to see it burning :eg:
But sustained CW power is what gets stuff ablaze at range, that and a steady aim.
It's tough for me to just do a little, I would recommend some fins and a fan so you can do a couple minutes of cw fun, but I hope you can make it work, I have really hit heat load amazement with multi diode attempts so far, even with heavy sinks.

 

[Encap]

Copper for sink material with Aluminum fins if using fins might be a better way to go.

If you are going to all the trouble and expense --a liitle bit more for copper is not going to make much difference.

How much better I don't know but copper has twice the thermal conductivity and the AL fins would lose heat quickly

Just a thought

 

[RedCowboy]

This has been talked about before, and copper draws heat away faster but seems not to shed it into the air as fast as aluminum, so your concept sounds right to me Encap.

I am not an expert, that's just how it seems and what I think I remember reading.

I know styropyro has run 8 diodes with a big fin but I don't know for how long.

I know the projectors run at a lower output that is more efficient and use heat pipe to fan forced radiators of fins, early models were fins with fans, but again not overdriven.

 

[ElectricPlasma]

Fans? Air cooling? Nobody gunna mention it?

Even one fan will go a lomg way, using fans I've easily doubled my runtimes, sometimes tripled. Use one for intake and one for heat output on each module and youll see a difference.

 

[RedCowboy]

Fans? Air cooling? Nobody gunna mention it?

Even one fan will go a lomg way, using fans I've easily doubled my runtimes, sometimes tripled. Use one for intake and one for heat output on each module and youll see a difference.

:thinking:
I recommended fans and fins in posts 2 and 4

 

[ElectricPlasma]

Oops, my bad :crackup:

Only read the first couple posts.

 

[RedCowboy]

I have done the same thing before and likely will again. :beer:

 

[Trevor]

From the diagram it looks like you're using a pair of PBS cubes to combine the two pairs of diodes. What optic are you using to combine those resulting beams?

Trevor

 

[AaronT]

From the diagram it looks like you're using a pair of PBS cubes to combine the two pairs of diodes. What optic are you using to combine those resulting beams?

Trevor

A Blue/Blur Dichro mirror.

Here is a labeled diagram.

 

[Trevor]

Have an already picked an optic? A dichro that will do what you describe is a pretty tall order... 100% transmission and 100% reflection ~10-15nm apart.

...then again last time I actually paid attention to people doing multi-multi-multi-diode builds, all that was happening was PBS + knife-edging.

Trevor

 

[AaronT]

Have an already picked an optic? A dichro that will do what you describe is a pretty tall order... 100% transmission and 100% reflection ~10-15nm apart.

...then again last time I actually paid attention to people doing multi-multi-multi-diode builds, all that was happening was PBS + knife-edging.

Trevor

http://laserpointerforums.com/f52/blue-blue-dichro-live-lasersystems-combining-445-462nm-98517.html

It does better than 95% reflect/transmission at only 20nm apart.

Each 26 Watt module would weigh about 1/2lb.

It would fit on one of the 6D Cell Maglite for a host.

I'm thinking about a Multi (Dual NUBM44) Multi (Paired NUBM06Es) Multi (Combine the 44 and 06E pairs) Multi (26 Watt single beam "modules" into a knife edge array)

 

[Trevor]

http://laserpointerforums.com/f52/blue-blue-dichro-live-lasersystems-combining-445-462nm-98517.html

It does better than 95% reflect/transmission at only 20nm apart.

...impressive. Will wonders never cease!

Trevor

 

[Light superglue]

Hi Aaron,

Just a suggestion:

For your purpose I would tightly wrap modules in as much as possible of some of these:
https://www.aliexpress.com/wholesal...=SB_20160919111605&SearchText=tube+clamp+12mm

and sandwich between those:
https://www.aliexpress.com/item/771..._2&btsid=1c72bb8b-ea1a-4b9c-a0f0-ca78debd6602

Do not forget the dichro has 3mm thickness (is not just a thin line).

And I would not keep to mainstream hosts, batteries and drivers but find something special.

Sorry I am not an expert in warp drives, but do you need collimation for it? Will an Nichia array alone or Phlatlight LED not be good enough?