20 Diode Modules

[planters]

I've been making steady progress with my projector over the last few weeks, but trying to incorporate extreme cooling, a two-D array and a so many diodes has added a LOT of steps. Except for the slightly different collimator thread specs.(for the O-like and the Optima lens mounts) the 445 and the LOC modules are identical.
The case grounding of the LOC diodes requires parallel (yuck) operation. In order to cool these diodes to -30C an isolation window is necessary and the cost of an optical window skyrockets beyond 75mmx75mm. and so the tight grouping of the diodes prevents practical thermal conduction and electrical isolation. I chose to use 10W, 30 ohm resistors to both balance as well as protect the other 3 diodes in any 4 diode strip where a diode to go dead short or open circuit. The setup allows easy replacement of a failed diode and so I am running them relatively hard at 350ma. The 4x100W TECs are actually able to bring the assembled head (minus the diodes) to -35C when 25mm of low density urethane foam sheets cover the surfaces and the stainless legs are wrapped in several thin layers of flexible PVC tape.
The 445 diodes will be run at a more conservative -10C but, will be driven at 1300ma to give a combined 28W output.

So far my only disappointment has been the retaining springs on the knife edge mirror array in that they bind as the 3mm set screws are adjusted beyond a fairly small range. Nevertheless, they are very effective and I have been able to adjust a diode spot by as little as 100u radians and it remains stable. Although, this projector is for a semi permanent installation and I wouldn't want to compete with a typical flexure mounting. Just today I've tried a pin and extension spring alternative to the retaining springs and this seems to be working much better!
And now some pics.

[dnar]

OMG that is BFG!

That is some resistor array! At 30 ohms, how is the driver performing and what is the driver input voltage? In theory, the resistor is now performing part of the drivers role, so I am keen to hear more about performance and modulation linearity.

I don't consider 350mA "pushing them hard" BTW.

[RedlumX]

Great mechanics!
But what's the point of running the 445nm diodes at -10C?

[flecom]

holy $hit!

[masterpj]

when I saw the title... 20 modules........I already was like woa...

This is beautifull.... please keep posting more pics this is a feast for my eyes.

[andy_con]

you have clearly put a lot of time, effort and work into building this and it look amazing. some fantastic workmanship

but i dont understand the purpose of cooling diodes that are rated to 70 degrees to -35?

[dave]

but i dont understand the purpose of cooling diodes that are rated to 70 degrees to -35?

To shift the wavelength down

[dnar]

To shift the wavelength down

Yep, and for that reason the 445's should be run hot, not -10c!!!!

[andy_con]

To shift the wavelength down

but how many nm would they drop at -35? maybe to a max of 650nm?

looks fooking awesome but seems a bit extreme

[dnar]

but how many nm would they drop at -35? maybe to a max of 650nm?

looks fooking awesome but seems a bit extreme

If I recall, it is in the order 0.3nm/C so could be around 642nm @ -35c.