Is my SF-AW210 dead?

I'm using a LM317-based current driver circuit from this post. Has worked great so far. Current and voltage are measured (as opposed to theoretical).

Maybe it's the diode. I bought that batch of 5 on eBay and the first one was DOA. :banghead:

Thanks for the link. I'll check it out.

My bad, I forgot. You even have a nice pic of your setup and everything. The "1-1.5W" threw me off, huge range. If it's the common 9mm diode use 1 ohm w/ the lm317 for 1.25A max power, 2.5 ohms for 500mA roughly 400mW and long duty cycles. Heatsink the lm317 at these high current levels because when it gets hot it turns down the power and could be damaged. If you measure the amperage you'll know. I have no idea how tough those diodes are/ how hot they can handle befor led'ing= duty cycle. Good module heatsink should be a good amount of time though

I'll take a pic and post of the new driver I built for this one (moved it off breadboard to perf board). I have a big-ass heatsink on the LM317 and the LD is pressed into a 9mm heatsink module. I'm using a 25 ohm rheostat for current control in series with a 1 ohm. I measured 1.5A max out. I do need a higher wattage 1 ohm though as the 1/4 watt resistor I used turned black from the heat (BTW, using 808nm-rated safety goggles at all times). I think I can stack a bunch of these resistors in parallel to get the power load rating I need.

It seems like I'm not getting full power from the LD. It should burn paper, right?

The more resistors you parallel, the lower the resistance. 2-1 ohms in parallel would be .5 ohm. I think 1.5A might be too much. Do you have the datasheet on the diode? I just burrnt through white paper with 250mW of 660nm, you should be able to roar through it with 1000mW of 808nm. Check your circuit, lm317 pins down: V+ right pin, limiting resistance across middle and left pins, LD+ left pin, LD- gnd. Probably turn your supply down to 6V as the diode drops ~2 and the lm317 ~2.5, less heat for the driver. Or better yet, use batteries. Even with the cap, you get a spike/s out of the supply and you poof your laser. Try at low power to get everything set up and focused and then crank up the amperage. Good job with the googles, IR is scary. You're making me want to duplicate this experiment so bad but I just spent my $ on a 12x sled

Here's a pic of my setup:

I can't tell what's wired to what in the pic. Looks like the rheostat is in series after the power supply + to the lm317 V in pin? That would be wrong. It just looks a little over-complicated for what you're doing. I think you should just go PS+ to V in, 2 ohm 1/2 watt resistor across V out and ADJ on the lm317 driver, laser diode + to ADJ and laser diode- to PS-. ~615mA no frills current regulator.You can add the reverse voltage protect diode and cap after you figuer it out. Also I still say use battery/s instead, at least for setup. That PS could have a ton of ac ripple

It is wired as per this diagram:
Pos is wired to the "in" pin on the LM317 while the resistors are in series between the 'out", the "adj" and the LD anode. cathode goes straight to PSU neg.
It's the same circuit as on the breadboard with the addition of the diode. I was concerned that some current fluctuations I was having before was related to the loosie-goosie breadboard. Same circuit on the perf board is stable. I need to get a higher wattage resistor. Good suggestion on the battery at least until I get teh results I'm looking for (although that PSUs' been fine on the other 12x and 405nm LDs).

I'd have added a 220 ohm resistor across the diode as well, in case the lead comes adrift.
This way at least the diode has a fighting chance.

I would also put a series connected blue and red LED to make sure you can see if something is wrong and turn off the power.

-A
Who fried an expensive HD-DVD diode due to not doing the above.

Thanks Conundrum. Where in the circuit are you suggesting connecting the blue and red LED's?

Between LM317 Adjust and Gnd.

i.e. LEDs in series between anode and cathode of LD, so if the voltage rises too high the LEDs turn on to let you know something is wrong.

-A