Lm317 for 445nm 1w laser

[Reonarudo]

Hello,
Is it possible to use LM317 to build a laser driver for the 1Watt 445nm laser?
If so, how?
My diode has this specs:
3.7 - 4.3 V
~1400 mA

With a normal driver and 12V input, I can only get ~300mW output from the diode.

Maybe I'm doing something wrong...

Thank you very much.

 

[ludwigfan]

What exactly is a "normal" driver?

Not a bad idea to take a quick look at the spec sheet once in a while:
http://www.national.com/ds/LM/LM317.pdf

 

[Reonarudo]

Sorry I was not explicit. I meant one of these drivers:
Laser driver - It can be done

 

[magonegro]

What resistor are you using? You'll need a ~1 Ohm @ 4Watt one. A LM317T can handle up to 1.5A but I recommend heatsinking 'cause it also has thermal cutoff. It has a drop of 2.0v so will be fine at 7.2v, warning the extra volts will be heat in the regulator.

 

[Reonarudo]

I have the LM317 heatsinked and I'm using a 1 Ohm @ 1Watt.

Thank you for your help.

 

[kiyoukan]

you do not need a 4w resistor for the lm317
for 1000ma you need a 1ohm resistor at at least 1.25 but recommended 2w
even when running at full power the max wattage resistor you will need is 1.88 a 4w is way to much.
Go to digi key and buy a bunch of 2w resistors and they are perfect for the 317s.
source
http://www.reuk.co.uk/LM317-Current-Calculator.htm

 

[jayrob]

Yep...

1 Ohm 2 Watt resistor with the LM317 should be about 1250mA's... (P.S. - with silicon diode and cap)

Here's another source for that resistor:
Cascade Surplus Electronics

P.S. I have not tried it yet, but I think that 2 X 18650's in series will work for a 445 using this set up.

If the diode is drawing 4.2 volts or so, it should stay stable until about 7 ish volts, and you should get pretty good run times between charges...

But 3 X 18650's would be better!

 

[Cyparagon]

But that would suggest the lm317 has a dropout of 2.8V when I think it's closer to 1.8V (or maybe as low as 1.25V as suggested by this image:

) . An extra battery in series would be wasted as the extra voltage would be thrown away as heat.

You could argue with 3, it will more fully discharge each cell (as low as ~2V, presumably?) but with two, it's ~3V which is a safer charge state than ~2V.

 

[jayrob]

Don't forget about the silicon diode and the cap...

 

[Meatball]

To be honest, I've been leaving the output cap off of my more recent LM317 circuits. It seemed like for a long time, I could short the cap with soldered leads, solder on the diode, remove the cap jumper and then turn on the circuit only to have the cap blow my diode instantly.

I don't know why this kept happening, so I've been leaving the cap off, and have had no problems at all so far. I think as long as your LM317 input PSU is stable, it will happily and safely drive a diode.

 

[Rafa]

Don't forget about the silicon diode and the cap...

The diode and the cap don't waste any power at all.
Well if you put the diode in series it will eat 0.7V but if you place it in anti-parallel it will be like if it's not there.

 

[jayrob]

Well for what ever reason, the circuit eats more than 1.25 volts...

The 1.25 volts is what is regulated at the sense pin, but the circuit uses more like 3 volts...

 

[Reonarudo]

Thank you very much for all your input. I don't know how many nW is it putting now, I'll measure it later but do you think it is safe to push it a little bit more than 1000 mW?

 

[Rafa]

Well for what ever reason, the circuit eats more than 1.25 volts...

The 1.25 volts is what is regulated at the sense pin, but the circuit uses more like 3 volts...

Yep, It eats 1.25V for the Vref resistor and about 2 other volts for the circuits inside it.

Reonarudo: You can push it safely up to 1300mA