Which is better?

Here is how my driver looks, works very well:



Dots:
Red = bc337 NPN transistor
Orange = LM317 voltage regulator (DDL style)
Yellow = CNY17-3 opto isolator
Green = RGB LED IC, common anode (+).
Blue = 1N4001 to 1N4007 diode
Violet = 1 kohm resistor
Pink = 470 ohm resistor
Brown = laser diode
Gray = +12V


Squares:
Orange = lasing threshold resistor (around 50 ohm, 2W for green, 560 ohm 1/4W for red/violet). Some experimenting needed.
Green = max current resistor for laser diode (use ohm's law to find your value)
Blue = 100 ohm pot
Violet = 47µF 16V electrolytic capacitor
Pink = 100nF ceramic capacitor Not needed on the "BigClive-kit" as it comes with its own filter. The laser diode will need however.
Brown = 10 kohm pot (sensitivity adjust)
Gray = ground (negative)


The "BigClive-kit" runs off of 12V, if you use it you wont need the 470 ohm resistor in series with it.


I also had a similar idea as Localghost, but I never tried it:



I'd like to know what's the smallest output capacitor could be, without risking the diode?


Good luck to your experiments and keep posting!

FireMyLaser said:

Here is how my driver looks, works very well:

Dots:
Red = bc337 NPN transistor
Orange = LM317 voltage regulator (DDL style)
Yellow = CNY17-3 opto isolator
Green = RGB LED IC, common anode (+).
Blue = 1N4001 to 1N4007 diode
Violet = 1 kohm resistor
Pink = 470 ohm resistor
Brown = laser diode
Gray = +12V


Squares:
Orange = lasing threshold resistor (around 50 ohm, 2W for green, 560 ohm 1/4W for red/violet). Some experimenting needed.
Green = max current resistor for laser diode (use ohm's law to find your value)
Blue = 100 ohm pot
Violet = 47µF 16V electrolytic capacitor
Pink = 100nF ceramic capacitor Not needed on the "BigClive-kit" as it comes with its own filter. The laser diode will need however.
Brown = 10 kohm pot (sensitivity adjust)
Gray = ground (negative)


The "BigClive-kit" runs off of 12V, if you use it you wont need the 470 ohm resistor in series with it.


I also had a similar idea as Localghost, but I never tried it:

I'd like to know what's the smallest output capacitor could be, without risking the diode?


Good luck to your experiments and keep posting!

Click to expand...

Do you need that static resistor next to the pot in series?

FireMyLaser said:

I'd like to know what's the smallest output capacitor could be, without risking the diode?

Click to expand...

You could go down to the nF range, but I don't know if I'd go much lower.. Maybe try 10nF.

ElektroFreak said:

You could go down to the nF range, but I don't know if I'd go much lower.. Maybe try 10nF.

Click to expand...

4,7nF, electrolitic?

Probably you mean microfarad .....

Anyway, remember that a capacitor always tend to equalize the level of the voltage you apply in pwm mode ..... and also, that more is high the frequency that you use, more the capacitor short part of that power to ground.

The "pure" pwm drivers don't uses capacitors on the LD, other than some 4,7 / 6,8 pF (picofarad) for the spikes generated from the commutation ..... better of any protection, in this case, is a 10Kohm in parallel with a fast diode, like 1N4148, and to provide a clear, well-shaped and the more possible spikes-free square wave to the diode, for the better results.

Also, take care about the maximum ratings in the datasheet of the diodes, if you know them, cause it's a lot easy go overlevels with pwm driving ..... as example: if the maximum rating of a diode is set at 1 W pulsed at 50% duty cycle, it mean that you have to always considerate this as maximum, also changing duty cycle, so 0,5W at CW, 1W at 50%, 2 W at 25%, and so on ..... but ofcourse, this means also that the complessive power is always the same (driving it CW at 0,5W, that is 0,5W/S, is the same that drive it at 2W at 25%, that is still 0,5w/S)

If you want to try a pair of my schematics (but are really basic ones) with 555, i drawed them on-the-fly ..... the first is an inusual way for use a 555, but still work, it's a fixed frequency, and change the duty cycle approximatively from 10 to 90% (you can also modify the output for use a normal NPN like mpsa42 or similar), and with these values, the frequency is approximatively 3 KHz ..... for increase frequency, decrease the 47nF capacitor ..... the second one is the classic version, and you can change the frequency, other than the duty cycle

only remember to decouple the power from circuit to LD (with a diode, as example), and use a large capacitor as filter. like 470 microfarad or more, with another one after the diode, for the IC part of the circuit

HIMNL9 said:

Probably you mean microfarad .....

Anyway, remember that a capacitor always tend to equalize the level of the voltage you apply in pwm mode ..... and also, that more is high the frequency that you use, more the capacitor short part of that power to ground.

Click to expand...

That's why I stated that the filter capacitor would limit the frequency at which the laser could be modulated. As the modulation frequency increases, more and more signal gets sent to ground. Using a significantly smaller capacitor across the diode should allow for higher frequencies to pass before being shorted to ground, but you don't want to go too low or you'll jeopardize the protection provided by the capacitor.

^ yes, sorry ..... i still not get what frequency you want to use for it, and the purpose of it ..... more efficency or more visibility ?

If more visibility, i got some results driving certain special leds at 100 KHz with 20 A for 1 uS and 999 uS off duty cycle (the total current is still 20 mA/S, so the power, but for persistence effect of the eye, it was looking more bright) through a mosfet, but they was leds, not laser diodes, and there was no capacitors in parallel .....

I know that in burners, LD can be drived also til 1 MHz and more (and there is NO capacitors, in parallel to LD's in burners), but i suppose this is not the case, also cause they use special IC's

The original poster was looking for a way to directly modulate a diode using an LM317 driver. The method you suggested (mosfet, use adj. pin) yielded what appears to be a straightforward circuit for doing just that. I really have no idea how fast they need it to go.. The only real reason to use a parallel cap on the LD is for ESD and transient protection.

I replaced the 47µF cap with 1µF and the phr diode was dead at start up. Coincidence?

The diode had a long life at 135mA, and while replacing the cap the diode was shorted...


The resistor next to the pot is to make sure you can't over do the pot.

^Maybe, but in truth the LM317 provides a very clean DC. The LM317 circuit will work fine with no protection cap and not kill the diode that is attached to it. Simply lowering this value should have no ill effects unless ESD or transients (although transients would be killed by the LM317 in your RGB circuit) are present. This capacitor is merely redundant protection to supplement the protection that the LM317 itself provides. Laser diodes can be expensive, so peace of mind is important.

nikokapo said:

4,7nF, electrolitic?

Click to expand...

That should work, but I'd check the output on a scope while powering or modulating the circuit up to see if any spikes occur. Either that or attach a cheap diode and see what happens. Obviously you wouldn't want to risk anything expensive..

FireMyLaser said:

I replaced the 47µF cap with 1µF and the phr diode was dead at start up. Coincidence?

The diode had a long life at 135mA, and while replacing the cap the diode was shorted...


The resistor next to the pot is to make sure you can't over do the pot.

Click to expand...

Wait... If it worked before, WHY did you change it?

Because the larger caps made the pulses 'diffused', especially the phr's. Now it looks much cleaner and brighter.. well at least the red and green.

FireMyLaser said:

Because the larger caps made the pulses 'diffused', especially the phr's. Now it looks much cleaner and brighter.. well at least the red and green.

Click to expand...

Oh. Weird how the phr died but the red didn't.

I replaced the diode with the one in my pocket pal, and it works now. I've must have been very unlucky.

FireMyLaser said:

I replaced the diode with the one in my pocket pal, and it works now. I've must have been very unlucky.

Click to expand...

Maybe a mysterious peak of current killed the previous one.

Current killed the Diode star!!