Diode intensity adjustments

[xanatos]

Has anyone here used anything like pulse width modulation to make a variable-power pointer? I was thinking about doing so (think "pointer", "stun" and "kill" ). My question probably most centers around how to merge a PWM circuit with a CCS so that I am delivering a constant current pulse with variable duty cycle. Anyone played with this idea before?

Also - diodes here seem to be pumped with maximum CONTINUOUS power - anyone know what the peak pulse power might be on some of these? Like 100mA for continuous, but you might be able to sink an amp for 10 mS (with a VERY low duty cycle) or something like that to generate huge power pulses...??? Seeking advice from the gurus please!

D

 

[Benm]

I've done some experiments with PWM into a current source using the group buy 16x diode. I set it up by using a PWM driver that pulls the LD cathode negative, while its anode is hooked to a current source (found elsewhere here, 2-pnp design).

From an electronics standpoint, this all works very well. The problem is that there is little to be gained in terms of average power output, as the diodes power/current curve already flattens around 240 mA, and you can drive them CW at that current.

Ramping the duty cycle from say 2 to 98% does provide a very nice brightness control though. One advantage is that the output pattern doesn't vary with power output as it does under linear control.

 

[xanatos]

Cool... the setup you mention is roughly what I was imagining - just need to mod the CCS circuit to make up for the V drop across the Q junction.

Thanks for the info about the power output leveling off around the 240 mark. So CW is really the way to go anyway... makes it easier for me!

I'll play with the PWM schemes and find something really simple that work good and post here. I just ordered 4 of the GB 16x diodes so I should have plenty to play with.

Thanks again,

Dave X

 

[Cxrazy]

sorry for the spam,

But cool ava xanatos.... Trippy...

To make this less of a useless post, can this same thing not just be achieved in ddl's circuit by turning the pot?

 

[danq]

To make this less of a useless post, can this same thing not just be achieved in ddl's circuit by turning the pot?

Not really... see above post by Benm - supplying the diode with less current is different from supplying it with the same current intermittently. For example, if you give it 10% of its max current, it probably won't be up to its lasing current so won't lase - right? But if you give it the max lasing current 10% of the time, it will lase, but your eye will only see it as 10% brightness.

DanQ

 

[Benm]

That's right, it'll lase even at the lowest of duty cycles. To achieve such power levels with CW control you'd have to drive it right on the treshold, which varies with temperature and would be dificult to drive reliably. Another benefit of PWM is that average power is perfectly linear with duty cycle, while the mW/mA curve of laser diodes isn't.

For my experiment, i used a rc oscillator with variable duty cycle based on a cmos 4093 ic. This doesnt allow full 0% or 100% (but 2-98% is no problem), and is easier to construct than 'proper' PWM drivers based on sawtooth oscillators and comparators.

 

[xanatos]

I'm lazy - I'm just planning on using a 555 set up as a PWM circuit to drive the transistor on the LD cathode leg! Not sure it's possible to get any simpler than that! Plus I'll use a switch to drive at CW for full power - that extra 2% can make a difference ya know!

D

 

[Benm]

A 555 will surely work just as good, guess the 4093 was just higher up in my component pile

 

[xanatos]

Plus I have a boxload of 555 SMDs!

D

 

[highvolt]

Ok, i am very interested in this as well.

What I would like to know: is it possible to take a NOVA X105, take the batteries out, and 'feed' it PWM instead?
(if not, can someone tell me how to mod the X105 and post a link to the PWM / Driver)

The advantage is also that if the PWM dutycycle is low enough, the X105 can be run without any 100s/10s timelimit.

 

[Krutz]

my 2 cents
(where did that expression come from anyway?)

you have to feed the pwm directly into the diode, with no filtering capacitor (being larger than "really tiny) in between. it would filter the chopped input, just like it would (should) filter spikes as well.

you propably wont be able to pulse regular laserdiodes very high. shortness of pulse doesnt matter, COD can happen in *very* short times. dont know if its in the ms or ns range though..
COD - catastrophical optical damage, turning the laserdiode in a metal-can led with small output. happens by too high internal laserpower (which is much higher than the output, 10 to 100 fold easily) that damages the internal mirror(-coatings). so i dont think you can pulse a diode higher than on CW. sure, it would prevent some of the degeneration and warming, but not COD!
there are pulse-diodes (IR), i would assume they just have more robust mirrors. so they cant automatically put more CW out than a comparable die with normal mirrors, even when capable of 100x stronger pulses.

another thing to remember:
you have a 100mw green, and use pwm to make it look like 1mw. if you shine that into your eye, it will still be 100mw, just pulsed. at least it wont be as "harmful" as a real 1mw laser.