Welcome to Laser Pointer Forums - discuss green laser pointers, blue laser pointers, and all types of lasers

LPF Donation via Stripe | LPF Donation - Other Methods

Links below open in new window

ArcticMyst Security by Avery

Feeding diode from capacitor for pulses

DrSid

0
Joined
Jul 17, 2010
Messages
1,506
Points
48
I'm thinking about 445nm laser gun for target shooting. I'd like to have short flash, as strong as possible. I'm thinking about simply feed the diode from charged capacitor. It might also be better idea to use current driver anyway, and to feed the driver from capacitor.
Estimating right capacity seems to be a bit complicated.
Also they might be unknown risks to it.
Also I want to push it a bit harder then usual, as it is only for pulse. But then again, if diode can run at 1W for second, does it mean it will handle 2W for 0.5 second ? Will it break by power, or by voltage ?
Anybody tried anything like that ?
 





Benm

0
Joined
Aug 16, 2007
Messages
7,896
Points
113
It will likely break from catastrophic optical failure if you push it too hard, even for a split second. There isnt that much experience with 445's failing yet, and considering their price people are still sort of careful with them (although i've seen reports of 1.4 amps or so).

Pulsed operation will deal with thermal problems, but that doesnt go very far with diode lasers... they are not something you can run at 10x rated power 1/10th of the time.
 
Joined
Feb 5, 2008
Messages
6,252
Points
83
You should research into a monostable circuit.

In short, you've got bistable (I'm using directly ported words from Croatian here... I dunno if it's right translation but you get the idea),
Bistabile has two stable states, 0 and 1. Used in memory circuits obviously.

You've also got astabile. That has no stable states, used as frequency generator because it's regularely switchin between two states.

Then you've got monostabile. That one is the one you want. When triggered, it remains semi-stable in one state. That means it's only stays in that state for short amount of time. Circuits are made to influence the time of stable state via external capacitor and resistor. You can make the time from dozens of seconds (think supermarket autoclosing doors), to microseconds... which is what you would want.

Have that circuit's output (5V) to the TTL modulation input of your driver.

How to make the switch would be kind of difficult because you cannot press and release a normal switch within 1 ms or shorter, depends on what you want. You'll have to improvise that one I think, I cannot think of any solution so far.

Another case is if you want it let's say, half second, you can click the switch and release it within one second. Holding it down would be "auto fire" haha :D
 

HIMNL9

0
Joined
May 26, 2009
Messages
5,318
Points
0
You can use a current driver set for 6A with TTL modulation, and a monostable that put out pulses of approximatively 180mS at 1 Hz, or 90mS at 2Hz, without damage the diode (heatsinked, ofcourse), having short and bright flashes ..... over this limit, i cannot say, this is at the moment the only trial i've made ..... ofcourse, single pulses are better for the diode survival .....

Just have a big capacitor (at least 1000uF) at the driver input, as energy storage .....

Discharging the capacitor in the LD without a current regulation is never a good idea ..... (*POOF* ..... byebye LD)
 

DrSid

0
Joined
Jul 17, 2010
Messages
1,506
Points
48
Just to be sure .. so you actually run it at 6A for 180ms ? That would be nice ! Anyway common portable drivers go nowhere near such currents.
That's why I think about direct capacitor. If I had good V/A plot for the diode, I could calculate exact discharging curve .. but no plot's I've seen went over 2A. Still some current regulation would be good, because capacitor looses voltage very quickly .. thus the pulse would be sharp triangle, and after dropping under plot 'leg' it the output would drop even sharper. Square pulses would be best, that's for sure.
As for monostatic (not sure about the word too, but it's same in my native Czech) controls, it would be good for precise control of the timing .. but driver alone would be enough the 'squarify' the pulse, and then it would be only matter of correct capacitor, which could even be done by trial and error. Still it's clear oscilloscope would be needed. Or maybe even 60fps webcam would suffice.
 
Last edited:
Joined
Feb 5, 2008
Messages
6,252
Points
83
Just to be sure .. so you actually run it at 6A for 180ms ? That would be nice ! Anyway common portable drivers go nowhere near such currents.
That's why I think about direct capacitor. If I had good V/A plot for the diode, I could calculate exact discharging curve .. but no plot's I've seen went over 2A. Still some current regulation would be good, because capacitor looses voltage very quickly .. thus the pulse would be sharp triangle, and after dropping under plot 'leg' it the output would drop even sharper. Square pulses would be best, that's for sure.
As for monostatic (not sure about the word too, but it's same in my native Czech) controls, it would be good for precise control of the timing .. but driver alone would be enough the 'squarify' the pulse, and then it would be only matter of correct capacitor, which could even be done by trial and error. Still it's clear oscilloscope would be needed. Or maybe even 60fps webcam would suffice.
Actually, monostable and other TTL logical circuitry only deliver square pulses, they are not analog devices, but digital.

Discharing a capacitor to the diode is by any common sense terrifically bad idea.
 

HIMNL9

0
Joined
May 26, 2009
Messages
5,318
Points
0
For shooting target practice, you need a shorter pulse (but it depend from what type of target you're using ..... electronic / active targets ? ..... or just a white paper sheet and bare eyes ? :D)

For electronic / active targets, 10mS are more than enough, if you instead want to do it "eyes only", i can say you a 100 to 150 mS is visible enough for left a "visual memory" impression about where it hit, without damage the diode ..... assuming you don't shoot more than a hit per second .....

You can use the schematic that i have on the album, as current driver, and made an easy monostable from a 555 IC, for the trigger signal (so the pulse have the same lenght all the times, regardless how much time you keep the trigger pressed) ..... the assembly don't fit for sure in an aixiz module, but for that what you want to do, i think it's enough to have it fit inside a plastic weapon, or a box connected to a weapon, right ?

Do you need a complete schematic ?
 

DrSid

0
Joined
Jul 17, 2010
Messages
1,506
Points
48
Come on guys .. you never tried something ? If bad idea, I ask why. You can limit the voltage (by charging capacitor to that voltage), by that you also limit the current (as it is depends on the voltage). Only risk would be thermal runaway, but that is unlikely with short pulses, and the rapid dropping of capacitor voltage. Also you can limit the power impulse by the capacitor capacity.
I don't like the idea either, but mainly because of the shape of the pulse.
 
Joined
Feb 5, 2008
Messages
6,252
Points
83
Well, trying to trigger it with monostabile circuit is trying. Nobody did that as much as I am aware of. Well have it your way, I'm only trying to help. Seems you guys like capacitors more than complex-ish circuitry. Fine .
:yabbem:
 

DrSid

0
Joined
Jul 17, 2010
Messages
1,506
Points
48
Not really .. I just don't like saying it's bad idea without also saying why. I find all reactions very helpful !

Do you need a complete schematic ?
Not at the moment. This is supposed to be just preliminary research. Generally yes, there is more room .. but I'd like to do without cables. Also I don't want light detecting target .. that is old and not fun at all.

Something like this: http://www.youtube.com/watch?v=aQjESISG6N0

But if you could compress the power of 1 sec pulse into 0.1 sec, it would be much better.
 
Last edited:
Joined
Jun 24, 2010
Messages
243
Points
18
If you charge up a capacitor and dump it directly into the diode, your current will be limited only by the ESR of the capacitor and whatever resistance is in the circuit. Most likely it will result in a dead laser diode. Without at least some sort of ballast resistance in the circuit I think it will be too hard to control the output.

Diode lasers are not generally very good for pulsed applications. The pulsed power they can handle without failing due to COD is typically not very much higher than the CW power they can handle with adequate heatsinking. This seems to be much less of an issue with the high powered ones, but it's still something to watch out for.
 

DrSid

0
Joined
Jul 17, 2010
Messages
1,506
Points
48
That is IMHO not correct. Capacitor is always charged to some voltage. Diode itself will limit the current, based on actual voltage.
The fact that it won't shine much brighter is on the other hand real issue. It seems the output does not increase after 2A, which is rather achievable with small drivers.

From here:
http://laserpointerforums.com/f52/1...in-ultrafire-rl-2088-handheld-host-56255.html

Btw. this Flexmod P3 looks perfect. It has TTL input, good power, size is acceptable for pistol build. I would tune it to 2A, add monostatic circuit to TTL input, end here I go. I could also use analog input for lower power continuous mode. Imagine safety lock with safe/pulse/cont positions :-D
 

Toke

0
Joined
Jul 25, 2010
Messages
1,099
Points
0
The main problem here is not making a pulse circuit or geting a diode to survive pulsing, they are made to pulse fast enough to put 1'es and 0'es on a spinning dish.

Diodes have a CW and a max pulse value, I am pretty sure we are running them CW at their rated max pulse value or beyond.

In order to get a noticeable effect on the target you must look at the Joule value of your pulses, 1W for 1sec gives 1 Joule.
Here is an IR LD rated for 70W, but at pulse widths of <30ns. :D
 

HIMNL9

0
Joined
May 26, 2009
Messages
5,318
Points
0
Well, the better way for have fast commutation on "edges" of the driving waveform (except using gas switches :p), is using mosfets, and it's decently easy to build it in a current driver (that is already based on a mosfet, after all ;))

Just be sure to use a decent "energy storage device" (a capacitor or an inductor :p), on the power line, so the power supply don't become influenced from the high current pulses ;)
 
Joined
Oct 26, 2007
Messages
5,438
Points
83
Yeah, having a mosfet circuit would be best. Then you could actually microcontroller it or use something like a 555 timer to control the short burst you want for target practice. Forget the capacitor. You're probably thinking of one of those awesome movie lasers with the big capacitor sound that sends out this awesome megapulse. A better idea is to set the mosfet to the maximum current you want it to drive, and then trigger its gate with the timer or microcontroller. That way you know you won't burn out your laser, but you can still tune it for high power. With a short enough duration, you could possibly go up to something like 1.8A through a 445nm diode, as that's what they're using in the X-140.
 
Joined
Oct 21, 2010
Messages
5
Points
0
keep in mind that capicators take up to 6 times as long to fully charge than to disscharge. like mentioned above a mosfet circuit could work well. and calculating the amount of V and I output from the cap to the diode so you are to not over drive it in watts or mW.
 




Top