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

Diode Degradation

E

ElektroFreak

0
Joined
Nov 7, 2008
Messages
5,725
Points
0
BShanahan14rulz said:
I am curious what the biggest factor that causes long term power degradation is, that we may focus our efforts on the big elephant in the room, rather than the marmoset in the corner.
Click to expand...

Nothing lasts forever, so obviously laser diodes are no exception. Simply applying voltage and passing current through any semiconductor causes "wear and tear" on the structure of the device. If you want to see a laser diode last for 10x it's rated lifetime, then you could reduce the current supplied to it by 10x. Not very fun though..


There's your elephant..
 
Last edited:





BShanahan14rulz

BShanahan14rulz

0
Joined
Dec 29, 2009
Messages
3,136
Points
63
Hahah, so maybe I AM more interested in the marmoset in the corner! If we hold that parameter as being constant across the board, assuming clean power, no esd events, then I guess I'd imagine the next one to be heat, then the next to be... clean-ness? heh
 
E

ElektroFreak

0
Joined
Nov 7, 2008
Messages
5,725
Points
0
Yessir!.. Lowering the heat buildup will help extend lifetime, but not indefinitely and probably not significantly in cases other than cases of extreme heating. Heat is an enemy of electronics in general, but it's effects vary greatly from device to device and from circuit to circuit.
 
Last edited:
M

miikkkllll

0
Joined
Feb 24, 2012
Messages
29
Points
1
The big difference is that in a projector, the diodes are actively cooled and C@Sio guarantees the diodes at 70% of their original power
 
ARG

ARG

0
Joined
Feb 27, 2011
Messages
6,772
Points
113
Bumping my own dead thread here, but I have some interesting things to add.

I have a diode in a block of aluminium with a TEC and 100W CPU heatsink on the TEC. I used this for calibration, and to do so it required to be run for 3 hours continuously to reach equilibrium so the power does not change at all and it can be used to calibrate.

I used this diode for about 30 hours. M140 diode at 1.6A. When I first started it was outputting 2.1W at 1.6A. Now it's struggling to maintain 800mW at 1.8A; pretty much a zombie.

I guess we don't see this as much in handhelds since the diodes aren't used as much. I take my 445nm handheld out once a week and have it on for about 10 minutes. About 10 hours a year, so the diode would last much longer.

Neat eh? Now I got to buy another 50$ diode and lower my driver current :p
 
Bionic-Badger

Bionic-Badger

0
Joined
Oct 26, 2007
Messages
5,438
Points
83
On another note, do you use feedback to adjust amount of power the TEC receives for temperature stability? It might cut down on that 3hr stability time.
 
ARG

ARG

0
Joined
Feb 27, 2011
Messages
6,772
Points
113
Nope, but I guess I should. Do you know of a driver with temperature feedback?
 
Bionic-Badger

Bionic-Badger

0
Joined
Oct 26, 2007
Messages
5,438
Points
83
Not an all-in-one package (unless you want to purchase a PID controller like the osPID); however, you could use a microcontroller attached to a temperature sensor and a low-side MOSFET attached to a PWM output. The MOSFET would provide the current gate for the TEC unit. To start with you could use something like an Arduino and then simplify it as needed.

While you're programming PID controllers, you should also consider regulating the output-power by providing feedback to a current controller. A semi-reflecting mirror to a photo sensor would work well for feedback; that's what they used on those B+W 473nm labbies that were for sale. That would definitely help with stabilizing it as now you'll have direct feedback to the regulating source. It'll can also compensate for diode degrading; though you'd need to clamp the output current just in case, so if it really dies it doesn't keep raising the current in order to compensate.

You could probably even just attach an analog PID circuit to the analog input of a driver such as the Flexmod (which has maximum current settings anyway).
 
ARG

ARG

0
Joined
Feb 27, 2011
Messages
6,772
Points
113
Bionic-Badger said:
Not an all-in-one package (unless you want to purchase a PID controller like the osPID); however, you could use a microcontroller attached to a temperature sensor and a low-side MOSFET attached to a PWM output. The MOSFET would provide the current gate for the TEC unit. To start with you could use something like an Arduino and then simplify it as needed.

While you're programming PID controllers, you should also consider regulating the output-power by providing feedback to a current controller. A semi-reflecting mirror to a photo sensor would work well for feedback; that's what they used on those B+W 473nm labbies that were for sale. That would definitely help with stabilizing it as now you'll have direct feedback to the regulating source. It'll can also compensate for diode degrading; though you'd need to clamp the output current just in case, so if it really dies it doesn't keep raising the current in order to compensate.

You could probably even just attach an analog PID circuit to the analog input of a driver such as the Flexmod (which has maximum current settings anyway).
Click to expand...

Thanks! I never considered optical feedback as well as temperature feedback.
 
You must log in or register to reply here.




Top