dicking around with a red diode

[sbk]

yes one of them sleds.

i think a fair test would be to just set it at 300mw with the tec running and see how long it lasts.

ElektroFreak is making a test like this, without TEC but with a very good cooling. I think the lifetime wouldn't change a lot with some degrees colder...

My only worry about my test is the fact that all diodes vary considerably in life expectancy when they are run above their maxmum ratings. Some people might have great luck, while someone else might have no luck. The ideal method for determining lifetime would be to run several tests simultaneously and average the results.

Well, if the worse diode lasts only 1500 hours, that's still very acceptable
But yes, it's impossible to tell how's the diode you're testing when testing only one. But if you take +-20% of the lifetime, you shouldn't be so far from reality!

[andy_con]

it would be nice to have the facilities to test more than one diode but i dont sadly

[ElektroFreak]

Me neither.. that's why we're each only testing one, I guess.

[sbk]

And it's quite like sort of wastage to let a lot of diodes running until they die, even if they are cheap

The important thing is to know that these diodes seems to handle a nice over power level for a sufficient time when properly cooled.

[andy_con]

to be fair they dont produce that much heat when running at 300mw

[ElektroFreak]

No, not nearly as much as you would expect, but heat only plays a small part in diode death. COD (catastrophic optical damage) to the diode facets can be caused without the diode ever even becoming warm. COD is usually caused by too much photon flux (too many photons bouncing back and forth between the mirrors) which destroys the coatings. Keeping diodes cool goes a long way towards lifespan, but it definitely isn't everything. If only they were like CPUs..

[mccarrot]

How many mA are you running the LOC's?

[andy_con]

cant remember off the top of my head, i would need to check