How Big of a Spike Can It Handle?

Hey everyone.

Out of curiosity, does anyone know how big of a current spike a cold-started 445 diode can handle on turn on? For how long?

I don't have an O-scope, so I can't test the drivers I have, but I'd like to know what kind of spikes these diodes can handle.

Ive driven one at 2.5A off of a dual LM317T for a grand total of about 15secs in 3 sec bursts. At that current the heat generated is enough to saturate the diode to the point where it output drops like a rock. After about 3 secs its putting out about as much light as another 445 driven at 200mA would be just crossing the laseing threshold.

But if I had 2.4A spikes for a few hundred microseconds, it should be fine? It won't go into COD immediately?

Not immediately, mine survived and has been happly running at 1.75A after my little experiment. After 3 secs i would let the heatsink soak up some heat and the diode would cool down enough to lase properly for another burn. I dont know how well the die will hold up to that kind of current in the long run, regardless of the spike duration. What do you plan on doing with short intense pulses?

It's not actually a planned pulse, it's just the by-product of a driver I am making. Everything so far works with the exception of these short pulses that I am trying to eliminate.

Any reason why a cap on the output won't work?

I tried it, but I may have done something wrong... I will try a sort of big cap on the output, see what happens. It needs to be able to handle 1ms spike duration. Or so.

Wolfman29 said:

But if I had 2.4A spikes for a few hundred microseconds, it should be fine? It won't go into COD immediately?

Click to expand...

It might, its actually very few people will know the answer to, if you can find any.

Although the power output of these diodes is large, they are multimode diodes, and the power -density- on the mirror surfaces and such is probably much lower than that in a singlemode 200 mW red laser.

COD is usually the result of the output coupler failing, which can be gradual if it receives a damaging blow every time the laser is switched on with a peak. At some point the faults add up to too much and it will fail and go LED. The 445s seem less prone to this problem though - the only way i've lost one so far is by accidentily unleashing the full output of a power supply (15V/5A) onto it. This resulted blowing the bond wires, although i'm not sure if the chip actually survived or not.

Hmm, that's interesting. What I may end up doing then is, if these ever do enter production (unlikely, but it may happen), I may donate some of these at no cost to people if they want to do some testing for me (which I will also be doing).