Dead 445 diode

[DZ]

Hey guys, just wanted to mention this in case you run into the same problem. I had just finished converting the second Maxy module I have and set it up just like the first one. Same current, nearly the same output power and it died rather quickly after being ramped up to full power. The first module was making about 770mW and the second one was making about 810mW. The only difference between the two modules, was the lens being used. The first one is using one of the 405nm AR coated 3 element lenses. Since I didn't have a second 405nm lens, I just used one of the cheap acrylic lenses. After discussing the failure with the folks on the IRC chat thing, it seems the failure may have been caused by back reflection. The first module has run for several hours at full current and has been fine. Can't say for sure if this is the true cause of failure, but it's highly suspect.

There is one more thing to mention about the acrylic lens used, this lens came from a pointer I purchased from LPF about a year ago. The baffle was cut away, I presume to allow more light to hit the lens and this may have also contributed to the failure.

[ElektroFreak]

To me it sounds like a nearly 100% chance that back-reflections killed you diode. At class 4 powers even a small percentage of reflected light can be a bit of power.

I've killed one of these myself, but I'm not 100% sure why it died... only about 85% sure. I'm using a flexmod, and I'm feeding it with a 5V source. For testing purposes, I've also connected the mod input to the same 5V source. I plugged in the 5V source and I heard a faint pop through my nearby speakers from the small discharge that can happen when you plug things in.. The diode was LED when I turned it on. I can only assume that somehow the voltage transient from the initial plug-in made it's way through the 5V PSU and through the Flexmod to the diode. I'm assuming the actual drive circuitry of the Flexmod should protect against transients from the power source so I think it somehow made it in through the modulation input..

[dream beamz]

To contribute to the cause. The one I have running constantly at the house at a watt has an acrylic lens (i think, it came out of a blank aixiz module)

[DZ]

To contribute to the cause. The one I have running constantly at the house at a watt has an acrylic lens (i think, it came out of a blank aixiz module)

Well when I assembled the first one, I had also used an acrylic lens, until I noticed that it quickly started to melt the baffle. Which is why I chose to use the other one with the acrylic that had the baffle cut out. I can only imagine that the melting baffle would leave residue on the output lens of the diode itself. You may want to check that.

[DZ]

I've killed one of these myself, but I'm not 100% sure why it died...

EF, do you have a lasorb installed? Not sure if it would stop a transient, but just curious.

[Laserman532]

To me it sounds like a nearly 100% chance that back-reflections killed you diode. At class 4 powers even a small percentage of reflected light can be a bit of power.

I've killed one of these myself, but I'm not 100% sure why it died... only about 85% sure. I'm using a flexmod, and I'm feeding it with a 5V source. For testing purposes, I've also connected the mod input to the same 5V source. I plugged in the 5V source and I heard a faint pop through my nearby speakers from the small discharge that can happen when you plug things in.. The diode was LED when I turned it on. I can only assume that somehow the voltage transient from the initial plug-in made it's way through the 5V PSU and through the Flexmod to the diode. I'm assuming the actual drive circuitry of the Flexmod should protect against transients from the power source so I think it somehow made it in through the modulation input..

just curious...were you using lasorb?

[ElektroFreak]

No lasorb.. I would imagine that the lasorb would have saved the diode. I do need to pick up a few of them.. I generally don't use them due to the low cost of most of the diodes I use regularly. It's not much trouble to replace a diode, and this one is actually the first I've had die on me in some months. I assumed that Flexmods protect against transients, and I'm sure they do, but maybe not on the modulation inputs.

[DZ]

I extracted the dead diode and cut the can off the top to see the inside. Thought I'd share the pics! I didn't resize them, so unless the forum software changes the resolution, these should be in pretty good detail.

[Stuka]

I extracted the dead diode and cut the can off the top to see the inside. Thought I'd share the pics! I didn't resize them, so unless the forum software changes the resolution, these should be in pretty good detail.

Wow -
Up to 1 watt of 445nm blue - from a splinter!!!

[mikkojay]

Hey David,
Those are some great pics! They made me wonder- would it be insane to try and cut the can off one of these, and snip 2 of the 3 conductors, to result in a 1/3 power single mode open can? Sure you would lose 2/3 power, but for apps that only need 200-250mw I would think it would be plenty. Just a random thought while looking at the pics, just after reading another thread about collimation of these (hot topic).
-Mike