R.i.p. Laser diodes experiences

[jors]

I think it could be interesting to start a thread about the 'magic moment' (irony) when a laser diode goes to a better place.
At least, we could comfort altogether, but the most important is....to share the CAUSES that killed our precious LD, to avoid others errors while keeping the wallet safe!!

Sooooo....I start the post with my pains .....

Last week I killed 3x 9mm Blues 445nm. I always test the driver with Dummy load before and it was fine.
However, it failed on power-on and killed 3x blues and 3x lasorbs...at a fraction of second!! After removing the driver and testing again...it was delivering constant 8A even with no modulation...
an yes...magic smoke on one Lasorb and no leded diodes...vaporised bond wires for sure WFT!
I still don't know why the driver failed

One month ago I killed 4 P73 (in series) and 4x Lasorbs...yeah very 'funny'. This time all 4 were leded...and again magic smoke on one LAsorB.
They worked good with a LAB linear PSU and a certified switching Meanwell. But when connecting to a chinese NO-Brand 'cheap' PSU, reds were killed.
It seems this PSU is defective...or LOTs of trash in DC line...it killed reds even with startup delay and driver protection....So my advice is ALWAYS GOOD quality PSU to keep safe our diodes.

[ZeroPoint]

I've been fortunate enough not to blow too many diodes, although that's largely due to the fact that I haven't been able to afford too many to play around with. Other than the countless LPC diodes I blew up years ago when I first started tinkering with lasers the most recent was a 405nm diode I was going to use in a lumia project. I was making my own constant current driver for it but ended up having a short between the voltage regulator and the base plate that the diode was mounted to. I never even saw the laser attempt to lase. it didn't like the unregulated 24 volts it got.

Other than that I think I've blown 1 M140 in the past as well but I don't remember how. Another M140 I've had for about 4 years and it's still chugging along, although at severely diminished output. I fell asleep one night while giving it a run in test when I got my first projector built but the cooling wasn't sufficient enough and when I woke up it was cooking and barely had any output. Once I let it cool down it came back to life but it has since dropped to about 500mW max output. Other than that, my NDG700 and 2x p73's are still going strong (although only one of the reds is currently used). I also recently bought a few single mode diodes that have been a lot of fun. Heaven help me when I have the money to actually start experimenting more, I'm sure I will lose my fair share of them. In the mean time though, I am careful as hell not to blow anything up xD

[Maltes]

I recently killed a NDB7875 diode.
Cause of death under investigation.

RIP

[VDX]

... until now I managed to kill one 9W@975nm and two 2W@445nm diodes - all three with overcurrents while developing LD drivers, so this was to be expected

Viktor

[JRGixxerfreaK1]

Killed 3x Mitsu G71 1 from overdriving, 1x esd when handling diode, 1 when I had a loose psu power connection while driver is modulating
1x Osram PL450B 80mW single mode, the pins shorted out on the brass mount, heat shrink tubing not properly in,when moving cables about.
1x Osram PL520 50mW single mode, overdriven till LED Was crying when this diode was 80 bucks back 3 years ago haha

My most regretting moment is when I killed 5 out of 6 red Mitsu G55 single mode 180mW wired all in parallel and one has got a loose soldering joint on the anode side flickering then LED all of them

[krazer]

Most annoying failure I had was in my singlemode RGB diode projector. I used a trio of flexmod's that I had got second hand, and for some reason the blue diode would randomly fail every once in a while. I carefully checked all of the pots and double checked the data sheet to ensure that they were set correctly but couldn't for the life me me figure it out. Eventually I borrowed the drivers from it for a different projector, and in the rebuilt projector the blue diode didn't turn on at all--turns out the trace leading to the solder pad for the diode connection had a tiny crack in it and was giving an intermittent connection the whole time.

[catalanjo]

This is the second time I have heard of an FB3 failure, and it bothers me a lot !
Many thanks for sharing this.
Now I know what to check on the ones I have out in the field, (sounds like a dodgy batch), do you remember the approximate purchase date?
Cheers

[kecked]

I had a boat load of problems with early reds when 35mw was big power and the cases were not isolated.
Had tons of p3 driver issues blowing diodes but I think it all came back to and this is my point...the power supply I used.
I had been using the power off my breadboard to run the drivers. Once I went to a solid power supply I never had an other issue.
The other issues all revolved around ground loops. It was not a ringing issue on the driver.
I was using a power conditioner from furman thinking it would clean up the power coming in and for some reason I never solved it spoiled my grounds and placed my unit above ground potential. It was soo back it caused my adat unit to periodically fail.

[VDX]

... my biggest concerns when building my first laser-diode driver around 1996 were for overcurrent spikes only visible with a scope, but enough to killing the diodes.

So from the first drivers on I've used some form of "stabilized" constant current sources with soft start and fast modulation abilities.

So no R.I.P.'s for spikes, ESD or such ... only for testing with overdriving current or input errors for the constant current regulators (e.g. 0-5V for normal use, erroneously caught 12V)

Viktor

[neobenedict]

In building my first projector I've lost 5 diodes:

2 100mw single mode green, one to random bullshit recently, another to an accidental short
3 250mw single mode blue, two died while turning the projector off (still figuring this out before I replace them), another due to my fault with undervolting.

I probably have the worst luck on this board.