Quote Originally Posted by Buffo View Post
Actually, there is a section of Sam's Laser FAQ that addresses this problem, though it's not labeled as "back reflection" per se. Still, if you read through the FAQ from start to finish, you'll find the above passage that talks about power changes in a pointer diode when the beam is reflected back onto the face of the diode. They also mention catastrophic failure due to over-power at the face in that section, and they warn you about back-reflection.
It's there now. I'm not at all sure it was there then. I did those tests about 5 years ago. The seller I bought them from didn't even know the phenomonon existed. He didn't lie to me. This was David McGuiness of Photonics Products Ltd in the UK. We exchanged many mails until we had enough for him to take to the makers during a visit to Japan. (Not sure it was actually him who went though). Anyway, Opnext wouldn't admit it, not even to them, and they were buying thousands.

Now it IS common knowledge, but it wsn't then. Diodes had only just reached 80 mW. It didn't happen to the lower powers at all, so it won't have been documented in the LaserFAQ I think. While there are stronger multimode diodes, were even then, their larger emitters and lower power densities meant the problem didn't happen. It's only the extreme densites in either single mode diodes, or narrow stripe multimodes driven strongly, that are really vulnerable so far as I know. If the problem really WAS more widespread, to the point where all it took was a bit of reading to prove it existed, it would have been widely known a decade ago, and it certainly was not. Most people aren't aware it exists even now unless the seller explicitly tells them. (As I do).