16X DVD RW Red Laser Diode Questions

Hello everyone. I'm still very new to making lasers, I have only built a few of them, and all of them have died due to some stupid noob mistake I have made. Anyways I'm gonna be straight to the point: I have extracted a Diode from a Sony DRU-820A (IDE) 16x DVD RW Drive and installed it into an AxiZ Laser Module. I originally built a LM317 Driver circuit set to output 250mA. I fired up the laser and it worked. I tested it and it popped balloons and ignited matches, and all that fun stuff. Anyway after about maybe 20 minutes collectively of the laser being powered on, I noticed that it would dim and sometimes not even light. I thought maybe I was overdriving the diode and I'm killing it. I then modified the driver circuit to output 200mA, and the diode works fine now. I have been trying to keep the on-time short at like a minute at most. It works fine now, but can't really light a match (and yes I am focusing and marking the surfaces black.) So now that you know the story, I need to ask these questions:

Was this dimming and turning off due to my power supply (I was using a computer PSU; 5v (red) and Gnd (black)) or because I was overdriving my diode?

Without some sort of power meter and some sort testing apparatus is it possible to estimate the mA required to drive any 16x diode? (Or even this one...)

Is it possible to estimate the mW output of this laser without a meter?

Is it safe to use USB 5v to power my laser for testing purposes?

Images: (Resolution too big for forum)
http://i57.tinypic.com/x3suq0.jpg
http://i59.tinypic.com/4lsp51.jpg
http://i60.tinypic.com/30k7gc9.jpg

Oh crap I think I have accidentally posted this in the wrong section. If someone can move it, I'll be really thankful.

it might be dimming because you have no heatsink....

The AxiZ housing alone isn't going to absorb the heat? I didn't mention this, but I turn it off when I feel the module is getting the slightest bit warm.

Not for any length of time. That little housing will heat up quickly. But if your monitoring the heat, it should be OK.

What do you think of the situation? Was this a problem with my circuit or power supply? I have killed diodes in the past, and it seems that once they dim, or fail to light, they don't come back from that. This seems like it works just as well, but like I said, I tuned it down from 250 mA to 200 mA and it still works well. I think I'm going to crank it back up to the original 250 mA and see if it lasts. I have no surefire way of knowing how long this diode can last or how much it can take, but if I keep the temperature under control, hopefully it will work just fine. I'm going to wait to do this though because I would like to see what other forum members have to say. Thanks for the replies.

Sometimes a diode set higher than its peak current will start to dim. I had a 405nm like that. When I fed it too much it would dim but not die. When I turned it down it actually got brighter.

Interesting. I will try to turn it up to like 210 mA or so. I know it's capable of more though because it was stronger than it is now, I just don't want to push it too far. Maybe the damage was already done and the diode is now partially broken? I don't know. Thanks though. I'd still like to hear what others say. Like I said, I'm still new to lasers and its easy for me to get excited and accidentally screw stuff up.

This is a bit off topic but here's a little story:
My first attempt at a laser build was a PHR-803T. I was trying to de-solder the PCB/Ribbon and I didn't take my time with it and I ended up just ripping the positive and case pin off. I was quite pissed, more at myself for not taking my time and, well, just doing things properly instead of doing them the way I wanted to. Anyway I'm trying to prevent these kind of mistakes from happening again so I'm trying to be careful and do things right. I just don't know what the limits of this unknown diode are. I am probably going to end up killing it, but if I get a diode that I know, or can find the information for in the future, I will probably run it safely

The more you get into lasers, the more you'll see the worth of just going to DTR Lasers and just getting one ready to go, pressed in a module, or even with a pre-wired driver all you need to do is give it power

Yeah I hear ya. Oh I cranked it to 210mA (5.9 ohms of resistance across the Adj and vOut pins of the LM317) and it lights matches once again! I run it for about 15 seconds and I feel the module getting warmish so I will try to get a heat sinking solution and/or use short duty cycles.

https://www.dropbox.com/s/fhwxx12ckxxnsdi/Red Laser Match.MP4?dl=0