PHR-803T batch testing.. Windowless 803Ts?

[Milos]

Big tap on the back to all of you guys, for doing all the testing and reports.

I have pushed my 803s before only as far as 220mA and was getting similar results as yours ( about 205-210mW ) but with cut out acrylic lens. They seem to be pretty consistent at the time (about 3 months ago) or maybe a good batch.

Anyhow, I have bunch more to build into portable lasers, so I will report if I find anything interesting.

cheers

 

[IgorT]

Well, i used the 211mW to test some glass lenses, and it survived being powered up for very long. But shortly after that it just popped and went completelly dark. Haven't seen that before!

 

[Zom-B]

Zom-B, what hardware do you use to make those graphs?

Does the efficiency calculation take the Vf into account? Did you see any abnormal Vf behavior, or does it always steadily increase with the current?

I make the I-P graphs with a Dell Vostro 1700 2.5GHz Dual core (you asked hardware, right...)
I make the U-I curves with a transistor curve tracer:

This is a dead diode and I'm testing it with about 350mA. It's totally uncalibrated though so I can only use it to roughly see the shape of the curve.
Afterwards the diode was so hot I could burn my fingers

My efficiency is the current-efficiency, and doesn't take Vf into account. I did measure the VF of the diodes and they all lie om the same trend:

Here, the Vf is exaggerated 10 times to make it somewhat visible on the graph. 10mW=1V

 

[Milos]

Well, i used the 211mW to test some glass lenses, and it survived being powered up for very long. But shortly after that it just popped and went completelly dark. Haven't seen that before!

do you mean that you brought it up to 211mW and left it to burn at that level for some time. Than powered it off, and it poped after you powered it on again?

if yes, than I say thats normal for every diode ps3, sony reds, 803t, and phasor reds I tested. Its just how they handle being over driven.

 

[Zom-B]

I don't think that's what it says.

It sounds like it's internally shorted like mine. The 8 diodes I tested here were from a batch of 10, and the other two diodes did this before I got to test them. They did not flickered like that one but were just black and had an internal short until I broke through it with a steady 5V from a bench supply. They all turned out to be COD. The first one died while testing it quickly >150mW and the other one broke even before it gave it's first light.

 

[ArdvarkOfDoom]

This could be a stupid question, so bear with me... but what if you tested in a TEC cooled set up? I know jayrob made some portable TEC cooled BluRays... but do you think that the cooler temp could help pushing the power or at least stabilizing?

 

[Zom-B]

Today is really not a good day to work with lasers

EVERY laser diode I touched today DIED :'( :'( :'( :'( :'( :'(

Just now, I wanted to extract a long open can from an AixiZ --the safe way-- with a tube of thin sheet metal. The tube bent and folded beyond believe, taking the entire laser cavity off the substrate WHILE THE LASER DIODE WAS ALREADY COMPLETELY LOOSE AT THE BACK OF THE AIXIZ HEAD!!! why would it curl up when it's already loose?...

 

[IgorT]

This could be a stupid question, so bear with me... but what if you tested in a TEC cooled set up?  I know jayrob made some portable TEC cooled BluRays... but do you think that the cooler temp could help pushing the power or at least stabilizing?

A TEC can help you push the power, yes, but for the same reason, it can make the diode die even faster.

For example a diode, that does 211mW at 24°C, you turn it on, then warms up and does 205mW. If you force cool it, the power will increase. Increased output power means an increased optical flux at the end facets (really really tiny mirrors). The higher the output power, the higher the strain on them. Too much, and they melt, resulting in COMD (catastrophic optical mirror damage).

They are always under strain at any power, the higher the power, the higher the strain. Small defects cause a point to stop emitting light, and heating up instead, increasing the strain on the area around it, the effect is exponential and when the areas becomes large enough, the stress becomes so high, that the effect looks instantaneos. The spot goes really dim. You can then actually observe whole areas going even darker, as more and more of them burn up.



Anyway, yes you can use a TEC to stabilise the power. But for this you need a constant temperature Peltier driver. A driver with a probe, that measures the temp of the diode, and adjusts the heat pumping power, trying to keep the temp. constant. That way you can eliminate the heat from influencing efficiency.

Without such a driver, the temp will always drift one way or another. Too much cooling is not necessarily a good thing. It doesn't make them live longer, but it does increase the power.

 

[IgorT]

[quote author=IgorT link=1214268330/100#113 date=1223755099]Well, i used the 211mW to test some glass lenses, and it survived being powered up for very long. But shortly after that it just popped and went completelly dark. Haven't seen that before!

do you mean that you brought it up to 211mW and left it to burn at that level for some time. Than powered it off, and it poped after you powered it on again?

if yes, than I say thats normal for every diode ps3, sony reds, 803t, and phasor reds I tested. Its just how they handle being over driven.
[/quote]


When i said that i haven't seen that one before, i meant the diode going COMPLETELLY dark. As in, no light at all. Usually a little light remains, and i can then observe the optical surfaces failing, whole areas at once..

Otherwise, it didn't die instantly at a powerup. It worked for a while, so when it went dark, i thought it was the battery.


But when you say "overdriven", what exactly do you mean by that? Pretty much every laser we make, the diode is overdriven. Do you mean over-overdriven?

And if so, what is your experience in such cases, if the diode is not left for a long time once, but only for very short times at once? Was the emphasis on the "for some time", or were you just describing how they die?

 

[Milos]

[quote author=Miloš link=1214268330/100#115 date=1223755990][quote author=IgorT link=1214268330/100#113 date=1223755099]Well, i used the 211mW to test some glass lenses, and it survived being powered up for very long. But shortly after that it just popped and went completelly dark. Haven't seen that before!

do you mean that you brought it up to 211mW and left it to burn at that level for some time. Than powered it off, and it poped after you powered it on again?

if yes, than I say thats normal for every diode ps3, sony reds, 803t, and phasor reds I tested. Its just how they handle being over driven.
[/quote]


When i said that i haven't seen that one before, i meant the diode going COMPLETELLY dark. As in, no light at all. Usually a little light remains, and i can then observe the optical surfaces failing, whole areas at once..

Otherwise, it didn't die instantly at a powerup. It worked for a while, so when it went dark, i thought it was the battery.


But when you say "overdriven", what exactly do you mean by that? Pretty much every laser we make, the diode is overdriven. Do you mean over-overdriven?

And if so, what is your experience in such cases, if the diode is not left for a long time once, but only for very short times at once? Was the emphasis on the "for some time", or were you just describing how they die?[/quote]

ok, i see. I had diode too go completely dead (no visible light at all) all of a sudden.

Yes, when I say overdriven, I do indeed mean over-overdriven ;D ! In case of 803s, i roughly consider this to be over 210mA with passive cooling. (just a heat sink)
I drove few om my first ones over 220mA and they lived but only cause i had used a potentiometer to increase current to that level. Couple lived for few min. They all would die (with small light remaining) when I tried to turn them on again and surge of 230+mA hit them suddenly.

i had this results a while ago.
200-205mA 180-192mW
212-215mA 195-205mW
225-230mA 205mW

i also know that at 200-210mA with passive cooling this diode will not live survive past 10min of ON time. (had 3 tested with almost identical death times)

best regards

Miloš

 

[Zom-B]

I've never heard of the multiple mirror story. As fas as I know, it's just one mirror on each side. When it fails, then the little remaining light is the pump LED. This one pumps the gas in the cavity and inverts the population. If the gas is gone or the mirrors dark, then you still see the LED part working.

I've also had a few completely dark diodes, these were internally shorted. Pushing them with 5V (with whatever cirrent needed) i usually broke through the short and it gave light again, although not lasing anymore.

 

[IgorT]

Yeah, but it was not shorted. Had it been shorted, the flexdrive would be dumping a huge amount of current through it.

Otherwise i still need to get it out, to see what happened..


But when you mention breaking the internal short.. I did something like that with dead PS3 diodes. They were LEDs, untill i attached them to my PSU and pumped 1.6A through them for a split second.

After that, they were lasing again, and were brightest at 230mA, where i could see the beam with some smoke.

 

[Zom-B]

230mA I don't think that's lasing, but just a really overpowered LED. was the beam nicely round TEM00 and containing lots of interference noise? I don't think the facets could even survive a split nanosecond of that amount of flux. Btw, isn't that like the same as attaching a charged cap without discharging?

---

I started testing the diodes for durability. I started with this diode:

At 225mA it emits around 199mW cold.

I attached a duty-cycle generator to the driver modulation input which I've set to 7.5 seconds on 22.5 seconds off. That's exactly 25% on and 2 times per minute, 120 times per hour, 15 seconds per minute, 15 minutes per hour. I'll let it run for 3 4 hours and see if it still lives. When it does, it's passed the test.

 

[IgorT]

Btw, isn't that like the same as attaching a charged cap without discharging

Indeed it is, but it was dead, so i figured what the heck.. And after that it made a beam again, which was visible, and brightest at 220-230mA. You can imagine how much it heated up. As it heated up, the beam went dimmer, but i cooled it. I was without a blu ray for months after that, so every so often, i hooked it up, just to see the beam.

Otherwise, that's not even a diode anymore. It passes current in both directions. That's why i called them zombie blu rays..

 

[Zom-B]

Well, my freak died. It was fun watching 300mW for a short time. I ran it at 250mA/263mW during a longevity test and it died at this flux. I did manage to get some shots while testing it. It lasted about 20 minutes with this (duty-cycled) test.

The upper spot is a 'regular' 160mW vliolet

The test procedure in action. The video is accelerated 4 times.
[media]http://www.youtube.com/v/_4xyLg0PHyo[/media]

 

[pullbangdead]

I've never heard of the multiple mirror story. As fas as I know, it's just one mirror on each side. When it fails, then the little remaining light is the pump LED. This one pumps the gas in the cavity and inverts the population. If the gas is gone or the mirrors dark, then you still see the LED part working.

We have a little misunderstanding here, that's not how a diode laser works. That would be optically-pumping a gas laser with an LED. A diode laser is completely solid state, and we run them completely electrically pumped. The whole thing is solid, no gases, just one monolithic hunk of material ("single" crystal - The whole thing is grown as a crystal, but different materials one-on-another epitaxially). The "cavity" isn't really a cavity, it's just the area inside the solid, monolithic chunk where the light is resonating. There's wave-guiding in place to keep most of the light in the middle of the diode, in the "cavity", but really it's just an area in the middle of the solid.

(Disclaimer: This is all brief and hand-wavey, just to get an image of what a laser diode "looks" like and how it lases).

Below threshold, the diode is just an LED, only spontaneous emission. The current going into the diode injects carriers into the active region of the diode, in the middle (also the same area as the cavity, ie the place where there is guiding to keep the light in, so the light is kept in the same area as the carriers that are injected when operating as a laser). The positive side injects holes, the negative side injects electrons, and the ones that make it there get trapped in the active region. The carriers, holes and electrons, being forced into and trapped within the active region is what gives the population inversion. Below threshold, these electrons and holes have a finite lifetime before they meet up with one another and recombine. Basically, this look like an electron falling through the bandgap into a hole, and when that happens it gives off its energy in a photon, in a random direction. All the electrons and holes doing this is an LED, all the light going in random directions. Of course, some of these photons will be emitted along the "cavity", hit the mirrors at the end, and bounce back and forth once or twice, but there aren't enough of these to do much stimulating emission since there is as much, or more, absorption as there is stimulated emission.

As you increase current to threshold, those photons in that correct direction along the "cavity" bounce back and forth, and there are enough of them that you start getting significant amounts of stimulated emission. Instead of an electron and hole getting injected and sitting there for some time before spontaneously emitting on their own, there are enough photons bouncing back and forth that the electron/hole pairs get stimulated to recombine BEFORE they spontaneously fall on their own. It's the exact same recombination as in the LED, except when there is enough light in place, stimulated recombination will happen before they can spontaneously do it. It's an LED, but when you have tons of light bouncing back and forth through your "cavity", the electrons/holes get stimulated to recombine so fast that they don't have time to spontaneously emit, instead they get emitted as light coherent with the light already in the "cavity", and you have a laser.

When the electron/hole pairs sit around on their own for a while in an LED, they'll eventually decide to just recombine and emit their photon in a completely random direction. In a laser diode, instead, a passing photon tells them "hey, recombine NOW and emit your photon in this direction, the same direction as me, in the same phase as me", and they do.

Any of that make any sense at all?