Back reflection killing maXYZ module?

I sent my 225mW 660nm maXYZ module back to Chuck because it just suddenly died one me. The beam got very dim and the module got too hot to touch. He claims it is caused by beam back reflections. What would cause this besides putting a mirror in front of the module? It never had any optics in front of it and it was always sitting on a 1/2" aluminum breadboard with nothing under it. Has this happened to any of you? I feel bad that it died, but I have to pay $120 to get it fixed and I really don't want that to happen again.

Because the module got too hot, I was thinking of a short-circuit between the diode's case and the ground of the driver (the maXYZ diode's positive pole (cathode) is the case).
Is it a chance that you connected the alu baseplate together with the case of the diode module (the brass part) ?

I don't see how I could have done that. It was just sitting on the bread board.

Was just a supposition... Maybe you adjusted the focus of the laser with a screwdriver, and touched at the same time the laser and the breadboard? :D
Or is it the module that had was defective...

Quote:

---

Originally Posted by Laser Ben

What would cause this besides putting a mirror in front of the module?

---

The beam reflecting off something shiny in the room and bouncing back into the module... Something as crazy as your metal belt buckle even. You walk in front of the laser and for an instant everything lines up. Poof!

Back-reflection can kill a diode in a matter of NANOSECONDS. No, scratch that. It happens in less than a nanosecond. Here, from Sams Laser FAQ:

Quote:

---

Another possibility is that the return beam from the mirror precisely hits the output facet of the laser diode. While this is a very small area, it only needs to happen for an instant. The result is an extended cavity which suddenly has a much lower loss due to the higher reflectance of the external mirror compared to the cleaved facet. The result is a virtually instantaneous increase in intracavity power and if the laser was running close to the COD (Catastrophic Optical Damage) limit, poof goes the laser diode. This would be more likely with a constant current driver but even in constant power mode, the increase in intracavity power would take place in less than 1 nanosecond - much less than the response time of the feedback circuit.

---

You have to be very careful with a direct injection diode. Especially if you're operating it close it it's absolute limit. I've heard of people killing laser diodes while trying to measure the power output with a lasercheck wand because the small reflected beam from the wand made it back to the diode face. The risk of failure is greatly reduced if you're not running at full power. But when you're near the max power limit, you are in the danger zone.

Quote:

---

I feel bad that it died, but I have to pay $120 to get it fixed and I really don't want that to happen again.

---

Yeah... I hear you. At least take solace in the fact that you bought a Maxyz module and Marconi is only charging you $120 to fix it. If you had bought a diode from anyone else, you'd be paying full price for a replacement. (Diode failure due to facet damage is easy to discern using a relatively low power microscope, so it's not something that you can hide.)

When you build your projector housing, always keep back reflection in mind. Even if the reflected beam has to pass through a couple optical components (dichros, pbs cube, etc) in order to get all the way back to the diode face, it can *STILL* cause it to die. So be mindful of anything that can cause a beam to be reflected back. I purposefully mounted my galvos such that it was impossible for them to rotate far enough to send the beam back to the lasers, even if the amps failed and sent the mirror shafts hard over.

Adam

Hm, so I think, marconi overdrives his diodes... Never heard about red laserdiodes over 150mW CW!!!

You CAN overdrive them, but 225mW??? ouch...

hmmm...
i don't own one of them but everyone here who does has only good things to say about them and marconi.
hopefully laser ben will get to the bottom of what happened but i don't see any complaints about any problems from overdrive.
peace,
wes

It was a great module. Nothing wrong with it up until it died. And yes you can have red diodes up to a couple watts. They have large emitters and are multimode but it can be done.

Hey Adam,

Yet another pearl of wisdom you pass on to us newbies..

When you build your projector housing, always keep back reflection in mind. Even if the reflected beam has to pass through a couple optical components (dichros, pbs cube, etc) in order to get all the way back to the diode face, it can *STILL* cause it to die. So be mindful of anything that can cause a beam to be reflected back. I purposefully mounted my galvos such that it was impossible for them to rotate far enough to send the beam back to the lasers, even if the amps failed and sent the mirror shafts hard over.

I was about to ask, given that I have one of Chucks dual Maxys whether or not one gains any added "protection" via the cube,mirror and dichro.
This has been a good wake up call for me to be careful as well.

Is this back reflection problem as bad with DPSS lasers ??

During setup, when "stuff ups" are more likely to occur does running the laser at very low output offer any degree of protection ??

Cheers:)

Ray
NZ

Quote:

---

Originally Posted by Laser Ben

It was a great module. Nothing wrong with it up until it died. And yes you can have red diodes up to a couple watts. They have large emitters and are multimode but it can be done.

---

Sure, I know that. Ok, you could have missunderstand me. There are diodes with a few watts red ;)
But you said it already: They have big (too big!) emitters, so you want be able to get a nice beam out of them :(

Marconis moduls looks nice, but I think, he uses normal 5.6mm or 9mm-diodes with 40-60um-emitter. In this case, there are no diodes over 150mW!
So I think, he overdrive them a lot! Sometimes it works, sometimes not, like your problem now... :(

Quote:

---

Originally Posted by Pitts

I was about to ask, given that I have one of Chucks dual Maxys whether or not one gains any added "protection" via the cube,mirror and dichro.
This has been a good wake up call for me to be careful as well.

Is this back reflection problem as bad with DPSS lasers ??

During setup, when "stuff ups" are more likely to occur does running the laser at very low output offer any degree of protection ??

Cheers:)

Ray
NZ
[/COLOR]

---

Anything that can return light will cause it, the path is perfectly reciprocal unless you use an optical isolator (expensive, and lossy), but you can at least avoid the cube and other parts reflecting by setting stuff at slight angles.

Temporary use at low power during setup will help a LOT. Best keep it right down though, 10 mW, (30 max). That's safer for eyes too.

DPSS lasers are pretty much immune because the various coatings and crystals form a strong barrier to the output light getting back to do harm. It might be possible to destabilise the cavity enough to notice a flicker, but no more than that.

Quote:

---

Originally Posted by lucerne

Sure, I know that. Ok, you could have missunderstand me. There are diodes with a few watts red ;)
But you said it already: They have big (too big!) emitters, so you want be able to get a nice beam out of them :(

Marconis moduls looks nice, but I think, he uses normal 5.6mm or 9mm-diodes with 40-60um-emitter. In this case, there are no diodes over 150mW!
So I think, he overdrive them a lot! Sometimes it works, sometimes not, like your problem now... :(

---

Hello my compatriot ;)

I don't know if he overdrive them, but for sure he take care to reach the maximum lifetime. He makes a lot of burn-in tests to find the ideal power to run the diode each times he try a new one... Now they are rated at 225mW.
With my module I can reach something like >510mW after the beamsplitter cube on my dual module, this is overdriving the diode (but I did that just for testing purpose).

Seb

Arg, I just tryed to use an old maXYZ 5.6mm diode to make a high power red laser pointer, I did a lot of tests to find a solution to focus it right, but suddenly the diode died http://www.calamarcru.net/Smileys/default/Grrrr03.gif
And nothing about back reflection, it died when I held it and had nothing in front of it... Now it emits a ridiculous amount of red light but still draw full current, and get hot after seconds.

I think it is the excessive heat which destroyed it, it hasn't big heatsink when I was testing...

Here is a solution to the back reflection problem. Buy an optical isolator for the diode laser to protect it. As long as the laser is polarized this is possible. One problem...the optical isolator will cost several diode modules. So the best solution is to use less than maximum power especially when doing initial projector alignments to keep the back reflections slightly off axis. It is true that back reflection can kill a diode 'instantly".

Quote:

---

Originally Posted by Pitts

Is this back reflection problem as bad with DPSS lasers ?? During setup, when "stuff ups" are more likely to occur does running the laser at very low output offer any degree of protection ?

---

Hi Ray!

DPSS lasers are not nearly as suceptible to this sort of damage. Remember that the reflected light is going to be a multiple of the original diode frequency. So in a green laser, for example, when the reflected light enters the KTP/Vanadate crystals the coatings will be all wrong for the 532 nm light going back towards the diode. The optics, too, will be focused for IR and not for visible light. Finally, the pump diode normally isn't running anywhere near it's maximum possible limit, so you have some headroom there. (This partly answers your other question too: Yes, operating at low power levels reduces the risk of facet damage due to back reflection.)

Bottom line: I don't personally know anyone that has killed a DPSS green laser through back-reflection. But I know 3 or 4 people that have killed direct-injection red diodes via back-reflection. (Including myself!)

As far as running lower power during setup, that is *absolutely* a good idea, and not just to prevent back-reflection. It's easier to align things when the power levels are lower.

After I built my optical table I connected all the blanking leads for the lasers together to a variable voltage bench power supply. Then I dragged the whole mess out onto my back porch at night and did the alignment. (Alignment is tricky, but if you have a really long throw it's easier to see where you're off.) I dialed it in up close using the adjustments on the individual laser tables, and then used the dichro mounts to tweak the far-field alignment. (Far field ended up being about 100 ft away from the projector.) At night you can see the spot 100 ft away even with the power turned way down, yet with the low power it's also easy to look at the dichro and make sure the beams are intersecting at the right point.

Adam

EDIT: D'oh! Just noticed there was a page two in this thread. Looks like the Doctor answered your question quite nicely. Sorry for the dupe.

Quote:

---

Originally Posted by The_Doctor

DPSS lasers are pretty much immune because the various coatings and crystals form a strong barrier to the output light getting back to do harm. It might be possible to destabilise the cavity enough to notice a flicker, but no more than that.

---

i hope not to drag the thread too far off topic, but i find this interesting because i had some flicker in the cni 532 i used.
i did have some shiny aluminum surfaces showing in the optics area ,so i blackened them and the flicker went away.

hmm what about back reflection off an output window.

I regularly see light thats reflect off our projector output window, hittng the galvo mirrors in turn and eventually end up on the lasers...

its very random, ie laser has to be at the perfect angle... maybe the solution here is to tilt the output window so its not facing perfectly back at the scanners ?

Quote:

---

Originally Posted by aijii

hmm what about back reflection off an output window.

I regularly see light thats reflect off our projector output window, hittng the galvo mirrors in turn and eventually end up on the lasers...

its very random, ie laser has to be at the perfect angle... maybe the solution here is to tilt the output window so its not facing perfectly back at the scanners ?

---

the same is with pol cubes... even if they are ar-coated a small amount of light will return back to the diode. You should be careful, setting the cube not exactly perpendicular to the beam. Pol cubes reflection is much worse than window reflection, because the beam is fixed so if the cube is perpendicular to the beam the diode will receive back reflections as long as it is powered! (i'm quite worried about that because i have 2 maxyz's red diodes with pol cube)

if you are worried about back reflections from the output window, you have to tilt the window more than half of the maximum scanning angle, otherwise you will only change the point from where the beam will be reflected back. I think the easyest solution is to mark that point with a permanent black marker, that absorb the light :D:D:D:D

I have killed 4 (yes, four) of them with back-reflection, so far.

I've stopped worrying about it. They have been infinitely nice about replacing the diodes for very cheap.

Quote:

---

Originally Posted by eyceage

I have killed 4 (yes, four) of them with back-reflection, so far.

---

Oh man... :( I think I'd be in tears after that. I recently managed to kill another one of my DVD diodes, but I think that was from static electricity, not back-reflection. (I've been careful to the point of paranoia ever since.) But back-reflection is always on my mind.

Adam

I killed 2 Maxys at the same time by rotating a test mirror mounted on a screw..
D'oh -- aint gonna do that again.. now I got a firmly planted output mirror that doesnt rotate.. until I get my galvos..
Then gonna be real careful still.. DI arent forgiving like dpss..

Damn... :eek: I didn't know there were that many dead Maxyz around. Is there anyway to protect them from this? :confused:

Yeah - don't reflect the beam back into the diode! :D Seriously though, while you're setting up the table and aligning everything it's smart to run all the lasers at low power. When I did my alignment I was running less than 1.5 volts to the analog rails of the lasers. This also makes alignment easier because the specular reflections from the optical components are quite dim at that low power so you can easily see where the beams are landing on the optics. With the power cranked up I find that the specular reflections are so bright I have trouble even looking at the dichros, much less trying to verify that the beams are overlapping.

Adam

Quote:

---

Originally Posted by steve-o

I killed 2 Maxys at the same time by rotating a test mirror mounted on a screw..
D'oh -- aint gonna do that again.. now I got a firmly planted output mirror that doesnt rotate.. until I get my galvos..
Then gonna be real careful still.. DI arent forgiving like dpss..

---

You got my full sympathy there. I remember when I first discovered retroreflection death. I'd just paid £530 for TEN diodes, they cost so much then that's all I got for my money, and they were only 80 mW too. It's why I never really followed through on my designs. I'd destroyed nearly HALF of those diodes before I'd established that retroreflection death was a real phenomenon, there was nothing in the LaserFAQ about this, and the makers and sellers of the diodes certainly weren't telling anyone. I had to push for weeks to even get them to admit it. (The seller was much more helpful than the maker though).

I won't be building anything grand, I decided. Not worth it. No-one will pay for the engineering required to make a red laser be able to change diodes easily like projector lamps, it would cost far too much money to do it, and it's why no-one with any commercial sense does it. The nearest to that is to charge money to refit a laser already sold. I've considered that too, as Marconi does, but ruled out building entirely, as I discovered I can get more money far more easily just selling diodes! As I have more diodes than I can use, and far less money than I need to finance laser building properly, I'm doing that. What that leads to I don't know. Maybe an FB3. If people DO decide it's worth paying serious money for designs that allow easy refit of diodes, I might build some, but I doubt it, to make it truly EASY to get the same kind of alignment and thermal coupling and avoid easy contamination of clean parts would be so difficult it will never be something for the casual end user, and it could drive the price up tenfold unless you had a serious commercial machine shop already owned and in your own pocket. It would be like the difference between a bit of DIY metalwork and a set of finest Newport optics mounts.

Quote:

---

Originally Posted by Laser Ben

He claims it is caused by beam back reflections.

---

How does he know this? Can it be proven??

Yes...

ALL Red diodes can be destroyed from back-reflection
I have not found one that couldnt..

You must be very careful during alignment or just playing around as it
is very easy to reflect some light back in an destroy the diode.

It is very unlikely that the diode will just die.. but that does happen too.
But we cant rule out static discharges and improper focusing techniques.

In returns that we get, It is very hard to tell if back-reflection actually was the cause of the damage.. thats why we use a very expensive microscope.
That way we can see if the facet was damaged.

Since we do not have a warranty on them because of the low price and because of the nature of laser devices and their use-
We must charge for a diode refit after a unit has been out in the field for some time..its just cheaper that way.

Otherwise the prices would be double to justify a warranty.

Thanks for reply. I have various diodes here that have been killed and a microscope. What do I look for to see if back reflection has seen them off or if something else has done it?

Well, There are several indicators of damage.

Either the facet face or a fracture in the die.
The face is easy to see in a microscope.. but the die takes a bit more to see.
Since the diode is basically dead. and we dont want to fire it up directly,
Sometimes we have to excite the die by an external light source (another laser) to see the damage.

Since this would take up a lot of time trying to explain.. I'll make this short.
I think a read in Sams laserfaq and maybe some googling may be in order since this has been done already.

For those who might be interested http://www.laser-diodes.net/techinfo...mg_prevent.pdf
has some typical senarios of how some diodes were damaged and what they think damaged them.
The example in section 5.2 is prettymuch how we look at diodes.

But the short of it is, this mostly shows up as measles on the output face of the facet or abnormalties in the active layer.

But it still is very hard to know whether the damage was optical overload or electrical without evaluation. This is a study in itself.
There is tons of info on the internet about this.

Of course you could purposely damage some diodes in different ways by expermentation an see the results.
But that could get expensive...

Thanks for that. The other problem I'd have dissecting them is that I've had to remove them from the lens and the only way I could do this is by tapping the diode back out from inside the lens so probably did it more damage anyway.

Quote:

---

Originally Posted by The_Doctor

I'd destroyed nearly HALF of those diodes before I'd established that retroreflection death was a real phenomenon, there was nothing in the LaserFAQ about this, and the makers and sellers of the diodes certainly weren't telling anyone.

---

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.

I often pick random sections of the FAQ and just start reading when I'm bored at work. (I've kept an up-to-date copy of the FAQ on my hard drive for the last 5-6 years or so, both at work and at home, and I'm *still* finding cool things in there.) It's amazing how many useful nuggets of information are buried inside some of the more mundane topic headings.

I agree that most diode manufacturers won't warn you in advance of the danger inherent in back-reflection, however.

Adam

Quote:

---

Originally Posted by Buffo

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).