combining beams

[stanwax]

Badger
the PBS works a treat however its out of a DVD drive not a burner and it was an old one (maybe 4-5 years old). Problem is as time goes on the cubes get smaller and smaller now they are <3mm whereas in the good 'ol dayz they were 5mm just right for laser diodes. if you use one from a DVD RW it will work but you will spill too much light over the sides

See if you can pick up an old drive - even better a dead one as it will likeley be free! Im pretty sure this cube came out of a Samsung drive - I have a model mean written on my whiteboard above me of SD616 DVD master 16E - if its any help.

Rob

[450nm]

Well I just quickly slung the little DVD optic block together with a couple of low power LD's and wasn't impressed. While the 2 did indeed combine the losses through the block seemed to negate the gain.

Maybe it will be better when the whole caboodle is set up a bit more robust?

One other thing. is it essential to mount the diodes side by side and use a mirror at 90 degrees? Would it not lead to less losses to have the diodes at 90 degrees to each other into the block? Seems a lot less work to me.

[stanwax]

450

No you dont need to mount them side by side I just did this for space reasons and its much easier to do the far field alignment if you have control of at least one of the beams. Plus that optic I have used makes a tiny tiny loss to the beam so dont worry about effeciency.

Also I was unimpressed with the process myself until I really got a handle on what was needed.

First of all I would concentrate on the laser that is shining through the cube. You will find that there will be some light going through and some bouncing at 90 degrees - turn the diode until the bounced light is minimal. Then try rotating the cube (see image) by a few degrees until this bounced light is diminished further - until you hit the sweet spot. Now you need to feed the second laser into the side of the cube but with its diode positioned so it is rotated at 90 degrees to the first laser - this is where the beam patterns in my photo come in (see below). This diode will also need to be angled slightly to the cube to max the ammount that is bounced to the output.

Then align the far field spots and wallow in nearly double the power. I get 350mW from a pair of DVD diodes.

You mentioned in a previous post about my photo - this does not indicate the polarity of the beams but just their relative orientations. These diodes produce a polarised beam and you do not need to know any more than that save the fact that the 2 diodes outputs must be at 90 degrees to each other in terms of rotation about the beam axis. My image is a simple and clear way of seeing the relationship between the two diodes. If you have a round housing for your didoe then you can easily rotate the module to max the throughput.

The confusing thing for everyone seems to be that there are 2 different 90 degree angles involved. One is that the beams hit the cube at 90 degrees to each other. Second is that the beams must be rotated so they are at 90 degrees to each other along the axis of the beam. (this rotates the polarisation)

[badger1666]

thanks stanwax you and the doctor have let me know alot of good stuff
would have been hard work without you guys i dont think i will rape dvd burners for the diodes i will proberly reck them getting them out the doctor is selling diodes on ebay i may well buy some from him the only worry i got is that they are rhome and here they are not to reliable
thanks again guys

[450nm]

Following you so far and assume tha the little cuts on 5.6mm's cases might there to indicate polarity then. So why is the beam oval and not round? Fascinating stuff.

[stanwax]

450

The cuts are so that you have a predicatable repeatable way to position the diodes but it is not related to the polarity as such. I dont profess to understand the intricacies of polarised light - but to be honest to do this you dont need to know about it either! All you need is the relationship between the 2 diodes you are using. The cut does not tell you that the polarisation is A or B unless that is stated in the data sheet. As long as them there diodes are 90 degrees wrt each other thats enough knowledge.

The beam is oval because of the divergent effect of the very narrow beam appeture on the laser diode chip. The emiting area is oblong and the narrow part - top to bottom - is often only a few micrometers in height. Passing light through such a small appeture causes it to fan out. As the emitter may be 3x100 micrometres you get more fanning out in one direction than the other so when you stick a lens on it the beam is always going to be fatter in one plane than the other. You will be able to focus the beam at some far point down to a spot but if you could look closely at that magnified spot it would still be a stripe - its just your eyes cant see it cos its so small.

Rob

[The_Doctor]

...the doctor is selling diodes on ebay i may well buy some from him the only worry i got is that they are rhome and here they are not to reliable

They're not significantly worse than any others. They're just the first most people tried because a lot of them appeared on eBay, so they were the first to cause widely reported disappointment, but not the only ones. People are still looking for diodes that will do 200 mW for long enough to be left to themselves for a few thousand hours, and only Marconi seems to have any, and I'm not sure if anyone's actually run them that long yet.

You might get marginally better from diodes meant for 18x or 20x burners but it's a small difference, that's why burner speeds aren't leaping like they did with CD-ROM's and CD writers. There are no single mode red diodes significantly better, no-one's made any yet, unless in some lab not yet ready for marketing. That's why there are people here trying to find diodes that will do 200 mW constantly for a long time, reliably.

The only ones anywhere near close are those Marconi is using in the MAXYZ modules, and he's not telling anyone what they are, exactly. All we know for sure is they are narrow stripe multimode types with 50 nm stripe width and maybe a longer cavity that makes them easier to get good performance from. Another possibility that occurs to me is they might have some compound microlensing built in. Whatever they do, they do it needing a taller can than any standard 5.6 mm diode.

If you're limited to DVD diodes, none will reliably hold at 200 mW for a long time, which is why people are still asking questions about what single mode diodes will do this. No-one's found one yet, or if they said they have, they've either found later that they weren't as good as hoped, or they aren't counting hours.. I tested one of the Opnext diodes at around 180 mW and it died within several days during some quiet moment while nothing was happening to it. Most DVD diodes will give a lot of effective life if they're not trying to burn their way uselessly through the wall all the time. For one thing, no analog mod show projector will be running them flat out all the time, and if you tried to run those for the thousands of hours, the galvos would break down first, probably.

More on red diodes here:
http://www.photonlexicon.com/forums/...ead.php?t=3237
and here:
http://www.photonlexicon.com/forums/...ead.php?t=2180

The Rohm's have a CW rating for around 120 mW, Someone mentioned others at 130 mW, the Opnexts possibly, though I thought they might be a tad higher, possibly 150 mW though if so they shouldn't fail well within two weeks at only 180 so I don't think they're that strong. They cost more too. Proportionately, the Rohm diodes are a good deal even at the full price I sometimes get of &#163;60 for 5. Opnext's are &#163;75 for 5, even if part of a larger quantity, which is why a group buy was arranged to get that price. No-one's complained to me about Rohm's bought from me, and two buyers have come for more. At least, one has, another is intending to, and a couple of cubes... And a third has bid again, but not yet won more.

If in doubt, look at the MTTF data sheet for the Rohm diodes I linked to on my listings. That shows some actual data. The reason so many Rohm diodes showed up on the net wasn't mass rejection, it was a single engineer, Fleming H Pedersen in Denmark, he bought many thousands of them for some large project, a kind of optical network scheme, and it didn't go ahead so he sold them, 500 at a time. I got lucky, I had his last 200 from a split box he'd already opened. Wanted 500. Was too late though, just. Guy in Germany bought 2000 of them, and he's doing a nice trade in lasers built from them. (Herr Nicolai Pusch). Florian Rotter was doing similar, with ABSee Laser, also of Germany, but he stopped doing that, I think he prefers to do specific work for individual clients, a lot of it involving gas lasers. That's how I got some cubes, he was selling stock from the small laser building exercise.

I think I learned WHY they stopped. I made a couple of lasers, and they sold for a pittance. Sure, they had no modulation, but they were built like brick shithouses, with smooth stable focussing, BNC power input, very tough metal casings, very good power supply accepting wide ranging inputs from crude wall warts and with good protection against surges, static, reverse polarity.. They can be updated for years to come. And they sold for LESS that the value of their parts!

I think this is the real reason why few decent direct diode lasers are made at low cost. People have come to expect them to compete with cheap tubes stuffed with epoxy and a couple of wires sticking out the back. Marconi sticks with it, he's managed to build a following, and has a source of diodes and optics that have made it economical to do that, but few have managed this. I'm now building lasers ONLY to order. Currently without modulation, but with manually variable output, for astronomers who've learned that green sucks if you want your eyes to stay adaptd to darkness. Their divergence usually sucks too... If I raise the money to finish getting the tooling I need to do this economically, I'll extend to a modulated version, and possibly a PBS paired version, but I doubt that will ever fetch enough money to justify the cost of parts and building it. Diodes will always need replacing, however good they are, and enginnering a laser well enough to stand repeated access COSTS, there's no way round that.

In short, the diodes themselves aren't really the problem. For people who want to build their own lasers, any of these diodes will do. They'll all do over 200 mW too, the question is how long for. The whole point of buying a hadful, cheaply, is it doesn't matter then, you just put together something that lets you change them. Until the ever-elusive 200 mW cw single more red diode stops being a pipe dream that people ask for every few months, this is as good as it gets. And not ONE of these diodes is immune from static or retroreflection or carelessness either.

[The_Doctor]

First of all I would concentrate on the laser that is shining through the cube. You will find that there will be some light going through and some bouncing at 90 degrees - turn the diode until the bounced light is minimal. Then try rotating the cube (see image) by a few degrees until this bounced light is diminished further

That's interesting, but if you optimise the diode orientation you won't get a lot more out. You'd be seeing maybe a little change in the reflected light with cube optimisation, but as a percentage of what's going through, it's negligible. It's more important to get the angle aimed to avoid retroreflection. If you can do both, great, but I'd put that avoidance as a priority, then you can push the diodes a bit harder, for longer. Certainly hard enough to more than make up the small loss in unoptimised cube orientation.

[p1t8ull]

I tested one of the Opnext diodes at around 180 mW and it died within several days during some quiet moment while nothing was happening to it

Doc, you were pushing that Opnext tester to 225mW: <- EDIT - 220mW
http://www.photonlexicon.com/forums/...6&postcount=53
Before it died:
http://www.photonlexicon.com/forums/...1&postcount=68

Whatever the failure time when overdriving any diode CW, I think your statement that in an analogue projector the diodes are rarely pushed to the max, is a worthy point to note. Most of these DVD diodes will pulse at well over 300mW so there's some headroom for overdiving them in this environment - as opposed to CW in a lab.

[stanwax]

That's interesting, but if you optimise the diode orientation you won't get a lot more out. You'd be seeing maybe a little change in the reflected light with cube optimisation, but as a percentage of what's going through, it's negligible. It's more important to get the angle aimed to avoid retroreflection. If you can do both, great, but I'd put that avoidance as a priority, then you can push the diodes a bit harder, for longer. Certainly hard enough to more than make up the small loss in unoptimised cube orientation.

Doc
I guess it may depend on the cube itself and my be a function of the DVD cubes I have used but I can get increases in the outputbut that are easily visible - ie the output beam going visibly brighter - by turning the cube a fraction. This is after optimising the position of the diode (rotation). This has come to light (no pun intended) right at the momnet as I am working on combining 2 diodes for a fellow PL member. He has square housings for his diodes (see my picture of the beams with no lenses) this makes it difficult to get the diodes turned optimally so this method will work in such an instance. Plus angling the cube a little will help to keep back reflections at bay - certainly off the first surface of the cube.

Rob