little question about laser building

[SJP]

ive seen a bunch of DIY lasers but do they build the laser from scratch or do they take the laser out of a dvd drive and add a power suupply, heatsink etc......?

 

[HaloBlu]

Build the laser from scratch? 100% from scratch you mean? Nearly every laser your going to run into here or online requires a laser diode or gas filled glass tube to produce the laser beam. Power regulating driver, heatsink, lens focusing module, body (host) can all be DIY. You can't build a laser diode from scratch. You also need a lens. (unless its a gas laser)
A TEA laser can be built 100% from scratch. Ok still not quite 100% since you need a 5kv transformer.

 

[Morgan]

I'd just buy one. Much easier! Saves time and limits useless posting. Of course you can try following some of the instructions about removing a DVD drive laser diode, adding a power , (sic), suuply, heatsink, and see if works out... I doubt anyone here really knows what they're talking about though. See how you go.

M

 

[SJP]

Okay...that clears things up...mods can close this now

 

[Cheech]

I'd just buy one. Much easier! Saves time and limits useless posting. Of course you can try following some of the instructions about removing a DVD drive laser diode, adding a power , (sic), suuply, heatsink, and see if works out... I doubt anyone here really knows what they're talking about though. See how you go.

M

Are you kidding, DIY is way better/cheaper. And not much experience with electronics is required to put the pieces together.
You can make a burning laser for under $20 usd

And I'm pretty sure many people on here know what they're talking about.

 

[HaloBlu]

I doubt anyone here really knows what they're talking about though.

Fairly sure that was a joke there Cheech.

 

[Cheech]

Seems like SJP's other posts are looking for diy answers, but we nolonger have MFO's guide to point to. But rog8811's Laser driver - It can be done is still up

 

[Morgan]

Yep, post #3 was a joke. Hmmm. Seven posts in but is sceptical anyone can use harvested diodes but knows what a 7 colour system and what beam splitters are about? And then knows the mods can close this thread? I smell troll.

M

 

[pullbangdead]

^He was asking if people actually MAKE the laser diodes that go into DIY builds, or if the laser diodes are just bought/harvested.

The answer is that hobbyists either buy the laser diodes or harvest them from consumer electronics. Things like TEA lasers can be made, but laser diodes require a heck of a lot more resources and know-how than your average hobbyist can afford. We're talking serious time, serious money, and serious know how. So no, hobbyists generally don't make their own laser diodes. But hey, anything is possible, technically speaking.

 

[Morgan]

^He was asking if people actually MAKE the laser diodes that go into DIY builds, or if the laser diodes are just bought/harvested.

Thanks PBD,

If I've misunderstood and that is the true nature of the question then I offer an apology to SJP. We can all sometimes be too quick on the draw.

To be productive, further the thread and delve deeper on the theme then, how easy/hard would it be to home produce a laser diode that output even the tiniest coherent light?

M

 

[pullbangdead]

Thanks PBD,

If I've misunderstood and that is the true nature of the question then I offer an apology to SJP. We can all sometimes be too quick on the draw.

To be productive, further the thread and delve deeper on the theme then, how easy/hard would it be to home produce a laser diode that output even the tiniest coherent light?

M

No problemo, dude. :beer:

And making a working laser diode is very difficult. Very, very difficult.

Really though, it depends how much of it you need to do yourself to feel like you made it. In the lab I work in, when we start we have a bare substrate, some liquids, and some gases and when it comes out the other end, it's a working laser diode. So I would personally consider that making a diode from scratch.

In some groups, they buy material that has already had the hardest part done, which is the crystal growth, and then they do the second half of the process.

You can also buy bare, unpackaged commercial laser diodes that are already fabricated and package them, and that may very well feel like making a laser diode to many people, as it goes from a little semiconductor chip to a packaged diode that can be mounted in a laser.

But having made laser diodes, I do not in any way consider it to be something a hobbyist would do at home. The labs I work in have millions of dollars in capital invested, have full-time staff dedicated solely to keeping tools running, have a staff of Ph.D-level scientists leading projects, and are headed by some of the foremost experts in the industry, and even in an environment like that, it's still difficult to make laser diodes "from scratch".

I'd be glad to try to answer any more questions on the subject of actually making laser diodes.

 

[Morgan]

I didn't think it would be an easy one. Not a kitchen table project then?!!!

I saw a link recently that showed how to mount a bare 808nm chip and that alone looked a tricky process.

There are probably a million more questions I can think of but just wouldn't know where to start to be honest...

... Okay, here's one that just pops into mind. What is it about their construction that determines wavelength? I think I know that substrate properties are similar for instance for 405nm and 445nm but what are the finer details? Is it possible to say, "Right, I need an output, (randomly), of 427nm", and mix up a batch knowing the outcome?

M

 

[pullbangdead]

If everything is working properly, you can anticipate the wavelength quite well, and indeed, devices very often come out exactly as designed. If things are not working properly, it can be a crapshoot. If something goes wrong, it can be pretty off.

The wavelength is determined largely by the bandgap of the semiconductor used in the active region. It is also dependent on other things with a laser, like the waveguide, but the biggest factor is bandgap. You mention the violet and blue being similar, and that is absolutely true, they are VERY similar. Start with the same substrate, grow a very similar structure, and you can make either wavelength. You "just" (haha, it's not simple) have to decrease the bandgap of the active region (the place the light is emitted) and then tailor the rest of the structure to fit the other wavelength. If you're starting from a violet laser structure and just change the bandgap to emit blue light, that basically screws up several other components of the structure, and you have to adjust the others in order to fit the new device and new wavelength. With LEDs it's much easier, there's much less to have to adjust, but all the pieces have to come together to make a laser diode.

In the violet/blue/green laser diode region, the active region is made of InGaN, indium gallium nitride. Basically, GaN has a higher-energy bandgap that would emit in the UV. Adding indium, which replaces the gallium in the crystal, lowers the bandgap, thereby changing the wavelength of light emitted. In a perfect world, adding indium would be easy, and it is pretty easy to add a little bit of indium, hence making violet lasers is much easier than blue, which require more indium. And even harder are green lasers, which require a LOT of indium. Basically, as you add indium, the size difference between In and Ga atoms creates strain in the material, and sometimes instead of staying "GaN with In evenly interspersed throughout", with too much indium the material would rather become "InN and GaN", instead of "InGaN", as the In and Ga will cluster together with atoms of their own kind. This can actually be a good thing sometimes for some applications, but it's generally not very good for lasers.

So yeah, in the blue end of the spectrum, it is currently possible to grow a laser diode with any wavelength between near-UV (~380nm, I believe) and green (highest reported 531nm). And you somewhat pick what wavelength will come out. Mind you, at the extremes of UV and green, VERY few companies or universities can make those, because they're very hard to make.

In the red end of the spectrum, some things are easier, but some things are harder. They worry a lot more about something called lattice-matching, and have different variables spaces to work in. That is to say, some things that are adjustable in the red region are not adjustable in the blue region, and vice-versa.

Another little detail is that the wavelength never comes out EXACT. Hence diodes always having the +/- 5nm, or even +/- 10nm, rating on the wavelength. There is just variability, and you can't get that precise. If you do everything right, the diodes will average a certain wavelength, but even within a single wafer, the wavelengths will vary from different places on the wafer.

Make sense?

 

[SJP]

What about gas lasers? Can those be built. I'm guessing not unless your at a lab

 

[Morgan]

What about gas lasers? Can those be built. I'm guessing not unless your at a lab

There is a great site that explains just about every nuance of constructing a homemade gas laser. It's called Sam's Laser FAQ. Here's a link to the Amateur Laser Construction section - http://www.repairfaq.org/sam/lasercon.htm#contoc - but there are many more sections to have a look at including, direct output diode pointer building, DPSS systems and power supplies for all types. It's considered pretty much as the bible round here. (pullbangdead is somewhat of a prophet too! )



pbd - I think I understand a lot of that. I can see why atoms of different sizes would create stress and tend to cause, 'clumping'. From what you say, a more homogenous material is required for laser diodes. Bandgap is a word I have to add to my vocabulary though. (I think), I understand the principle at a basic level but there are ideas and a few words surrounding the description that need to be understood to get it completely. I've briefly been through the P-N junction at college describing holes and free electrons and it rings bells with that, correct?

Just to see if I have some sort of handle on it... To shorten the wavelength, (or increase the frequency whichever way you prefer to see it), you reduce the bandgap? This is to do with energy to cross the gap and therefore you'd need less of it?

If that's case, would we expect the forward voltage of any future greens to be less than current blues and BRs, (assuming construction methods and materials remain similar)? We already see a reduction from 5+V for BR to 4+V for 445nm. I may be way out here but to my tiny mind it seem sort of logical. Would their optical output also be expected to be higher?

I'm speculating, I know, but interesting none-the-less. You describe your science very well! :beer:

M

 

[Kevlar]

@morgan, I was just going to link sam's FAQ. You beat me to it...