Can you use a beam splitter to combine beams?

[The_Doctor]

Their beam profile is not round or square still. LOL that build didn't stay like that for long. I also gave up using LPC as they kept giving up on me! Now using Optnext 638nm diodes that punch hard with a tight more visible beam. Even the camera picks up red better at this wavelength.

What's the difference in beam width in each axis for the new single mode diodes you're using, for 10cm downrange, and 10m downrange? While single mode diode beam cross-sections aren't exactly round, the price of a halfwave plate suggests you hit the law of diminishing returns. Accepting the rotation of one diode might give a closer-to-round beam for less money, and most of the beams of each diode will overlap. The strong core certainly will, after the first few inches.


A bigger improvement would be had by spatial filtering or even just by tight control of coaxiality of diode emitter and lens. That;s probably where I'd spend extra, to reduce spill by good centering and keep a distinct hot core, that way the non-overlapping parts of a combined beam will have much less energy. (The beam ought to look good even without spatial filtering, based on the Opnext diodes I have used, so I would deal with diode and lens positions instead.)

If I ever build a laser again it will be VERY simple. New, stronger diodes will become cheap before I'll ever want to make a complex laser module. Even the 120mW multimode green allegedly comes close to a single mode's beam quality, and I imagine other colours' specs will follow. If these new multimode diiodes can look as good as single mode, they'll be less prone to retroreflection death than single mode diodes despite a fairly high energy density. Likely less prone to ESD too. If they can be treated optically like single mode diodes, I'll do exactly that. The new red from Oclaro (new name for Opnext) is said (by Planters who asked about it at a trade show recently) to be stronger AND have a narrower divergence than the P73 reds, so I think that a lot of the expensive and complex, fiddly-alignment builds of recent years will give way to much simpler systems. If the beams keep getting better, people won't want to spend on waveplates that won't add a lot to the quality.

[TechJunkie]

I'll be using 2 of these diodes. http://laserpointerforums.com/f50/ne...ml#post1064590

Unfortunately there is not much info on these. I need to fire one up to check the beam profile. But these should work without a waveplate, correct?

[The_Doctor]

I strongly suggest: AVOID. By the time you have handled those little bitty diodes, found them a bitch to mount accurately, have contaminated their facets (despite the best possible efforts you'll ever manage) and seen them die early again and again having never been accurately mounted well enough to give a really nice beam, you'll wish you'd tried something, ANYTHING, else.

So look elsewhere now. Don't reconsider those diodes until you're damn sure you haven't found a better shot. I think you'll find several. While the low price of those diodes may be tempting, the problems will make you pay through the nose anyway. At least find somethign with minimum of 5.6mm can and a proper can at that, with a little glass window. You'll be a LOT happier in the short AND long term. Most lenses expect that little window, and aspheres are usually optimised for it, so you might as well get it because it helps the beam and protects the diode chip. Omitting it might pay if you're mass-producing millions of DVD drives for people want them yesterday for less than a tenner because they don't care if the drive fails before its diode does, and have an entire ultra-clean production line to handle them with, but they'll suck in a DIY build.

[TechJunkie]

I strongly suggest: AVOID. By the time you have handled those little bitty diodes, found them a bitch to mount accurately, have contaminated their facets (despite the best possible efforts you'll ever manage) and seen them die early again and again having never been accurately mounted well enough to give a really nice beam, you'll wish you'd tried something, ANYTHING, else.

They are already mounted. I had custom cut module heads made from aluminum especially for those diodes. They are the same diameter, and lens thread size as the AiXiZ. The 2 diodes and modules cost me $6 shipped.

[The_Doctor]

Ok, well if they give you nastiness squared, you have been warned. I'd never use a bare diode chip. People do, but I think many of those people post stuff asking why their P73's die too.

[swamidog]

the 3.8mm diode mounts that DTR sells works quite well.

https://sites.google.com/site/dtrlpf/home/diode-modules

I strongly suggest: AVOID. By the time you have handled those little bitty diodes, found them a bitch to mount accurately, have contaminated their facets (despite the best possible efforts you'll ever manage) and seen them die early again and again having never been accurately mounted well enough to give a really nice beam, you'll wish you'd tried something, ANYTHING, else.

So look elsewhere now. Don't reconsider those diodes until you're damn sure you haven't found a better shot. I think you'll find several. While the low price of those diodes may be tempting, the problems will make you pay through the nose anyway. At least find somethign with minimum of 5.6mm can and a proper can at that, with a little glass window. You'll be a LOT happier in the short AND long term. Most lenses expect that little window, and aspheres are usually optimised for it, so you might as well get it because it helps the beam and protects the diode chip. Omitting it might pay if you're mass-producing millions of DVD drives for people want them yesterday for less than a tenner because they don't care if the drive fails before its diode does, and have an entire ultra-clean production line to handle them with, but they'll suck in a DIY build.

[The_Doctor]

Dave sells one too. I bet they work ok, I just think that once people start to push diodes they need a lot of heat removal. Remember that Vf goes up too, not just amps, so P goes way high. Got to get rid somehow. I'd stay with 5.6mm and a nice windowed can, to make heatsinking easier, and get better and more reliable beams from them. Fortunately Oclaro seem to agree, so I can do this sometime if I want to try building again.

[TechJunkie]

Ok, well if they give you nastiness squared, you have been warned. I'd never use a bare diode chip. People do, but I think many of those people post stuff asking why their P73's die too.

Unfortunate budget takes precedence right now. I want better parts, but the more I have to save for them the longer my PJ will take to complete. Its been almost a year since I started gathering parts, and I still dont have galvos, drivers, power supplies, baseplate, case, connectors, and some odds and ends. I did just get a great deal on a DAC, but I had to sell one of my backup laser modules to afford it.

Modules are easy to swap out. So there is always room for future improvement. Right now I just want to get this thing running. You have no idea how disheartening it is to be here and nothing to show.

Another question please? Can I run the 2 red diodes off the same Flexmod driver?

[The_Doctor]

I don't know the Flexmod but I think it was designed to do TEC control as well, so the odds are it can handle Vf greater than that for a couple of diodes in series. Beware that green diodes alone maybe have more Vf than three reds in series, so it's a question of how much headroom for volts the Flexmod will give you. Probably quite a lot given that it aims to control current, not voltage.

A thought on galvos.. If you go for 20K with the best accuracy you can get, you'll get a good return for your money. Can always speed it up later. The demands on torque, speed, power are so much lower for 20K than 30K. Good optimisation of scan route will get the best out of them, and teach stuff you can use later too.

Can you find a broken PA amp for free? If so it might get you a lot of useful projector bits in one go. Heatsinks, fans, cabinet, some switches and lamps.. You can buy blank 3U or 3U panels to fit on the front, drilled and cut to make the projector. Big aluminium shelf in there for baseplate...

[TechJunkie]

A thought on galvos.. If you go for 20K with the best accuracy you can get, you'll get a good return for your money. Can always speed it up later. The demands on torque, speed, power are so much lower for 20K than 30K. Good optimisation of scan route will get the best out of them, and teach stuff you can use later too.

Can you find a broken PA amp for free? If so it might get you a lot of useful projector bits in one go. Heatsinks, fans, cabinet, some switches and lamps.. You can buy blank 3U or 3U panels to fit on the front, drilled and cut to make the projector. Big aluminium shelf in there for baseplate...

1. Can you recommend a good 20K galvo?

2. Speed up the galvo later? Do you mean replace with a higher speed galvo later, or physically turn up the speed? I have no idea how to tweak galvos.

3. You said "Good optimisation of scan route will get the best out of them, and teach stuff you can use later too." I have no idea what you are talking about. How do you optimize a scan route? Are you talking about beam alinement on the galvo?

No, I dont have an amp, but that a great idea. I'll keep my eye out for one.