Mitsubishi ML501P73 500mW 638

[Stoney3K]

i think you're being way too optimistic.

I think it's interesting to hear all the solutions from people who are trying to get the best lasers possible for show use. It's always possible to go the easy route (quad or 8x LPC826) anyway.

Those Mitsubishi diodes are 500mW with a huge emitter. If you were to stack them into a quad next to each other and PBS them with another quad, you can get a square, 4W beam out of it.

Now that beam is going to be a huge flashlight since you've essentially stacked an image of big emitters together, and the laws of physics say there's nothing you can do about that. The only thing we can do to tighten the beam is reduce the size of the image -- effectively chopping the edges off the beam to make it round and concentrate the power in the center.

This would imply shooting the whole lot through a circular aperture after it's been collimated. Yes, that means losses, but you'd have those losses anyway if half of your flashlight spills over the galvo mirrors. I'd personally be more happy with a tight 1-2W of 638 coming from the same assembly (which would be about $700 worth of diodes, plus mounts and the lot, still coming in at about $1/mW) than a 4W flashlight which is worthless half the time.

[kiyoukan]

My gut feeling is that with the chopping of the beam plus a prism pair the beam would be usable for scanning.
I would love someone who has these tools to do the testing but im getting them soon.
not to bad only a 300$ for possible high powered decent beam 640.
if this works out even if i use high loss optics i should still get 300mw and if i use good optics and push the diode a bit i should get from 450 to maybe 600mw (pushing it more than a bit to get that high tho)

[swamidog]

i have experimented with these diodes and a prism pair. at *extreme* correction, i managed to get a spot that was 1"x.3" at 40'.

i did not measure the beam size exiting the prism, but it was fairly large and there was a lot of loss from the prisms.

in theory you could prism correct two beams, then pass them both through a cube and then a pinhole filter to trim the cross into a splotch. you will then have to worry about fitting the beam on your scanners and matching the goofy divergence of your beam to the other lasers in your rig.

i did this with two 445's and it worked really well, but the emitter size and the divergence of the 445's is smaller than those red fire hoses.

i don't want to tell you you're wrong because:

1) i'd like you to learn why it doesn't work.
2) i'd like you to make it work then teach all of us.

ii'm hoping for option 2.

c.

My gut feeling is that with the chopping of the beam plus a prism pair the beam would be usable for scanning.
I would love someone who has these tools to do the testing but im getting them soon.
not to bad only a 300$ for possible high powered decent beam 640.
if this works out even if i use high loss optics i should still get 300mw and if i use good optics and push the diode a bit i should get from 450 to maybe 600mw (pushing it more than a bit to get that high tho)

[flecom]

my 2x 650's PBS'd make a giant +, it works fine for beam shows and ok for graphics, its like theres many choices

and as far as 4x 826 setups, my 2W of 650 isn't that bright... so I'm guessing 1W of 660 is going to be pretty dim... only once I add in 1.2W of 642 is the red kind of bright...

if you want crap tons of red you have to accept that either

1) its going to cost a crap ton of money
2) its going to be a flashlight

[p1t8ull]

My gut feeling is that with the chopping of the beam plus a prism pair the beam would be usable for scanning

That depends on your definition of usable. Recently steve-o managed to get a beam from a 250mW Opnext HL6388MG diode down to 9x4mm and just over 1.1mR here but losses are quite high at 35mW through a single prism.
So if Steve cut/folded/combined his beam to get about a 4.5 x 4mm beam with about 1.1mR, hopefully that would still be producing something over 200mW after optical losses.

The thing is, that HL6388 diode has a 20x1µm emitter, which is half the size of this P73 series diodes wide stripe emitter of 40µm.

if you want crap tons of red you have to accept that either

1) its going to cost a crap ton of money
2) its going to be a flashlight

Or source some HL63133DG diodes along with the other parts you can't make, and put it together yourself.

[steve-o]

Or source some HL63133DG diodes along with the other parts you can't make, and put it together yourself.

Yep, that may be the way to go seeing as how it is single-mode and typ I= 250mA but according to the spec sheet could be pushed to max 320mA which might result in close to 200mW output at the expense of mttf. Not knowing the emitter size off-hand, it might be easier to tame too .. experiments need to be done

[flecom]

Or source some HL63133DG diodes along with the other parts you can't make, and put it together yourself.

that would meet #1?

[p1t8ull]

that would meet #1?

I'm not saying you can homebrew a watt of bright red with a nice beam for under $200 (bring that day on ) but depending how resourceful and capable at manufacturing you are, versus the best deal you can get from a distributor for a complete module.. you should be able to save somewhere between a few and many beer tokens.

[steve-o]

I like beer tokens. What are they worth in megaWatts?

[catalanjo]

What's all this about homebrews and tokens


Cheers