Mega RGB via KTP / Dye / 445... Discuss...

[goninanbl00d]

assuming 55 watts of green?????

in a perfect world, with a perfect system, while standing on one foot, patting your head with one hand, rubbing your belly with the other hand while facing Mecca...

generally speaking...only the best of techs can get 40 watts with the gold cavity...lets say you are one of the best and you do get 55 watts...well, that is a very short list of individuals...now you can start on the red...40% conversion (assuming nothing gets damaged) should be a reasonable...target...

never mind.... goes back into hiding

So I guess the earlier figures with 40W total power would be more 'reasonable'?

It'd be hard, but it shouldn't be impossible.

[Photonbeam]

I like to be able to vary the amount of pump beam sent to the dye. That way if I want the highest power red only, I'm using all of the 532 for pump. Or, (for whatever reason) full tilt green, none for pump. Anyone have a 20W blue to trade for a dye?

[goninanbl00d]

I like to be able to vary the amount of pump beam sent to the dye. That way if I want the highest power red only, I'm using all of the 532 for pump. Or, (for whatever reason) full tilt green, none for pump. Anyone have a 20W blue to trade for a dye?

That could easily be done with a half-wave plate in a rotational mount, and a polarised beam-splitter cube, with one of the outputs feeding the dye, and the other feeding the scanners.

With all the PBS cubes I've seen (albeit crappy ones), the output slowly 'transfers' from reflection to transmission, until it's either completely transmitted or completely reflected.

If this works, it'd just be a matter of finding an actuator capable of rotating the polarising element fast enough (analogous to 10k analog modulation, or thereabouts).

[Doc]

That could easily be done with a half-wave plate in a rotational mount, and a polarised beam-splitter cube, with one of the outputs feeding the dye, and the other feeding the scanners.

With all the PBS cubes I've seen (albeit crappy ones), the output slowly 'transfers' from reflection to transmission, until it's either completely transmitted or completely reflected.

If this works, it'd just be a matter of finding an actuator capable of rotating the polarising element fast enough (analogous to 10k analog modulation, or thereabouts).

I don't have any experience of these vaginal wleding devices so don't have me assassinated if I am talking shit, but wouldn't a modulated beam pick off mirror do this?

[dsli_jon]

If this works, it'd just be a matter of finding an actuator capable of rotating the polarising element fast enough (analogous to 10k analog modulation, or thereabouts).

Mmm, not really gonna werk, cause you can't *vary* the attentuation, with a 'spinning' wave-plate - the speed-changes needed to keep-up with modulation signal, and 'sync' with yer scans, would not be possible, with this method, AFAIK - this is really what 'retro-reflector-style' blanking is all-about / for...

in a perfect world, with a perfect system, while standing on one foot, patting your head with one hand, rubbing your belly with the other hand while facing Mecca...

...Obi Wan has taught me well..

generally speaking...only the best of techs can get 40 watts with the gold cavity...

..you are correct, Sir....

But, seriously, he's totally right - it's really pretty-rare to see these puppies, even with the hi-power 'upgrade', doing 50+W... best to plan on more 'conservative' #s...
but, I still think this is a 'worthwhile' project...

never mind.... goes back into hiding

Nooooooo! Actually, Pat, nothing would please me more than to be able to come-up, with a briefcase full o' cash, and do this all based on the 'Cobra' platform - much more room to 'play', and certainly could bank-on a foundation of your solid 52+W... plus all the other brilliant engineering in that sys...

Buuuut, alas, between fickle-Clients and wimpy-budgets / bunk payers / etc, it's a lot just to be able to keep the head above water, these days...

...I guess here's always the lotto / inheritance...

peas...
j

[goninanbl00d]

Mmm, not really gonna werk, cause you can't *vary* the attentuation, with a 'spinning' wave-plate - the speed-changes needed to keep-up with modulation signal, and 'sync' with yer scans, would not be possible, with this method, AFAIK - this is really what 'retro-reflector-style' blanking is all-about / for...


...Obi Wan has taught me well..

Yeah, the only remote way you would get fast actuation is if you drilled a hole through the waveplate, and mounted it like a wheel on a galvo. How well that'd work, I have no idea. But I'm guessing there has got to be a better solution than that, so, it won't do.

And without an AOM, galvo-blanking is the only way to go. Using a stepper is too slow, it won't be able to keep up.

I don't have any experience of these vaginal wleding devices so don't have me assassinated if I am talking shit, but wouldn't a modulated beam pick off mirror do this?

Nah, like this:



So when you rotate the waveplate (using the default stepper-driven mount), the polarisation changes, and so does the percentage of light being split off by the PBS.

Or, at least that's how I think it works.

[DMills]

Err, a half wave plate changes the polarisation 90 degrees irrespective of the rotational orientation of the plate itself, so that does not work.

A pockles or kerr cell followed by a PBS would do it, but remember you are dealing with very high peak powers here, that green is many tens of KW during a pulse. This is complicating an already complex and fiddly laser past the point of diminishing returns in my view.

Personally I would probably go with a simple 3:1 beam splitter with a bank of GM20s dumping the waste beams via a standard retro reflector blanking arrangement, simple, effective and it is not going to screw up the already piss poor beam quality of the pump beam any worse then it already is.

It is interesting to speculate as to what scanning the thing would actually look like as the red and green will be pulsed while the blue will be CW.

Regards, Dan.

[mophead]

It is interesting to speculate as to what scanning the thing would actually look like as the red and green will be pulsed while the blue will be CW.

Could you not PWM the blue via the pulse train from the q-switch driver? At least they would all appear to scan the same.

[DMills]

But then the blue becomes very low average power (Probably, nobody has tested what those diodes will do QCW or pulsed).

If we can tolerate a 10us pulse from the blue (The red and green will be less then 1% of that length), and the thing is switching at say 10Khz then the effective duty cycle for the blue will be 10%, if running QCW it turns out that those diodes can reliably hack say 2.5W each for 10us pulses at 10Khz, then to make say 15W CW equivalent brightness you need 60 diodes....

Personally I suspect that such a beast would be very much a slow scan sort of beam source so the difference between the Q switched and CW blue may or may not matter that much, but it will certainly have to be factored into what you try to do with it.

Regards, Dan.

[dsli_jon]

...so the difference between the Q switched and CW blue may or may not matter that much, but it will certainly have to be factored into what you try to do with it.

Indeed. I'm thinking, however, to experiment with a hi-speed, variable 'chopper' on the blue-side... that-way, you could easily 'custom tune' the blue 'pulses' to what is going to be fairly 'predictable / consistent' pulse-trains from both the green and red... (not sure what sort of 'delay' may-be introduced thru the red-process, if-any, vs the greens' pulse-timing.. dunno, no data on that at the moment... )

...and, like you say, it might not really 'matter', since, even-if you somehow got the three sources to 'blend' perfectly, timing-wise, it's all gonna change once you blow them thru retro-style blanking and scans... I've had some scans look very-nearly perfectly-cw, and some look quite 'dashy', depending on the 'interfernce' from the scan-rate, point-count, etc...

Sometimes, however, you can use this to your 'advantage' - one of the fx-modules we developed for our 'conversion-systems' a few-years back, was what we dubbed the 'sky-scan' - basically, a large-mirror 'cone-scanner' with a variable-speed motor, that could be attached to the chassis / projector, and when you scanned even simple-graphics onto it, and 'played' the scan-rate off the motor-speed rate off the pulse-train, you could create some mind-blowing fx, that you could vary on the-fly - like, turn 5 simple dots, in a 'star formation', into 'dancing cheetoes' ... made for incredible aerials that would-otherwise take hours to program / animate, ... ie:

...and then, 'play' w/ scan-rate vs the Q-Sw pulses..

...especially-fun, was 'deconstructing' stills, ie, a simple test-pat, (...recognize this? ...and watch it become a 'living' aerials super-star...

Nothing 'new', really - I file it under "Lessons Learned from Laserium / the 6B" playing all these 'variables' off-each other can produce some simple to-achieve, yet jaw-dropping fx... chop, in-particular, is your very, very cool friend...

But again, personally, I envision this a mostly-strictly aerials / fx projector... but *really* bright, and - color!, vs just the ho-hum green... And Steve (photonbeam)'s suggestion of using the wave-plate attenuator is a great-idea for controlling red vs green % - even trying to 'color-balance' a bit... hah, now that's funny...

cheers..
j