Page 3 of 38 FirstFirst 123456713 ... LastLast
Results 21 to 30 of 372

Thread: The Big Green Thread (I'm gonna regret posting this in the morning)

  1. #21
    Join Date
    Jun 2009
    Location
    Nebraska
    Posts
    168

    Default

    Quote Originally Posted by steve-o View Post
    Ok I looked up the TI PTH12040W and they look like neat little modules. They recommend a "loosely" regulated 12V @ xxAmps .Thx Meatball. They appear to be voltage regulated, not current regulated tho .. ?
    --Edit--
    PS ---- I just thought this would be easier.. a bigger 808 pump diode and a bigger KTP sandwich super-heatsinked -- I guess it's not going to be that simple.... I was aiming at 1 Watt.. not 5..
    I guess leave it to the experts... That's why the price goes up so much .. I'm not giving up tho ...........
    Here's the complete circuit: https://dl.dropbox.com/u/19038241/Circuit.jpg
    assembled: https://dl.dropbox.com/u/19038241/Assembled.JPG

    You send a voltage proportional to the diode current back to the tracking pin - and there, constant current mode. I've seen this drive a pt-121 phlatlight at 30A no problem. You can add extra diode protection if you like, probably not ultimately necessary if you have a cap soldered to the diode already. Driving pump diodes: Done.

    I think you're right Steve-o. 1W is good to start with. The maybe a polarized 2W from there??

    PS - I like your video. Wear eye protection next time you goof.

    Planters - I could research some thermal control circuits if that's needed? I've done some PID stuff in the past - maybe something with such a faster response is better to start with for best results?


  2. #22
    Join Date
    Feb 2011
    Location
    New Hampshire
    Posts
    3,513

    Default

    Meatball,

    Yes do that. I'm not too sure what the nature of the thermal load will be. Just as an example, say you had a huge thermal sink which is produced with a mix of actual mass x specific heat backed by some active control. Firstly, I assume the IR diode merely needs to remain sufficiently cool and so tight thermal coupling is all that is needed here (remember HUGE sink). Secondly, the YAG also needs to remain stable and tightly coupled to prevent Fx and lensing, but does it need specific temp control? Finally, the doubler Xtal; now that needs a specific set temp, however I doubt that if held by the sink to whatever the optimal temp is determined to be that it will be necessary to get active on the order of milliseconds as the laser is modulated.

    If I am right then the effort needs to be to achieve the tightest practical thermal conduction and an easily adjusted temperature control loop with more than sufficient power to exceed a reasonable temperature range. The second laser...and beyond can trim the overkill components once the requirements are better known

    As I see it, the big hurdles are:

    1. A practical optical design including layout dimensions, FL of mirrors, OC % reflectance and an etalon for polarization
    2. Pump options as in end vs side
    3. IR diode drive and this should include high speed modulation capability
    4. Thermal control circuit
    5. Power supplies
    6. Case/deck for mounting components
    7. Sources for the critical components IR diode/stack, YAG and doubler

    I think 1 and 7 are the most challenging and will probably affect each other again and again. Although there will undoubtedly be money wasted pursuing some dead ends I think that needs to be expected. I also think a working system has to be reproducible over and over and not be a cobbling together of lucky e bay finds... we all want one (or four).

    Approaches to consider. 1. Pot luck- each part of the project is "volunteered" to an interested person or group and the result is quick assembled at an LEM or by one of the volunteers.
    2. One primary builder with a mix of skills ( Steve R. comes to mind). As an example, my optical and mechanical skills are pretty good, but I'm not too impressive electronically. In this case, components are sourced, bought, fabricated and sent to the builder.
    3. Everyone builds simultaneously and shares progress reports/ideas. The risk here is that a dead end is visited more than once.

    I'm ready to start.

  3. #23
    Join Date
    Jan 2007
    Location
    Florida
    Posts
    4,382

    Default

    I was going to keep this secret because I wnted to buy it but I've been auto-outbid twice and I know in the last 3 seconds of an auction all hell breaks loose in an automatic bidding frensy. I'm no expert on this so whoever can get this - go for it if wanted .. It isn't our target goal of 2W and an involved member here questioned me about buying from this person .. so I don't know.. Is this a good source of 532 or not? Its definitely (right now) cheap enough ..

    http://www.ebay.com/itm/500mw-532nm-...item1c2ca279e7

    According to the finer print below it can produce 1 watt according to the seller.
    And ..the auction ends tomorrow.
    Last edited by steve-o; 11-03-2012 at 08:49. Reason: added info

  4. #24
    Join Date
    Jan 2006
    Location
    Charleston, SC
    Posts
    2,147,489,446

    Smile

    Hey Eric;

    Sorry I'm a little late to this party, and admittedly Mixed-Gas probably has better solutions that I could ever come up with for a most of the items on your hurdle list, but concerning the one issue of heat management for the pump diode: Why not consider a liquid cooling solution using something like Flourinert? It's more or less non-toxic, completely non-corrosive, and despite it's high thermal conductivity it's a great electrical insulator.

    If you use it to remove the waste heat from the pump diode, that frees up lots of heat capacity on your baseplate for the more precise control of the doubling crystal. (Which is more temperature sensitive anyway.)

    Regarding the modulation circuit, I have that covered. Mo (Daedal) has a design that he worked with Bill Benner on last year, and the prototype is almost ready. The key is the output stage, which is scalable depending on which FET you use. I think the original design allows for a range of maximum output current from 100 ma to 60 amps. This entire project is sort of on the back burner, but if you want, I can push it up if you think you're ready to move forward with this green project. (Note that estimated cost for the driver is under $100, making it more than cost effective vs 3 flexmods in parallel.)

    Finally, I have a contact in China that may be able to supply the housing, once we have a set of dimentions. I've been working with them for a standard projector case to be used for the "PL-Spec Projector" project, but I'm pretty sure they could supply a module housing as well. Prices are very reasonable, and they've been an absolute joy to work with. (Considering they are in China, that alone is saying something...)

    Adam

  5. #25
    Join Date
    Feb 2011
    Location
    New Hampshire
    Posts
    3,513

    Default

    If you think that the waste heat from the diode will be a problem then I'd say go for it.

    Here's some of my thinking. IF 2W is a target then @ 15% conversion efficiency you will need approx 12W of IR. That is close to the thermal dump of two very conservatively driven 445 diodes and here an Al based air cooled heat sink works for this kind of load. At 2V the 12 W should be around 6A. If you look at some of the IR diode stacks available on eBay as the PO rises they assume you will use forced liquid cooling. When the transition becomes an advantage...I'm not sure.

    Now, I say 2W because I really want to do this and don't want to frighten anyone into thinking "He's crazy and this is a waste of my time", But scaling up is exciting and whenever a component can be sourced with the kind of flexibility you're describing for the driver and it isn't a "pie in the sky/probably never going to see" solution then this is a definite winner vs the flex mods.

    On the case...yes go for it . And if they turn out to work for your first application then when the details are worked out for the head and the driver/PS case then that could be a slick way to tie it together.

  6. #26
    Join Date
    Feb 2011
    Location
    New Hampshire
    Posts
    3,513

    Default

    Steve-o,

    Sorry, I forgot to add that your find is very interesting. Do you think you could source more than just this one? Who makes it? Can such an assembly be obtained in other dimensions ( I don't mean other universes)? What would the beam quality be like?

    This implies that side pumping might be the easier route because it avoids a highT808/highR532 input optic and opens up various coupling configurations.

  7. #27
    Join Date
    Jan 2007
    Location
    Florida
    Posts
    4,382

    Default

    I have no idea Planters. Unless somebody here can win the unit, ID it, find the mfr, p/n, etc see if it's still made, find the current cost .. All I know is I stumbled upon it on ebay. I'm still going to try to get it. Who knows how to use the auto-bid function on ebay ?

  8. #28
    Join Date
    Jan 2006
    Location
    Charleston, SC
    Posts
    2,147,489,446

    Smile

    Quote Originally Posted by planters View Post
    IF 2W is a target then @ 15% conversion efficiency you will need approx 12W of IR.
    Your numbers sound in line with what I've read elsewhere for DPSS greens, though admittedly that was for Vanadate, not YAG. Figure something between 45% and 60% on the 808 pump to 1064 conversion, and something between 30% and 40% for the KTP doubling from 1064 to 532. So yeah, a 12 watt pump sounds like a good starting point if the goal is 2 watts of green out the front end.
    That is close to the thermal dump of two very conservatively driven 445 diodes
    I'm not following you here. Surely the heat would be close to 12 times that of a single 445? (Assuming it's running at around a watt, that is...)

    Most of the high-power IR diodes I've seen run between 25% and 50% efficiency. None of the 445 nm blues are much different from this. So if you scale the waste heat from a 1 watt 445 nm blue up to a 12 watt 808 nm, you're going to need to get rid of 12 times the heat, right?

    Put another way, looking at the data sheet for a typical 10 watt, multimode single emitter 808 nm pump diode (link - last column on the first table), it's clear that it runs at a forward voltage of 1.9 V and a current of 11 amps. So using watts law, that's 20.9 watts of input. You're getting 10 watts of output, so the rest has to be dumped as heat, or 10.9 watts (close enough to 11) of heat. (Insert spinal tap joke here about how these go to eleven!)

    But 11 watts into a tiny C-mount diode is a lot of heat, and that's just for a 10 watt diode. Add another 20% on top for a 12 watt diode, assming everything scales lineraly...
    At 2V the 12 W should be around 6A.
    My research suggests that the pump diode current will be closer to double that figure for a 12 watt IR pump diode. Do you have a datasheet for the diode you're thinking about using?
    Now, I say 2W because I really want to do this and don't want to frighten anyone into thinking "He's crazy and this is a waste of my time"
    I hear you. Actually, 2 watts of green output is probably a good starting point, because below that figure the cost of building it yourself really doesn't show a benefit vs buying at retail. (Example, a 1 watt overspec green from DHOM can be had for around $650 these days, but a 2 watt unit is easily double that.) It would be worth building one as a proof of concept at the 2 watt level before trying to build something really powerful (say, 5 watts).
    But scaling up is exciting and whenever a component can be sourced with the kind of flexibility you're describing for the driver and it isn't a "pie in the sky/probably never going to see" solution then this is a definite winner vs the flex mods.
    Definitely not pie in the sky. Bill signed off on the design last year and released it to us to use as we saw fit, but a combination of job pressures and other factors forced the three people working on the project (Daedal, DZ, and me) to shelve the idea for a while. But it's literally sitting there, waiting to go. All we really need to do is have DZ layout the final board in Pad-2-Pad and then we can order a sample run of them to play with.

    I'll get in touch with Mo and see if I can get the ball rolling on this again. (What the heck... It will be useful for the PL-Spec projector project too, so I may as well get it going anyway...)
    On the case...yes go for it .
    My work on the case for the projector is already going. In fact, I've got a prototype sitting in my front hallway right now. If (make that when) we nail down the final dimensions for the module housing, I'll send that to my contact in China and see what they say.

    Adam

  9. #29
    Join Date
    Feb 2011
    Location
    New Hampshire
    Posts
    3,513

    Default

    Yep. I did make an error and forgot the IR diode conversion efficiency. So. yes I agree that the INPUT electrical power will probably be 2V x 12A =24 W. Now I just want to make sure I understand the liquid cooling thing. The suggestion that Steve R. made, I believe, was to immerse the components including the IR diode in a free bath of liquid coolant whereas the typical use of liquid coolant for high power IR laser stacks is the forced, flowing, even micro-channel circulation thing. The second option I have done several times with good success, but it does add some complexity.

    Steve-o

    I was thinking you might F/U with the seller. His comments suggests he has worked with this assembly and would work with the buyer. The info might be available without the win or may not be available even with it. If it is a one off then it won't be worth engineering around it as no further systems could be built.

    Steve R. said he might be able to pull up a standard Asian 3-4W design. That might be a good starting point. Another would be a fried 2-3W Vashio or DHOM that could be opened up, reverse engineered ( improved?) and largely copied if no plans to market the result are anticipated.

  10. #30
    Join Date
    Jan 2006
    Location
    Charleston, SC
    Posts
    2,147,489,446

    Smile

    Quote Originally Posted by planters View Post
    I just want to make sure I understand the liquid cooling thing. The suggestion that Steve R. made, I believe, was to immerse the components including the IR diode in a free bath of liquid coolant whereas the typical use of liquid coolant for high power IR laser stacks is the forced, flowing, even micro-channel circulation thing. The second option I have done several times with good success, but it does add some complexity.
    Agreed - submerging the whole mess in a free bath doesn't sound like a great idea to me. For one, if you use Fluorinert, you want to keep it sealed, because otherwise it will evaporate. (Vapor pressure for Fluorinert is ~ 15 times that of water, and just above that of MIBK!) Second, without a radiator of some sort, you'd eventually end up with nothing more than a hot bath of Fluorinert, which isn't going to help much. So we'll probably need some sort of forced circulation coupled with a radiator.

    Microchannel circulation sounds complicated, but if you think you can tackle that problem, cool! (No pun intended.) I'm assuming it's something we could have machined into the diode mounting block? The circulator and radiator components should be easy to source using off-the-shelf products designed for water cooling rigs for overclocked CPU's.

    Adam

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •