Imalent        
Results 1 to 17 of 17

Thread: Next Generation PMS-K2

  1. #1
    Flashaholic* modamag's Avatar
    Join Date
    Aug 2004
    Location
    Bay Area, CA
    Posts
    2,101

    Default Next Generation PMS-K2



    First off, special thanx to IsaacHayes for the K2 Preliminary datasheet.

    As many of you already know by now Lumileds have pre-release their spec for the next generation. So beeing stuck in Atlanta airport due to layover & flight delay, I had nothing to do but this ... Yes there are quite a few flavors.

    === PSS2 ===
    The traditional single Lux heatsink is modeled after my good friend hotbeam's hotlips with minor changes to the D-size to further reinforce the side wall. I was thinking of pulling the emitter base deeper into the body to allow future throw optimized reflector design but didn't have a chance to do it yet.




    === PTS2 & PQS2 ===
    The design for these were inherited from the PTS/PQS. The major change occur in the D size version where it now host an E-Can. This will increase the mass of the design and improve heat transfer to the body of the light.




    === PTS2-TypeX & PQS2-TypeX ===
    This is the extreme edition where increase mass & surface area will allow the maximum heat removal from the light. For those who wish to run their LEDs at the coolest temperature possible could also incorporate an active cooling fan to further remove the heat from the light. The machining time for this will be rediculous (3x) but that's why it's the extreme edition.




    All of these design utilize the self centering mechanism the reduce installation headache. Originally I planned of using a hex for centering and even smaller pores for the TypeX but that would further increase the machining cost with little performance gain.

    Now these design are far from optimal, this is just my view. I'm open to any suggestion, comments or criticism to create the "perfect" one.

  2. #2
    Flashaholic* modamag's Avatar
    Join Date
    Aug 2004
    Location
    Bay Area, CA
    Posts
    2,101

    Default Re: Next Generation PMS-K2

    Oh Isaac, I didn't forget about your recommendation. Here's one PTS2-C-TypeXU.

  3. #3
    *Flashaholic* wquiles's Avatar
    Join Date
    Jan 2005
    Location
    Texas, USA, Earth
    Posts
    8,459

    Default Re: Next Generation PMS-K2

    Dude - good job!. These look kool [img]/ubbthreads/images/graemlins/happy14.gif[/img]

    Will

  4. #4
    *Flashaholic* IsaacHayes's Avatar
    Join Date
    Jan 2003
    Location
    Missouri
    Posts
    5,876

    Default Re: Next Generation PMS-K2

    Came here from the Copper PQS/PTS thread. Thanks for the credit, but I didn't find the datasheet, just merely pointed you towards it..

    Anywho, as far as active cooling, a fan won't help inside a sealed flashlight. There is no means for cool outside air to come in, and hot to get out.. So I'd say that's a waste of effort/machining...

    I like the extra mass for the D sized ones especially since they have to be so thin. The raised "walls" with a taper are cool, and I see you used something like that for the D single emitter too. Also the extended bottom part is cool.

    But you need to make sure the tolerances are good for the walls in the head. My PTS-D doesn't make very tight contact on the sides of the mag head, but it doesn't have much area to contact there anyways... Same goes for the part going into the body tube of the mag.. It would need to be pretty tight to do any good or use a ton of thermal paste... And I don't know how well this would work if the tolerances were tight on the head part walls of the sink with the mag light. Example if you screwed it down all the way, slid the HS in, it might go smoothly into the barrel, and be offset enough to snag when it goes into the head, or vise versa. The fit could be perfect on both the head and tube, but when both are together on the mag light they may not be in true alignment from the maglight threads...

    So I'd rely on thermal contact area in the head firstly, and leave the space in the battery tube just for holding an e-can or whatever. Good contact to the head will allow the heat to spread over it and then to outside air/your hand. If one were to spread thermal paste on the maglight body threads then the heat from the head would transfer to the body tube naturally.

    So that's my analysis. No "active" cooling holes, tight tolerances on the walls going up the mag head, no extended peice to contact mag battery tube, unless it's just an area to hold an E-Can/driver.

    As far as the emitter centering, it should work granted the K2's match the specs. With pegs/without not sure which is best, probably without for simplicity, but I always worry that the deppresion could be too deep and the case would bottom out before the slug gets a good metal-to-metal flush contact. So the deppresion would not be needed at all if you use pegs... I just don't trust the emitters to have the same distance of case to slug bottom distance on each one.. Call me paranoid!! [img]/ubbthreads/images/graemlins/smile.gif[/img]

    (the single emitter ones look great, no improvements I can think of, unless you were to go to pegs, but like I say, either one is just as good probably)

    Hope this helps... [img]/ubbthreads/images/graemlins/wink2.gif[/img]

  5. #5
    *Flashaholic*
    Join Date
    Apr 2004
    Location
    Lost In Space
    Posts
    5,441

    Default Re: Next Generation PMS-K2

    ""PTS2-TypeX & PQS2-TypeX ""
    i dont know about a cooling fan [img]/ubbthreads/images/graemlins/confused.gif[/img] , but the increased height around the edges of this thing , look like it could increase the heat transfer to the exterior mag head.
    especially in a loose fit, and via the slower transfer of heat with thermal compound stuff.

    do we get a 3 or 4 star thing to for us people who couldn't mount an emitter without damaging something? or do these new wizz bang leds come on bases?

    i like the air holes, or if there is ridges for heat/air transfer and all, and the idea of a fan, BUT getting the heat OUT of the light to the exterior seems to be the only way to remove it. unless were going to remove the lens, or put air holes in the rest of the light somewhere.
    Any additional 'IN CASE' air conduction will only travel out the glass face , most of it still is going to go through the head, and threads and all.
    and the heat going to the batteries, doesnt sound so great either (for me), so its through the head/body that it really needs to depart.

    it wouldnt hurt if the raised sides are also at an appropriate angle for sending any lost light (optics) out of the front, which might make the machining easier, if the high side edges were just ONE ramp downward, instead of the ridges.

    improving the mag head, with deep fins round or even verticle, is the only thing left to try and improve the remove. like that huge 3" one , but in 2 inch size .

  6. #6
    *Flashaholic* idleprocess's Avatar
    Join Date
    Feb 2004
    Location
    dfw.tx.us
    Posts
    5,558

    Default Re: Next Generation PMS-K2

    If you had some sort of external radiating surface, heatpipes would make sense, but as is, the head and body represent a decent thermal mass.

    A larger head with fins such as 5Mega's 3" head would help, but unless you have an exposed "lower chamber" that's sealed on both ends and your fan is also splash-tolerant, I don't see a benefit to forced-air in a flashlight the size of the typical maglight. The airflow requirement to be of much benefit also mandate a powerful fan that will consume no small amount of energy relative to the light sources themselves.

  7. #7
    *Flashaholic* IsaacHayes's Avatar
    Join Date
    Jan 2003
    Location
    Missouri
    Posts
    5,876

    Default Re: Next Generation PMS-K2

    Looking forward to more discussion on these... [img]/ubbthreads/images/graemlins/smile.gif[/img] [img]/ubbthreads/images/graemlins/popcorn.gif[/img]

  8. #8
    Flashaholic* modamag's Avatar
    Join Date
    Aug 2004
    Location
    Bay Area, CA
    Posts
    2,101

    Default Re: Next Generation PMS-K2

    IsaacHayes
    The body extrusion does have very tight tolerance (unlike PTS-D). It is 0.2mm smaller than the Mag body to allow for the inconsistency in the Mag production. The head extrusion is much tighter 0.1mm tolerance since it's just a cylinder.

    Regarding the depression it's 0.25mm deep, and the K2 spec is 0.25mm. This means that the space btw emitter base & HS is practically none. All the epoxy do is filling in the machining gap.

    As for the cooling holes. I did make one light with a small 20x20mm fan. The air was drawn thru the tailcap (holes), cooled the batteries and channeled thru the switch. The head was milled so the exhaust would exit thru the side of the head.

    It consumed 100mA which was almost neglectable considering the light draw 2A from the batteries.

    VidPro
    No spec have been release for the star version yet.

    As for profiling the side similar to reflector. It's a great idea. It will add to cost because It requires both milling and lathe.

    idleprocess: Agreed.

    CONCLUSION #1: Good-bye to the cooling pores.

  9. #9
    *Flashaholic* IsaacHayes's Avatar
    Join Date
    Jan 2003
    Location
    Missouri
    Posts
    5,876

    Default Re: Next Generation PMS-K2

    Hmm. Not sure if we are thinking about the same thing on the deppresion.. If I were to go with depression, I would want it just barely there so you can center the led, but the black case of the luxeon would have no chance of touching the HS and holding the emitter slug away from the HS. That way less space for thermal compound to fill in, and more metal-to-metal contact. That's what I mean..

  10. #10
    *Flashaholic* wquiles's Avatar
    Join Date
    Jan 2005
    Location
    Texas, USA, Earth
    Posts
    8,459

    Default Re: Next Generation PMS-K2

    I also fail to see the advantage of the cooling holes for a fan IF the body is sealed. I have done my fair share of CPU overclocking and the problem we have is thermal steady-state. A fan that would only circulate air within the body will only distribute the heat, but not get rid of it.

    The same applies to CPU's and computers. A copper heat sink will remove heat away from the CPU, but only if you have a fan can you keep the temperature from rising and cooking the chip. Even then, to achieve the best/lowest temperature, you have to have a second/3rd/4th fan to get colder outside air to flow inside the computer case so that the fan on the heat sink can work efficiently. And this only works if the room can suck-in the heat from the computer (like if you have AC) - otherwise hot air goes in, hot air goes out, and you again achieve thermal steady-state, usually at a point where overcloking fails to work reliably.

    With a sealed body as we have in flashlights, you would need something with many fins - lots of surface area and good air flow "outside" the flashlight in order to remove heat from the body.

    I also agree on the fit of the heatsink to the body. The beter the fit the less need for thermal compound (which is really terrible in terms of transfering heat). Thermal compound use should be kept to a minimum since its only purpose in life is to fill the tiny voids between the two metal pieces - you still want as much metal to metal contact and don't want to rely on thermal compound alone - this is why any good CPU overclocker knows to lap a heat sink before it is applied, and only uses the absolute smallest amount of thermal compound as he/she can.

    Even after all of this, at best we are only distributing the heat within the body of the light. True, this lowers the temperature under the LED's die, but very quickly you achieve termal steady-state and that's it. Without active external cooling the temperature will not go any lower. That is why holding a flashlight helps since the human body is actively absorbing heat from the light.

    Sorry for my rambling [img]/ubbthreads/images/graemlins/icon15.gif[/img]

    Will

  11. #11
    *Flashaholic* idleprocess's Avatar
    Join Date
    Feb 2004
    Location
    dfw.tx.us
    Posts
    5,558

    Default Re: Next Generation PMS-K2

    Ah, but with most flashlights you get some secondary liquid cooling ... namely the blood in your hand circulating throughough your body. That's one reason that it's not advised to leave many high-performance LED lights running at full power unless you're holding them - warm blood transfers heat better than air.

    Also, re:steady state. Rarely will you be operating a flashlight (or computer) in an enclosed environment without air circulation or some other means of heat exchange. It's also unlikely that your environment will be anywhere near the critical temperature of your equipment, so simply conducting it into a thermal mass that can absorb and radiate it away usually works assuming you don't overload the heatsink.

  12. #12
    Flashaholic*
    Join Date
    Dec 2004
    Location
    Cesshole on the Potomac
    Posts
    534

    Default Re: Next Generation PMS-K2

    You should incorporate some pin holes to make alignment easier.

  13. #13
    Flashaholic* Kryosphinx's Avatar
    Join Date
    Jun 2004
    Location
    NW suburbs, Chicago
    Posts
    791

    Default Re: Next Generation PMS-K2

    So when are these avaliable?

  14. #14

    Default Re: Next Generation PMS-K2

    Hi Jonathan,

    to my modest suggestion, you could implement the same "Threaded portion to screw down into the Mag head" as the PTS2 for luxeon star, for increment the heat transfer.


  15. #15
    Flashaholic* Nitroz's Avatar
    Join Date
    Jul 2004
    Location
    Monroe
    Posts
    3,241

    Default Re: Next Generation PMS-K2

    Waiting.

  16. #16
    Flashaholic* Nitroz's Avatar
    Join Date
    Jul 2004
    Location
    Monroe
    Posts
    3,241

    Default Re: Next Generation PMS-K2

    Are these available yet?

  17. #17
    *Flashaholic* IsaacHayes's Avatar
    Join Date
    Jan 2003
    Location
    Missouri
    Posts
    5,876

    Default Re: Next Generation PMS-K2

    I belive the PMS (tri/quad) recent thread/contest should hold K2/any luxeon slugs.

    Not sure about the single one, if that one will be developed or not?
    Mags: 2C: R/O, True Cyan, UV. 3C:Penta-XR-E Q5, Single SSC P4. Nitecore D10, Fenix LOD-CE, ArcAAA

Posting Permissions

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