Improving on HeatSyncing STARs (an IDEA)

I always buy stars because i cant keep emitters alive.
then i mount them on various aluminum things to heat sync them.
They Have thier own heat sync, and i sync that sync to another usually.
this time i am limited for space, and the sync the star was going on, was only 2x bigger than the stars own base.
then i had a revelation why not use BOTH.

because this time i did not have much sync area, i thought HMMM, how could i use the stars OWN, AND the base sync TOO.
all that would require is a slight Lift in the center of the star, so air can pass by the stars sync.
so
i am thinking of putting a small copper base in between the star sync and the main sync, say like 1/4th the size of the star itself. the copper would transfer Some of the center of the star heat , to the next heat sync, AND still provide a gap, allowing air to pass by the stars OWN sync.
basically improving the air transfer by almost 2X.

EDITED IN:
the whole thing is open air, not encased.

what think ye?

SINK. It's a heatsink. Clever idea, but I doubt that the tiny exposed area under the star will do much of anything. What TIM are you using between the sinks and your LEDs?

Hi VidPro,

I had thought another Idea too... in fact built a few of them (for Newai QIII Lights)

Have a look

its Heat Sink /ubbthreads/images/graemlins/smile.gif

[ QUOTE ]
LEDependent said:
SINK. It's a heatsink. Clever idea, but I doubt that the tiny exposed area under the star will do much of anything. What TIM are you using between the sinks and your LEDs?

[/ QUOTE ]

that is what i was wondering also. the small gap might not activly accomplish anything usefull.
i did insure that the small gap would have the ability for convection of air, by orienting the gap vertically, because i am not actually using a round, but a strip.
It might be acting more like fin serrations, than a whole secondary sinking.

this particular project has the advantages of being 100% exposed TO the air, as opposed to contained in a casement.

for thermal epoxy i have taken to using J&B weld, as it seems to work as well as that overpriced silver stuff, and it is mostly non conductive too. it acts very much like metal does, when you touch it, your fingers get cold fast, like metal.

[ QUOTE ]
ViReN said:
Hi VidPro,

I had thought another Idea too... in fact built a few of them (for Newai QIII Lights)

Have a look

its Heat Sink /ubbthreads/images/graemlins/smile.gif

[/ QUOTE ]

those are great, but that is the opposite of what i was saying. with those you form a perfect union, for full contact, i was trying the reverse of that. there is nothing to transfer the heat TO after it gets to my sink, the rest of the thing is plastic.

i should have posted pictures, my Bracket IS my heatsync, and no more added weight or space is desired.
i was just trying to increase the air to sink space.
when i stick the star Down, i lose the stars own air transfer.
by lifting it a bit, i was hoping to increase the stars own capacity to contact the air. (in between the star sink, and the heat sync its attached to)

if the heatsink is enclosed and no air will be blowing by it, like if it was inside a plastic light, then just make it solid metal as big as possible. You wont' have any air flow inside a light just hot air that isn't moving.

[ QUOTE ]
IsaacHayes said:
if the heatsink is enclosed and no air will be blowing by it, like if it was inside a plastic light, then just make it solid metal as big as possible. You wont' have any air flow inside a light just hot air that isn't moving.

[/ QUOTE ]
I was just going to post something along these lines. If it's inside the head of a flashlight you'll just have stagnant air that isn't helping to get rid of heat. Air is a terrible heat conductor so if you make the gap to the heatsink bigger you'll just waste that part of the heatsink.

If you have air constantly being pushed over that area I still don't think it would be a good idea. Just look at something like a CPU (a processor in a computer) the heatsinks for those are made to fit as perfectly as it's practically possible to insure that the heat is directly transferred to the heatsink.

I feel you should put your efforts in getting a good contact between the heatsink and the flashlight body to draw the heat away from the LED.

I'm just an amateur though and if I'm posting bad information please correct me.

its not a flashlight
its a headlight, with no casing, a loose star open to the air, with a open to the air bracket of aluminum.
so i dont have the problems of casement and stagnant air.

please advise based on that.

You're still better off using as much of the star heatsink as possible to provide a thermal path to the larger mass (bracket) to use as a heat sink. The star has less than 1 square inch of surface area, whereas the bracket it's attached to will have 10 or more times the surface area. That bracket will definitely do a better job transferring heat to the air than the fraction of a square inch of surface area you will partially expose to the air given your proposed change.

I'd say just stick with attaching the star directly and using the entire surface area of the star to transfer heat to the larger bracket.

the bracket is only 2X bigger than the stars own heat sync.
that is when i tried to think how i could increase the air to sink ratio.
once the star covers the bracket , and the gap between them is sealed, there goes a lot of the base sink.
and of course the top of the star does not look to efficient at heat removal, for this situation, a emitter on a air exposed piece of metal might have been even better still. but i cant deal with emitters.

here is another question:
driven at SPEC, is the star base supposed to be sufficient heat dissapation for a 1W star? , , For a 3W star?
or is the base they put these things on just a starter for assembly to something else.

I must say that I don't quite get how you are thinking about this. If I may return to a common CPU once more. If I was to go with your idea I would just put a thin slab of aluminium on top of the core and be done with it. That config would kinda be equivalent to the luxeon star config. But we all know that the surface area on a flat piece of metall is very small. And thus it transfers heat much less effectively to the air than a standard heatsink with fins.

In my mind it feels like I want to get the heat from the rather wimpy heatsink of the star as quickly as possible to another sink that is more capable of handling the heat. That is, give the air an area large enough so that it compensates for its terrible heat conducting properties. And that means connecting the star sink to the larger sink as directly as possible (flat against each other with a thin layer of the TIM of your choice).

I hope I made some kind of sense here. It's reheheaally late and I'm mostly rambling... hehe.

""I must say that I don't quite get how you are thinking about this.""

only because i didnt put in pictures
but it is so close to finishing, i was going to do that AFTER it actually works without melting the plastic.

a CPU is 40-80W of pure heat, in a packaging no bigger than a star, at spec, this is a mere 3W TOTAL power, with the alegation that most is being converted to light, of course i will probably go a bit beyond that.

how many watts are wasted as HEAT in a luxeon thing anyways?

i have run 3X3W lux on as small as a 3" sheet of aluminum for a sink (never over spec) and they have operated for years. a cpu even a 233 would die the moment i turned it on.

A Luxeon I running at 350mA on the star board with no heat sinking in free air will stabilize to 60C board temperature. This puts the junction temperature at 80C. Increasing the current will both increase the stabilized temperature, and the junction temperature (both by increasing the stabilized temp, and increasing the thermal differential between junction an board due to increased power dissipation).

It sounds like you just need more heatsinking period.

The vast majority of power in an LED is converted to heat. Consider that the best lab samples hit 25% efficiency. Only about 10-15% of the power dissipated in a Luxeon gets turned into light. So 3W of power in means 3W of heat to dissipate.

take it from me, stay away from air. especially dead air as you have inside a flashlight.

here's why in terms of thermal conductivity:
http://hyperphysics.phy-astr.gsu.edu/hbase/tables/thrcn.html

see Air at the bottom of the chart? (BAD) see solid silver at the top (GOOD)

frankly, you're better off using wood. than still air. LOL.

here's my *minor* improvement on the QIII design. works like a dream.



Leon

Currently, the best is to use the whole star surface.

The data presented below was designed for worst case scenario for 1 LuxIII. Calculations based upon LUXEON™ Application Brief AB05 & Datasheet DS46.

Junction Temp: 135°C, Forward Voltage: 3.9v, Rθ Junction-Ambient(°C/W): 17

At(°C) Current(mA) Pd(W) Rθ B-A (°C/W) Horizontal HS (sq. in.) Vertical HS (sq.in.)
85 .7 2.73 1.315 14.274 7.137
70 1.0 3.9 -.0333 14.933 7.467
70 1.4 5.46 -5.095 16.838 8.419

Rule of Thumb: LUXEON™ III(700mA DC) - Maintain T board below 85 °C in all ambient conditions, LUXEON™ III(1000mA DC) - Maintain T board below 70 °C in all ambient conditions.

See why you can't use part of the star as a HS device.

[ QUOTE ]
Leeoniya said:
take it from me, stay away from air. especially dead air as you have inside a flashlight.

Leon

[/ QUOTE ]

ok so you have a 85* hunk of copper in 30 minutes
and the value of that is??? you MUST dissipate the heat to somwhere eventually. or only have 10 minute run times.
usually that is through the flashlight body, and to (none other than ) the AIR.

this is NOT IN a flashlight, its exposed to the air, nobody is going to be conducting the heat away with thier hands on it.
if it doesnt conduct to the air (in this situation) where do you propose it go , into my forehead /ubbthreads/images/graemlins/smile.gif

[ QUOTE ]
n_den said:
Currently, the best is to use the whole star surface.

The data presented below was designed for worst case scenario for 1 LuxIII. Calculations based upon LUXEON™ Application Brief AB05 & Datasheet DS46.

Junction Temp: 135°C, Forward Voltage: 3.9v, Rθ Junction-Ambient(°C/W): 17

At(°C) Current(mA) Pd(W) Rθ B-A (°C/W) Horizontal HS (sq. in.) Vertical HS (sq.in.)
85 .7 2.73 1.315 14.274 7.137
70 1.0 3.9 -.0333 14.933 7.467
70 1.4 5.46 -5.095 16.838 8.419

Rule of Thumb: LUXEON™ III(700mA DC) - Maintain T board below 85 °C in all ambient conditions, LUXEON™ III(1000mA DC) - Maintain T board below 70 °C in all ambient conditions.

See why you can't use part of the star as a HS device.

[/ QUOTE ]

so via those stats, are you saying that a good sync would have 7 square inches of to the air dissipation?

[ QUOTE ]
evan9162 said:
A Luxeon I running at 350mA on the star board with no heat sinking in free air will stabilize to 60C board temperature. This puts the junction temperature at 80C. Increasing the current will both increase the stabilized temperature, and the junction temperature (both by increasing the stabilized temp, and increasing the thermal differential between junction an board due to increased power dissipation).

It sounds like you just need more heatsinking period.

The vast majority of power in an LED is converted to heat. Consider that the best lab samples hit 25% efficiency. Only about 10-15% of the power dissipated in a Luxeon gets turned into light. So 3W of power in means 3W of heat to dissipate.

[/ QUOTE ]

well that answers 2 question.
so is a 80* junction temp a bad thing?
would you say that something like these 3W drop-ins (pr2 replacements) run low enough to achieve 80*?

does 3 LUXIIIs in a Mag body, with a 2" heat sink stay at 80* ?

does anybody have lights that are achieveing the 80* when they are encased , and 100% sealed, when they are run for a full hour?

better question, would lots of these lights melt plastic if it was put on the heat syncs they use?

OK so because of what your all saying, i have put in a 5 level switch, so it can run at 5 different brightnesses, that way if its getting hot, i can cool out my use of it.

also, because its direct li-poly drive, i can use these levels to reduce the juice when the battery is hot off the charger, and still go direct drive.

i think reducing the power, will keep the heat down adequitly.

[ QUOTE ]
VidPro said:
[ QUOTE ]
n_den said:
Currently, the best is to use the whole star surface.

The data presented below was designed for worst case scenario for 1 LuxIII. Calculations based upon LUXEON™ Application Brief AB05 & Datasheet DS46.

Junction Temp: 135°C, Forward Voltage: 3.9v, Rθ Junction-Ambient(°C/W): 17

At(°C) Current(mA) Pd(W) Rθ B-A (°C/W) Horizontal HS (sq. in.) Vertical HS (sq.in.)
85 .7 2.73 1.315 14.274 7.137
70 1.0 3.9 -.0333 14.933 7.467
70 1.4 5.46 -5.095 16.838 8.419

Rule of Thumb: LUXEON™ III(700mA DC) - Maintain T board below 85 °C in all ambient conditions, LUXEON™ III(1000mA DC) - Maintain T board below 70 °C in all ambient conditions.

See why you can't use part of the star as a HS device.

[/ QUOTE ]

so via those stats, are you saying that a good sync would have 7 square inches of to the air dissipation?

[/ QUOTE ]



A MINIMUM of 7 sq. in. of vertical surface area, both sides exposed to air to stay within the operating life of the Lux III. If placed horizontal then you would need about 14 sq. in. of surface area, one side exposed to air.

However, this may not apply to flashlights cause you won't be able to hold it for long. But if it's hanging on the wall or something else where you won't be touching it, it's good.