****Copper Eagle Eye X6 heatsinks****(design)

150 posts / 0 new
Last post
nitro
nitro's picture
Offline
Last seen: 1 month 3 weeks ago
Joined: 05/22/2011 - 13:32
Posts: 2706
Location: ohio

Post #78 ? Your talking about ?

hank
hank's picture
Offline
Last seen: 1 week 3 days ago
Joined: 09/04/2011 - 21:52
Posts: 9599
Location: Berkeley, California

I hate to say this, but it’s probably worth checking more than one of each flashlight.

I’d bet there’s some variation.

I understand the BLF light uses a thicker (copper) star, and the stock light uses a thinner aluminum star.
And so on the BLF the body was cut deeper — so the result is, the LED is at the same height relative to the reflector in both cases.

(and do we know about the “Host” light yet?)

But — for lights modded without reflectors, with triple and quad optics — I agree it shouldn’t make a difference.

I think.

DB Custom
DB Custom's picture
Offline
Last seen: 3 weeks 6 hours ago
Joined: 01/13/2013 - 22:28
Posts: 20733
Location: Heart of Texas

Lackluster performer? From the quad? Huh? 4400 lumens in a small light is lackluster? Man, I want to see some of YOUR builds!

The quad I built has XP-G2’s in it because of a few reasons, not the least of which is because that’s what I had 4 of. XP-L’s change the game considerably. Either way, I was asked what I would do to improve the sink I built, and this is what I’d do. Optimized thermal path through close attention to contact surfaces. The absolute most mass obtainable by filling all available space.

A top cell is around $10. The top emitter’s for a quad will cost about $25. If you have a quad board. Why not build the optimum sink to accommodate full bore use of your own driver?

wight
Offline
Last seen: 1 year 8 months ago
Joined: 11/27/2013 - 16:40
Posts: 4969
Location: Virginia, USA

nitro wrote:

Post #78 ? Your talking about ?

I was really referring to Post #66. Smile

Still fine, still on a break. One day I’ll catch up with you folks! previous wight catchup Wink
list of my drivers & variants (A17DD, FET+1 stuff, WIP stuff, etc)

DB Custom
DB Custom's picture
Offline
Last seen: 3 weeks 6 hours ago
Joined: 01/13/2013 - 22:28
Posts: 20733
Location: Heart of Texas

The problem with that Wight is that if indeed the original stock emitter shelf is 1.5mm then a copper star will fill the gap. But if the new one is 2mm, a star won’t fill the gap under the sink. For ease of use to accommodate those that aren’t fully modders, that 2mm on the end can be fairly crucial, it’s the direct link to the fins on the outside. A .5mm gap would be detrimental to the entire process, kind of like only one small mouse turd in your coffee…it’s not much f’d up but still f’d up.

nitro
nitro's picture
Offline
Last seen: 1 month 3 weeks ago
Joined: 05/22/2011 - 13:32
Posts: 2706
Location: ohio

I would love to have more non blf lights measured but i only have one on hand and only one member posted in my other thread asking for help ?? And if it don't fit perfectly more copper is better than less

 

B42
B42's picture
Offline
Last seen: 9 months 1 week ago
Joined: 03/07/2013 - 21:03
Posts: 1876
DBCstm wrote:
The problem with that Wight is that if indeed the original stock emitter shelf is 1.5mm then a copper star will fill the gap. But if the new one is 2mm, a star won’t fill the gap under the sink. For ease of use to accommodate those that aren’t fully modders, that 2mm on the end can be fairly crucial, it’s the direct link to the fins on the outside. A .5mm gap would be detrimental to the entire process, kind of like only one small mouse turd in your coffee…it’s not much f’d up but still f’d up.

^ Yup, I don’t have the equipment to mill it or work metal where I am, that’s why I’d want to buy one. So if it didn’t fit, I wouldn’t want to buy in the first place, if the question of it fitting was not guaranteed or at least returnable.

DB Custom
DB Custom's picture
Offline
Last seen: 3 weeks 6 hours ago
Joined: 01/13/2013 - 22:28
Posts: 20733
Location: Heart of Texas

If it were 0.5mm too long, the head wouldn’t screw on quite all the way. Half a millimeter.

wight
Offline
Last seen: 1 year 8 months ago
Joined: 11/27/2013 - 16:40
Posts: 4969
Location: Virginia, USA

DBCstm wrote:
The problem with that Wight is that if indeed the original stock emitter shelf is 1.5mm then a copper star will fill the gap. But if the new one is 2mm, a star won’t fill the gap under the sink. For ease of use to accommodate those that aren’t fully modders, that 2mm on the end can be fairly crucial, it’s the direct link to the fins on the outside. A .5mm gap would be detrimental to the entire process, kind of like only one small mouse turd in your coffee…it’s not much f’d up but still f’d up.
nitro wrote:

I would love to have more non blf lights measured but i only have one on hand and only one member posted in my other thread asking for help ?? And if it don’t fit perfectly more copper is better than less

 

I don’t have a light in hand to measure, but I simply can’t see where you two are coming from. Exactly how is that area critical? Is the large, 31.7mm wide shelf not also directly connected to the fins?
  • That larger shelf surface alone provides 1900 square millimeters of contact area by itself. The 20mm shelf provides only 1250 square mm of contact area.
  • The outer surface of the 31.7mm cylinder provides 1110 square mm of contact area. The outer surface of the short 20mm cylinder provides 250 square mm of contact area.

In other words the top section provides 3010 square mm of contact area, which is quite a large amount. [More than Old Lumens uses on his current MT-G2 Maglite heatsinks IIRC?] The bottom protrusion provides an additional 1500 square mm… and causes a potential fitment issue!!

B42 wrote:
[snip] I don’t have the equipment to mill it or work metal where I am, that’s why I’d want to buy one. So if it didn’t fit, I wouldn’t want to buy in the first place, if the question of it fitting was not guaranteed or at least returnable.
DBCstm wrote:
If it were 0.5mm too long, the head wouldn’t screw on quite all the way. Half a millimeter.
This is exactly what I’m talking about! This is where that extra protrusion on the bottom seems like a problem: it provides little benefit that I can see, increases cost, and appears to cause a fitment issue.

Still fine, still on a break. One day I’ll catch up with you folks! previous wight catchup Wink
list of my drivers & variants (A17DD, FET+1 stuff, WIP stuff, etc)

bugsy36
bugsy36's picture
Offline
Last seen: 3 months 3 weeks ago
Joined: 07/11/2014 - 18:15
Posts: 2475
Location: Florida USA

 

I am not sure what the issue is. If the X6-SE indeed does have a lower emitter shelf than that if a stock X6 and the difference is is only (.5) mm that that would equal 0.019685 inches and could easily and quickly be sanded down with a flat surface and some 80 grit sandpaper - aluminum or copper.

On a different note...Dowel/alignment pins should also be unnecessary and just create a manufacturing expense.

It's the simple things that we take for granted that cost us the most

Ευκαιρία λέει πιάσε με από το μέτωπο γιατί μόλις έχω περάσει δεν θα με πιάσειs

DB Custom
DB Custom's picture
Offline
Last seen: 3 weeks 6 hours ago
Joined: 01/13/2013 - 22:28
Posts: 20733
Location: Heart of Texas

What is the cost on the quad sink? $10? How much do you think a machinist would knock off for NOT performing that one operation? $1?

I won’t bother to check your math cause you’re pretty dang brilliant, but if I can add 1250 sq. millimeter to 1900, then by gosh I’m all for it!

That vertical area that you are referring to is under the threads of the bezel. I’ve always been told that a lot of thermal path is lost in threaded sections. So perhaps depending on that section almost exclusively for your thermal path is not the best of ideas? Do you have an X6 in hand? Look at it. The head of the light, the part where the emitter sits normally, the part in question on the sink, that’s right at the finned portion of the outside that is contiguous with the emitter shelf, a solid one piece milled head. The area above that is threaded, sits under the bezel. See?

Mine was built from a single bar of 1.25” diameter 6061. I bought that for making a triple sink. So it doesn’t touch metal above the head of the light. This makes that emitter shelf a vital area. Adding contact to that, under the bezel or not, can only help. But the vital part is still the emitter shelf. The heat travels into the metal in much the same conical radiance pattern that the light leaves the emitter. I like having deep thermal sink directly under the source to catch it.

Maybe I’m wrong. Maybe I’m not so tight I can’t part with a dollar.

wight
Offline
Last seen: 1 year 8 months ago
Joined: 11/27/2013 - 16:40
Posts: 4969
Location: Virginia, USA
bugsy36 wrote:

 

I am not sure what the issue is. If the X6-SE indeed does have a lower emitter shelf than that if a stock X6 and the difference is is only (.5) mm that that would equal 0.019685 inches and could easily and quickly be sanded down with a flat surface and some 80 grit sandpaper – aluminum or copper.

On a different note…Dowel/alignment pins should also be unnecessary and just create a manufacturing expense.

  • I disagree with the first part. The whole point of having that 20×2mm protrusion on the bottom is to create a good thermal interface on that round 20mm face. Once the average user sands it down it’s just not going to be flat enough for that. That’s a big part of why a lathe is employed here in the first place…
  • The second part makes it clear that I haven’t adequately demonstrated my point, despite writing volumes upon volumes. Sad I brought up the pins specifically as something that was unnecessary and should not be included. Just like the 20mm protrusion is unnecessary….

The issue is that: 1. I’m stubborn. 2. I have a valid suggestion which improves compatibility and reduces cost with what I strongly suspect will be zero performance penalty. 3. Either nobody cares or nobody understands, or both. What can you do, eh? <shrug>

Still fine, still on a break. One day I’ll catch up with you folks! previous wight catchup Wink
list of my drivers & variants (A17DD, FET+1 stuff, WIP stuff, etc)

nitro
nitro's picture
Offline
Last seen: 1 month 3 weeks ago
Joined: 05/22/2011 - 13:32
Posts: 2706
Location: ohio

 

Will all see when the sample is test fitted , under $20 for a copper custom heatsink is a darn good price i think

bugsy36
bugsy36's picture
Offline
Last seen: 3 months 3 weeks ago
Joined: 07/11/2014 - 18:15
Posts: 2475
Location: Florida USA

 

Clearly miscommunication on my part.

I do agree with the protrusion to reach the emitter shelf. almost seems silly not to have. I just wonder the actual need to machine at 1.5mm or 2.0mm depth. I can take off an excess but I cannot add for a shortage. Even without precision tooling the shelves I have seen are not all that perfect anyway so a dab of paste after machining or sanding would not worry me. Hand lapping is performed on custom PC's everyday and they achieve good thermal transfer.

In reality, especially with the quad adapter the added cubes alone will more than help with heat dissipation.

Dowels? Well my Dremel will take care of those pesky things. LOL

It's the simple things that we take for granted that cost us the most

Ευκαιρία λέει πιάσε με από το μέτωπο γιατί μόλις έχω περάσει δεν θα με πιάσειs

wight
Offline
Last seen: 1 year 8 months ago
Joined: 11/27/2013 - 16:40
Posts: 4969
Location: Virginia, USA

DBCstm wrote:
What is the cost on the quad sink? $10? How much do you think a machinist would knock off for NOT performing that one operation? $1?

I won’t bother to check your math cause you’re pretty dang brilliant, but if I can add 1250 sq. millimeter to 1900, then by gosh I’m all for it!

That vertical area that you are referring to is under the threads of the bezel. I’ve always been told that a lot of thermal path is lost in threaded sections. So perhaps depending on that section almost exclusively for your thermal path is not the best of ideas? Do you have an X6 in hand? Look at it. The head of the light, the part where the emitter sits normally, the part in question on the sink, that’s right at the finned portion of the outside that is contiguous with the emitter shelf, a solid one piece milled head. The area above that is threaded, sits under the bezel. See?

Mine was built from a single bar of 1.25” diameter 6061. I bought that for making a triple sink. So it doesn’t touch metal above the head of the light. This makes that emitter shelf a vital area. Adding contact to that, under the bezel or not, can only help. But the vital part is still the emitter shelf. The heat travels into the metal in much the same conical radiance pattern that the light leaves the emitter. I like having deep thermal sink directly under the source to catch it.

Maybe I’m wrong. Maybe I’m not so tight I can’t part with a dollar.

I do not have one in hand. I’ve been waiting for the host since I wasn’t interested in the BLF Edition driver or stock single LED config. (and now I’m scratching my head since I may as well purchase a complete light w/ the host being a bit pricey)

What if you could add 5000? What about 10000? Would it do anything?

As far as shaving a buck off… that’s a matter of strategy. If you come back to the same machinist with the proposed change and he’s a normal sort of person then I would not expect much of a discount.. [After all, if he’s marking up all his work then the more work he does for you the more money he makes… in order to maintain the same profit he’s already set his sights on for the job he’s got to mark up a lesser quantity of work to a larger extent] If you reset the encounter to one where you are asking a fresh, untainted individual to produce a stack of 5.7mm thick disks w/ two clean faces, a specific precise diameter, and a center drill… then I think you’d be getting somewhere. (based on the reasoning I posted back in post #78… we’re actually talking about a lot less work IMO.)

[realizing that I’m not actually a machinist] I’d expect to get around 15-20% more heatsinks out of a copper bar based on the 3 sections: 5.7mm top section, 2mm protrusion, 1/8in for what’s wasted by a saw or parting tool.

Between the copper savings and reducing the number of operations I’d expect to reduce cost significantly, not just $1.

I could be wrong about the whole thing. I’m certainly not going to build two lights and send them off to have somebody stick them in a sphere and compare, so maybe it’s best I let the subject drop.

Still fine, still on a break. One day I’ll catch up with you folks! previous wight catchup Wink
list of my drivers & variants (A17DD, FET+1 stuff, WIP stuff, etc)

nitro
nitro's picture
Offline
Last seen: 1 month 3 weeks ago
Joined: 05/22/2011 - 13:32
Posts: 2706
Location: ohio

Where would be the fastest place to buy a non blf x6 ???

DB Custom
DB Custom's picture
Offline
Last seen: 3 weeks 6 hours ago
Joined: 01/13/2013 - 22:28
Posts: 20733
Location: Heart of Texas

His hosts aren’t complete for some reason beyond American understanding, but it could be assembled with parts on hand. For example, the lens is available through various sources, I have several taken out of previous X6’s and replaced with UCLp versions. The switch has already been linked, o-rings can easily be found.

If Richard wanted to or had time to, he could measure 75 lights for you. Wink

I would re-measure mine, but it’s sitting under a hard earned quad heatsink and I’m not feeling like taking it apart. Silly

wight
Offline
Last seen: 1 year 8 months ago
Joined: 11/27/2013 - 16:40
Posts: 4969
Location: Virginia, USA
bugsy36 wrote:

 

Clearly miscommunication on my part.

I do agree with the protrusion to reach the emitter shelf. almost seems silly not to have. I just wonder the actual need to machine at 1.5mm or 2.0mm depth. I can take off an excess but I cannot add for a shortage. Even without precision tooling the shelves I have seen are not all that perfect anyway so a dab of paste after machining or sanding would not worry me. Hand lapping is performed on custom PC’s everyday and they achieve good thermal transfer.

In reality, especially with the quad adapter the added cubes alone will more than help with heat dissipation.

Dowels? Well my Dremel will take care of those pesky things. LOL

Ugh. Sigh sigh sigh! OK: When you hand lap a CPU and a heatsink, let’s assume you lap them against either each other OR against a “true flat” surface. Whatever, in our hypothetical scenario they mate perfectly afterwards, right? Great. What doesn’t matter in that case is that the CPU’s top surface (top of the IHS / lid) is no longer square with the sides. [Because you’ve been lapping it freehand of course, you don’t have a machine which can keep your grip perfect enough!] Why doesn’t it matter you ask? Well, we’re clamping another flat surface across it with springs [the heatsink w/ it’s retention bracket], so that other surface will simply tilt until it mates correctly with the perfectly-flat-but-no-longer-square-with-anything IHS!

Now think about what happens when you place a rod (the heatsink is a stubby rod) into an empty cylinder (flashlight head) where it fits tightly… and the bottom is no longer square to the rest. See the problem? The tight fit around the circumference requires that the faces be square or you’ll have a tiny, tiny contact patch.

Furthermore, in order to achieve contact on both surfaces the bottom post/protrusion must be of exactly the right height. Too much and you only get one surface… too little? You only get one surface…

Still fine, still on a break. One day I’ll catch up with you folks! previous wight catchup Wink
list of my drivers & variants (A17DD, FET+1 stuff, WIP stuff, etc)

DB Custom
DB Custom's picture
Offline
Last seen: 3 weeks 6 hours ago
Joined: 01/13/2013 - 22:28
Posts: 20733
Location: Heart of Texas

Isn’t it amazing how seemingly simple things get complicated so fast?

DB Custom
DB Custom's picture
Offline
Last seen: 3 weeks 6 hours ago
Joined: 01/13/2013 - 22:28
Posts: 20733
Location: Heart of Texas

Free hand? Why? Doesn’t the sink have a hole through the middle? A bolt through that hole allows it to be mounted in a drill so it can be spun, a file, sandpaper, rotary tool, S30V Spyderco, lots of ways to cut it once it’s spinning, not so difficult then to keep it square… not perfect, perhaps, but much closer than freehand.

bugsy36
bugsy36's picture
Offline
Last seen: 3 months 3 weeks ago
Joined: 07/11/2014 - 18:15
Posts: 2475
Location: Florida USA

 

Alex Smile I completely understand and this the need for the paste. My whole purpose of purchasing rather than making is time vs. money. My toolpathing would take me more time than it would be worth to make "one" and the quad could be a pain since I do not have the parts to check fitment....guess better go to RMM's site Smile

It's the simple things that we take for granted that cost us the most

Ευκαιρία λέει πιάσε με από το μέτωπο γιατί μόλις έχω περάσει δεν θα με πιάσειs

DB Custom
DB Custom's picture
Offline
Last seen: 3 weeks 6 hours ago
Joined: 01/13/2013 - 22:28
Posts: 20733
Location: Heart of Texas

I had to cut a stem on the table saw that would chuck into the drill press. Then I literally used a broken HSS planer blade clamped in a drill press vise and swung the table by hand for the cross cuts. To cut the cup, much of it was blind. The table and vise were in the way. Between squatting in front of that bench top drill press that’s over 30 years old and from Harbor Freight in the first place and my shoulder knifing me, it was quite fun! Smile But I don’t hold any corner on stubborn, do I? If the wife, a CNA, saw even a little of how I do things out in the shed, she’d have conniption fits for sure!

Oh, by the way, did you know that if you clamp a high speed steel 70 yr old planer blade into a 100 yr old floor mount vise and hit it with a large metal object it will cleanly break like a piece of glass? And then you have a sharp new cutting surface! Smile

bugsy36
bugsy36's picture
Offline
Last seen: 3 months 3 weeks ago
Joined: 07/11/2014 - 18:15
Posts: 2475
Location: Florida USA

 

Be careful man. The wife could be reading LOL.

Just think. First you ask her to throw things at you at high velocity and now you are talking about attempting to change physics. She's going to take the meds away! LOL

It's the simple things that we take for granted that cost us the most

Ευκαιρία λέει πιάσε με από το μέτωπο γιατί μόλις έχω περάσει δεν θα με πιάσειs

Rufusbduck
Rufusbduck's picture
Offline
Last seen: 7 months 2 weeks ago
Joined: 04/04/2012 - 15:34
Posts: 10389
Location: Golden state

It’s your god given right to be stubborn wight. Right or wrong everyone else has it too so what can you do? Being the prick that I am, I just had to do a math check though.

Area of a shelf is pi*r^2 minus emitter hole or 3.14*15.85^2 – 3.14*10^2 or 3.14*(15.85^2-10^2) or 475mm^2.

Area of shelf plus emitter pad area is 789mm^2.

Area of 5.6mm high rim is 2pi*r*h or 6.28*15.85*5.6 or 557mm^2

Area of 2mm lip is 126mm.

Other than to agree with wight about the difficulty in getting both bottom surfaces to mate at the same time I do tend to also agree that the emitter area does seem to add significantly more contact surface area to the host. Whether that additional contact area is needed is left as an exercise for the reader. Me, I’m just throwing marbles on the dance floor. Silly

Three Tanna leaves to give him life, nine to give him movement. But what if he eats the whole bag?

Scott

DavidEF
DavidEF's picture
Offline
Last seen: 1 week 2 days ago
Joined: 06/05/2014 - 06:00
Posts: 7699
Location: Salisbury, North Carolina, USA

If the best place to mate the heat-sink is the emitter hole (where the MCPCB would usually sit), then it doesn’t make any sense to not use that area. Add to that the math showing how much added contact you are getting by mating both the hole and the shelf. As for the possibility of not getting one or the other mated due to differing height, well let’s think about this. If the protrusion on the heat-sink is machined ~0.5mm too long, you don’t get a mate between the heat-sink and the shelf. Put some thermal paste on the shelf and forget about it. That would be better than having the protrusion too short and using some thermal compound at the most critical point for best heat transfer to the fins. As for leaving out the protrusion entirely, I think I’ll just leave this here:

DBCstm wrote:
But the vital part is still the emitter shelf. The heat travels into the metal in much the same conical radiance pattern that the light leaves the emitter. I like having deep thermal sink directly under the source to catch it.
(Emphasis mine)

The Cycle of Goodness: “No one prospers without rendering benefit to others”
- The YKK Philosophy

wight
Offline
Last seen: 1 year 8 months ago
Joined: 11/27/2013 - 16:40
Posts: 4969
Location: Virginia, USA

Rufusbduck wrote:
[snip] I just had to do a math check though.

Area of a shelf is pi*r^2 minus emitter hole or 3.14*15.85^2 – 3.14*10^2 or 3.14*(15.85^2-10^2) or 475mm^2.

Area of shelf plus emitter pad area is 789mm^2.

Area of 5.6mm high rim is 2pi*r*h or 6.28*15.85*5.6 or 557mm^2

Area of 2mm lip is 126mm.

Other than to agree with wight about the difficulty in getting both bottom surfaces to mate at the same time I do tend to also agree that the emitter area does seem to add significantly more contact surface area to the host. Whether that additional contact area is needed is left as an exercise for the reader. Me, I’m just throwing marbles on the dance floor. Silly

Sure enough, I was mistakenly using D instead of R. (!) The figuring works out similarly either way: you’re getting more contact area from the big shelf than the little one, and the big one appears to be at a known, non-variable height whereas the little shelf – not so much.

I agree with RBD and anyone else who’s posted it: yes, if we could implement a heatsink which adequately interfaced with both shelves [eg no bridging a gap with TIM, this is not adequate] that would reduce thermal resistance a lot. I don’t think that the reduction would be significant, despite being a lot.

IF the 20mm shelf was at a known height I’d recommend interfacing with that rather than the 31.7mm shelf. (I’d establish a loose fit on the 20mm diameter cylinder, tight fit on the 31.7. I’d do a rough face on the 31.7mm surface and a perfect face on the 20mm surface.)

Still fine, still on a break. One day I’ll catch up with you folks! previous wight catchup Wink
list of my drivers & variants (A17DD, FET+1 stuff, WIP stuff, etc)

DB Custom
DB Custom's picture
Offline
Last seen: 3 weeks 6 hours ago
Joined: 01/13/2013 - 22:28
Posts: 20733
Location: Heart of Texas

Wish we could do a pour, but then machining a new head altogether would probably be the way to go. lol

DavidEF
DavidEF's picture
Offline
Last seen: 1 week 2 days ago
Joined: 06/05/2014 - 06:00
Posts: 7699
Location: Salisbury, North Carolina, USA

DBCstm wrote:
Wish we could do a pour, but then machining a new head altogether would probably be the way to go. lol

Maybe that’s what nitro should be working on instead of just an insert? Or, maybe have all three options?

The Cycle of Goodness: “No one prospers without rendering benefit to others”
- The YKK Philosophy

Rufusbduck
Rufusbduck's picture
Offline
Last seen: 7 months 2 weeks ago
Joined: 04/04/2012 - 15:34
Posts: 10389
Location: Golden state

A few different choices here. I don’t think milling an entire head from copper is a viable one. A simple 5.6 × 31.7 disc is easiest. Adding a 2 × 20 post adds some unknown to the capabilities and looks cool sitting on a desk. If the cost stays reasonable then I’ll get one but if not then I’ll just ghetto some scrap aluminum. It’s a bit unreasonable to ask Nitro to come up with more than one option but once he settles on one there’s no reason someone couldn’t produce another.

Three Tanna leaves to give him life, nine to give him movement. But what if he eats the whole bag?

Scott

DB Custom
DB Custom's picture
Offline
Last seen: 3 weeks 6 hours ago
Joined: 01/13/2013 - 22:28
Posts: 20733
Location: Heart of Texas

Well, in all fairness, a few things changed on the X6 for the Group Buy…the tail cap is not interchangeable for example. So they’re two different lights. We run into that a lot don’t we?

The target here is for the Group Buy Special Edition light. Looked at solely in that regard, and with over a thousand of them out there, then it makes sense to spec the heat sink to that specific set of measurements.

Pages