BLF recoil über-thrower

384 posts / 0 new
Last post
Jerommel
Jerommel's picture
Online
Last seen: 11 min ago
Joined: 01/04/2014 - 13:18
Posts: 4734
Location: the Hague, Netherlands

Nicolaas wrote:
Even if you guys could find a suitable host and reflector, won’t the placement of the led with big limitations on heat transfer limit the Amps the led will be able to endure?
And thus limit the beam intensity – throw of the light?

If a light like this can use 100% instead of roughly 75% ot the light for the spot, but the led can be driven at 75% of maximum, where’s the benefit? (other then beam characteristics).

The heat transfer will indeed be the big limiting factor, so just one arm would probably not suffice. Maybe the “Mercedes” star design would be best?

Interesting idea, bu waiting to see whether it can be turned into a practical design.

Grtz
Nico

I think some of you underestimate the thermal conductivity of a copper boom.
But we could opt for a bar across the reflector, so it goes both directions to the head.
A ‘mercedes star’ will definitely look nicer but is much more complicated to construct from one piece.
I think i will change the drawing to have a copper bar in stead of a boom, doubling the thermal conductivity.
But you should also consider that an XP-L can run 3 Amperes on a ‘normal’ aluminum MCPCB, which is absolutely not ideal.

..waiting for parts..

Still looking for 5” parabolic reflector (for recoil light)

Enderman
Enderman's picture
Offline
Last seen: 2 days 8 hours ago
Joined: 11/03/2016 - 22:42
Posts: 3105
Location: Vancouver, Canada

Jerommel wrote:
I think some of you underestimate the thermal conductivity of a copper boom. But we could opt for a bar across the reflector, so it goes both directions to the head. A 'mercedes star' will definitely look nicer but is much more complicated to construct from one piece. I think i will change the drawing to have a copper bar in stead of a boom, doubling the thermal conductivity. But you should also consider that an XP-L can run 3 Amperes on a 'normal' aluminum MCPCB, which is absolutely not ideal.

It's honestly not that conductive due to the distance the heat has to travel, liquid cooling work work far better but cost far more.

Maybe if you can find some heat pipes from china that would work a lot better? They transfer heat a lot better than just solid copper.

Jerommel
Jerommel's picture
Online
Last seen: 11 min ago
Joined: 01/04/2014 - 13:18
Posts: 4734
Location: the Hague, Netherlands

Enderman wrote:

Jerommel wrote:
I think some of you underestimate the thermal conductivity of a copper boom. But we could opt for a bar across the reflector, so it goes both directions to the head. A ‘mercedes star’ will definitely look nicer but is much more complicated to construct from one piece. I think i will change the drawing to have a copper bar in stead of a boom, doubling the thermal conductivity. But you should also consider that an XP-L can run 3 Amperes on a ‘normal’ aluminum MCPCB, which is absolutely not ideal.

It’s honestly not that conductive due to the distance the heat has to travel, liquid cooling work work far better but cost far more.


Maybe if you can find some heat pipes from china that would work a lot better? They transfer heat a lot better than just solid copper.


Hmmm… I guess i overestimate the copper boom…
What’s in those heat pipes anyway?
I assume you mean those copper tubes used in laptop cooling etc..

..waiting for parts..

Still looking for 5” parabolic reflector (for recoil light)

Lexel
Lexel's picture
Offline
Last seen: 1 hour 55 min ago
Joined: 11/01/2016 - 08:00
Posts: 4058
Location: Germany

Heat pipes have some sort of liquid inside

If I remember right a bar of 3 heatpipes of about 5×6mm could cool my 85W Laptop CPU they were like 15cm long

They are used in big cpu coolers because they can transport heat a lot better than way thicker solid copper

I am sure a bar of 3 smaller heatpipes can easily handle 20-40W of LED power

8×2.5mm Heatpipe
https://www.aliexpress.com/item/1PCS-LOT-YT257-Flat-Copper-Heat-pipe-100...

Jerommel
Jerommel's picture
Online
Last seen: 11 min ago
Joined: 01/04/2014 - 13:18
Posts: 4734
Location: the Hague, Netherlands

Cool! Thumbs Up
(pun not intended)

..waiting for parts..

Still looking for 5” parabolic reflector (for recoil light)

pommie
Offline
Last seen: 7 hours 15 min ago
Joined: 08/28/2013 - 03:56
Posts: 932
Location: Corio, Victoria, Australia

It will only need one pipe to handle a 50w light as roughly half is converted to light and half generated as heat, a 8mm x 2.5mm pipe will handle that with room to spare, however I would suggest two for rigidity and also the cooler a led is the more efficient it is.
Have a read of this pdf from Digi-Key.

Cheers David

Nothing to see here folks, move along...

Nicolaas
Offline
Last seen: 1 day 5 hours ago
Joined: 10/31/2012 - 15:38
Posts: 423
Location: Netherlands

pommie wrote:
It will only need one pipe to handle a 50w light as roughly half is converted to light and half generated as heat

Cheers David

Still that much?
If an XP-L HI is used at 6A @ 3,9 V, would that not mean a total energy of 23,4 Watts of which the majority is transformed into light?
Please correct me if I’m wrong, these kinda calculations are not my strong suit.

So maybe one small copper bar with one small heat pipe inside?

Grtz
Nico

djozz
djozz's picture
Offline
Last seen: 2 hours 27 min ago
Joined: 09/07/2012 - 17:04
Posts: 12746
Location: Amsterdam

The reflector that I’m checking out atm is a cheap plastic reflector used in those optical illusion discs, like this one: http://www.ebay.com/itm/Toysmith-3-D-Mirascope-6-Inch-/162165285686?hash... The focal length of this reflecor, as EasyB noted, is short, just as needed for this kind of set-up, just a bit above the edge of the reflector. I have a 6 inch version that I’m trying out, but there is a 9 inch version that should work better and have better throw. But here in The Netherlands I can not buy one for cheap, in the States they are plenty available. I think that optically the reflectors in these toys are not bad (or they would not work for the optical illusion) but I suspect that the surface does not reflect optimally, there’s more light loss there than is desirable.

In a test set-up with a XP-G2 S3 3d dedomed and direct drive (I used a 14500 zoomie flashlight without lens, so as a mule), and blocking the outer edge of the beam so that a 90mm diameter section of the reflector is used, I get 310kcd. But it is running from an Efest 14500 and the flashlight and clamp are blocking quite some light, in a better build I expect over 400kcd from this. Considering that 90mm of the reflector is used, there is room for improvement, probably with a better reflector, but it is fun anyway to see the principle working so well.

link to djozz tests 

“I used to think that top environmental problems were biodiversity loss, ecosystem collapse and climate change. I thought that thirty years of good science could address these problems. I was wrong. The top environmental problems are selfishness, greed and apathy, and to deal with these we need a cultural and spiritual transformation. And we scientists don’t know how to do that.”   (Gus Speth)

emarkd
emarkd's picture
Offline
Last seen: 6 hours 10 min ago
Joined: 04/14/2015 - 22:04
Posts: 1526
Location: Georgia, USA

Alright, if you guys manage to build a recoil thrower with active liquid cooling I’m definitely in Cash

joechina
Offline
Last seen: 1 hour 4 min ago
Joined: 03/05/2016 - 08:23
Posts: 1028
Location: Germany

Would a heat pipe behind the LED a good idea?

Jerommel
Jerommel's picture
Online
Last seen: 11 min ago
Joined: 01/04/2014 - 13:18
Posts: 4734
Location: the Hague, Netherlands

joechina wrote:
Would a heat pipe behind the LED a good idea?

Apparently, yes.
I ordered one today, hope it will arrive soon so i can check it out.

..waiting for parts..

Still looking for 5” parabolic reflector (for recoil light)

Jerommel
Jerommel's picture
Online
Last seen: 11 min ago
Joined: 01/04/2014 - 13:18
Posts: 4734
Location: the Hague, Netherlands

emarkd wrote:
Alright, if you guys manage to build a recoil thrower with active liquid cooling I’m definitely in Cash

…you mean with a pump and a radiator? Big Smile

..waiting for parts..

Still looking for 5” parabolic reflector (for recoil light)

alpg88
Offline
Last seen: 1 month 3 weeks ago
Joined: 10/11/2013 - 12:35
Posts: 533
Location: usa

Lexel wrote:

I am sure a bar of 3 smaller heatpipes can easily handle 20-40W of LED power

8×2.5mm Heatpipe
https://www.aliexpress.com/item/1PCS-LOT-YT257-Flat-Copper-Heat-pipe-100...


heat pipes by themselves can not handle anything, they only move heat, not dissipate it.
there is always a heatsink and fan on the other side of heatpipes, also heatpipes work well only when hot end is lower than cold end, with something stationary like a computer, you can have radiator\heatsource mounted with height difference, but a flashlight that is not stationary it is not the best solution at all.
EasyB
Online
Last seen: 11 min 53 sec ago
Joined: 03/09/2016 - 15:24
Posts: 1436
Location: Ohio

alpg88 wrote:
Lexel wrote:

I am sure a bar of 3 smaller heatpipes can easily handle 20-40W of LED power

8×2.5mm Heatpipe
https://www.aliexpress.com/item/1PCS-LOT-YT257-Flat-Copper-Heat-pipe-100...


heat pipes by themselves can not handle anything, they only move heat, not dissipate it.
there is always a heatsink and fan on the other side of heatpipes, also heatpipes work well only when hot end is lower than cold end, with something stationary like a computer, you can have radiator\heatsource mounted with height difference, but a flashlight that is not stationary it is not the best solution at all.

I think a heatpipe is probably a good solution for getting heat away from the LED. Even if it is not oriented optimally, it’s probably a lot better than a solid piece of Cu of the same size. I could be wrong about that, but heatpipes in laptops, for example, are not oriented vertically.

I think everyone is aware that the heatpipe is just a way to transfer heat and not dissipate it. Of course it would be connected to a heatsink.

Lexel
Lexel's picture
Offline
Last seen: 1 hour 55 min ago
Joined: 11/01/2016 - 08:00
Posts: 4058
Location: Germany

Heatpipe is the best solution to cool a led above 15W

djozz
djozz's picture
Offline
Last seen: 2 hours 27 min ago
Joined: 09/07/2012 - 17:04
Posts: 12746
Location: Amsterdam

I’m going to make my test light without active cooling: the XP-G2 will be producing about 14W of heat, let’s see if I can get that guided away fast enough, I think I can Smile

link to djozz tests 

“I used to think that top environmental problems were biodiversity loss, ecosystem collapse and climate change. I thought that thirty years of good science could address these problems. I was wrong. The top environmental problems are selfishness, greed and apathy, and to deal with these we need a cultural and spiritual transformation. And we scientists don’t know how to do that.”   (Gus Speth)

EasyB
Online
Last seen: 11 min 53 sec ago
Joined: 03/09/2016 - 15:24
Posts: 1436
Location: Ohio

To make this a bit more quantitative:
The thermal resistance of a piece of material with cross sectional area A, length x, and thermal conductivity k, is
R=x/(A*k)

So for a copper bar 0.5cm x 1cm and 7cm long, with k=385 W/m*K, the thermal resistance is 3.6 K/W. This is quite high when compared to other flashlight shapes/designs. For example, I did some thermal testing of a couple flashlights and found their thermal resistances (from MCPCB shelf to outside of flashlight head) were almost 10x less. Whether it’s too high depends on the power of the LED and the size of the heatsink and how hot we allow the heatsink to get.

Jerommel
Jerommel's picture
Online
Last seen: 11 min ago
Joined: 01/04/2014 - 13:18
Posts: 4734
Location: the Hague, Netherlands

alpg88 wrote:
Lexel wrote:

I am sure a bar of 3 smaller heatpipes can easily handle 20-40W of LED power

8×2.5mm Heatpipe
https://www.aliexpress.com/item/1PCS-LOT-YT257-Flat-Copper-Heat-pipe-100...


heat pipes by themselves can not handle anything, they only move heat, not dissipate it.
It’s supposed to transfer the heat from the LED to the head (and body) of the light in this case.
Quote:

there is always a heatsink and fan on the other side of heatpipes, also heatpipes work well only when hot end is lower than cold end, with something stationary like a computer, you can have radiator\heatsource mounted with height difference, but a flashlight that is not stationary it is not the best solution at all.
I didn’t know that it uses height differences.
If so, the drawing in opening post is upside down. Smile

I think i should have ordered more than one heat pipe maybe…
(edit) Just ordered a 2nd one.

..waiting for parts..

Still looking for 5” parabolic reflector (for recoil light)

Jerommel
Jerommel's picture
Online
Last seen: 11 min ago
Joined: 01/04/2014 - 13:18
Posts: 4734
Location: the Hague, Netherlands

djozz wrote:
I’m going to make my test light without active cooling: the XP-G2 will be producing about 14W of heat, let’s see if I can get that guided away fast enough, I think I can Smile
Dang, i wish i had a mirror i could use…
…but when you say the XP-G2 will produce 14 Watts of heat, you’re apparently gonna push some 7 Amperes through it..??

..waiting for parts..

Still looking for 5” parabolic reflector (for recoil light)

EasyB
Online
Last seen: 11 min 53 sec ago
Joined: 03/09/2016 - 15:24
Posts: 1436
Location: Ohio

I think the height difference is not necessary, it just helps. If the hot side is lower, the liquid that recondenses on the cold side then doesn’t have to diffuse/wick back to the hot side against gravity. Just a guess.

Enderman
Enderman's picture
Offline
Last seen: 2 days 8 hours ago
Joined: 11/03/2016 - 22:42
Posts: 3105
Location: Vancouver, Canada

djozz wrote:
The reflector that I'm checking out atm is a cheap plastic reflector used in those optical illusion discs, like this one: http://www.ebay.com/itm/Toysmith-3-D-Mirascope-6-Inch-/162165285686?hash... The focal length of this reflecor, as EasyB noted, is short, just as needed for this kind of set-up, just a bit above the edge of the reflector. I have a 6 inch version that I'm trying out, but there is a 9 inch version that should work better and have better throw. But here in The Netherlands I can not buy one for cheap, in the States they are plenty available. I think that optically the reflectors in these toys are not bad (or they would not work for the optical illusion) but I suspect that the surface does not reflect optimally, there's more light loss there than is desirable. In a test set-up with a XP-G2 S3 3d dedomed and direct drive (I used a 14500 zoomie flashlight without lens, so as a mule), and blocking the outer edge of the beam so that a 90mm diameter section of the reflector is used, I get 310kcd. But it is running from an Efest 14500 and the flashlight and clamp are blocking quite some light, in a better build I expect over 400kcd from this. Considering that 90mm of the reflector is used, there is room for improvement, probably with a better reflector, but it is fun anyway to see the principle working so well.

That's really high. Could you post a picture or link of your setup plz? Smile

 

emarkd wrote:
Alright, if you guys manage to build a recoil thrower with active liquid cooling I'm _definitely_ in $)

OP is not going to do that for costs reasons and convenience, but when I do in the future, I'll definitely put a build log on the forum Wink

 

Nicolaas wrote:
Still that much? If an XP-L HI is used at 6A @ 3,9 V, would that not mean a total energy of 23,4 Watts of which the majority is transformed into light? Please correct me if I'm wrong, these kinda calculations are not my strong suit. So maybe one small copper bar with one small heat pipe inside? Grtz Nico

The actual light efficiency of an LED drops when you give it more current, so even if the LED is 150lm/W at stock it might only be 100lm/W or less when running near max.

https://en.wikipedia.org/wiki/Luminous_efficacy

So you get about 15-20% efficiency.

I think people overestimate how efficient LEDs are, there is still a ton of energy being turned into heat.

Jerommel
Jerommel's picture
Online
Last seen: 11 min ago
Joined: 01/04/2014 - 13:18
Posts: 4734
Location: the Hague, Netherlands

Enderman wrote:

I think people overestimate how efficient LEDs are

I think people don’t realise how totally inefficient incandescent lights are. Smile

..waiting for parts..

Still looking for 5” parabolic reflector (for recoil light)

Enderman
Enderman's picture
Offline
Last seen: 2 days 8 hours ago
Joined: 11/03/2016 - 22:42
Posts: 3105
Location: Vancouver, Canada

Jerommel wrote:
I think people don't realise how totally inefficient incandescent lights are. :)

That too.

10x better efficiency than 2% is still only 20% Silly

Jerommel
Jerommel's picture
Online
Last seen: 11 min ago
Joined: 01/04/2014 - 13:18
Posts: 4734
Location: the Hague, Netherlands

Enderman wrote:

Jerommel wrote:
I think people don’t realise how totally inefficient incandescent lights are. Smile

That too.


10x better efficiency than 2% is still only 20% Silly

I thought LEDs did better than 20% though, Djozz’s XP-G2 will probably survive then. Smile

..waiting for parts..

Still looking for 5” parabolic reflector (for recoil light)

djozz
djozz's picture
Offline
Last seen: 2 hours 27 min ago
Joined: 09/07/2012 - 17:04
Posts: 12746
Location: Amsterdam

The led will be reflowed on the middle of a copper bar, 3mm thick, on average 15mm wide, and 90mm long, so the heat can go in two directions 45mm far. After that it is spread out over an aluminium plate. I’m ok with the led running at 100degC I guess….

Sounds like it is made already but I have done nothing yet Silly

link to djozz tests 

“I used to think that top environmental problems were biodiversity loss, ecosystem collapse and climate change. I thought that thirty years of good science could address these problems. I was wrong. The top environmental problems are selfishness, greed and apathy, and to deal with these we need a cultural and spiritual transformation. And we scientists don’t know how to do that.”   (Gus Speth)

Enderman
Enderman's picture
Offline
Last seen: 2 days 8 hours ago
Joined: 11/03/2016 - 22:42
Posts: 3105
Location: Vancouver, Canada

djozz wrote:
The led will be reflowed on the middle of a copper bar, 3mm thick, on average 15mm wide, and 90mm long, so the heat can go in two directions 45mm far. After that it is spread out over an aluminium plate. I'm ok with the led running at 100degC I guess.... Sounds like it is made already but I have done nothing yet :P

 

But 95mm isn't even close to the diameter of the reflector?

A 95mm heatpipe would be good from the center to the edge, but not the full diameter, if you're talking about the 6" one.

BTW thanks for the info on these cheap mirascope reflectors Smile checking some out right now.

djozz
djozz's picture
Offline
Last seen: 2 hours 27 min ago
Joined: 09/07/2012 - 17:04
Posts: 12746
Location: Amsterdam

I’m not going to use the full 6 inch, but a 9mm section (~3.5 inch). I have a feeling that that is a sort of optimal size/throw combination for this reflector.
So for the 9inch version the optimal size would be 135mm, and that may even get close to the 1 Mcd (well, 900 kcd). But as said, can’t get that one for cheap, let’s first try this one.

link to djozz tests 

“I used to think that top environmental problems were biodiversity loss, ecosystem collapse and climate change. I thought that thirty years of good science could address these problems. I was wrong. The top environmental problems are selfishness, greed and apathy, and to deal with these we need a cultural and spiritual transformation. And we scientists don’t know how to do that.”   (Gus Speth)

Jerommel
Jerommel's picture
Online
Last seen: 11 min ago
Joined: 01/04/2014 - 13:18
Posts: 4734
Location: the Hague, Netherlands

djozz wrote:
I’m not going to use the full 6 inch, but a 9mm section (~3.5 inch). I have a feeling that that is a sort of optimal size/throw combination for this reflector.
I hope you’ll try if more will work too, because the idea is to collect / catch as much light as possible / sensible / worthwhile.

..waiting for parts..

Still looking for 5” parabolic reflector (for recoil light)

EasyB
Online
Last seen: 11 min 53 sec ago
Joined: 03/09/2016 - 15:24
Posts: 1436
Location: Ohio
Jerommel wrote:
djozz wrote:
I’m not going to use the full 6 inch, but a 9mm section (~3.5 inch). I have a feeling that that is a sort of optimal size/throw combination for this reflector.
I hope you’ll try if more will work too, because the idea is to collect / catch as much light as possible / sensible / worthwhile.

Collecting that last bit of light (the less intense light) will not add that much extra light (lumens), but the extra area of the reflector (going from 90 to 120mm) will add a lot more throw, assuming the reflector is shaped well enough to be focused. The light reflected from the outer portions of the reflector would make a very small hotspot.

djozz
djozz's picture
Offline
Last seen: 2 hours 27 min ago
Joined: 09/07/2012 - 17:04
Posts: 12746
Location: Amsterdam

Jerommel wrote:
djozz wrote:
I’m not going to use the full 6 inch, but a 9mm section (~3.5 inch). I have a feeling that that is a sort of optimal size/throw combination for this reflector.
I hope you’ll try if more will work too, because the idea is to collect / catch as much light as possible / sensible / worthwhile.

I did try what the complete reflector did, but as I was sort of playing around I did not write it down. I did give more throw but not much more, and the light will be a whole lot bigger.

link to djozz tests 

“I used to think that top environmental problems were biodiversity loss, ecosystem collapse and climate change. I thought that thirty years of good science could address these problems. I was wrong. The top environmental problems are selfishness, greed and apathy, and to deal with these we need a cultural and spiritual transformation. And we scientists don’t know how to do that.”   (Gus Speth)

Pages