5360 Lumens from a LED array

Amonra

Enlightened
Joined
Jan 18, 2005
Messages
779
Location
Malta
Looks like roithner-laser are using Lamina boards with their own ( or some other )chips with much better efficiency (40Lm/W) other than Lamina's (23Lm/W). /ubbthreads/images/graemlins/grin.gif

The 7-cavity array from Lamina is rated at 108 lumens @ 4.7W typical ( no maximum is stated )whilst roithner ones are rated at a whopping 840 lumens @ 21W max. /ubbthreads/images/graemlins/smirk.gif

that's more than 8 times the flux at less than 5 times the power.

They also have a 134 cavity array delivering a massive 5360 lumens @ 134W max

Now i don't know if these figures are obtained from running it in some cryogenic bath or something but that's bloddy bright for a LED device.

Although there is very little info for the 134 cavity one and it is not for sale yet, the 7 cavity one is, for 115.85$ for 1-9pcs.

compared to the 18.15$ for the lamina one that's F@#%$* expensive but it's 6 times the price for 8 times the light which sounds like a good deal to me.

Oh it's also flashlight friendly (apart from the optics bit)since it runs at 4V not 11.3V like Lamina.

here are some links :

Lamina 7 cavity data sheet
Roithner-laser 7 & 134 cavity datasheet
roithner price list

enjoy !!
 

HarryN

Flashlight Enthusiast
Joined
Jan 22, 2004
Messages
3,976
Location
Pleasanton (Bay Area), CA, USA
Very interesting. So at the higher powers, the 7 cavity is running at 3 watts / die. I wonder whose LED die they are using ?

They must be very good at Vf bin matching their LEDs. Running them in series is not such a big deal, but running that many LEDs in parallel at those power levels is very impressive.
 

Amonra

Enlightened
Joined
Jan 18, 2005
Messages
779
Location
Malta
Well actually there's 6 dies per cavity which make it 0.5W / die for approx 125mA / die.
 

jtr1962

Flashaholic
Joined
Nov 22, 2003
Messages
7,506
Location
Flushing, NY
Actually I got 350 lumens from a stock 7 cavity array at a power input of 18.5W. The output curve was still rising even at that point so at 21W I probably would have gotten maybe 375 lumens. Therefore, roithner-laser is using dies which are maybe 2.25 times as efficient as the ones Lamina is using. And my part far exceeded the typical of 108 lumens at 0.42A, giving me 160 lumens. Another one which was bluer (~5000K instead of about 4500K) gave me about 135 lumens.
 

Amonra

Enlightened
Joined
Jan 18, 2005
Messages
779
Location
Malta
after stumbling upon the roithner today i'm just pissed at myself for ordering a lamina bl-2000 last week
/ubbthreads/images/graemlins/Christo_pull_hair.gif although 375 lumens is respectable 840 is much better /ubbthreads/images/graemlins/buttrock.gif

i guess ill have to reach into my pocket AGAIN /ubbthreads/images/graemlins/banghead.gif
 

Amonra

Enlightened
Joined
Jan 18, 2005
Messages
779
Location
Malta
[ QUOTE ]
cornkid said:
Thank you for the information. Very intersting. When do you think the new generation of LED will come out?

-tom

[/ QUOTE ]

this IS it ( for now )
 

enLIGHTenment

Enlightened
Joined
Feb 5, 2005
Messages
814
Location
Ottawa, Canada
Anyone else thinking that the seven cavity array--provided the price drops far enough--could be revolutionary enough to kill the 100W Edison? Two seven cavity arrays put out slightly less light than a 100W bulb but would use only 42W. Even with 25% power supply losses, that's only 56W for a light that won't need replacing for a very long time.
 

Mike Painter

Flashlight Enthusiast
Joined
Sep 16, 2002
Messages
1,863
[ QUOTE ]
Amonra said:
after stumbling upon the roithner today i'm just pissed at myself for ordering a lamina bl-2000 last week
/ubbthreads/images/graemlins/Christo_pull_hair.gif although 375 lumens is respectable 840 is much better /ubbthreads/images/graemlins/buttrock.gif

i guess ill have to reach into my pocket AGAIN /ubbthreads/images/graemlins/banghead.gif

[/ QUOTE ]

Maybe. Note the Roithner site says "up to" and does not list lumen values separately as does Lamina. It could be that the Amber puts out up 840. If so the white output would be much less if it's in the same ratio as on the Lumina site 290/108 the white would be "only" about 312.
 

jtr1962

Flashaholic
Joined
Nov 22, 2003
Messages
7,506
Location
Flushing, NY
[ QUOTE ]
Mike Painter said:
Maybe. Note the Roithner site says "up to" and does not list lumen values separately as does Lamina. It could be that the Amber puts out up 840. If so the white output would be much less if it's in the same ratio as on the Lumina site 290/108 the white would be "only" about 312.

[/ QUOTE ]
I was thinking the exact same thing, and if they're getting 312 lumens at 21W input then they're not doing any better efficiency-wise than the standard Lamina array.
 

jtr1962

Flashaholic
Joined
Nov 22, 2003
Messages
7,506
Location
Flushing, NY
[ QUOTE ]
enLIGHTenment said:
Anyone else thinking that the seven cavity array--provided the price drops far enough--could be revolutionary enough to kill the 100W Edison? Two seven cavity arrays put out slightly less light than a 100W bulb but would use only 42W. Even with 25% power supply losses, that's only 56W for a light that won't need replacing for a very long time.

[/ QUOTE ]
Yes, I was thinking exactly that. In my opinion, what Lamina is doing makes more sense than worrying about increasing the power per emitter. Unless there is some real need for the light to be tightly focused there is really no need for a single high power emitter. In fact, with general lighting being a point source is actually a detriment since it creates all sorts of distracting shadows. I'd love to see a room lit with Lamina arrays distributed across the ceiling. It would probably seem very natural like skylight.
 

OddOne

Enlightened
Joined
Mar 5, 2003
Messages
495
[ QUOTE ]
Amonra said:
after stumbling upon the roithner today i'm just pissed at myself for ordering a lamina bl-2000 last week
/ubbthreads/images/graemlins/Christo_pull_hair.gif although 375 lumens is respectable 840 is much better /ubbthreads/images/graemlins/buttrock.gif

i guess ill have to reach into my pocket AGAIN /ubbthreads/images/graemlins/banghead.gif

[/ QUOTE ]

I just got a BL-3000 last week. /ubbthreads/images/graemlins/icon23.gif /ubbthreads/images/graemlins/broke.gif Still, if it's as "overclockable" as the 2000 is, I might can get some serious light out of it anyway. /ubbthreads/images/graemlins/grin.gif

Wonder if the thing can take 23 AMPS... (That'd be 100mA per chip times the 234 chips.) /ubbthreads/images/graemlins/faint.gif

oO
 

jtr1962

Flashaholic
Joined
Nov 22, 2003
Messages
7,506
Location
Flushing, NY
I just want to point out here that yes, while the Lamina Ceramics LED arrays are very "overclockable", you need excellent heat sinking if you want to get more output at the higher levels. For the BL-3000 you would need a heat sink with a thermal impedance of about 0.1°C/W or better if you want to overdrive it to 100W levels. Basically, this means something fan forced about 4"x4" with 1.5" tall fins, and with fins spaced every 1/8", or an extrusion about 10 times that size if you want passive cooling (i.e. no fan).
 

Amonra

Enlightened
Joined
Jan 18, 2005
Messages
779
Location
Malta
[ QUOTE ]
Mike Painter said:
[ QUOTE ]
Amonra said:
after stumbling upon the roithner today i'm just pissed at myself for ordering a lamina bl-2000 last week
/ubbthreads/images/graemlins/Christo_pull_hair.gif although 375 lumens is respectable 840 is much better /ubbthreads/images/graemlins/buttrock.gif

i guess ill have to reach into my pocket AGAIN /ubbthreads/images/graemlins/banghead.gif

[/ QUOTE ]

Maybe. Note the Roithner site says "up to" and does not list lumen values separately as does Lamina. It could be that the Amber puts out up 840. If so the white output would be much less if it's in the same ratio as on the Lumina site 290/108 the white would be "only" about 312.

[/ QUOTE ]

that's possible but i dont think so since there is that BIG difference in price between it and the lamina. i dont think it would be good business practice to price something 8 times the original selling price for something unless something has been changed in it

edit: sorry six times the price
 

idleprocess

Flashaholic
Joined
Feb 29, 2004
Messages
7,197
Location
decamped
[ QUOTE ]
jtr1962 said:
For the BL-3000 you would need a heat sink with a thermal impedance of about 0.1°C/W or better if you want to overdrive it to 100W levels. Basically, this means something fan forced about 4"x4" with 1.5" tall fins, and with fins spaced every 1/8", or an extrusion about 10 times that size if you want passive cooling (i.e. no fan).

[/ QUOTE ]

Wow - like almost any other 100W device that can't sink heat with its packaging alone?

Dang. Next thing you know, you'll be telling us that there's no such thing as a free lunch either.
 

jtr1962

Flashaholic
Joined
Nov 22, 2003
Messages
7,506
Location
Flushing, NY
[ QUOTE ]
idleprocess said:
Wow - like almost any other 100W device that can't sink heat with its packaging alone?


[/ QUOTE ]
I'm just trying to give people some idea of what an appropriate heat sink for 100W would be like. Based on some of the posts I've read that doesn't seem to be common knowledge around here except among the engineering/science types. For these arrays, especially at high levels of overdrive, fan forced cooling is definitely better unless it must be absolutely silent. The fan may cause a loss of a few percent in efficiency but you get to use a much more compact heat sink. For the BL-2000 something like a Pentium 4 cooler ought to be more than sufficient to keep the device within 5°C of ambient, even at 21W power levels.

[ QUOTE ]

Dang. Next thing you know, you'll be telling us that there's no such thing as a free lunch either.

[/ QUOTE ]
There isn't? /ubbthreads/images/graemlins/grin.gif I don't know. Any number of people on the taking side of government entitlement programs might disagree with that. Then again certain things aren't common knowledge. I've had people working with thermoelectric coolers who think the heat absorbed from the cold side just magically disappears somehow, and doesn't have to be added to the power the device uses when you calculate what size heat sink you need.

BTW, thinking about this some more, since Roithner's data sheet says "up to", only one of the colors needs to reach 840 lumens. Their white may perform no better than the ones from Lamina. I personally wouldn't buy anything without getting the specifications for it, especially at the price they're charging. 840 lm at 21W is probably an impossibility with white at the current level of LED development. This scales to ~350 lumens at the 4.7W nominal power rating, or nearly 75 lm/W. As far as I'm aware, nothing that efficient has made it out of the lab.
 

NewBie

*Retired*
Joined
Feb 18, 2004
Messages
4,944
Location
Oregon- United States of America
Roithner numbers are have been subject to debate in the past...no surprises here.

On their web page datasheet, they are using the numbers for the ***RED*** LED not the white.

Another interesting thing you will notice is the ***up to*** numbers on the Roithner site...humm

For the Lamina part:
http://www.laminaceramics.com/products/bl3000.aspx
http://www.laminaceramics.com/docs/BL_3_Red.pdf

<font color="red"> Red:</font> 2045 lumens @ 104 Watts and --->25C LED die Junction Temperature<---



The junction temperature 25C thing is the same game LumiLEDs plays, which is highly unlikely in use, unless you are outdoors in Alaska in the winter...

It is nice that the white part dropped to 81 dollars:
http://www.mouser.com/index.cfm?&handler=data.listcategory&D=*BL32D00133*&terms=BL-32D0-0133&Ntt=*BL32D00133*&Dk=1&Ns=MfgrPartNumber%7c%7cSField&N=0&crc=true
 

OddOne

Enlightened
Joined
Mar 5, 2003
Messages
495
[ QUOTE ]
jtr1962 said:
I just want to point out here that yes, while the Lamina Ceramics LED arrays are very "overclockable", you need excellent heat sinking if you want to get more output at the higher levels. For the BL-3000 you would need a heat sink with a thermal impedance of about 0.1°C/W or better if you want to overdrive it to 100W levels. Basically, this means something fan forced about 4"x4" with 1.5" tall fins, and with fins spaced every 1/8", or an extrusion about 10 times that size if you want passive cooling (i.e. no fan).

[/ QUOTE ]

Yep. Fortunately I'm very familiar with high-performance cooling - in my earlier days I was a contributing writer to Overclockers.com, and built my own active (TEC/water) cooling system that allowed a 50% overclock on a Pentium II 700. (It'd run at 1,100 MHz, which at the time was astounding.) More recently I built a 120-watt-per-channel "gainclone" amplifier, and put two Xeon processor heatsinks/fans onto it, connected to temperature sensor circuit that switched the fans on when the heatsink reached 90-ish degrees F.

Dealing with a hundred or so watts of heat isn't really that big of a problem if one knows how thermal dynamics works in small radiant sources on heatsinks. I'm already designing a heatsink "stack" that should handle quite a bit more heat than an overdriven BL-3000 would generate.

oO
 
Top