Lumileds uping the ante on the Rebel

saabluster

Flashlight Enthusiast
CPF Supporter
Joined
Oct 31, 2006
Messages
3,736
Location
Garland Tx
I wonder why the Rebel doesn't have a larger heat pad? Now it looks like it have 1/3 of it chopped of, or in other words, it could be 50% larger.
There would be very little if any benefit to making the heat pad larger. The heat does not travel laterally through that ceramic. It would be better to make it thinner or of a material that conducts heat better.
 

HarryN

Flashlight Enthusiast
Joined
Jan 22, 2004
Messages
3,941
Location
Pleasanton (Bay Area), CA, USA
Hi, comparing the ES data sheet (D61) I downloaded in March, to the current version, the die to pad thermal resistance has dropped from 10 C / watt to 6 C / watt. That may not seem like much, but it is a huge gain in practice at these power levels.

Interestingly, the spec Vf has stayed the same, but now is spec at 700ma vs 350ma on the old sheet, indicating an improved efficiency.
 

Gryloc

Enlightened
Joined
Jan 20, 2006
Messages
596
Location
Cincinnati, Ohio & North Lewisburg, Ohio
I am now thinking that Lumileds had two separate parts with the name "Rebel ES". I remembered a white Rebel that was guaranteed to be 100lm/W that appeared to have either the same, or slightly better specs than the Rebal 0100 bin. I recall that it was binned at 350mA. I always thought it was a premium Rebel white (it had a smaller CCT range that maxed at 4500K I believe) that used a 1mm^2 die.

Was this earlier Rebel that was guaranteed to be 100lm/W minimum called the ES for sure, and was it the same part? Was it ever available for sale? If so, that has to add to the confusion of customers that bought this earlier product.

If that old ES was this newer version with a 2mm^2 die, then it was not clear, so were we really waiting ages for this new part? I wonder if names of some parts were swapped because a part failed to come out; done just to save the ES name and/or to make sure something special was released.

Did I recall things correctly about a different part from the past with the same name? Anyone know what really happened? I am just speculating.

-Tony
 

saabluster

Flashlight Enthusiast
CPF Supporter
Joined
Oct 31, 2006
Messages
3,736
Location
Garland Tx
Hi, comparing the ES data sheet (D61) I downloaded in March, to the current version, the die to pad thermal resistance has dropped from 10 C / watt to 6 C / watt. That may not seem like much, but it is a huge gain in practice at these power levels.

Interestingly, the spec Vf has stayed the same, but now is spec at 700ma vs 350ma on the old sheet, indicating an improved efficiency.
The lower thermal resistance is a result of the larger die. That said it is not really an improvement over the small die version, as far as surface brightness is concerned, as you will not be able to increase the energy density of this new Rebel part.

I am now thinking that Lumileds had two separate parts with the name "Rebel ES". I remembered a white Rebel that was guaranteed to be 100lm/W that appeared to have either the same, or slightly better specs than the Rebal 0100 bin. I recall that it was binned at 350mA. I always thought it was a premium Rebel white (it had a smaller CCT range that maxed at 4500K I believe) that used a 1mm^2 die.

Was this earlier Rebel that was guaranteed to be 100lm/W minimum called the ES for sure, and was it the same part? Was it ever available for sale? If so, that has to add to the confusion of customers that bought this earlier product.

If that old ES was this newer version with a 2mm^2 die, then it was not clear, so were we really waiting ages for this new part? I wonder if names of some parts were swapped because a part failed to come out; done just to save the ES name and/or to make sure something special was released.

Did I recall things correctly about a different part from the past with the same name? Anyone know what really happened? I am just speculating.

-Tony

You are correct that they are calling two different LEDs by the same name. It is very annoying and is why I complained about it earlier in this thread. You just don't do that. The easy way to tell is to look for the externally mounted esd diode. The old small die ES did not have that.
 

Black Rose

Flashlight Enthusiast
Joined
Mar 8, 2008
Messages
4,628
Location
Ottawa, ON, Canada
You are correct that they are calling two different LEDs by the same name. It is very annoying and is why I complained about it earlier in this thread. You just don't do that. The easy way to tell is to look for the externally mounted esd diode. The old small die ES did not have that.
OK, that's good to know.

I was on the LuxeonStar site the other day and saw that they were listing the Rebel ES as being discontinued, which I thought was odd since I thought it had just shown up.

Very poor product naming.
 

HarryN

Flashlight Enthusiast
Joined
Jan 22, 2004
Messages
3,941
Location
Pleasanton (Bay Area), CA, USA
Hi, I confirmed through Future that the new Rebel ES part is 1.4 x 1.4 mm, vs the "normal" Rebel which is 1 x 1 mm. As you might expect, it is intended to compete with the XP-G in the low CRI marketplace for streetlights.
 

Nil Einne

Newly Enlightened
Joined
Apr 15, 2010
Messages
82
Location
New Zealand
Interesting from the http://www1.futureelectronics.com/FLS-Doc/LuxeonRebelESLaunchFAQ.pdf Philips Rebel ES FAQ:

The new chip is 1.4mm x 1.4mm and is the same size as the Cree XPG chip but delivers better performance in actual operating conditions


As for two parts having the same name?

These parts will remain available for the time being although we expect our customers to focus on the new parts and therefore our marketing materials will emphasize the new parts and reduce visibility of the other LUXEON Rebel ES emitters.

:p
 
Last edited:

HarryN

Flashlight Enthusiast
Joined
Jan 22, 2004
Messages
3,941
Location
Pleasanton (Bay Area), CA, USA
That is quite a bold statement. It is rather general with nothing to back it up though.

You are right. I guess it is time to buy some of each of the "typical bin" and send them off to JTR1962 for comparison under real world conditions thermal conditions, not super cooled like some of the testing. I need to buy a few of those Rebel ES parts anyway, and I have some decent bin XP-Gs already.
 

saabluster

Flashlight Enthusiast
CPF Supporter
Joined
Oct 31, 2006
Messages
3,736
Location
Garland Tx
You are right. I guess it is time to buy some of each of the "typical bin" and send them off to JTR1962 for comparison under real world conditions thermal conditions, not super cooled like some of the testing. I need to buy a few of those Rebel ES parts anyway, and I have some decent bin XP-Gs already.

I was thinking the same thing. Hope he has the time. But what aspect of performance are they claiming to be better in? That is the vagueness I was referring to. Color consistency, droop, output, reliability?:shrug:
 

HarryN

Flashlight Enthusiast
Joined
Jan 22, 2004
Messages
3,941
Location
Pleasanton (Bay Area), CA, USA
I was thinking the same thing. Hope he has the time. But what aspect of performance are they claiming to be better in? That is the vagueness I was referring to. Color consistency, droop, output, reliability?:shrug:

All good questions.

As a practical matter
- Both Cree and Lumileds are well known for their reliability. I have no interst in funding or waiting around for a 100K hour test.
- I am not able / willing to fund the purchase of enough LEDs to test color consistency, and since the target market for these is outdoor streetlights, which will have "many" LEDs in it, maybe it doesn't matter?
- Droop and output - JTR can test this, so that is what I am hoping to see a comparison of.

No matter how it comes out, should be fun and informative.
 

jeffosborne

Enlightened
Joined
May 15, 2007
Messages
252
Location
southern Indiana
Good thought HarryN, I too would like to see JTR1962 do his analysis on this part. I have saved and scrutenized his previous LED test reports. Has anyone asked him about this? I read that he has been busy, overheated and distracted with his pedaling lately. I would volunteer test parts. Jeff O.
 

jtr1962

Flashaholic
Joined
Nov 22, 2003
Messages
6,558
Location
Flushing, NY
Good thought HarryN, I too would like to see JTR1962 do his analysis on this part. I have saved and scrutenized his previous LED test reports. Has anyone asked him about this? I read that he has been busy, overheated and distracted with his pedaling lately. I would volunteer test parts.
I will eventually get back to testing. I've taken a break from it these last few months both on account of the heat, and also my desire to take better care of myself. I need to make a regulated thermoelectric hot/cold plate in order to automate any testing of LEDs at various heat sink temperatures. Once that's done, I can easily test LEDs for HarryN at "real-world" temperatures. I had done some testing of that type for him in the past but it was quite tedious trying to stabilize the plate temperature by hand. I'll give a heads up in my lumen testing thread when I've built the regulated temperature plate. It's actually not a terribly hard project by my standards. It's just that a combination of heat and humidity has left me pretty drained. By early October hopefully my normal energy levels should return.
 

MichaelW

Flashlight Enthusiast
Joined
Dec 8, 2007
Messages
1,789
Location
USA
Any thoughts?

Will this large die Rebel be applicable to the PC Amber Rebel?
 

HarryN

Flashlight Enthusiast
Joined
Jan 22, 2004
Messages
3,941
Location
Pleasanton (Bay Area), CA, USA
Re: Any thoughts?

Will this large die Rebel be applicable to the PC Amber Rebel?

Hi, I don't know what the long term will bring, but the Future Electronics guy I spoke with (Lumileds Lighting specialist) was pretty insistent that the focus of this is on high brightness, 70 CRI outdoor white lighting.

The PC amber LED I saw was pretty bright (for amber) already. How much do you need ?
 

MichaelW

Flashlight Enthusiast
Joined
Dec 8, 2007
Messages
1,789
Location
USA
Re: Any thoughts?

How much do you need ?
How much you got...
I like low pressure sodium,[efficiency & color] but dislike the highly monochromatic output. [who doesn't]
I am thinking 200 lumens @ 1 amp drive, perfect in large groups for street lights.
Pros: very little light shorter than 530 nanometres-Helps with melatonin, and I think bugs too.
Cons: it may not be trendy enough for hipsters who want to bring daylight realism to night.

Maybe parking garages...
 

IMSabbel

Enlightened
Joined
Dec 4, 2004
Messages
921
Re: Any thoughts?

How much you got...
I like low pressure sodium,[efficiency & color] but dislike the highly monochromatic output. [who doesn't]
I am thinking 200 lumens @ 1 amp drive, perfect in large groups for street lights.
.

But what would be the point? 200lumen at 1A is MUCH MUCH MUCH worse than sodium, to you completely use the main advantage.
The second advantage (color) is also your main disadvantage (color.... er... rendering).
 

uk_caver

Flashlight Enthusiast
Joined
Feb 9, 2007
Messages
1,406
Location
Central UK
Are there figures for useful lumen output per Watt (or area adequately illuminated per watt) for various complete LED Units and sodium units?
 

MichaelW

Flashlight Enthusiast
Joined
Dec 8, 2007
Messages
1,789
Location
USA
I don't know a Vf of a hypothetical 'big' PC amber Rebel, I would guess 3.0 as a bottom (a 10% improvement). It may be possible to drive this hypothetical 'big' PC amber to 1.5 amps, producing 250 lumens. But efficiency really suffers.

The spectral half-width of the phosphor converter amber is 80 nanometers [565-645, I think], way better than direct amber [20 nanometers, 580-600] LEDS, and infinitely better than LPS.

Efficiency is lower than LPS, but if LPS is completely unacceptable in the marketplace, that efficiency advantage is moot.
 

hank

Flashlight Enthusiast
Joined
Apr 12, 2001
Messages
1,558
Location
Berkeley CA
> Pros: very little light shorter than 530 nanometres-Helps with melatonin

Good to see this getting attention. I'd been wondering how wide the spread of the phosphor type amber LEDs would be.

I'd like to get one somewhere and send it to Craig at LEDMuseum to put on his spectrometer and get an honest output measured. I went looking for "low-blue" lamps sending him samples a few years back and all the "white" phosphor lights he checked for me have a big spike emission right in the band that suppresses melatonin. That's why I went with "turtle safe" amber for home evenings.
 
Top