Recent Blog Posts

Is UL certification required to sell LED lightng in USA?

The bsest answer to this frequent question.

Jeff Walters via LinkedIn

UL is not needed to sell to utilities in the USA. I have been designing and selling to this market since 1985. UL is not written into any IOU (investor owned utility) specs. Conformance to ANSI C136 standards is required.

UL or ETL or CSA is needed to sell to retail, commercial or industrial market. Needed in the DOT market. Helpful in the municipal streetlight market

UL products are not required for insurance companies to process claims. Nor does the use of UL listed products protect against liability.

UL is a private company. Its use is not required by and Federal, state or local laws.

Additive effect improves polymer yellow LEDs

A group of Dongge Ma from Chinese Academy of Sciences has reported an efficient way to increase the thickness of solution processed small molecule based electron-transmiting layer in yellow PLEDs. Solution-phase processing would be a highly desirable method of multilayer diode manufacturing as it's much easier and cheaper to perform. They have noted that commonly chosen electron-transmitting materials ( TmPyPB, TAZ and TPBL) can only produce a maximum of 15 nm thick electron-transmitting layer when processed in solution. That makes corresponding fabricated diodes reach their efficiency lim it quickly. Interestingly, when mixture of various compositions of these small molecules were tested, ideal combination was discovered. The TPBI:TnPyPB:TAZ mixture (1.3:0.7:1 w/w) allowed to obtain 40 nm thick layer and reach external quantum efficiency of 41.7 cd/A and 12.7%. The maximum brightness was 23926 cd/m2.

The original article and supporting information were published by ACS Applied Materials & Interfaces:
Efficient Phosphorescent polymer Yellow-Light-Emitting Diodes Based on Solution-Processed Small Molecule Electron Transporting Layer
Tengling Ye, Shiyang Shao, Jiangshan Chen, Lixiang Wang, Dongge Ma

Ion motion controls LEDs time response

Research group of Thuc-Quen Nguen at UC Santa Barbara has published a study aimed to design anionic conjugated polyelectrolyte (CPE) that could enable improved temporal response through facilitation of ion motion. Such an improvement would be greatly beneficial for the development of better PLEDs based displays. They had designed their polyelectrolyte to contain ion-conducting polyethylene oxide pendant (PFPEOCO2Na) as electron injection layers (EILs). Pristine PLEDs containing PFPEOCO2Na exhibit luminance response times on the order of tenths of seconds. This delay is attributed to the formation of ordered structures within the CPE film that slow ion migration and therefore result in a longer temporal response time. The researchers have found out that presence of the ethylene oxide units on PFPEOCO2Na did not improve the luminance response times of as cast PLEDs (46s). Yet it is possible to accelerate the response by a combination of thermal and voltage treatments that “lock” ions within the interfaces adjacent to PFPEOCO2Na resulting in the response time of 200 micro seconds (105 fold enhancement). This study may lead to the development of efficient multilayer solution-processable PLEDs with stable high work function cathodes and fast luminance response time.

Original article is published by the Journal of American Chemical Society DOI 10.1021/ja106268w

LED Production in Europe

Recently I was asked: "Are there any companies with LED manufacturing facilities in Europe?"

Here is what I've managed to find so far:

  • Osram Opto (factory in Regensburg)
  • Optogan with facility in Dortmund, Germany. Next year they'll start LED manufacturing in Russia.
  • Tridonic (Austria)
  • Vossloh-Schwabe Optoelectronic GmbH & Co. -- Factory is located in Kamp-Lintfort, Germany.

Do you know other companies manufacturing high brightness LED in Europe? Please post in comments.

Strategies for 50% Energy Savings in Office Buildings

The U.S. Department of Energy and its National Renewable Energy Laboratory (NREL) have released two technical reports that provide recommendations on how to achieve 50% energy savings in large office buildings and large hospitals.

Technical Support Document: Strategies for 50% Energy Savings in Large Office Buildings evaluates the potential for new large office buildings to achieve a 50% net onsite energy savings compared to a baseline standard (as defined by ANSI/ASHRAE/IESNA Standard 90.1-2004). The report found 50% energy savings can be achieved in both low-rise and high-rise office buildings in a broad range of U.S. climates. The analysis was conducted in 16 cities that represented different climate zones, such as hot and humid, hot and dry, marine, cold and humid, and cold and dry. The following energy-efficiency measures helped researchers reach the 50% energy-savings target:

•Lighting power density was reduced in office spaces and occupancy sensors were used in infrequently occupied spaces.
•High-efficiency boilers, chillers, air distribution units, and service water heating equipment were installed.
•Plug loads were reduced by purchasing high-efficiency electronic equipment and using special controls that shut off equipment when not in use.

Large Hospital 50% Energy Savings: Technical Support Document details the technical analysis performed and the resulting design guidance that will enable large hospitals to achieve whole-building energy savings of at least 50% over the above standard. The large hospitals report also documents the modeling methods used to demonstrate how the design recommendations will help institutions meet or exceed the 50% energy-savings goal. This report found 50% energy savings can be achieved in large hospitals across all eight U.S. climate zones. Energy savings range from 50.6% to 61.3%, with the smallest savings in humid climates and extremely cold climates. The highest energy savings were achieved in marine climates, with relatively high energy savings achieved in dry climates. In general, for each climate type (humid, marine, and arid), savings were seen to decrease as the climate became progressively colder.

More light usage with LEDs in place?

In past years we all heard that efficient solid-state LEDs would allow to decrease the need for new power plants.

In paper published Thursday in the Journal of Physics D, researchers from Sandia National Laboratories argue cheap efficient lighting can increase consumption.

"Presented with the availability of cheaper light, humans may use more of it, as has happened over recent centuries with remarkable consistency following other lighting innovations," said Sandia lead researcher Jeff Tsao.

Comment: America is bathing in cheap electricity, compared to many European countries. To prevent excessive light usage energy tariffs should be increased gradually. It will make migration to energy efficient appliances faster and economically reasonable. It's still hard to convince consumers to buy expensive LED lighting in order to save $10 on electricity bill.

What do you think?

Scottsburg picks LUXIM's Light Emittiing Plasma over LEDs for street lighting

  • Some US cities are turning OFF their streetlights to save costs.
  • Scottsburg, Indiana is switching to Light Emitting Plasma [LEP] from LUXIM.

  • City to save $70,000 a year AND reduce energy consumption by 50%.
  • City selected LEP for this high-output application, not LEDs.
  • Fixtures to be made locally by Stray Light Optical Technologies on site of former gasket factory

Will new sockets for LED non-retrofit emerge?

Will new sockets for LED non-retrofit lights emerge? What do you think?

52 LED Lights Will be Designed in 52 weeks

Kevin Willmorth, founder of Lumenique, started an interesting project. He will design 52 new SSL products in 52 weeks for 2010.

Kevin writes in his blog:
I will be developing a few wall and pendant products as the inspiration leads me. Of course in the course of one week, engineering documentation will be only what I need to execute the prototype, using Rhino CAD to lay out 3D models before making chips, and there will be no UL listing. However, every design will be completed by me including design, engineering, machining, fab, and finishing - which is a lot of fun.

The first weeks design is already complete and posted with details here.


24"L x 20"H x 6"W. Utilizes Lynk Labs 12VAC Tesla LEDs on 12" 12W SnapBrite strip with SnapDriver power supply. The reflector/heat sink extrusion was also provided by Lynk Labs. Custom on-off switch. Black wrinkle body, tumbled aluminum with clear coat, and matte white finishes. All finishes are powder coat.

All of the designs will be made available for sale to fund future work on the project through the Lumenique Product Center for anyone interested.

Nice New Year Resolution, Kevin. Good luck!

4-inch Substrates Slowly Gain Momentum

Aixtron reported delivery of their CRIUS® MOCVD system in a 7x4 inch configuration to SemiLEDs. It will be the first 4-inch LED chip line at SemiLEDs' production facility in Taiwan.

Do you know other companies already using 4-inch wafers for HB LED chips?

OLED Train Delayed

Following Philips, Osram recently announced commercial availability of its ORBEOS CDW-031 OLED panel .

Diameter of the light output area is 79.00 mm (approx. 3"), the luminous efficacy at 186 mA is typically 23 lm/W. At 186 mA, the time to half luminance (L50) is 5,000hrs. Lifetime is strongly dependent on operating current: at 93 mA the L50 is 10,000 hours, and at 47 mA, 15,000 hrs. The tile is a little over 3 in. in diameter (80mm). Data sheet.

With significantly lower efficacy than HB LEDs and exorbitant prices I liked someones note on the subject:
Is OSRAM planning for these OLEDs to be sold in the organics aisle of grocery stores?

FTC Proposes New Output-Based Labels for Light Bulbs

The Federal Trade Commission has proposed new labels for light bulbs based on light output rather than energy consumption.

The Lighting Facts label would provide information about brightness, energy cost, the bulb's expected life, color temperature (for example, whether the bulb provides "warm" or "cool" light), as well as wattage. The label also would require disclosures for bulbs containing mercury. The bulb's output in lumens—and a mercury disclosure for bulbs that contain mercury—would also have to be placed on the bulb itself.

The Notice of Proposed Rulemaking was published Nov. 10, and comments on the change are being accepted through Dec. 28.

I'd love to hear your thoughts on the proposed changes. Do you think it will make it easier for consumers to pick out energy-efficient bulbs or will it just confuse them more?

Intensity profile, Lambertian emission and... Candela artifacts

After having learned the theoretical definition of light intensity, let's see how to apply it to LEDs.

1 - intensity profile

It's simply the graph showing the light intensity vs. the angle. It can be displayed either in Cartesian or in polar coordinates. This graph usually appears in all reliable LED's specifications.

2 - intensity profile of LEDs dice

The emitted light fr om the die itself (without any optics) is usually very close to a Lambertian emission. Nice word, but what does it mean?
A Lambertian emission ( refers to a source emitting the same quantity of light in all directions. When you look at this source, the quantity of light (intensity) you will see is proportional to the seen surface. And this seen surface is equal to S.cos(teta) where S is the total surface of the source, and teta the viewing angle.

Now we know why the intensity profile of a die without optics is a cosine curve, as simple as that!

Another interesting equation to know is the power emitted by a Lambertian source within a given cone:

Where P0 is the total power emitted by the LED and phi the half-angle of the cone.

Finally, let’s calculate the intensity of a die for small angle cones, normal to the surface:

3 – adding optics to the die…

The optics integrated in the encapsulated LEDs is usually intended to collimate the light (send all the rays towards a parallel direction). Light that should have gone to the sides is now redirected to the center. The intensity graph is therefore changed so that the intensity is stronger at the center (teta=0), but at the price of a reduced intensity on the sides (the integrals of both curves, before and after the lens, are of course equal).

Let’s calculate now the intensity of the system LED + optics, wh ere the optics collimates 50% of the energy within a +/-3deg cone:

And for a system collimating 30% of the light within +/-1deg cone:

which is 1000 times (!) stronger than the original intensity without lens!

Thus, a significantly weak LED can exhibit a much higher intensity value than real power LEDs, only with the use of a simple lens. It sometimes seems to be a method to artificially enhance the power (and the sales to those who are not “skilled in the art”) of old and low-power LEDs.


Luxeon I users -- Avago has a replacement for you

Luxeon I emitters are obsolete now. LED lighting pioneers either have to redesign their products or they can turn to Avago which offers cross-reference parts.

Luxeon I emitters were the top technology few years ago. But I think it's time to move on.

the candela: the weapon of the poors

Now that we all know the difference between Watt and Lumen, it’s time to learn about another unit you can sometime find in LEDs’ specifications: the candela.

We have first to remember an old story from school: the solid angle. Roughly, it can be compared to “3D angle”; Wikipedia ( describes it as a “measure of how large an object appears to an observer”. And in the same way that “2D angles” do not have physical units but are still measured in degrees or radians, the solid angle has no physical unit and is measured in steradian (abbreviation: sr). The most useful equation to remember is, by far, the solid angle of a cone:

And the 2 values to remember are:
- the solid angle of an entire sphere:

- the solid angle of a half-sphere:

And now, back to our candela: the “light intensity” is the quantity of photons passing through a given solid angle. The units are W/sr in radiometric system, and lumen/sr in photometric system. The latter is also called candela.

In a LED’s specification, this value is intended to tell you how much light is directed in the LED-axis direction.

Seems easy, isn’t it? Next time, we will speak about optics and intensity diagram, and you will understand why this parameter is so much misleading if you don’t use it properly…

Watt, Lumen, Candelux and Steralux…

The most important item in a LED specification is also the most obscure one: this is the “Optical Power”. It seems that the LEDs’ manufacturers do it on purpose to confuse their customers! How can you compare the emitted energy when it’s called sometimes “Watt”, other times “Lumen” or “Candela”??? Isn’t it like searching the most beautiful woman with deforming glasses???

Let’s put all this in order, and let’s remember:

1 – there are 2 scales for optical power measurement:

a) radiometric scale (units: WATT): this is the absolute and objective scale (my preferred one!). It’s simply proportional to the number of photons and their energy!

b) photometric scale (units: LUMEN): this is the physiological and subjective scale, the sensitivity of your eyes. In other words, how strong YOU will see the light.

2 - The sensitivity of the eyes does not depend on their color, but on the wavelength of the light reaching them. As you may know, the eyes are the most sensitive to green, less sensitive to blue and red, and blind to UV and IR.
So, how can I compare radiometric and photometric scales? The following graph is the conversion from Lumen to Watt. Let’s take the red dot as an example: 350 lumen at 610 nm are equal to 1 watt. Easy, isn’t it?

3 – Quiz
question 1: how does it make sense that the UV that burns your skin and your eyes is as strong as 0.0 lumen?
question 2: Take a blue LED that emits 1 lumen. Cover it with phosphor. This LED emits now 10 lumen. Did we discover the optical perpetuum mobile (source of perpetual energy)???

Molex Transcend RM2 and PM3 samples

More samples to test. This time it's Transcend RM2 and PM3 modules based on Acriche AC LED. Read our preview published in June: Since then, specifications got better, especially efficacy and CRI. I like these modules for simplicity and build quality. Read full review later this month.

Transcend RM2

Transcend PM3

What you don’t have the right to ignore about LEDs

My goal here is not to bombard you with frightening physics equations and terms, you can find them by yourself in many sites. You can also read the Wikipedia 10 pages article ( I only want to compile some essential concepts that will allow Newcomers to understand the “basics of basics”.
Here is a small list of what you don’t have the right to ignore. The real world is of course a little bit more complex than that, but this list is a good starting point:

L.E.D. is for Light Emitting Diode. It’s a small die that emits light when crossed by a current. This die needs to be electronically connected and mechanically protected, and that’s why it’s usually sold as an encapsulated device. By extension, this encapsulated device is called LED too.

about the die:
- usually square, between 250 and 1000 microns. The larger sizes are called High Brightness (HB)
- there are roughly 2 families of materials, one for the higher part of the spectrum (yellow to IR), the other one for the lower part (UV to green). Those 2 families differ in their electrical behavior.

electrical facts:
Like usual electronic diodes, the current can travel only in one direction inside the LED, and is blocked in the other direction. But, even in the right direction, there is a voltage threshold:
- below the voltage threshold, no light is emitted
- above the voltage threshold, the LED emits light, and the optical power goes with the current (and NOT with the voltage)

optical facts:
- the spectrum emitted by a die is close to a Gaussian curve, FWHM is between 25 and 40 nm
- you can find LEDs in numerous wavelengths and optical powers (but not all) from the UV to the near IR. There are even some LEDs in the 200+ nm region
- the die itself emits a Lambertian beam (half-sphere emission)
- the encapsulation can act as a lens, and the emitted beam is narrowed
- a “white LED” is either an encapsulated LED composed of 3 RGB (Red Green Blue) dice, or a blue / UV LED covered with phosphor.

What did I forget in this list?

my first blog

I am completely new in this world of “blog”, I never read them, I of course never wrote any, and in fact I even don’t know what it really is… And that’s exactly the reason why I decided to accept this challenge (Yaroslav, thank you for this opportunity!).

My name is Elie Meimoun, I am a physicist (does “electro-optical architect” sound better?), and my hobbies are 3D CAD optical design (in other words everything except standard lens design), plastic optics, illumination technologies, and …… LEDs.

I grew up in France and have a diploma of “Physics Engineer” from “Ecole Centrale de Marseille”. After my military service near Verdun (a really depressing place), I decided to emigrate to Jerusalem.
With such background, you will probably forgive me for my low-level English. You will not be able to correct my heavy accent, but you are invited to correct the grammar.

I asked Yaroslav to let me write an “educational blog”. This is a kind of “back to basics” about LEDs and their interfaces with the surrounding (optics, thermal, driving….). I am open to any comment and question.

Meet us at LEDs 2009

I'll be attending LEDs 2009 event this week. If you want to schedule a meeting please email me at

LEDs 2009 Exhibit Hours:

Tue, Oct 20 - 5:30 PM - 7:30 PM
Wed, Oct 21 - 7:00 AM - 7:30 PM
Thu, Oct 22 - 7:00 AM - 3:30 PM

LEDs 2008

San Diego Skyline
Post 1 - 20 of 27
First | Prev. | 1 2 | Next | Last

© 2002 Битрикс, 2007 1С-Битрикс