December 23, 2011 at 01:09Inside view: LED industry in Russia
Hello to all Habra users!
The publication of an article on Habré on the analysis of LED bulbs from different manufacturersraised a significant wave, both from one of the leading “domestic” manufacturers of LED lamps, Optogan, and from ordinary users (almost 200 comments as a joke). Much has been said, both good and bad. Immediately after the publication of that article, a forum on nanotechnology was held under the patronage of RosNano, where we managed to talk with Optogan representatives and get some comments. A little later there was the Interlight exhibition, which I could not attend, then the LED conference in Skolkovo ("The second scientific conference of the cluster of energy-efficient technologies"). In a word, there are many events, and, mindful of the previous article, I hasten to share with everyone who will be interested in thoughts about the LED industry in Russia.
The publication will be divided into two parts: an interview with Vladislav Bugrov, executive director of Optogan, and some generalization of the material accumulated over the past month and a half.
I know that many hawkers like to delve into the problem until the very end - this is how Optogan began .
Part one. Long-awaited comments of Optogan representatives
Vladislav Bugrov, Executive Vice President, Optogan *
Why is Optolux E27 used polycarbonate rather than glass? Despite the fact that glass is about 2.5-3 times heavier than polycarbonate, it takes a lot more material to make a “bulb” (it is made of polycarbonate of large thickness) than a glass diffuser, and thus the total weight This part of the lamp is comparable. In addition, due to the large bulb, and the need for a separate platform for its “gluing,” the size of the aluminum radiator has been increased, which also affects the cost and weight of the lamp.
We made a diffuser for a lamp made of frosted polycarbonate for two reasons: firstly, this material is unbreakable, and secondly, it is cheaper.
And how much cheaper can you quote the prices for glass and polycarbonate used by Optogan?
We, like any other manufacturer, do not disclose the cost structure of our products, polycarbonate costs us about 20% cheaper than glass. Although, of course, it all depends on the required characteristics and shape of the diffuser. I must admit that glass has an advantage in some ways, in particular, it has better transmittance. For example, frosted polycarbonate, which gives uniform light scattering, absorbs about 15%, while frosted glass with similar characteristics - 9-10%. In principle, in the future we plan to use glass for some fixtures, and for some polycarbonate. While it was important to make the bulb unbreakable, because as it turned out, the first thing that everyone (including you) checked was how it would beat.
The housing of the LED modules manufactured by Optogan OLP-X5050F6 * is similar to that of Sveta LED (SVL03P1-FX-XX). Are these cases purchased from third-party manufacturers or is it the development of Optogan?
In addition to the production of Chip-on-Board modules, Optogan company (at the recently concluded Interlight exhibition Optogan introduced the second generation of COB), similar to what is used in the Optolux E27 lamp, is engaged in the sale of LEDs, for which it purchases housings from third-party manufacturers, so there may be a visual coincidence. And not only at Optogan and Svetlana, but also at the overwhelming majority of world manufacturers. The difference in this case is which LED chip is used inside the standard case, as well as in the silicones and phosphors used (in this case, Mr. Bugrov means “gel” -filling based on silicone and a phosphor).
I repeat that in the Optolux bulb, the light source is not a board with discrete LEDs, but a single LED module. In this module that we developed, there is no plastic case, and just such a module is an example of an integrated solution that we are actively promoting on the market.
Standard cases with LEDs from Optogan before packing in tape *
How does the company plan to reduce the price of finished products by 2 times?
Very simple: through economies of scale. Although this is not always enough. Now I am advocating everywhere (at the round table, in many interviews, at Interlight I told you) - you need to go into integrated solutions. The first step you saw. If you disassemble the bulbs of different manufacturers, then in the vast majority you will see individual LEDs, as well as an individual driver, consisting of individual electronic components. And whatever one may say, but it turns out that a large number of discrete elements are used. And we took the first step towards an integrated solution - we developed an integrated light module (COB).
The next step is that we will develop into an integrated driver, and then we will try to combine different modules into one solution. This is similar to switching to microchips in due time. Once there were computers that occupied entire rooms and even buildings. Then they invented semiconductor chips. Now the cost of microcircuits varies from hundreds of dollars per processor to units of dollars for some simple solutions. In lighting technology will be about the same. Here is our philosophy.
Perhaps a COB solution with one large LED is not justified (for example, the heat sink is deteriorating, which could potentially lead to premature “wear” of the LED module)? Maybe it was worth taking the path of world market leaders?
Because COB, i.e. an integrated solution will help to reduce the price in the future. As for the deterioration of the heat sink - this is an erroneous statement.
The thermal resistance of the most widely used 0.5 W LED in the 5630 Rjs (junction-solder point) package is approximately 40-60 K / W. When this LED is mounted even on a metal board with a small thermal resistance (typical values of 1-4 K / W), due to the presence of passivation layers separating the contacts from the base, the total resistance Rjb (junction-board) will be the sum of the board resistance and LED, i.e. approximately equal to the thermal resistance of the LED (at best 7K / W, in typical applications> 40K / W).
In the COB module, the chips are planted directly on a metal base, which allows Rjb (junction-board) <1K / W for large chips and 2-3K / W for small ones. Accordingly, a lower thermal resistance between the recombination zone of the chip and the external environment is provided in the SOC than in a board with LEDs, which obviously provides improved heat dissipation when compared with the widely used technology.
Lamp Optolux E27, in which the COB-solution is applied *
That is, the aluminum substrate copes well with the heat sink?
Yes, moreover, it does not have to be made of aluminum or copper. For example, ceramic substrates may be used. In the simplest case, it is a ceramic based on aluminum oxide or nitride. Many manufacturers (CREE, OSRAM OS, Nichia, LumiLEDs) have recently started using it in the manufacture of their LEDs. For example .
The difference is in the detailed compositions of ceramics, which make up the technological know-how of material manufacturers. At Interlight, our COBs on ceramic substrates were also presented. It is worth noting that the thermal conductivity indicators for them were better than for metal substrates.
But does the driver not get warm during operation, because it is so tightly “packed”?
The driver, of course, heats up during operation, however, the main heat is generated by the light source. The driver efficiency is approximately 85%, the light source efficiency is less than 50%. Strategically, we are moving towards greater integration. This is not just about reducing the size, but about reducing the number of components due to the high degree of integration of the latter. This is our philosophy, the direction of further development.
So, the philosophy of Optogan can be described as follows: every year a new iphone or htc appears, which costs about $ 1,000. A year later, it already costs $ 500, but there is a new model for $ 1,000. Translating to lamps. Today we released a lamp for 1000 rubles with 1% ripple, and in a year we will make a "nano" lamp (claims for the most part precisely in the amount) with 0.5% ripple for 1000 rubles, and we will sell this for 500 ... Well, microelectronics it works ... Or do you still want to make chip-on-board + driver-on-board = optolux ?! Tell me, did anyone calculate the transition to "integrated" solutions? In my opinion, purely subjective, not always integrated solutions provide significant advantages and reduce the cost of production ...
The size of the radiator will decrease automatically as soon as the efficiency of the light module increases - the size of the radiator is determined solely by the area required for the dissipation of excess heat. The more efficient the light module, the more energy is converted to light and, accordingly, less to heat, and there is no need to use a large radiator.
Otherwise, our strategy is aimed at reducing the cost of replacement lamps and improving their parameters (such as dimmability, color rendering and light emission angle). We do not plan to produce light bulbs that will be more expensive than existing ones, which is why we are looking for ways to reduce the cost of the product while improving its characteristics.
If we talk about integrated solutions in microelectronics, then they did not always lead to cheaper production, but always led to cheaper final products, which is our goal.
You can understand the economic component of an integrated solution using a simple example, comparing the light source with LEDs and the Chip-on-Board (SOV): if we leave aside the issues of economics of scale and the availability of existing production lines, then at the level of LED modules we observe the following.
At the level of elementary components, a board with LEDs is:
1. the board itself,
2. LED chips, which
3. are packed in plastic cases
4. each chip is connected to the contacts of the case
5. the cases are mounted on the board.
There are no plastic cases in the COB module, the chips are attached directly to the board, so points 3 and 5 disappear. In fairness, it should be noted that the COB uses more silicone with phosphorus mixed in it than in a board with soldered LEDs, but in the end the cost OWL is still lower than boards with LEDs (depending on the design, the difference can be up to 50%).
The peculiarity of the current moment is that in the world there are too many lines for the production of LEDs and too few for the production of COB modules. Given the crisis of the overproduction of LEDs designed to illuminate TFT monitors and TVs, manufacturers of these LEDs staged a real dumping war, selling LEDs at the cost level. This situation is temporary and will change after the ruin of a large number of small manufacturers of such LEDs and market consolidation, which will happen over the next year.
During this same time, manufacturers will increase capacities for the production of COB modules (this trend is clearly observed, just look at the total number of companies offering COB modules a year ago and now) and everything will fall into place, that is, COB modules will become cheaper than the modules on assembled LEDs.
Will hope. Do you plan to conduct R&D in the field of minimizing the driver and reducing the size of the lamp? Why is your driver the largest?
We already carry out R&D partners in the field of integrated solutions for the driver, which will lead to a decrease in its size, and with an increase in the efficiency of the light-emitting module and a decrease in the size of the entire lamp. At the moment, our driver is not only the largest, but also the best, we provide such a level of ripple that no other manufacturer can do.
This is good, but are there any studies that confirm that the pulsation of 1% and 3% differ significantly in their effects on the body?
Between 1% and 3% the difference is not significant, however, between 1% and 10% is already significant. Studies on the effects on the body are enough. Here, for example, is one of the reports .
Installation for mounting individual LEDs in modules. Judging by the photo - large crystals measuring about 1 mm by 1 mm *
Is the phosphor used in Optogan lamps produced in Russia?
We worked with phosphors of many manufacturers and now settled on products manufactured in Germany and the USA. There are also interesting companies in Russia, but so far we do not use their products in our production. First, strangely enough, but it turned out that domestic phosphors are more expensive than imported ones ...
Have you analyzed why such a strange trend appeared?
“Why” domestic phosphors are more expensive, this is probably a question for manufacturers of phosphors. If we compare the parameters, then, from an objective point of view, they should be cheaper. Apparently, the turnover is not large enough to offer goods at a lower price.
Although the price difference is not decisive. The most important thing: domestic phosphors cannot fully satisfy our needs. Domestic companies have some successful phosphors, but for the production of a large set of different lamps with different light characteristics, a fairly wide line of phosphors is needed. Unfortunately, we do not find such a wide range of products from any domestic supplier. Although, I see a very noticeable development in the production of domestic phosphors and, if in the near future the situation will change for the better, then we will actively use them in our production.
Isn’t it easier to buy a phosphor in China, because it is a rather expensive structural element?
The quality is worse. This component is expensive, but now all prices are such that it does not determine the price of our products. But the quality of Chinese products is still noticeably worse. In general, while American and German are the best that we have found for ourselves. It was more important for us now to ensure precisely the quality of the products. But we periodically test various phosphors and it is clear that Chinese manufacturers are quickly pulling themselves to the leaders.
However, it is worth noting that the production of phosphors is the industry that is likely to be created in the near future in Russia, because it does not require large investments in equipment. We need knowledge, the necessary skills, the chemistry of processes - we all have it, so the forecast for the development of such domestic industries is positive.
As far as I know, problems with phosphors in rare-earth elements (REE), which are used for their production. And in Russia there are few deposits where REE would be mined, mainly they are concentrated in China.
This is not entirely true: in the CIS countries (mainly Russia and Kazakhstan), 19% of the world's REE reserves are located, in China - 38% (for details here ).
By the way, the REE “accessibility” problem is not always the main one.
What else ?!
For example, the purity of the materials used and the size of the granulate are important.
If possible, tell us more about phosphors, quantum dots, rare earths and silicates ...
Let us return to phosphors.
As you know, to convert the blue light emitted from LED chips into white, various phosphors (phosphors) are used, which are divided into different classes.
So widely used in our production YAG phosphors ( Yttrium aluminum garnet - yttrium aluminum garnet Y 3 Al 5 O 12 ) and silicates contain rare earth elements. YAG contains yttrium, which is a rare-earth element, and is doped with Ce ions. In addition, YAGs and their analogues in some manufacturers include other rare earth elements along with or instead of yttrium.
The composition of silicates with REE includes only Europium. Compared with YAG phosphors, the REE content in silicates is lower, because there they are not part of the matrix, but are only a dopant (with a concentration of <2%).
Rare earth metals and their compounds are low-hazard and non-toxic substances. For all the phosphors we use, there are safety certificates (MSDS) confirming their safety for the human body.
But phosphors on quantum dots of rare-earth materials do not contain. Typically, these are sulfur, zinc, cadmium, selenium compounds (for example, CdSe, CdS, ZnCdSe and others). We do not use such phosphors in our products, however, we are actively developing in this direction.
How many MOCVD installations have the company already installed, and how many plans to put into circulation in the coming year in the Russian Federation?
It is no secret that we have only five plants, and all of them are located in Germany. In Russia, we are not yet starting production, because the country has problems with the supply of pure materials necessary for the epitaxial growth of semiconductor structures at a reasonable price. In particular, pure ammonia costs 10 times more than in Germany. And the price pressure is already enormous due to strong competition from Chinese and Taiwanese manufacturers.
While in Russia we are manufacturing LEDs, electronics and lighting fixtures, we are also gradually localizing the production of LED chips. We plan that it will be launched in Russia next year. While it is economically more profitable.
Do you initially buy substrates somewhere?
Yes, we buy sapphire substrates. Further we buy gases and organometallic compounds containing gallium, aluminum and indium. And, accordingly, we ourselves grow semiconductor structures on substrates in MOCVD reactors.
Here they are the same sapphire substrates ... *
Deliver to Russia and cut it here?
So far, and cut in Germany. Next year we will cut in Russia.
Is the company ready to demonstrate a factory in the Russian Federation with a full production cycle (from processing sapphire plates to packaging LED modules)? Do you invite us to the factory?
Of course, we invite, believe me, there are many interesting things.
If we talk about the full cycle, then we will be able to show the processing of plates in St. Petersburg next year. Epitaxial growth is a separate issue in which we are still looking for a compromise.
Holy of Holies of the domestic LED industry *
Part two. Fly in the ointment: everyone drags a blanket over himself
In order not to constantly refer in the course of the narrative to the decisions of two conferences of the 7th and 8th “Gallium, Indium, and Aluminum Nitrides: Structures and Devices,” I would like to immediately give references to them. Conference proceedings .
From the interview with Vladislav Bugrov, at least two disappointing conclusions can be drawn :
1. The absence of MOCVD plants in Russia is explained, first of all, by the inability to purchase at the reasonable prices the clean gases and reagents required for deposition.
2. Despite the reserves of REE elements necessary for the synthesis of high-quality phosphors in Russia and Kazakhstan, we are not able to ensure their "conversion" into cheap and affordable materials for subsequent industrial use in the LED industry.
As for the first conclusion, here you can only shrug. Between a producer and a consumer, a whole network of intermediaries is often built, each of which wants to receive its own from 5 to 30 percent of profit. Although our country is often called the "raw materials appendage" of the global economy, in fact, it turns out that we are even worse. To date, we (the state or frequent investors - this is not so important) have not been able to create the infrastructure in our country that allows us to produce the same clean gases. But the energy efficiency program was adopted by the party (cross out) and the government at the end of 2009. Suppose that in 2 years we were able to build only an enterprise that, in fact, is engaged in screwdriver assembly, but, nevertheless, let us say a lot to the Optogan and Svetlana companies! - This is really a huge achievement.
About phosphors, I think Mr. Bugrov conveyed the situation very, very well. In Russia there is no normally applicable legislation regarding public procurement (for some reason, it seems to me that most of the companies mentioned by Bugrov are LLCs that grew up in business incubators from some research groups and partly live off state grants). Just the other day, a law was just signed that simplified all these procedures related to FZ-94, i.e. the bill that roamed the courts for almost a year.
By the way, this is what was said in the decision of the 8th conference: “In general, the problems of the industry, identified one and a half years ago at the previous Conference, did not change or even worsened. First of all, this refers to the institutes of the development of science existing in the country, the 94th Federal Law, strategic instability in the Russian market of high-tech materials, personnel shortages aggravated by demographic problems, and extremely low decision-making speed . ”
All these, of course, are mostly solvable problems, and within a very short time. But besides them there are also personnel problems, which in the current conditions of the widespread introduction of the Unified State Examination and the degradation of the education system, it is becoming more and more difficult to solve every year. Moreover, these are serious, structural and extremely time-consuming problems, and yet it - time - works against us. In the decision of the 7th conference (see page 5), it was said in some detail about this.
To my deep regret, all of the above has been voiced - “little things in life” compared to the fact that in Russia there is still no single system integrator. There are ministries, statements, companies pulling a blanket of budget financing on themselves, but each one having an approach is incompatible, not compatible with “competitors” from a neighboring ministry. And by the way, the example of Optogan in this sense is quite revealing: the company could not find a supplier (well, for the first time only for one MOCVD installation) of gases and some necessary materials, which made the production in a more suitable place - in Germany.
There are, of course, positive examples. Perhaps the only organization that has succeeded in this field of integrator is JSC Russian Railways, which began installing lighting at stations and semaphores based on LEDs. It is quite natural that the Russian Railways daughters from semi-finished products are engaged in production, and in general, this system is rather slow, but still there is some progress and forward movement.
And until there is such an integrator in the country that can use the administrative resource where it is needed (“crush” dealers in the good sense of the word, lower raw material prices, albeit by dumping, etc.) , to effectively develop a “collaboration” between various ministries and departments, we will solve only tactical tasks. Those. to produce products created outside our homeland, but under the domestic brand and proudly write "Made in Russia". If we want to write this truly historical inscription with a clear conscience, then the entire technological chain MUST be carried out within our homeland.
However, in order to understand what consequences the “tactical” approach can lead to in the field of LED industry, it is enough to recall the end of the 80s and the Elbrus series processors (as one of the brightest examples). In literally one decade, our country has moved from being “backward” in the field of development and production of computer equipment to the category of “heavily dependent on imports” of this equipment.
I emphasize once again that the Optogan and Svetlana companies are great fellows , which, at least partially, were able to move the problem off the ground! However, until they bring technologies, techniques and equipment that are really hi-tech (namely, MOCVD) because of the “bourge”, all this will be only populist slogans and statements ...
But you, as always, draw your own conclusions (my concept of a system integrator using an administrative resource is utopian in essence, maybe you yourself will offer a solution ?!), readers, and everything else will be shown by time ...
Afterword
1. If someone wants to, he can buy Optogan lamps in the Eldorado network ( proof ).
2. On the Internet, I found a very interesting version of the lamp (for example, here ).
3. But in general, do we, consumers, need an Ilyich lamp on LEDs ?! Indeed, in fact, it is only important for us that the light comes from somewhere above, replacing the sun, is it not better to change the paradigm of lighting our homes, offices, enterprises, etc. and make the lamps planar (2D), mounted directly on the ceiling (it is important to note that the shape of this ceiling will then not be important at all), and mount the driver itself in a switch. The problems associated with the purchase of chandeliers, polycarbonate and glass will disappear, and the ceiling will turn into a clear, sunny and warm sky ...
4. The other day in the December issue of GQ magazine an article was published in the “Made in Russia” section with a rather thorough interview with the president and CEO of Optogan Maxim Odnoblyudov.
5. Despite all the pressure from the Optogan press service (almost an hour of telephone conversations), which was extremely huge, I did not change almost anything in the final version of the publication at my own risk. Forgive me, Optoganovites, I can only write the truth: what I consider important, interesting, as well as what I see with my own eyes!
PS: I thought and searched for a while in which thematic blog to publish. First I wanted to post it on my personal blog, but, unfortunately, in this case it will be available only to registered users, however, I would like this article to be publicly available, so let it remain on the Internet blog
PPS : A lot of time was spent on restoring the sincere equilibrium, preparation of the material and so on, but in the near future there will be another material devoted, as I promised, flash-memory and DRAM-memory. It will be my gift to all thirstyfor the New Year (by the way, just my year is coming). And after the holidays, probably, Christmas gift number 2 will be published - an article devoted to ordinary objects under an electron microscope, various biological objects will also enter there, so stay tuned! Well, if my work is completely interesting, then buy the January issue of Gambling, there we will analyze the device of the nanocosuit from Crysis 2.
* Photos are kindly provided by the Optogan press service.
** Photo from portalnano.ru .
First , a complete list of published articles on Habré:
Opening the Nvidia 8600M GT chip , a more detailed article is given here: Modern chips - an inside view Inside
view: CD and HDD
Inside view: LED bulbs
Inside view: LED industry in Russia
Inside view: Flash -memory and RAM
Inside view: the world around us
Inside view: LCD and E-Ink displays
Inside view: digital camera arrays
Inside view: Plastic Logic
Inside view: RFID and other marks
Inside view: graduate school in EPFL. Part 1
Inside Look: Graduate School at EPFL. Part 2
Inside view: the world around us - 2
Inside view: the world around us - 3
Inside view: the world around us - 4
Inside view: 13 LED lamps and a bottle of rum. Part 1
Inside view: 13 LED lamps and a bottle of rum. Part 2
Inside view: 13 LED lamps and a bottle of rum. Part 3
Inside Look: IKEA LED Strikes Back
Inside Look: Are Filament Lamps So Good?
and 3DNews:
Microview: comparing displays of modern smartphones
. Secondly , in addition to the HabraHabr blog , articles and videos can be read and viewed on Nanometer.ru , YouTube , and Dirty .
Thirdlyif you, dear reader, liked the article or you want to stimulate writing new ones, then proceed according to the following maxim: "pay what you want"
Yandex.Money 41001234893231
WebMoney (R296920395341 or Z333281944680)
Sometimes you can read briefly, and sometimes not so much about the news of science and technology on my Telegram channel - we ask you;)