Dr. Nisa Khan

Chipmakers, telecom, patents/litigation, registered investment advisor
Dr. Nisa Khan
Chipmakers, telecom, patents/litigation, registered investment advisor
Contributor since: 2013
Company: IEM Asset Management, LLC
My advice is the same as that from Warren Buffett: never invest in something you don't understand. That being said, it is not that LEDs aren't the future for general lighting. There are physics and technology problems to be solved and implemented that none of the public or private companies are yet engaged in. Remember, Einstein's general theory of relativity wasn't understood or accepted until someone like Arthur Eddington came forward and declared Einstein was correct! It will take some time for the wider audience to figure out where the real LED lighting IP is!
I'll keep it short.
1) There are references and work done by the Americans and the Dutch on damage done by blue content light at night. For starters, you can try "J. Roberts" and go from there.
2) LED light distribution problem is not that it doesn't put light above it. The light distribution problem comes from the fact that current LED light distribution is a Lambertian radiation output which is not what curved light sources generate. For example - the sun, an incandescent and a fluorescent tube lamps don't generate a Lambertian.
3) The definition of 'coherence' is in any reputable physics or EE communication book. Mixing distinct color lights cannot maintain coherence.
All of this explained here in short requires quite serious studies as well as thinking. It is up to you all to figure these out.
The reference you quote says the light source simultaneously lases in red, green and blue. This is nothing new and has been done by my coworkers many years before these authors in "Nature". When you combine the 3 individual lasers to produce white light, even simultaneously, the combined white light itself is not lasing! Their white light laser definition is incorrect.
Here are a few things you should know:
1. Since the 1980's I have manufactured, tested, and used lasers (in Fortune 50 and 100 companies) to develop high-speed fiber-optics and microwave technologies which are vastly employed for making the internet work today.
2. I was a student of two laser inventors who worked at IBM and ATT Bell Labs in the 50s and 60s and beyond.
3. I myself have worked at Bell Labs to further develop laser and fiber optics networks for many years.
4. I do know the definition of a laser and didn't invent that. Charles Townes and Arthur Schawlow did and I learned from people who worked alongside them!
There is no such thing as a 'white light laser'. Laser light must be coherent and white light is not. It is true that you can create white light using a blue laser and combining it with yellow or similar color phosphor. But what comes out as white light after going through phosphor is no longer light that can be called 'lasing'. Mixing tri-color light from 3 lasers is also no longer 'lasing'. Both lasers and LEDs in semiconductors were discovered more than 50 years ago. So creating white light with phosphorescence or RGB mixes are nothing new. Besides Cree doesn't own these technologies - other companies have patents on these and have advanced technologies.
Aside from all of this scientific and technological brouhaha - filament bulbs do have many lighting advantages which may come back to fashion when people realize what these advantages actually are. We may invent dazzling technologies and get wowed by them - but nature still beats us with fundamentals that cannot be violated and its beauty!
Some of the benefits you are describing are correct. These include the facial recognition argument and perhaps in certain cases energy efficiency calculation at the unit level - taking only the lumen/W figure. However, there are adversarial effects to LED street lighting also. These include energy savings not being calculated at the end-end system level which is more than looking at the lumen/W figure. Also there are health issues involved with using too much non-yellow light at night. These are not part of the common knowledge portfolio at this time. LED problems also include high glare as well as needing too many lamps or luminaires for creating a total lighting design compared to designing for the same lighting application using existing lamps and luminaires. This has to do with LED light distribution problems.
Just because huge cities are converting to LED lighting does not mean they are doing the right things. Ask yourself who complained about fast food 30 years ago. It was fast, cost efficient and people liked eating burgers, fries, and fried chicken, and yes - sugary soda! Why are these being flagged now when doctors and nutritionists should have raised flags about these 30 years ago?
Let's see what happens to our eye sight and biological rhythm 30 years from now from LED lamps. Not to mention in the next 30 years we are sure to find out which LEDs actually last and which ones die and for what reason. Rigorous scientific and ethical work are not that common. Making money in any which way people and businesses can is.
Thanks for your feedback. I am factually backing up my scientific findings and not my belief in numerous peer-reviewed and media publications. I cannot present them all here. But my work is public and should be easy to find.
Lighting should not be judged by the way lamps look when illuminated as objects. In order to judge lighting, you need to measure luminance and illuminance accurately with photometers in 3 dimensions. When you do that you will find that for large space illumination like indoor commercial lighting, warehouse lighting, and street lighting - LED products illumination is not the best. There are scientific reasons for this. Unfortunately general people including many lighting professionals end up judging lamp by looking at them as illuminated objects. The purpose of lamps is not to be illuminated objects - but to illuminate space we live in in a high quality manner.
AYI is a lighting company by historical measures. Cree and Osram are not. Osram belong (soon to change) to a larger company that have Sylvania (also soon to change I understand) who is a lighting company however. However within the lighting industry - there is a food chain like other industries. Acuity is a luminaire company meaning they create the system lighting using Sylvania or GE lamps for example. Sometime during a revolution, a system company can look aggressive and can do better integration of lower food chain elements and look like the king. But that can change too. Right now - most people have some feeling that LED lighting will make out big - but the truth is that beyond where LEDs are a no-brainer, like in backlighting for TVs, laptops, billboards and in some small general lighting market, LEDs currently don't make sense. Many people in lighting and general consumers don't understand that and just are blind or in it for the money. Just like 30 years ago, people (or even medical/nutritional professionals) didn't think inorganic food like high-fructose corn syrup was bad and thought it was the same as natural sugar. There are plenty of examples like that. Another trouble with LED lamps for general usage is that LED lamps of real good quality are not cheap. They are subsidized and companies are taking a loss on them and they don't want to publicize it. To me - real high quality LED lamps are not available now - period. It will require new IP that none of the LED or any lighting companies know or understand. When they hear what I say or write/publish, they ignore it because it demonstrates what they don't know. Time will tell how this 'new IP' will play out. Stay tuned!
I am afraid some complex changes are already happening in the LED lighting industry. Osram plans to sell off its lamp and lighting business and it appears that a Chinese buyer will jump on this. Meanwhile Philips Lighting may be for sale and Lumileds (formerly Philips) hasn't found a definite buyer. Toshiba pulled out of LED lighting business altogether. LED lighting has been overhyped and Cree's retail lamp business is not sustainable. LED module and future lamp business will go through major overhauls in the next 5 years and in the meantime many will drop dead like flies. We saw this similar hype followed by death for the telecom in the 90's and the real-estate/MBS industries in the 2000's. The writing for LEDs as I describe it is on the wall. The smart lighting aspect of the industry will not survive, by and large. A small part of it will. Time for Cree and others like Cree to get smart and get the intellectual property from the right places.
Agree with you on all. I believe analysts from well-known financial companies showing a success record of less 50% is the reason contrarians exist and often the opposite turns out to be a better bet. Note that on average, throwing darts should get one to a 50% success!
Also agree on the possible GS tactic. It seems to happening more often these days than the past. Look at WFM. Frenzy selling after hours a couple of days ago is now a buying opportunity.
For private equity, Theranos comes to mind. Ms. Holmes should either defend herself thoroughly with facts and actual data or admit she has been a hoax. If she doesn't do it herself, the media will do it for her in less than 5 years.
I wonder which is the real truth. Whether she is guilty or not, she should have published real data and should have provided actual blood test results that commensurate with traditional blood tests. This isn't that hard to do provided her claims are valid.
So called "scientists" and "engineers" like her (with no advance degrees whatsoever) give bad names to some of us. Unlike Theranos and Ms. Holmes, I and my company actually publish in peer-reviewed journals and other media journals about scientific and technical things that we claim.
It is amazing how investors and analysts (from private equity and Wall Street) create and destroy wealth of many companies with their inadequate comments!
It is amazing how investors and analysts create and destroy company wealth by their inadequate comments.
I agree and disagree with Brian Lee's assessments. Mid-range and even lower-Mid-range LEDs will become important in the future. But smart-lighting is going to be short-lived as this is a hype/trend and irrational exuberance; there is plenty of important work that needs to be done to improve LED lighting and the focus on smart lighting shows a lack of understanding of lighting quality in general as well as a deliberate deviation from solving real LED lighting problems which happen to be quite difficult. Cree does certain aspect of LED lighting quite well and they may be able to solve real LED lamp modules issues in the future better than other companies. But that is a big "IF".
LED lighting market is growing rapidly because of mostly global growth. While it is also growing in the US, it is not 40% of the lighting market and I doubt the CAGR of 25% will happen or if it does, will not likely sustain. The energy efficiency pitch will not last too long because it is the end-end efficiency and utilization that matter. It is like saying let's go with solar and wind energy and do away with the grid system. The grid is the distribution system that is necessary to provision power where needed. LED lamps, while energy efficient, do not distribute light in space very well. There are too many dark areas and spots in between lamps and one ends up wasting too much energy in running single LED lamps at too high a current or use too many lamps in a room to get the same illumination one gets with far fewer incandescent of fluorescent lamps. This challenge is not well understood by even many lighting professionals and LED engineers. But the education process has just begun and it is just a matter of time before most people realize that LED lamps need to be designed differently. From where I am standing, any of the lighting companies mentioned in this article, do not yet know how to effectively design LED lamps for more effective illumination with fewer and more long-lasting LED lamps.
Many changes are taking place in the LED lighting industry as of now. LED lighting is being hyped up based on some chip efficiency improvements and the craziness over smart lighting. Smart lighting is nice to have - but it is the basic lighting quality that must be designed in first and it is still not achieved. I wrote about this in my June column in Signs of the Times magazine: find it here:
http://bit.ly/1eTKVff
Also my full review of this year's Lightfair will be published soon. For now, I'll continue to recommend reading my book, Understanding LED Illumination, especially Chapter 6 to understand why LED's have glare, why the light distribution is not suitable for general illumination and how to design LED lamps with the highest quality. All LED companies will need to come to terms with these issues and so far, I still don't see any signs from any LED designers or manufacturers that they understand the basics of LEDs and lighting. Stay tuned....
I applaud you for finding this gem. DAKT's value is unique and will be very long lasting. Its origin is from a Midwestern university's electrical engineering department. I know a thing or two about this kind of origin and I know a thing or two about LEDs on these billboards. Investors like Peter Lynch also knew a thing or two about investing!
All things being equal, if you get more energy from the sun on earth, that system would be considered more efficient. The total problem has to do with total radiant power incident within a particular bounded space measured over a finite period of time. That would give you total energy that one can measure properly. It is important to distinguish power and energy measured over finite space and time. The problem has to do physics as well as engineering, I am afraid.
With respect to efficacy vs. efficiency, first thing one has to realize is that efficiency is only appropriate when you compare what goes in vs. what comes out using the same unit of measurement. If you send a watt of power into a system, you need to measure what wattage of power that comes out to get a number that can always be represented in percentage; here the units must be the same in order to compare the same power.
Whenever you send some unit of power that doesn't match the unit of power coming out of the system, it is correctly called 'efficacy'. A man-made detector measures radiant (optical) power in watts; so if you measure the LED radiant power (in watts) with a detector, you can find out how efficient it is when you know what input electrical power (in watts) was sent into the LED. You will get an efficiency number that can be represented in percentage.
However, when our eyes see optical power, we don't see watts; the lighting industry has specified it to be what is called lumen - that is visible optical power seen by the human eye. So when we characterize a visible LED for its power conversion, we use the term efficacy - because it is lumen over watt; and not watt over watt. A lighting system's efficiency can be also defined when you are comparing total lumens in vs. total lumens out. In short, watt in/watt out and lumen in/lumen out - both give you efficiency; watt in and lumen out would be characterized by efficacy. To further understand all this, I would refer you to my book, "Understanding LED Illumination"; find it in amazon.com.
It is important to realize that efficiency comes at the cost of many other characteristics important to lighting and thermal management, which also relate to how long an LED lamp will last. Another important thing to realize about LED lighting is that a single device brightness is much too high for the so called "high-brightness" LEDs that Cree and Lumileds bet on earlier. These high-brightness LEDs are neither as efficient as the "mid-power" LEDs, nor does one need all that brightness concentration on a single spot. That is the problem.
Finally, color rendition is much more important for general lighting than automotive. In summary, lighting products (automotive included) has much more to do with getting the most radially efficient LED; other factors include (but not limited to) brightness balance, light distribution quality, thermal management, longevity, color quality, and finally cost.
Hope this helps.
Dr. Nisa Khan
A very good question indeed!
I am an engineer among other things and 'efficiency' is my middle name. Having said that, what most people understand about LED efficiency is not the whole story. The so called "undefined" midpower in most cases actually has better lumen/W efficacy (a better word than efficiency) because GaN has a droop phenomenon which says that at several hundred mA of injected current, the light output of a blue LED is actually less at per unit of injected current. For the same reason, thermal issue becomes more of a difficult problem for high-power LEDs, generally speaking. What injected current and LED light power level are the best, you ask? The answer lies in multi-dimensional design optimization process; a resource many of these well-known LED companies don't have!
The most important reason for going midpower is that in most applications, light concentration needs to be lowered. Even for automotive, the extremely high luminance from good quality LEDs is an overkill.
While people get hung up on very high efficiency, think about this:
If the sun were closer to us, we would get solar energy more efficiently. But it would also tip off the solar and the universal balance which is why we are able to exist in the first place. So be careful about asking for the highest efficiency without thinking about the whole problem. This is especially true about extremely bright light!
One of the important issues concerning risk is the following - which I posted yesterday:
By Cree's own admission, they were late in the game with the mid-power LEDs. But I have written about the importance of lower power LEDs for lighting over 4 years ago. In 2011, I pointed out to a speaker from Philips Lumileds in a conference in Austria that chasing the high-power LED is a mistake. Although the speaker acknowledged it was a very good point, unfortunately the executives at such companies failed to see the benefit of going lower power.
Mid-power LEDs are not magic and it is not a guessing game of what power LEDs are optimum. It is a matter of understanding the science and technology of LEDs and being able to design, simulate, measure and take a slew of tradeoff issues into account to zero in on the optimum LED input electrical power.
By Cree's own admission, they were late in the game with the mid-power LEDs. But I have written about the importance of lower power LEDs for lighting over 4 years ago. In 2011, I pointed out to a speaker from Philips Lumileds in a conference in Austria that chasing the high-power LED is a mistake. Although the speaker acknowledged it was a very good point, unfortunately the executives at such companies failed to see the benefit of going lower power.
Mid-power LEDs are not magic and it is not guessing game of what power LEDs are optimum. It is a matter of understanding the science and technology of LEDs and being able to design, simulate, measure and take a slew of tradeoff issues into account to zero in on the optimum LED input electrical power.
Recently a number of LED companies including Cree have realized the value of mid-power LEDs.
I brought up the value of lower power LED for lamps publicly in 2011. Also wrote about it in my LED Update Column in Signs of the Times magazine as early as 5 years ago. What is now the buzz word called "mid-power" LEDs could go down in power even further for better lighting performance. One has to do the engineering problem holistically to optimize electrical power input, optical radiation, thermal issues, cost and reliability issues, and others. The interested people can read my latest publication in LpS 2014. You may need to pay a few dollars to get the whole paper available online.
Would like to add a new comment relevant to LED lighting as of late.
It is exciting that this year's Nobel Prize in Physics went to 3 physicists that paved the way for white LEDs by discovering the blue LED. But it is still surprising that the world including the Nobel committee and the NYTimes still don't see the challenges behind LED lighting. While LEDs are very efficient, they are not 20X and 4X more efficient than incandescent & fluorescent lamps if one makes apples-apples comparison. When efficiency/efficacy is addressed for LEDs, the whole end-end system should be considered; this includes color properties, light distribution and spreading. LED lighting will not become the mainstream in the 21st Century unless its glare, cost, reliability, reproducibility, and safety problems are resolved. I hope the Nobel committee, NYTimes, and the general population benefit from the work I have done on LED lighting; much of it is published and more is on the way.
I brought up the value of lower power LED for lamps publicly in 2011. Also wrote about it in my LED Update Column in Signs of the Times magazine as early as 5 years ago. What is now the buzz word called "mid-power" LEDs could go down in power even further for better lighting performance. One has to do the engineering problem holistically to optimize electrical power input, optical radiation, thermal issues, cost and reliability issues, and others. The interested people can read my latest publication in LpS 2014. You may need to pay a few dollars to get the whole paper available online.
It is exciting that this year's Nobel Prize in Physics went to 3 physicists that paved the way for white LEDs by discovering the blue LED. But it is still surprising that the world including the Nobel committee and the NYTimes still don't see the challenges behind LED lighting. While LEDs are very efficient, they are not 20X and 4X more efficient than incandescent & fluorescent lamps if one makes apples-apples comparison. When efficiency/efficacy is addressed for LEDs, the whole end-end system should be considered; this includes color properties, light distribution and spreading. LED lighting will not become the mainstream in the 21st Century unless its glare, cost, reliability, reproducibility, and safety problems are resolved. I hope the Nobel committee, NYTimes, and the general population benefit from the work I have done on LED lighting; much of it is published and more is on the way.
There is a very good article on Daktronics in the NYTimes on July 19, 2014:
http://nyti.ms/1r3Vigs
The 8-mm pixel pitch on large electronic display boards is a great achievement. And it appears that this along with other high performances are now achieved cost effectively. Glad to see DAKT is getting the attention for what it does.
Glad to be of help in providing valuable information here.
To the general audience:
Want to learn about LED glare, photometry challenges and solutions? Come to the GL Optic Booth (Booth #832) at Lightfair in Vegas next week (June 2 - June 5) to buy my book "Understanding LED Illumination" and my latest LED articles that answer crucial questions on LED lighting, photometry, and much more.
Dear LED lamp user,
Thank you for your post. You are not alone in liking LED lamps, particularly from Cree. I have heard exactly the same feedback from a number of users that are EEs who are by nature technical, energy conscious, and good judgment holders. But lighting, unlike the exact scientific fields, is subjective as well as much more of an in-depth subject involving science, art, human psychology and evolution than most people realize. This included me 7 years ago as well.
The reason you don't see glare in CREE and other lamps is because they are using a thick diffusive material in front of the LED modules. The light distribution also appears acceptable for most applications involving limited spatial volume. The ~3000K color temperature also makes them look warm. These are good improvements over the LED lamps from a couple of years ago. But, for industrial/commercial lighting users in buildings with high ceilings, LED replacement lamps will be challenged. And for very picky photographers and cinematographers, LED lamps' light distribution will not suffice. This is hard to explain here. And don't forget, many picky women like a full-spectrum color distribution from the lamps because facial complexion as well as particular color of attires are fully rendered only under sunlight and incandescent lamps. Light distribution also matters here.
But aside from these picky issues, thermal management and lifetime issues are important. There LED chip and package yields are non-uniform and over time many lamps will fail because not all chips have the same quality. This is a compound semiconductor wafer quality variation issue. In the last several weeks, I have seen practically brand new Audis with one headlamp already burnt out while the other one is working! In other cars, including Infinity, I have seen a number of modules in the whole backlight red LED lamp ensemble burnt out. Observed the same in many LED traffic lights. So yes - over time, people will notice - not all - but a good number of LED lamps will show much shorter life spans. These need to be corrected as much as possible and there is denial (from many LED companies) as well as lack of technologies that can solve these problems well.
If you talk with very knowledgeable lighting/vision scientists, they will say that other LED problems include blue light hazard, flicker, stroboscopic effects, shadows from striations, nausea/seizure-induced dot patterns. (CFLs have some of these issues as well.) But I attempt to only solve a few at a time with LED lamps, being mindful that these are real.
LEDs for lighting was a breakthrough technology in the 1990s when the first blue followed by the first white LEDs were invented by Dr. Shuji Nakamura at Nichia. After that, LED technologies developed just like other electronics and optoelectronics technologies over the years following a similar improvement dynamics as Moore's Law. This allowed lumen/W to increase and cost of chips and packages to come down impressively. These were done by many more companies than just Cree. But alas, lighting is more than just producing bright, harsh, and energy-efficient visible radiation. The quality of light that the sun and incandescence process produce is simply beautiful and good for living beings in many ways. They have the color spectrum and light distribution balance like no other. LEDs to mimic this light quality would be a breakthrough for lighting and LED industries. As for giving examples, you can check my work in the published domain. More to come there in this arena from my company soon.
Cree's tech, for the most part is a commodity at this time. I have tested many bulbs from Cree, Philips, Osram, Samsung, and others. Many of these results are in my book, "Understanding LED Illumination". More will be published soon. Don't get me wrong - not everyone can make LED chips. But there isn't just one player in town either. I have been directly involved in making LED-type optoelectronic devices for much longer than Cree has.
Cree is surely a leader in SiC substrates; this is good for LEDs and power amplifiers. However, LED lighting is much more than just building chips. Cree has learned that somewhat. But the entire LED industry including Cree has still a long way to go.
LED lamps and components at this stage of maturity, or lack there of, are surely a commodity industry. So companies like CREE as well as GE, Philips, Samsung, Osram Sylvania and others need to be careful about how much they are investing in LED lighting. When real breakthrough technologies become available, all such companies will likely take major losses; some will consolidate and some will disappear altogether. We have seen this landscape before with IT, telecom and other tech arena. And yes they did cost a lot of people some serious bucks!
Here is another link to begin understanding the current LED hype:
http://bit.ly/QFi2H4
There is a lot more going on in the lighting industry than meets the eye. In order to provide substance for the readers in an easily readable fashion, I use the list below:
1) Cree bulbs now found in Home Depot have limited uses. The color rendering is only 80 and light distribution is still not ideal for many users. The users who insist on high light quality will still buy incandescent because its color rendering is 100 and the light distribution is beautiful. Understandably, Hollywood is a great proponent of incandescent lamps, it seems. So are many lighting experts who will try to persuade and educate people to not ban the incandescent lamps, but limit their uses.
2) Current LED lamps have many problems that many people don't realize. In order to solve them, more innovations are necessary; and neither CREE nor GE, Samsung, Philips or other well-known companies have all the patents or techniques that will solve real LED lamps' problems. One such problem include excessive glare from LED lamps. Cree just puts an inefficient diffuser around their LED modules to get rid of the glare in their bulbs. This is not a long-lasting solution.
3) Moreover, Cree lamps have thermal management issues that will only surface after longer uses. Cree seems to have inadequate inventions there as well.
4) LED lamps and luminaires are not yet suited for large-space illumination. In industrial environments, fluorescent tube lamps do a much better job with illumination and efficiency and they don't have glare. LED replacement lamps for these are not evaluated properly and people are making buying decisions just by looking at the lamps. Thorough measurements of wattage, lumen distribution, and color quality must be done for the entire space that needs illumination before making the decision to switch to LED replacement/retrofit lamps. A good number of those who switch to LED replacements lamps, switch back.
To learn more about these issues in detail as well as new LED lamp innovations, I recommend reading my book, "Understanding LED Illumination". I'll have more publications in the way this year to address LED lamp design challenges.
http://amzn.to/1iLqtey