Face, it, we all like it when someone shakes things up. Whether it’s in politics, our own families, or in technology—we rely on brave individuals to say things we feel we can’t, or do things we wouldn’t dare do. One such person is Ed Rodriguez, who I’ve had the pleasure to work with for the past several years as he stormed his way into my writing when I penned LEDs—Hotter Than You Think, based on his EDN contribution: True or false: High-power LEDs don’t generate IR heat in the forward direction like a filament lamp. Over the years, Ed continued to provide interesting articles that were a combination of, “Let me tell you the real story,” and here are difficult-to-find facts to rebut just about everything.
I recently talked with Ed again and asked if he’d let me interview him to get his take on the LED industry as it stands early in 2017. Read on and naturally, feel free to comment, whether you agree, or don’t.
Mathas: Ed, thank you for taking time out to talk today. Let’s start by looking at what everybody gets wrong about LEDs, and why?
Rodriguez: Before we talk specifically about what’s wrong, let me provide some insight. First, is the near total disappearance of technical publications as we knew them 10 years ago. For decades, there were thoughtful, lengthy articles monthly in 5-6 major electronics publications. Half of them were written by editors and half by application engineering types in corporations. These were well thought out, well edited pieces that had a long shelf life. That has disappeared. They represented fundamental checks and balances against all the nonsense that is everywhere today.
Second, there has been a disappearance of the role of the classic manufacturer’s sales rep among all makers and sellers of LED components. All major LED component people are essentially selling through distribution and not using reps. No longer does the sales rep stop and talk with the chief component engineer or design engineer every few weeks, providing valuable information from the manufacturer to the customer. That level of attention does not exist.
Even though distributors claim that they have in-house applications people, they’re little more than technicians or junior sales people with a business card that calls them a lighting solutions person. I've dealt with hundreds of reps over the years that were great for feedback and dialogue. If I have a serious question today and I am with a startup company, the best I can hope for is that the distributor sends an email to someone who should be able to answer my question. Yet, if they do, most likely there will be no response. When I would talk to reps, no matter if I was with a startup or a large company, I’d have a response within 24 hours that made sense.
Third, you can no longer call a company regarding application support. Even the best companies have stopped answering the phone when you call technical support for your application. Again, all you could do is email and someone might get back to you. Usually they do not.
The fourth thing is the obsession with intellectual property and patentability and the fear of Chinese knock-offs. Everything now is, “we don’t want to give away any of our trade secrets or any of our techniques.” If you go back 10-15 years, companies like National Semiconductor and International Rectifier, for example, constantly wrote and published application notes on how to use their products. Some companies wrote whole books on the subject—300-page technical application manuals.
All of this has now disappeared. If you try to find anything major published by the top LED manufacturers today on how to use their products in creative ways, you would come up dry. Bean counters and those pushing the price of stock are creativity killers. That reality is mixed with the fact that the money is now in software and in social media and not in hardware—further killing incentive to add this level of assistance.
There is one more thing in addition to these four. The wild, wild West of maximum baloney is the grow light industry. Every Tom, Dick and Harry used to put cheap grow lights in their garage or basement to grow pot. Now there’s an attempt to translate grow lighting into a legitimate agricultural lighting. There are now so many companies promoting utter nonsense and making false claims. This is an area that needs to be cleaned up, and 10 years ago that would have happened by way of several serious technical people in engineering publishing articles in trade publications debunking outlandish claims. Everyone would get smarter in the process. Now, there’s no mechanism to do that.
Mathas: So, the research that’s being done right now for agricultural purposes, are you saying it’s legitimate or not legitimate?
Rodriguez: There are four to five competing philosophies as to how to go about it. For example, there’s a startup that is raising money. Everyone knows that plants have been growing for millions of years under sunlight, but all plants don’t want full sunlight, but instead want red at a certain time of day or phase of growth, or blue, or green. So, these companies announce that they’ve put together a team of software computer experts that don’t know anything about plants, but they know computers. They’ve created algorithms to deal with the variety of light colors, and these magical algorithms will deliver big, bad and beautiful plants and flowers.
There are people raising millions in venture money based on fooling mother nature. Then there’s another group of people that claim to use magical algorithms to blend the light spectrum. Still someone else says that full spectrum lighting should be used with LEDs so that the light truly replicates the sun. Others argue as to the form factor used. A California company says they want to have the equivalent of a ceiling fan with LEDs on the fan—they call them spinners—so that they spread light everywhere like a water sprinkler.
This is what the industry is doing, and it is madness and one of the biggest frauds in the industry. Years ago, a company would write about their approach, causing dialogue. Saner heads would say, “everyone is worried about algorithms and the optimization of spectrum when the fundamental way of shining light down is insane. This all comes back to the lack of dialogue to expose these things.
Mathas: Let’s go back to, what does everyone get wrong about LEDs and why?
Rodriguez: In the grow-light industry, what everyone gets wrong is that they don’t understand the optics. But LEDs in general is a new industry, so there is no expertise. Everyone in this industry was doing something else 10 years ago. It isn’t like 10 years ago in the integrated circuit industry. The people designing ICs 10 years ago were experts. In LEDs, however, everyone is a Johnny-come-lately. And, they are getting a lot wrong because of the lack of expertise. There is no such thing as a degree in LED lighting that is based solely on the technology, its focus instead is on lighting/architectural design.
LED lighting is a combination of optics, semiconductors, thermal management and power supplies. Where does one go to become sufficiently educated in knowledge of four disciplines? The person that knows the semiconductor industry, knows nothing about thermal management and optics, etc. There are too many amateurs and one-trick ponies. So, to your question, what do they get wrong? How can they sell their product without that basic knowledge? They don’t appreciate the interdisciplinary nature of LEDs.
Mathas: One of the things you’ve written about in the past, involved how can you trust what is being published by the companies as to their specs?
Rodriguez: That’s true. They are throwing numbers out there, for example, the 25,000 or 50,000-hour life, and no one is willing to stand up and ask why we should believe them. LEDs haven’t even existed this long. When companies say their bulbs will last 50,000 hours, why can’t they explain it in English? Everyone is confused about the 50,000-hour life span. I know the large companies don’t want to talk about it too much because they’re concerned about their competitors seeing how they come to do their calculations. There’s no standardization.
Mathas: Do you think standardization will ever come to this industry?
Rodriguez: Probably never. Ironically, everyone thinks that standardization is good, but it is an absolute that the creation of new companies depends on the ability to not be standardized. That may seem like a contradiction. The only way for new companies to be successful is to come along and do things a differently. It takes disruptive technology to break the mold and enable growth. Entrepreneurs and venture capitalists look for companies that don’t do things the standard way.
The one area where standard is good in this industry is that lighting manufacturers must now put their actual specs on bulb packaging as to color, temperature and lumens. Fluorescents did not have to do this and any time you wanted to buy a fluorescent in the store, it was hard to get different bulbs delivering the exact same thing—it would never match. Some standardization is helpful, too much kills innovation.
Mathas: When you look at published specs, what do you not trust?
Rodriguez: The delivered lumen figures. What a lot of companies do is they make a lighting product and they buy an LED from Cree or others that says 100 lumens, and they insert it in their product. By the time they put a lens on it and reflector, etc., the actual amount of light output is nowhere near the 100 lumens. A lot of games are played with delivered lumens figures.
Mathas: Is this innocent or malicious?
Rodriguez: I’m involved in a project with a special light bulb. Everyone talks about the LED and lifetime but the lifetime is extremely sensitive to the temperature you operate at. However, if you buy an LED bulb, and you put thermocouples on it and measure it, you will find that 90 percent of the LEDs are running red hot. I wouldn’t even put it in my house unless I get thermal measurements.
There is a tremendous variation between bulbs, they look the same, they may be called a 30-watt equivalent or 60-watt equivalent, but some are running hot and they know that you don’t have any way to measure it. The user doesn’t know what it means, and if the bulb survives three years, they won’t even remember where they bought it.
Every single company makes their LED bulbs in China and now there’s a price war. Because LED bulbs are so heat sensitive, even the very best companies are pushing the limits. One of those is GE. They came out with a commercial LED called GE Bright Stik and they flooded Home Depot and other places. I tested some of those and you can’t put them in certain light fixtures. I would never use them in my house. I see some LEDs that, if you use them upside down in an enclosed fixture, it’s an accident waiting to happen. The companies are pushing limits and getting away with it.
Mathas: You’re talking about fire danger?
Rodriguez: Fire and sparking. About three years ago there was a huge recall because one company’s bulbs were smoking. So far, the companies must feel the danger is worth the risk.
Mathas: One of the things you’ve said in our discussions is that people in the know at large companies, for legal and general secrecy reasons, aren’t at liberty to reveal the facts about their products.
Rodriguez: They can’t talk about it publicly, they’ll get fired.
Mathas: Of course, but who is in the know?
Rodriguez: If I want to find out something in any company, I want to talk to the marketing manager. I want to talk to them more than the engineering person. Marketing tends to have a more complete perspective than engineering. I don’t go to marketing for the nitty gritty on thermal management, but the smart marketing person’s perspective and understanding is usually much broader than engineering’s. I still want the technical person to answer my tricky questions, but marketing is better for legitimate perspective.
OLEDs, for example, if I want to find out if OLEDs is still just a bunch of hype, I’m not going to the OLED Ph.D. to find out if it’s a real market or not. There are a lot of things that you can do in technology that are wonderful. They may still be wonderful 10 years from now, but they won’t sell. It might look great in the annual report and the stockholders love it. You are always going to have to show that R&D is working on the latest technology just like every automotive company has a self-driving car project.
Mathas: What are the real gaps that you see in the industry? What’s missing, still a barrier, etc.?
Rodriguez: I was at a sales organization in 2006 and sat in on a meeting. The sales manager said, pointing to the ceiling fluorescents, “Within two years, all of them will be replaced by LEDs.” That was 11 years ago. You can go to many office buildings and you still won’t see too many 2-by-4-foot ceiling fixtures that are LEDs. Mainstream automobile headlights have not gone LED. It’s been dragging out now for five to six years, but for the most part, isn’t happening.
LEDs were said to be on the same track as Moore’s Law, so that the efficacy would double every three years, going up to 300, 400, 500 lumens per watt. I’ve never believed it and it hasn’t happened. You can read about lumens per watt breakthroughs in the lab, but you must be able to replicate them. The test is, when can I go to the suppliers and buy one? There’s a huge disconnect between the lumens per watt available at the store and the breathtaking R&D predictions and claims that have been made over the past several years. Have we hit a wall on chip technology or a wall on phosphor technology? There’s nothing written on where the fundamental barriers are today?
It’s also important to look back at technology. If you were going to put a bulb into your house, an incandescent bulb versus an LED bulb, and if they were the identical price, there is no question whatsoever, that the incandescent bulb will be more reliable—there’s only one part in it, a piece of wire. Maybe there are some cases where other technology will never truly replace it.
Mathas: Where, in your opinion, has the press missed the boat?
Rodriguez: The press isn’t asking the questions. Nobody is asking the questions. I can probably come up with 50 questions that I’d love to have an answer to.
