Way back when I was just a pup, I did a series of diaries on energy and lighting.
That was when oil was in the $50's, Obama was an unknown senator who we only heard about from NPR's "Wait, Wait don't tell me", and most people still had jobs/health insurance/hope.
Well that's all changed, but the laws of physics and modern energy management haven't changed. So let's return to those thrilling days of yesteryear, and update the world.
This is a reprise of my diary series on the subject of artificial lighting, and how it gets delivered to your home. I'll use my 24 years of experience in the energy management industry to describe the basic physics, technology, and impact of the past, present and future of lighting. Before I get started, let me say that I'm not in the best of moods today (3/31/09) and will not be able to support this as much as I'd usually do. I'm awaiting a phone call at 9 EST, otherwise I'd be in bed by then... last diary I did I stayed up way too late to support (I get up 4:45 AM) and it really killed me the next day.
In other news, there will be more diaries like this (but not nearly as long...) on other forms of renewable/alternative forms of electrical generation, and motor vehicles in the next few months when I find the time. I used to have a lot more time for blogging, but the last year has been crazy. There have been a lot of diaries on these subjects, some brilliant, some not so. There are a lot of misconceptions on this subject, and people may not know exactly what are the implications of these technologies.
The basic ingredients of artificial light are electricity and excitation. You take an electric current, or a high-voltage electrical field and you use it to excite atoms into a higher atomic state, which they then depart, casting off photons.
One way to induce atoms into that higher state, is simply accomplished by heating. Running a high electrical current through a conductor will heat that conductor, causing it to glow. The first attempts of passing current through wire predated Edison by three-quarters of a century, but Davy's platinum wires were far too expensive to be practical. Carbon and natural fibers failed due to their fragility, and the constant need to adjust the distance of the two elements of the carbon arc. Literally dozens of partially successful attempts predate Edison, he didn't succeed because he had the better material, but because his pumps could achieve a far better vacuum. It wasn't inspiration, but perspiration... and his vacuum prevented the electrode from burning itself out in seconds.
The problem is, incandescent sources of light give off their energy in a bell-curve of frequencies, based on their temperature. The radio-spectrum of wavelengths transmitted by this heat is all over the map, most of it in the infrared. We can't see more than 90% of the energy produced by incandescent lights!
The vast majority of the energy used in artificial lighting for the last century is invisible.
Shortly after the introduction of incandescent lighting, experiments were made to pass current through an evacuated tube, filled with mercury vapor. Hewett's first lamps gave off an astonishing amount of light in comparison with incandescents, but they were huge affairs, and prohibitively expensive. But they pointed the way to perhaps a four-fold improvement in light efficiency. The problem was, the original mercury vapor lamps were very rich in UV, and had a ghastly green glow. It had its commercial uses, and was supplanted by higher-pressure models in the mid 30's, which were much smaller, and are a precursor to modern high pressure sodium lamps used in street lighting today.
Then they hit upon the idea of adding phosphorescent chemicals to the bulbs, which would absorb the UV light and emit visible spectrum light. These chemical phosphors are fluorescent sources of light, with several peak wavelengths instead of a broad spectrum.
In order to improve upon the color of the light, as well as the industrial look of the long tubes, they experimented in the early 70's with different shapes, and configurations. Today we see the cumulation of that research and development with CFL or Compact Fluorescent Lamps. By adjusting the frequency of electrical discharge to the ultrasonic, and by changing the mixture of gases and phosphors they've been able to tailor the color, as well as the cold-starting characteristics of the lamps. Early fluorescents were unable to start at cold temperatures, and had a sick orange glow for the first few minutes.
There have been a few other methods of generating light that have had limited success. Instead of using voltage to excite the gas, there have been inductive methods where the gas either sat atop or was in the near presence of high frequency electromagnetic coils. These techniques are very efficient, but they have drawbacks which have prevented them from wider distribution. They may in time have further penetration into the market, but they are still very expensive, require very expensive ballasts and do interfere with other electronic equipment.
The future of lighting seems to be drawing towards LED based lamps. LEDs are very efficient at converting electricity to photons, although they have relatively spiky energy spectra which causes them to have a diminished color spectrum. They too can be coated with phosphors to produce a broader spectrum or a "warmer" light. Early white LEDs were in the 7000°K and up color range, they are now down to as low as 3000°K.
Somewhere out in the distance, a generator is turning rotational energy into electricity. A series of wires are being turned in the presence of a magnetic field, turning mechanical energy into electricity. This is totally invisible to all of us, except for the once a month we get our electrical bills, and have a minor stroke but the mechanics and the delivery system is as important to our daily lives as anything else the modern world can offer. Without power, most of our daily live would grind to a halt, and we'd be forced to actually talk to each other, instead of using email and text messages.
For the sake of clarity, I will assume you are all American, and are receiving a line voltage of 120VAC, with a frequency of 60 Hertz. Those of you with various forms of two or three phase power might get 208VAC, or 240VAC, and some of you might even get industrial voltages of 277, 480, or 600. And some of you foreign readers get a lower frequency of electrical generation, 50 Hertz (Hz).
It was discovered early on that direct current, or DC was not capable of traveling over long stretches of wire. There was this massive feud between Edison and Westinghouse/Tesla over which was the preferred source, with Westinghouse and Tesla's alternating current or AC winning out.
The main advantage of AC power, is that it is easily converted from one voltage to another. In essence, the way you generate an AC voltage is to induce a fluctuating magnetic field in the presence of a coil of wires. This can either be done by moving the wires, moving the magnets, or generating a magnetic field that varies with time inside of a coil.
This brings us to the most important electrical component of all power distribution systems (other than wires!) and that is the transformer. What a transformer does, is it takes an electrical charge at one voltage, and transforms it into a charge at another voltage. In simple terms, a transformer is two loops of wire wrapped around an piece of iron that has the ability to conduct a magnetic field, transferring the magnetic energy from one loop, called the primary to the other loop, called the secondary. The simple rule of thumb is, you take the number of loops of wire in the primary, divide that by the number of loops in the secondary, and that turns ratio gives you the transfer capability of the transformer from one voltage to the other. So if you have 1000 turns in the primary, and 100 turns in the secondary, the voltage in the secondary will be 1/10th the voltage in the primary.
Now for some technical mumbo-jumbo. What is a "Watt"?
A Watt is the amount of power generated by a current of one Ampere, running through a resistance of one Ohm, which generates a voltage of one volt.
Power = Current * Voltage
Current = Voltage / Resistance ... Voltage = Current * Resistance so
Power = Current^2 * Resistance
It is very important that I now give you the real reason we use AC instead of DC: All wires with the exception of superconductors (more on this perhaps in part 4) generate heat, and waste energy. For argument's sake, let us say that the total resistance of the wires that go from the power station to your home is 1 ohm. If we can reduce the current going from the power station to your home, we reduce the waste heat by a factor that is the square of the reduced current! Now there are two ways to reduce the waste, one is by building thicker and thicker wires, which clearly is very expensive. The other way, is to generate a much higher voltage. So if your home needs 6000 Watts, we can either get it to you by sending a 120 Volt potential to you, at a current of 50 Amps, or we can use a big transformer on the pole next to your house to send you a voltage potential of 12,500 Volts, at slightly less than 1/2 Amp. Using that theoretical 1 ohms resistance (makes the numbers come out easier) this means that the waste heat is only 1/2 * 1/2 * 1, or a quarter of a Watt, versus 50 * 50 * 1, or 2500 watts! Using high voltage transformers and power transmission reduces the lost energy in this case by a factor of 10,000.
You can't do this with DC! In fact, the most efficient way to convert one DC voltage to another is to "chop" the DC into an on and off stream of voltage (AC), run that through a transformer, and turn that new AC back to DC. There are other ways, but they always have significant loses and when you want to have high efficiency, you don't want to have high losses. Now it is easy to turn AC to DC, all you need to do is rectify the signal using a rectifier, which is a series of diodes. Diodes are one-way switches which only permit the flow of electricity in one direction. Without getting too technical (and if you want, you can always ask questions which I will gladly answer) the "quality" of this DC is dependent upon how throughly you design the rectifier circuit.
We've had diaries recently on the topic of CFL's and LEDs (the "D" stands for diode) and both forms of light differ dramatically from the old standard incandescent light in the way they use that electrical energy to generate light. Incandescents heat up to generate light, so they are perfectly happy to have any form of electricity, either AC or DC. They actually prefer to start with a low voltage, and then stay at a fixed DC but that's not what we have at home. When you turn that switch ON, (my boss always likes us to capitalize ON and OFF, forgive me I'm not shouting) you can never tell what voltage the bulb will initially receive.
A 120 Volt AC line can have any voltage from 120 times the square root of two, at any time in the cycle. The difference in voltage between the neutral leg and the hot or line can be anywhere from about "-"173 volts to +173 volts, and it varies in a perfect sinusoid in exactly 1/60th of a second. That's where the 60 Hz comes from, in each 1/60th of a second that difference from one wire to another goes from 0 to 173, back to 0, then to "-"173, and back to 0 again. Is isn't a "ramp" (that's a triangle wave) and it isn't a near instantaneous jump (that's a "square wave") but it is based on the sine of the phase angle.
(This is where I get worried- if I become too complicated in my descriptions, I may lose my already thin audience! But I want to help educate and "enlighten" people, so I want to be as precise and in depth as possible. Again, if any of you have any questions, please post them and I'll go into greater detail.)
CFLs work by igniting a shock of higher frequency and higher voltage electricity into a thin mixture of gas, and having part of that light converted from ultraviolet light into visible light through phosphorescence. Obviously, this requires a lot of components and this adds to both the cost, and the size of these bulbs. The amount of light they disburse depends upon surface area, which is why the new "ice cream cone" designs have become so popular. They can now be tailored to give off light in many different temperatures (colors) making them far more pleasant than in the past.
For street lighting purposes, sodium or mercury vapor lights are very efficient, even more believe it or not than LEDs. But they are really horrible at "color rendering". I've got a lovely red car that under some street lights turns russet brown. And my green car turns almost blue. These "HID" or high-intensity discharge lights run very hot, and require a great deal of protective shielding to prevent injury to users, or even fire.
LEDs operate by converting DC current into light. They are very efficient, giving off more light per watt than any other source, but to get the power to the device is not as easy, and severe losses can occur.
Light Emitting Diodes turn DC current into light. When first invented, they were limited to red, and then in time green, and amber colored light. Very poor in terms of lumens per Watt, they spent most of their first 35 years as either instrumentation, or a way to display information. As their efficiency improved, and their colors were broadened to blue and then finally white, their ability to produce light that did not change color with respect to intensity made them almost unique in terms of lighting technology. But they still have drawbacks which continue today, in terms of their being a part of mainstream lighting.
Why would you want to use LEDs?
Light output is measured in lumens . The efficiency of a lighting source can be measured as lumens/Watt )"lm/W", and the efficiency of a 100W incandescent bulb is about 17 lm/W. The bigger the bulb, the more efficient, but since most people don't use 1000 W bulbs (I do, but that's for testing switches) they pay a huge penalty. In fact a 40 W bulb is generally about 12.5 lm/W.
Some of the LED's I'm describing claim to have an efficiency of 160lm/W, which would permit you to (in a lossless system) use about 10% as much energy to achieve the same light! Theoretically there can be green-colored lights that would be able to put out about 683 lumens per watt, but that's Star Trek.
1. LEDs are semiconductors. So their "Current vs. Voltage" curve isn't simple like a light bulb, or a resistor- in order to get even the most microscopic amount of light out of them, you need to go beyond their initial operating voltage. For most LEDs, this is in the low 2 volt range, but some require as much as 4 volts before they turn ON. And once they are on, it isn't a simple equation, such as twice the voltage giving you twice the current and four times the output. Their current goes up disproportionately with respect to voltage. To put is simply, they operate on a very small range of voltages.
2. Putting multiple LEDs in parallel, that is connecting them together where both leads are tied to each other, causes imbalanced light output. You will never get the same amount out of each, because one will always give off slightly more light than the other. In some cases, as when the batteries are dying on your LED flashlight, one or more "bulbs" might die out on their own, just because the voltage is too low for them "fire".
3. So if you try to use them on household line current, you get a few little problems. First they'll burn to a crisp as the high voltage turns them from semiconductor to smoking heaps of melting plastic, possibly scattered all around the room. Second, they'll only conduct electricity half of the time. (That's not what the "semi" in semiconductor means, but in some ways it is an adequate description. They will only conduct in one direction, as long as their threshold voltage is met.) Now we can put two LEDs in series with each other, so the current flowing through one flows through both, and the voltage needed to make them work is twice as much as the earlier voltage. Put 50 LEDs in series (making sure they are all lined up in the same direction, and that they all work- like the old Xmas lights, one dead one prevents them all from working!) and you would be able to plug it into the wall socket, and see a flickering light.
Now there are ways to make them work. Remember in my earlier diary I discussed how you can turn AC voltages into DC; to recap, this is called rectification. The best way to do it is to take 4 diodes, arrange them in what is called a bridge and that will give you a choppy sine wave, with two positive half cycles, instead of one positive, one negative. Now the LEDs will turn ON 120 times a second, instead of only turning on 60 times a second, for 1/2 the time. But this is still not what you want- we still have to convert the ~173 Volts to a voltage low enough to keep the LEDs from melting. And we'd like some way to turn that sine wave into something easier on the eyes, a flat line.
To the rectifier we add a voltage regulator. This is now a ubiquitous device, inside of almost every device known to mankind. They come in several different electronic shapes and sizes, not to mention physical ones! Obviously, we want it to be small, and we want it to work for a very long time. And we also want it to be cheap- we're all unreasonable people here! I'll skip all of the variations and instead stick to the most popular and smallest, a switching voltage regulator.
This consists of a few small high voltage capacitors, an inductor, and a chip, along with a smattering of resistors. Capacitors store electrical charge, which is a good thing it you want to turn your sinusoid into a straight line. The inductor stores magnetic energy, and it doesn't like to have any changes in current. So we turn the current which passes through that inductor on and off a few tens of thousands of times a second, and have the resulting current try to keep the voltage on that capacitor the same, no matter how much we draw out of it.
I can see this being very confusing, so I'll try to make an analogy. The chip and the inductor act like a spigot and a hose. The capacitor is like a bucket, with a hole in it. Our load which in this case is the LED lamp, takes the power coming out of the hole in the bucket. The chip and the inductor try to make sure that the level of the bucket is the same at all times, to keep the flow going through the LED to be roughly the same.
The bigger the LED, the bigger the flow. Now if we want to build one big enough to replace a 50 watt bulb, you will need about 6 or 7 watts of LED light, which requires a power supply that can at least handle that much power. As of now, that's going to be bigger than 4 or 5 quarters stacked together. And you will need heat sinks to draw the heat out of the regulator (not to mention the LED, they aren't 100% efficient, and trying to generate that much light will also generate a lot of waste heat.)
Now very recently (this is where the new stuff begins!) companies have started to work with very low-cost and efficient tiny chip-sized AC-DC converter systems. They are permitting you to plug a small LED lamp directly into your household socket! How big are these bulbs? They are smaller, believe it or not, than the smallest CFL bulbs sold! But they still cost an arm and a leg, along with several teeth, a hip joint and a large section of skin. But in time, they will clearly drop to a point where they will be the dominant light source. I've seen LED arrays that can put out as much light as a 1000W tungsten bulb, but they cost as much as a cheap used car. Last time I'd checked, these light generators were in the $2K range.
But standard street-lighting systems, the "cobra heads" that you see wherever you drive are now going for the $500 range. That's about double the cost of sodium, but the real kicker is they are so much smaller, so much lighter, and can be installed into the architecture in ways that are simply amazing. Since they can weigh so much less, they don't need to be ungainly. And because they are so light, they can be cantilevered further out into the roadbed, permitting lamps to shine directly down on both sides of the street at the same time. We still can not get more light for the same amount of energy
Well that's about it for lighting, except for magnetic induction that isn't yet huge, but might be one of these days.