Monday, January 16, 2012

Ending incandescent bulb production: Are LED bulbs the answer?

 ∙ ∙ ∙

I got an email today from a non-tech friend with this link in it.
http://www.asiapacificnews.net/story/202667941#.TxO7Bn2Wej0.email

It gave me the opportunity to think about and compose an answer for that audience about my own major doubts about the state of LED bulb production as a successor for the ubiquitous incandescent illumination bulb. U.S. law caused the "phasing out" of all production of standard household incandescent bulbs beginning 1/1/2012.
http://www.usatoday.com/money/industries/manufacturing/story/2011-12-28/light-bulb-rules/52256656/1

Here is my reply to him...and to you:
Thanks for bringing that up. I first have a quote for you:
Quite a bit of the GeoBulb’s light is directional. It’s very bright directly underneath it, but if you move far off to the side, it’s not producing nearly as much incidental light as the incandescent bulb. For many situations, this isn’t a problem at all – recessed lights and lamps won’t skip a beat. However, for other uses (such as a single bulb that lights a small room), you may want to wait for next generation LEDs.


That is exactly the problem with LED bulbs. The physical function of how LEDs GENERATE light is behind the problem. Incandescent bulb filaments squirt out light as if from a point source. (Imagine a point in space that emits the same light as a 100W bulb in all THREE dimensions.) You can see the shape of the light intensity in this diagram. It is the large set of lines...just extend them all the rest of the way around the bulb.



The way an LED works is that the light is instead generated out from the SURFACE of the "semiconductor junction" that creates it. So, in the same mental state, imagine the light of a 100W bulb squirting out instead from the SURFACE of a tabletop straight up from it with some spread, but mainly only UP. Now think about how many tiny LEDs it would take to accomplish sending light in all the directions we use light from a generic home light bulb to illuminate! Being drastically efficient at squirting light out of a tiny hole (too simple, I know) is NOT the same as sending it effortlessly in all directions.



Part two of the problem is that in order to WORK, incandescent bulbs have to incandesce or get hot enough to glow white hot. So heat is part of what makes 'em work and the distribution and dissipation of that has been solved over the last 120+ years. But the heat generated by the LED function is the BANE of their existence. And dissipating it is one of the major problems standing in their way of being successful. The hotter they get, the WORSE they work. That is why the bottom part of most of the LED bulbs is a metal heat radiator. Having that metal all over the base is also one of the things that means they eat up lots of resources (metal production and crafting of the heat sink) to create. See the image below. That may be partially done away with by the top developmental quote where a company has new bulbs with liquid cooling.



In order partially overcome the part one above, the makers put many, many LEDs inside the bulb to attempt to squirt light out in several directions. But it takes a whole lot of "several" to equal "all", so they compromise by only squirting light vaguely less than that above some horizontal plane surface. That is the part of the quote about not having light off to the side of straight out the end of the bulb. This is shown by the 60º beam angle shown below where the light has already dropped off to only half what it is straight out the end of the bulb. This makes the bulbs seem more like spotlights instead of general light bulbs. If you are viewing really wide items, you might have to move the item you're viewing around to stay in the bright part of the light pattern. I do not like this.



Part three of the problem is that there are several common methods out there for powering LED's. Why all the fuss? It boils down to this:
1) LED's are very sensitive to the voltage used to power them (ie, the current changes a lot with a small change in voltage)
2) The required voltage changes a bit when the LED is put in hot or cold air, and also depending on the color of the LED, and manufacturing details.
Incandescent bulbs work over a fairly wide range, but their target brightness and effective color temperature is only achieved in a few volt range of the 120 volts we have from household power. But the above numbered statements say we can't power the LED bulb the same if it is in a 140º attic as we can when it is dangling outside in -10º blizzard. The circuits have to be very careful how they arrange this, which means NO direct application of the household power: there have to be electronic control circuits containing many components built into the base of the bulbs, thus adding complexity and room for failure of one of the tiny parts to ruin your whole $30-40 bulb, as there is no replacing parts inside the bulbs. They are manufactured so the parts are NOT replaceable by the consumer.

See how hard it is to explain the idea of what the makers of the WONDER LED BULBS gloss over in their gushing rush to get ya to buy their bulbs? These makers are trying really hard to get you to buy these darned bulbs at $30-40 a pop, but THIS EE is not havin' any o' dat. They have to get mucho better-o before I bite. And they darned well better get a heap cheaper.

Hope I didn't lose you in the discussion. I tried hard to keep it on a non-EE-level basis. But feel free to ask comment.