New Posts  All Forums:Forum Nav:

Incandescent bulb ban

post #1 of 66
Thread Starter 
I'm just reading this in the news today:
Ontario to ban incandescent bulbs, in favor of energy-efficient CFL ones.
Australia and at least one state (I think it was California or Washington) have similar legislation in the works or passed already.

This is a huge mistake for multiple reasons, such that it should be up to the market not forcing it down people's throats, but I'll concentrate on the technical reasons. Rod Elliott covers reliability and safety concerns, and some efficiency myths about CFLs, here:

My biggest concern is with quality of the light, which I think fits here nicely.

The human eye has three types of color sensitive cone cells, covering three different frequency ranges, with response peaks near red, green, and blue parts of the spectrum. Here's what it looks like:

(the black line is the response of rod cells which are not associated with color perception, but only brightness)

Now, a display device which produces its own light, such as a computer monitor or a TV, obviously needs only three color types to produce any combination of visible colors (technically, it's a bit more restricted than that due to the specific emission frequencies of the phosphors used in display devices).

It would seem to follow that a light source with three or more narrow spectral emission regions, such as a fluorescent light (they use phosphor coatings on the glass tube to produce light from the UV the plasma inside generates), that was properly white-balanced, would be all that's needed. Unfortunately, that is completely false. I'll try to explain why that is, now, and why only an incandescent bulb can fully be made to mimic visually sunlight.

The problem is because the light you see is not what comes directly from the light source, but what has been reflected from the object. That is, the light you see is the product of the light source spectrum with the spectral reflectance of the object you're looking at. Finally, your eyes respond not to narrow spectral locations for RGB, but with the complex curves shown above. What this means is the following: take two lights that have been white balanced to appear white (excite your R, G, and B cones the same amount), but have different spectra.

Now, while some objects may appear to have the same color under one light (while having different reflectance spectra), they can have different colors under another light. This is known as illuminant metameric failure. That is, the fact that the light source is white balanced doesn't guarantee that object colors will appear the same as under a different white balanced light. Indeed, in general a CFL or other fluorescent light cannot be made to have illuminated objects appear the same color as when lit with sunlight, whereas an incandescent bulb can.

The remaining thing left to explain is why an incandescent bulb can be made to perfectly match sunlight. The reason is that, unlike a fluorescent one which has narrow emission peaks, an incandescent bulb is, like the sun, a blackbody radiator, producing a smooth and continuous spectrum. By running an incandescent filament at the right temperature, the spectrum can then be adjusted to match sunlight by using a filter on the reflector behind the bulb. There are a few commercial light bulbs that implement this. CFL and other fluorescent, even so called 'daylight' ones, simply cannot match the solar spectrum and thus a number of objects will necessarily appear the wrong color.

The following shows typical spectra between 'daylight' CFL and incandescent bulbs:
post #2 of 66
They souldn't be banned, but I can live with a little less perfect color if it means I'll save a lot of energy.
post #3 of 66
Thread Starter 
The point is, it should be up to the consumer, not forced down his throat by the government.

Also, it's not about "a little less perfect color". The difference can be enormous. You live in the same city as I do, so you know there's very little sunlight in winter here. A daylight incandescent actually makes a great substitute for it. It's also very easy on the eyes when one is reading, DIYing, etc., so it's not just something that matters for photographers and artists.
post #4 of 66
Hmm, I don't think to many people care if their lights are perfectly white balanced. I use those fluorescent bulbs and they work fine. The light has a slight yellowish tint to it, but other than that, it is just fine.
post #5 of 66
Originally Posted by 003 View Post
Hmm, I don't think to many people care if their lights are perfectly white balanced. I use those fluorescent bulbs and they work fine. The light has a slight yellowish tint to it, but other than that, it is just fine.
what about when you set them to dimmer? oops you cant.

how does the municipality deal with the disposal of these bulbs when "spent"?

what to do about overheating issues?

the statement that not many people care is like the statement that not many people care about good sound. while this is true, once you notice the differences it is hard to go back... the differences in color between a fluorescent and incandescent lit office when coming in on a nice spring day (to name one) are ASTOUNDING.
post #6 of 66
Thread Starter 
I use Solux myself, the 50 W version. They're 12 volt so use a switching supply. I can't be sure it's related, but I felt much better this last Vancouver winter when I had this daylight incandescent than previous winters, so it may be the light. Object colors appear almost exactly the same way as on a sunny summer day.

If BC follows Ontario with its own incandescent ban, I may just have to get into a black market selling incandsecents
post #7 of 66
I actually like fluorescent lights better than sunlight and incandescent light... It is just softer and easier on my eyes. Your right though, you can't dim them, I forgot about that. I can see where that could be annoying.
post #8 of 66
As long as they don't ban vacuum tubes....
post #9 of 66
Thread Starter 
As I said already, 003, there are far more problems than the visual appearance. There have been cases where CFL bulbs have started fires, for example. Their efficiency isn't nearly as much as it seems when one factors in energy cost of production for the three-times heavier CFL (not to mention lead content). These issues and others are discussed in the Elliott article I mentioned in the first post. (Though I'm not surprised most people don't read posts before replying to them)
post #10 of 66
Another concern might be what kind of electronic ballast that the CFL uses.

If they're poorly designed, they wind up having the same sort of power-factor as a rectified power supply. That is, they have a "unity power factor", which is normally a good thing and means that there is no lag between the Voltage and Current waveforms. HOWEVER, the current waveform is extremely distorted. There is a sharp rise in current consumption which occurs around the peak of the voltage input. This makes it very difficult for the electric grid, since every half cycle there is a massive spike in current.

Example of what the current input looks like

Compared to the voltage input...

Obviously the peaks should overlap, but I couldn't find a webpage which showed both waveforms.

Originally Posted by nikongod View Post
what about when you set them to dimmer? oops you cant.

how does the municipality deal with the disposal of these bulbs when "spent"?

what to do about overheating issues?
Actually, you can dim fluorescent lightbulbs. I didn't find any good webpages about it after looking for a few seconds, but i've personally done it in a power electronics lab.

The disposal of these lightbulbs is definitely important, but so is the disposal of computers, cellphones, and all manner of gadgetry we take for granted. I don't have much to say about this aside from "Yes, it sucks... and I hope we figure out how to deal with it soon."

As to the overheating, I really don't know. These kinds of lights certainly won't work in certain situations (tightly enclosed areas).
post #11 of 66
Originally Posted by 003 View Post
I actually like fluorescent lights better than sunlight and incandescent light... It is just softer and easier on my eyes. Your right though, you can't dim them, I forgot about that. I can see where that could be annoying.
Well when I want to paint, I would never use just a fluorescent light Also, incandescent light bulbs are not inherently white either (as they are not changing the lighting element, but are tinting the light that is being output). The daylight or white ones get closer, but they still don't quite get there. Some really expensive artificial light fixtures for artists have a combination of incandescent and florescent. CIE's definition of white daylight is 6,500 K: LEDs are actually the most readily available at this color range

personally I do like living with "daylight" or "white" tinted incandescent, and don't like the cast of fluorescent (at least the ones I've seen). It would just make me barf if I had to do any artwork under that icky green cast But the idea that incandescent is more white isn't right either.
post #12 of 66
CFLs are getting to be quite good. High quality ones have nearly as good quality light as incandescents IMO. It's also much whiter.

I agree with this though. The market is not going to self regulate in this case. First of all, the effects are far too removed from the cause for a user to even notice. Individually, lighting your apartment with incandescents might cost you $5/month more than using CFLs. This is not something anyone is going to notice, and nor are they going to be motivated by this to use more efficient technology. I doubt most people even think about this. A light bulb dies and you go get a new one that's the same from the hardware store. That's how it's worked for 100 years, it's hard to break that habit, especially when the initial outlay is much more and the marginal benefits won't be realized for some time. Are you advocating that the utility company increase rates several fold to encourage people to reduce their usage so they don't have to bring more capacity online?

The government is within their bounds here, in my opinion. Bringing more capacity online to fill the needs of the consumers is detrimental to the entire populace in a real and dramatic way. You probably wouldn't balk at the government deciding to deny a request by the utility to dam a river and ruin an ecosystem in a park you regularly hike in, nor would you likely complain if they denied a request to build a multi-GW nuclear power plant near your home. Okay, so we'll build it out in the middle of nowhere instead - but massive power lines are far more inefficient than incandescent lamps are, and aren't much less unsightly than nuclear plants.

The reality is that our usage of energy is increasing at a pretty crazy rate. We either need to slow down, or will need to bring large amounts of new capacity online in the next decade. They're anticipating this and are trying to stave it off for a myriad of rather obvious reasons. I don't see the problem. I'm sorry, but you cannot destroy my favourite parks and devalue thousands of people's life investments because you want to waste electricity so that you can have ideal black body light in your home.
post #13 of 66
Alright, I took some time to think about this and read the posted references. Most of my last post was a kneejerk reaction to the idea that the government shouldn't be regulating the energy industry. I don't necessarily think that an outright ban on incandescents is appropriate (and it's definitely not totally feasible), but I do believe that there should be some artificial incentive added to bring market forces to bear on this. It's a major issue, and it's not going anywhere. Much the same as gas for your car, the price of energy is ever-climbing and people bitch and moan about it - but don't actually do anything about it. In the same way that in these days of $1.20/L gasoline I see more brand new SUVs on the road than I did 10 years ago when it was $0.30/L, people will continue with wasteful practices like using incandescent bulbs everywhere unless some force convinces them to do otherwise. Apparently the market itself is not strong enough to do this.

Now, on to the technical merits of this proposed ban. There are definitely some concerns that need to be carefully weighed. Rather obvious is the initial investment and waste to bring everyone up to speed. I don't think this is as drastic as the linked article makes it out to be, but it is significant. There will be a lot of waste here, but I'm not convinced that everything will end up in landfill immediately. People will inevitably stock up on incadescent bulbs if such a ban were announced, and there would be vasts stocks in warehouses everywhere that would be exhausted first. I doubt a ban like this would be placed on the *use* of said products, but instead on their sale, which would allow this. With a stock of spare bulbs to last a while, this waste and cost could be spread over a significant time, perhaps to the point where it doesn't make much of a dent in the infrastructure at all.

Further, much of the apparently necessary waste in requiring the replacement of fixtures might be mitigated by new technology. LED-based lights don't have these heat issues, and contrary to the article I don't see why it's necessarily impossible to run such a lamp at the temperatures quoted. 58C seemed to be the temperature that his test rig stabilized at, and this is well within the SOA of most electronic components. Adjusting the circuit design a bit could improve this by potentially getting rid of the electrolytic cap, and that's before any real technological change. Not to mention that this heat is primarily waste heat generated in the power supply. Using appropriate components it should be possible to make such a supply extremely efficient and keep the heat generation almost entirely in the lamp itself, which doesn't seem to be the case in the devices he tested. Once LED-based lighting becomes more available (and this may be the small push it needs to get into economical quantities of production), these issues will be all-but eliminated - aside from use in very high temperature environments. Finally, I doubt such a ban would truly be a 'blanket' ban on the technology, as he points out there are a number of applications where incandescents are ideal or absolutely required and I would think that a sane legislative body would provide for these cases.

I also take a bit of issue with the way he describes the power factor issue. I'm not 100% sure my physics is correct, but he speaks in absolutes as if VA is the only important parameter for the effects on the power distribution system. What he fails to mention is that the majority of the components in the system have their maximum loads defined by the maximum temperature they can safely run at. There are of course absolute maximums of current, but for the most part current limits are defined because the inherent resistance of a part of the system will cause it to get too hot with a given current. What he really glosses over is that while the instantaneous current has large peaks, the average current directly corresponds with the wattage. Since we're talking about physical systems here with significant thermal mass, they are not going to instantaneously get hot enough to fail because they are subject to a high current for a few milliseconds each cycle. Instead, the thermal mass of the components will easily average this out, and the output heat will be defined by the average current. The critical exception might be in the transformers, where core saturation could be a major problem as their efficiency plummets at excessive load current, causing them to get too hot due to the non linear effect. There may also be mechanical constraints at the generator end due to increased instantaneous torque required during the heavy load portion of the cycle, though this should be trivial to solve with a flywheel or other energy reserve to smooth things out. There are still issues with power factor, but I don't think they're nearly as pronounced; the article implies that capacity will need to be added such that available watts = consumed VA. I don't believe this to be the case, though I do expect that some not insignificant amount of loss is caused by the poor power factor. This is easily made up for with the increased efficiency. Despite the poor power factor, even the worst-case losses he cites are still a factor of two better than an incandescent.

It would be nice if they actually did some relevant case studies...
post #14 of 66
Thread Starter 
Originally Posted by Davesrose View Post
The daylight or white ones get closer, but they still don't quite get there. Some really expensive artificial light fixtures for artists have a combination of incandescent and florescent. CIE's definition of white daylight is 6,500 K
The 6500 K figure is an average. Many different spectra can have that average, such as an LED or monitor that emit at a single wavelength.

However, it turns out that the eye's perception of color temperature is dependent on light levels--it is nonlinear. So if you have a 6500 K light indoors, it will appear way too blue. 6500 K appears daylight color only if it's at daylight brightness.

How much it should be shifted can be calculated mathematically. In my use, I find that 4700 K is optimal. The following shows adjusted daylight spectrum at 4700 K, and the 4700 K light bulb I use, showing it's a pretty close match (as well as some other spectra):
post #15 of 66
Thread Starter 
Originally Posted by error401 View Post
LED-based lighting
LED's are even more narrowband than fluorescents. They are excellent for display, and horrible for lighting! Please refer to my explanation in the first post of why you need an approximately solar spectrum to have proper color rendition.
New Posts  All Forums:Forum Nav:
  Return Home