This post is a mix of blatant commercialism and a request for advice. The two links are from an affiliate program (i.e. blatant commercialism). The advice request is: Do any of you readers any experience with LED lighting? I’d like to replace light fixtures in our guest bedroom, and outside. We have a very nice patio we use all summer, and I’d like some more energy efficient lighting. Also, when I visited the sites below I saw bulbs for the chandeliers lamps; those could work in the dining room.
I know these use much less energy, but how well do these hold up? What should I look for? Any tips welcome.
I have.. White LED’s use InGaN chips tuned to a blue spectrum with a phosphor to convert to white.
In 120 volt fixtures there are options, the problem is heat. Most of the stuff on the market uses Korean LED’s and don’t sufficiently remove the heat from the devices shortening the life and causing the color to shift to the blue as the phosphor looses properties.
Also, most of the devices run about 50 lm/watt which isn’t that great. There are a few options around that use 100+ lm/watt Cree, Osram, Lumiled or Nichia LED’s and are decent but more expensive and you really have to work to find them. These companies will usually have a warm white option which drops the output down to around 80 but actually need very special heat sinks to operate correctly.
Well, I certainly like the idea of LED lighting. I’ve had some good experience with the little bulbs in various funny projects (Pop Sci had a tutorial on how to make an ordinary flashlight into an LED one, and I made it ridiculously bright and it still works years later.)
But as far as indoor functional lighting goes, I have no experience (and none with Alternating current anyway.)
But LEDs definitely rule! 🙂
JeffId–
I definitely don’t want a “blue” tinge indoors. That’s for letting me know that.
I’ve found quite a few online sources of bulbs. So, I suspect I would get replacement bulbs that way. (Adding “shop for bulbs” to a list of periodicly repeating chores isn’t toooo bad, but I have to admit that I tend to try to find solutions that don’t involve adding yet another “shopping” chore. I hate shopping!)
Janet and I live on a boat. Although we have an inverter for 120vac, only things which are either too pricey in 12volt DC, or unavailable in DC run on the inverter.
Last year we replaced the 12 volt 15 watt bayonet reading light bulbs with led clusters that had the same base. light is whiter. We like them. I’m working my way through the boat making similar replacements. Lighting isn’t the big amp draw on the boat. Fridge, TV, and big computer are. I changed power supply on big computer to DSATX DC-DC type and cut battery consumption in half. but that’s not what you asked.
John
Lucia,
I have worked in my business with near-UV LEDs (what the white LED’s are based on) for a long time. Their life expectancy really can vary a lot, depending on the quality of the lamp, how hard it is pushed to increase light output (higher current = hotter running = shorter life) and how well it is heat sinked. Earlier LED’s of this type often lasted only a couple of thousand hours, but I guess good quality recent versions last in excess of 20,000 hours. You must expect that light output will gradually fall during the lamp’s life; the projected life is usually based on the time to a 50% drop in light output (though this 50% specification may not be provided… beware if it is not).
As JeffId notes, the lm per watt is often not that good. This is in part due to the losses associated with the AC/DC conversion that is needed for a screw-in replacement for a 120 volt bulb.
I have looked several times at LED’s for replacing the 12.5 volt DC halogen house lamps on my boat, but could never come close to justifying the very high price for the LED’s. Easier for us to just run the generator for a couple of hours while preparing dinner, eating, and cleaning up the kitchen area; this charges the batteries with plenty to spare for the night.
I’m also not wild about the spectral properties of the LED’s, although the warmer color versions are better (just like warm-color fluorescent lamps). Warm lamps are less efficient than blue-tinted ones.
If I were you, I would just buy some warm-color fluorescent replacements. They use much less power than standard bulbs (almost as good as LED’s) last a very long time, and cost MUCH less than LED’s. The only real disadvantage is that it takes them a couple of minutes from turn-on to reach full brightness.
SteveF-
Thanks for the informatin.
I have warm flourescents and ordinary flourescents in lots of spots. But I’m trying to explore options.
As SteveF said, just go with CFLs. Its a good example of the 80/20 rule in action; why pay $20 for a 60w –> 5w switch from incandescent to LED when you can pay $2.50 for 60w –> 13w incandescent to CFL swap?
Lucia: I had all LEDs and CFLs in my house. LEDs were blueish in colour, but I could use them outdoors. CFLs don’t play well with -40 outside, so that’s where the LEDs went. I tried low voltage incandescent bulbs first, but they never made it through one season. The LEDs stood up up much better.
All my LEDs were low voltage, which was also advantageous outside, and only used about 10 watts in total. I could also string a lot of LEDs on one line, on a photo-sensitive switch.
I agree with Zeke. CFLs are cheaper, at least inside, and still save significant power. Low voltage LEDs worked better for me, in the cold and wet.
I have never had either a LED or CFL burn out yet. In that regard, they are much better than conventional.
Its like most of this type of technology, though. Payback is a while in coming, if ever, and you can use both types of lighting only if you can afford the initial outlay.
The CFLs are kinda worrisome, in that they are as fragile as ordinary light bulbs, but they use mercury. Read the safety info on the box.
Zeke-because the “true cost” is born by the hobos who croak from the mercury in the dumpster.
More seriously-because LEDs are the way of the future.
Andrew FL/Les,
Actually, in virtually every state there is -less- mercury released to the environment via CFLs than incandescents given that the single largest anthropogenic environmental source of mercury is coal-based generation. That said, there is a difference between mercury released via a smokestack and in your home. If a CFL does break (and its happened to me a time or two), make sure to vent the room for awhile and mop up the mess. Still, the exposure level is minimal (the vast majority of active mercury remains bound to the glass), and as long as you don’t lick the mess you will still ingest less mercury from broken CFL fumes than you would get from a sushi dinner.
If you don’t believe me on the power plant vs. bulb claim, here is my math:
To determine the net mercury emissions for each zipcode, we using the following calculations:
Equation: CFLm * R% – (INCw – CFLw) * CFLl * HGef(zipcode)
Where:
• CFLm is the average mercury per bulb, based on data from EnergyStar
• R% is the average percent of mercury released to the environment from CFLs
• INCw is the wattage of the incandescent light being replaced
• CFLw is the wattage of the replacement CFL
• CFLl is the expected lifetime of the bulb in hours
• HGef is the final mercury emission factor [http://www.epa.gov/cleanenergy/energy-resources/egrid/index.html] per watt-hour of electricity generation, which differs by NERC sub-region generation grids and state-level transmission losses. It is calculated for every zipcode in the country
Specifically, we assume that:
• CFLm is 4 milligrams mercury per bulb based on data from the EPA [http://www.energystar.gov/ia/partners/promotions/change_light/downloads/Fact_Sheet_Mercury.pdf]
• R% is 14 percent of mercury contained in the bulbs escaping, as a conservative assumption based on data from the EPA [http://www.energystar.gov/ia/partners/promotions/change_light/downloads/Fact_Sheet_Mercury.pdf]
• INCw is an 100 watt lightbulb
• CFLw is a 23 watt lightbulb
• CFLl is 8000 hours of use
SteveF
It seems like you’ve done a lot with LED’s. The brands I listed are in excess of 50K hours with 70% drop from initial intensity. Also, in the past two years the color quality has exceeded that of fluorescent lamps for decent brands. While I also prefer the warm white appearance for interior use, the color rendering index isn’t any better than cool white in quality LED’s.
You’re completely right about power supply losses. My point was that most LED’s for indoor use are by Chinese companies who use the Sol Korea LED’s at 50lm/watt not including the supply. Dollars per lumen fluorescent cannot be beat but if you’re paying the local union to change your bulbs you should find a quality LED lamp.
In the next year or two the difference in efficiency of LED’s and fluorescent will become substantially wider.
Lucia:
I greatly appreciate your excellent site and the efforts of the site operator.
I have been sight challenged for most of my life and as a side effect of that, I have issues with different lighting scenarios. What I can clearly see in sunlight for example, may be very difficult for me to see under, say fluorescents. I too have been waiting for the improved LED devices for years just so I can eliminate fluorescent lights. That personal goop aside; the advice I offer is make the best use of your money and labor that you can.
LEDs are much like CFL lights were when they were first available, (everyone bought CFLs when they were $17-$20 an 8W bulb right?) that is, they are pricey for less than stellar results.
The good news is, LEDs last much longer than CFLs and use less juice, though the light is not as great right now. The better news is that with increasing demand, better technology will arrive, maybe not next year, but soon enough. personally, I’ve been waiting for “better LEDs next year” for almost ten years now.
So where’s the advice? It’s a little buried; buy LED lights for those fixtures where you burn bulbs near continuously especially if they are a pain to replace. Use CFLs and plain fluorescents for general lighting purposes and throw in incadescents where you need to augment the CFLs and FLs (like I do in the kitchen around the cutting board). To offset the cost burden, purchase the bulbs in small amounts infrequently. The bulbs last so long that even if purchased annually you’ll slowly convert over. Purchased quarterly and the conversion will be surprisingly quick.
A great site for general LED info is http://ledmuseum.home.att.net/.
The LED museum site has detailed information on many LED products; that info can be hard to find though.
The links (LEDLighting.com) you provided are interesting, if you get a terrific discount, maybe even good. But as an example; I found this bulb @ $31 each: http://ledlighting1.com/Silver-3w-JDR-1LED-Medium-Base-White-Bulb-6881-prod.htm?category_id=2911.
The LED museum describes this bulb @: http://ledmuseum.candlepower.us/eleventh/jdrbulb.htm. (Please note the spectrographic analysis provided which will indicate color influence of the light)
Following the link provided by the LED museum the same bulb can be found for (if I got this right) $22 and with their 2for1 sale gets you 2 bulbs for $22.
TedK
Jeff Id,
Well, I sure wouldn’t define 70% loss of intensity as a practical end of life level for room lighting. Heck, I think the commonly used 50% loss is already pushing the practical limit.
In a tightly controlled feed-back system, where the absolute light output is held to better than 0.01% by increasing current to compensate for lost intensity, the deep violet/near UV LED’s (core of all the white LED’s) give up the ghost pretty quickly (much less than 20,000 hours) unless you set the initial intensity way below what you get from running near the nominal current rating of the LED. What the LED manufacturers do not tell you (at least I have not heard it) is that the life-time efficiency of the white LED’s is much lower than the initial lumen values suggest. The power draw stays the same, but the light output falls continuously, following what appears to be very close to an exponential decay. The efficiency falls most quickly when the LED is new and efficient.
LED’s are no doubt great for some applications, but I don’t think they are a panacea for most space lighting. Their economics are crappy, probably much worse that hybrid autos. Will they get better? Yes, but they are a pretty mature technology, so expect a slow evolution, not ‘quantum’ leaps. More movement will likely come in price than in performance. I think of LED lights much like Freeman Dyson thinks of hybrid cars… toys for the rich, at least for now. 😉
TedK– Thanks for the links! I always end up shopping around for price.
I’m sensing that most people think it might be wise to wait a year? I know choices for CFL’s have increased dramatically over the past few years. Still, there are some lights to be replaced! So, I’ll mull this all over.
Whoops:
I didn’t read the description carefully (my shame). The linked bulb at Donsgreenstore is a two pin GU-5.3 base and not a medium (e27) screw in bulb. I only noticed as I was shutting down, my apologies, the price shopping still awaits.
TedK
Another fun thread Lucia.
Steve F.
I wrote it poorly, I’m sorry. The light intensity drops to 70% of initial after 50K hours. The human eye can barely tell the difference – if at all.
The near UV chips you talk about are the core of Cree manufacturing and a few others – it’s nice because UV chips prevent visible color separation. Quite a few manufacturers like Nichia use blue (which is relatively near UV but still considered blue). The data I’ve seen shows a much longer life for the main chip, the phosphor is the weakness these days.
The power draw stays the same, but the light output falls continuously, following what appears to be very close to an exponential decay.
I wonder if you’re keeping the die temp in the proper range. Overheating will cause migration at the junction and your chip will loose performance. This is the main problem with most LED replacement lamps – especially those that insert into old style incandescent sockets. Test data of a properly temperature maintained LED is quite flat for output over long periods of time.
I agree that in most area lighting applications it’s early for LED, there are some areas where it’s cost effective though.
I have just bought a couple of LED bulbs to replace CFLs. In the UK there’s not a lot of choice if you want bayonet cap fittings, though there is quite a range for replacing halogen mini spot lights.
Do make sure you buy warm colour versions, and even those seem bluer than most CFLs. As people have said the power saving does not yet justify the cost (in the UK CFLs are effectively free due to some bizarre government schemes). The only obvious advantage is that they turn on instantly to full brightness.
At the moment, they are more suitable for the smug feeling of being an early adopter than any serious use.
As usual, I am amused to note that this nest of lukewarmers and other fossil-fuel-funded climate vipers is packed with experts on energy saving technologies. Having just spent a small fortune on reducing my house’s carbon footprint by about 35% I am more than happy to look most warmists in the eye!
For any UK readers interested in direct replacement LEDs, the best site I have found so far is http://www.electricity-monitor.com/ which sells a reasonable range of bulbs in addition to their wide range of wireless energy monitors.
HI,
Only minimal experince and only outdoors in the form of solar powered garden lights. Ours are cheap and cheerful, very, very blue, and seem to work as a deer repelent. Honest that is what they are used for. Site them near anything you don’t want to be eaten.
Alex
Jonathan–
It seems like loads of lukewarmers are engineers. What engineer hasn’t wanted to get more bang for the buck out of energy even since the time of Watt? (My understanding is that Watt’s advisor, Black was funded by Whiskey distillers because they wanted to distill liquor using less wood. Anyone who makes and sells anything wants to do it cheaper, and energy costs are a big factor!)
Alex–I don’t have a deer problem. But my friend in Glencoe does. I’ll tell her!
Zeke-The EPA says you need to do much more than that.
Andrew_FL,
The EPA is a bit overly cautious on that front. http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2535642/ summarizes some lab research on the subject, and confirms that sweeping up and mopping the broken bulb bits and ventilating the room reduces mercury exposure well below the ambient occupational safety hazard levels as well as below the recommended limit for children. If you leave the broken bits on the ground in a poorly ventilated room, it might be more problematic for children or women who may become pregnant, but so is eating tuna more than once a month. :-p
Lucia,
I’d suggest going with CFLs for the interim even if you want to swap to LEDs in a year or two. You’d be surprised how much you will save. For example, if you swap a single 100w bulb that you use 4 hours a day for a 23w cfl you will save about $13 per year in Chicago (given electricity priced at 11.41 cents per kWh), well over the initial $2.50 or so cost. In fact, as long as you will have the CFL for over 2.3 months before buying an LED, its still a good investment.
Does the savings calculation take into account the heating budget? The IR emission is an additional cost if air conditioning has to cool the room, but a bonus if you need heating. Nights are longer in winter…
I suppose someone in a room with a strong IR source would on average prefer (and set the thermostate to) a lower air temperature than without the IR. I don’t know the actual numbers though.
Lucia,
I can guarrantee nothing deerwise, adjacent to the garden is a hunting reserve so perhaps they are frightened of their own shadow. Put it this way it does not take much to scare them.
Alex
I usually illuminate the back porch just with the high beams from an idling 1968 Ford F250. Seems to repel deer as well.
Low energy light bulbs are dubious in energy saving. For outdoor lighting they definitely provide an energy saving. In a air conditioned home where bulb heat must be pulled, they save energy.
Outside of southern California, however, most homes are heated for a large portion of the year. This makes all light bulbs 100% efficient (light bulb heat losses are input to home). If one cuts 1Kw from the lighting load, the furnace must fire 1 Kw harder. We have merely shunted electrical heating to some other (gas, electric, wood, oil).
If you are heating your home, changing light bulbs save negligible energy. It is mainly another way to make profit on GW hysteria and allow for the warm fuzzy feeling of doing something to save the planet.
50 to 100 K hours? That’s nothing. This bulb has been on for 108 years! http://www.centennialbulb.org/
Zeke: I totally agree with you on coal emissions; both mercury and radioactives. One plant in Canada, Nanticoke, is the single largest emitter of both mercury and radioactives in Canada, and probably North America.
I was more concerned with the micro-environment; my house, and to larger extent in landfills, where they will accumulate. Because very few people will dispose of CFLs properly.
I agree the risk is minimal, but extra care needs to be exercised, especially if there are pets, children, or pregnant women in the house. And they need to be disposed of correctly.
I bet the deer rifle in the front seat is what scares them off….
Now that governments have gotten into the act and made illegal the sale of products using certain forms of inexpensive technology (incandescent bulbs), it would of course follow that someone would find a way of making that technology far more efficient. Take a look at what University of Rochester researchers have been toying with:
http://www.rochester.edu/news/show.php?id=3385
The fact they they can do this ‘through the glass’ of an existing bulb would indicate to me that as good or better performance will be readily achieved when this technique is incorporated into the filament’s manufacturing process.
Mike
G. Karst repeats the common sophism that the heat loss from switching to Incandescents to CFLs results in negligible overall energy savings due to higher heating loads. This is actually a rather fun problem to model, so if you guys would forgive a bit of a lengthy digression on my part:
There is indeed a non-trivial heat loss in switching from Incandescents to CFLs. The actual benefit or cost of this heat loss to any given household will depend on a number of factors, including their heating requirements, their cooling requirements, their heating fuel, the efficiency of their furnace/boiler, the efficiency of their AC unit, and the times at which they use the light. For the sake of proving my point that the overall effect is rather negligible, lets use the Chicago example above, as its a case where there is considerable annual heating energy required and not that much cooling energy.
First, lets list the necessary assumptions:
A 0.8 AFUE gas furnace (the most common space heating type both in Chicago and the country as a whole)
Uniform light usage of 4 hours for each day of the year
Heating usage for 8 months during all lighting hours
Cooling usage for 3 months during all lighting hours
All light heat as “useful” as furnace heat
Transitioning from a 100w incandescent –> 23w CFL
0.735 kg per kWh electricity for the Chicago area via EPA’s eGRID
11.41 cents per kWh via the EIA
6.469 kg CO2 per therm nat gas via the EPA
$1.15 per therm nat gas via the EIA
100 w incandescent luminous efficiency = 2.56%
23 w CFl luminous efficiency = 11.7%
10 SEER AC –> 10.55 mj / kwh cooling provided
Now, luminous efficiency isn’t strictly the percent of energy not lost as heat, but it should if anything provide a conservative overestimate.
Lets start by calculating the strict electricity savings from the light replacement:
CFL electric savings = (100w – 23w) / 1000 w/kw * 4 hours * 365.25 days = 112.5 kWh per bulb
Next lets calculate the heat difference between the incandescent and CFL lights for the 8 heating months:
Heat difference = (100w * (1 – 0.0256) – 23w * (1 – 0.117)) / 1000 w/kw * 4 hours * 365.25 days * 3.6 mj per kWh * (8 months / 12 months) = 270.5 mj less heat per bulb
Assuming an 80% efficient gas furnace (0.8 AFUE), producing this 270.5 mj would require:
Extra heating gas = 270.5 mj / 0.8 AFUE / 105.5 mj/therm = 3.2 therms
Now, we can’t just penalize for extra heating energy needed without rewarding for reduced cooling needed, so lets calculate the cooling energy reduction over the 3 months of cooling (note that heating and cooling months don’t add up to 12, since we are leaving one Goldilocks month where neither heating nor cooling are required).
Cooling difference = (100w * (1 – 0.0256) – 23w * (1 – 0.117)) / 1000 w/kw * 4 hours * 365.25 days * 3.6 mj per kWh * (3 months / 12 months) = 101.4 mj less cooling needed per bulb
Lets convert this into kWh of energy needed for our 10 SEER AC unit:
Extra cooling electricity = 101.4 mj / 10.55 mj/kwh = 9.6 kWh
Now that we have that out of the way, we can compare both the carbon and dollar savings from installing the CFL taking all of the above factors into account.
Lets start by calculating the dollar and carbon savings from the CFL installation, ignoring heating and cooling effects:
Carbon light only = 112.5 kWh * 0.735 kg / kWh = 82.7 kg CO2 per year
Dollars light only = 112.5 kWh * 0.1141 cents / kWh = 12.84 dollars per year
Now, lets penalize for extra heating needs:
Carbon light and heat = 82.7 kg CO2 – 3.2 therms * 6.47 kg CO2 / therm = 62.0 kg CO2 per year
Dollars light and heat = $12.84 – 3.2 therms * $1.15 dollars / therm = $9.15 dollars per year
Finally, lets take these results and add in the modeled cooling-related benefits:
Carbon light, heat, and cooling = 62.0 kg + 9.6 kWh * 0.735 kg / kWh = 69.0 kg CO2 per year
Dollars light, heat, and cooling = $9.15 dollars + 9.6 kWh * 0.1141 cents / kWh = $10.25 dollars per year
So, for somewhere cold like Chicago, there is a 16.5% carbon penalty and a 20.2% dollar savings penalty due to the extra heating requirements of the house with CFLs. For somewhere warm like Florida, California, Texas, Georgia, etc. there would probably be net benefits due to the cooling reduction. Its a fun problem to examine!
Well, no sooner did I make the preceeding post than Rochester’s website started having problems. If you need to, you can view Google’s cached version of the news release here:
http://74.125.95.132/search?q=cache:H-x9rXrLYzUJ:www.rochester.edu/news/show.php%3Fid%3D3385+femtosecond+laser+site:rochester.edu&cd=9&hl=en&ct=clnk&gl=us
Or, you can read a Cnet news article here:
http://news.cnet.com/8301-11128_3-10256065-54.html
For those who have access, Physical Review Letters:
http://scitation.aip.org/getabs/servlet/GetabsServlet?prog=normal&id=PRLTAO000102000023234301000001&idtype=cvips&gifs=yes
I just want to stretch that high voltage LED’s, which may hold about 20 individual LED’s, have a big problem: the individual LED’s are wired in series: when one blows, the whole lamp is dead. You might read here:
http://bmb.lcd.lu/science/230VAC_LED/index.html
Zeke Hausfather (Comment#22231) October 21st, 2009 at 11:53 am
South Chicago is mostly nuclear energy so not much carbon for electric. I’m going to leave my lights on all night this winter to save the planet.
jeff id,
The composition of your electricity is actually a rather tough problem, because it does not always reflect your proximity to specific generation facilities. The finest grain analysis you can reasonably due is at a NERC subregional grid level, where subregional grids are defined by portions of larger grids where the majority of energy produced within the subregion is also consumed within the subregion. Chicago is part of the RFCW subregion, which as the following generation mix:
72.8% coal
2.7% gas
22.3% nuclear
So realistically the only places where the heating effects are really relevant in a carbon analysis are certain Canadian provinces where 95%+ of generation is hydroelectric and heating requirements are quite large. :-p
I’ve been trying LED lighting. Mostly several different kinds through eBay. After finding that light bulbs from China don’t have U.S. customs duty, I’ve tried some from there also
Some small low-illumination yellowish multi-LED spotlights have been suitable for accent lighting. I have a couple of single-LED spots which have a bright white light and are suitable for hallway replacement of 40 watt incandescents. Their big brother, with four LEDs, has been serving me as a white-light reading floodlight in a draftsman’s lamp.
My first LED lamp was a red-with-blue LED growing lamp in a floodlamp-sized mounting. It kept two plants happy last winter (they have single broad leaves on each stalk, so they’re probably used to shady areas). I’m trying more plants this winter, so I just added an inexpensive 17W one-foot-square inch-thick panel studded with white LEDs. I don’t know the type of LEDs, but from the side I can see blue highlights. I know red-with-blue is better for plants, but decided to see if the bluish tinge from these will keep the plants satisfied (and of course there is red somewhere in that white). If the plants don’t like it that will become a bright light elsewhere.
A small spotlight in a hallway might require a suitable diffuser, depending upon the desired effect. There are less directional designs but I haven’t tried them yet.
Lucia:
I recently bought a package of 2 “Lights of America 2004LEDDL-35K-24 LED 4 Watt Standard Base Light Bulb” at Sam’s Club for about $14.00.
I use them in place of 40 watt tungsten lamps recessed in the ceiling of our hallway.
They are claimed to consume 3.5 watts and put out about the equivalent light of a 35 watt tungsten lamp. Their expected life to 50% output is supposedly 20,000 hrs.
So far, they’re fine replacements at a good price.
Len Ornstein
Zeke Hausfather:
You seem to have a valid model. There are parameter value issues, but that is what a model is good for.
Other factors that alters results is the practice of minimizing lights when cooling a home, or homes that only have heating not A/C.
Lucia,
LEDs are clearly all the rage, but I’m not sure they are “the answer”. Not even for me, who lives off the grid. I read Zeke’s calculations, which are interesting, but unless I missed it, I did not see any calculation on the thread of the cost of writing off the CFLs or other lights in when shifting to LEDs. It is important to include that cost to the cost of the new LEDs and run the total against expected savings. It adds up to a lot of kW hours.
Also, no matter what the marketing folks tell us, as a photographer I have to say that LEDs are a very poor source of “artificial” light and in many cameras the automatic white balance function does not know how to deal with colour temp, even of the so called “warm white” variety.
As far as LEDs as a solution to deer predation in the garden, 80 grains of .243 will do the trick much cheaper and it’s permanent too…. It also moves the deer from munching beside your favourite roses [yes, they eat those too] to where they really belong, beside the mashed potatoes. But I digress..
🙂
Zeke Hausfather:
“Uniform light usage of 4 hours for each day of the year”
Is this a realistic assumption? How would your numbers change when this is replaced by, for example, dusk to midnight? I think taking this into account should decrease the undesired effect in the summer while increasing the positive contribution in winter.
“All light heat as “useful†as furnace heat”
That is another interesting point to consider. Most power from a lightbulb is emitted in the IR, which heats directly whatever it is absorbed by. I think thermal comfort can be achieved at lower air temperature when the person is illuminated by IR (compare sunny vs cloudy days at the same air temp). There is also a good correlation between lights and people – the unoccupied rooms usually don’t get IR-heated.
Les Johnson:
“CFLs don’t play well with -40 outside”
Les, it that -40 F or -40C?
😉
Isn’t that a “throwing good money after bad” calculation? Is the money already spent on a specific solution a justification for not using a better solution? If a certain LED meets your need, your choice is only how much longer to spend more on electricity than the new solution. Note that if you have any case where a CFL is best, you can use removed CFLs as replacements for the needed CFL.
The time factor has an upper limit: when the existing light burns out you either replace it or you don’t. Until then, time-dependent factors include how much cash you have, the changing prices of all the lights, and what new lights might appear.
Children: According to the historians, we say a light has “burned out” because light used to come from a device which often caused a flame or a quick flash of light when it failed. Flames are a mythical god of burning, which historians say had large steel temples scattered around the homeworld.
G. Karst,
Thanks. In a home without cooling, my “light and heat” numbers would apply.
Kusigrosz,
When the light is used does matter, though its harder to model since you need to know both the degree days (based on daily averages) and some model of diurnal variability, or good hourly temperature data, to model exactly how many light-heating-hours and light-cooling-hours you would have per year. Generally there would be more light usage in the evening, when heating use is greater, and less during the day when cooling use is greater. Still, the practical effects would be limited to most places to the “edge months” when heating/cooling are only needed for part of the day.
Also, the IR point is interesting. Also relevant might be the position: most bulbs are near the ceiling, while heating vents tend to be near the floor.
I do residential energy use modeling as my day job at the moment, so this gave me a good excuse to update our CFL model to provide an initial approximation of second-order heating and cooling effects.
Lucia, here is an LED with a Cree engine and 1000 lumen output 83 lumens/watt offered in warm white (80 CRI) or cool white(75 CRI).
CRI = Color Rendition Index
http://store.earthled.com/collections/earthled-best-sellers/products/earthled-evolux-s
Consider staying with incandescent bulbs in your main family room. When I get home every evening I turn on the plasma TV and some table lamps in the family room to cheer the place up and to add a couple hundred watts of heating. I live in Minnesota so most of the year the house is being heated and by using the spot heating I can keep the rest of the house cooler. It would be silly to spend a lot on expensive bulbs and then need to use an electric heater. In the summer, its light until nine anyway so the lights are not used as much.
Consider using low brightness amber led lighting for the patio. Amber leds do not need a phosphor so they should be more efficient. Also they will not attract as many bugs. I am not finding an on-line source for 110v bulbs but there must be some out there. Mouser electronics has a pretty good selection of leds if you want to make your own. An old laptop power supply and a load resistor would work well with a number of leds in series to minimize the power lost in the resistor. A power transistor and a potentiometer can make a nice dim-able power supply. The main power savings would come from the ability to much less light distributed much better than a single high power single light source.
About 4 years ago we started a conversion from incandescents to CFLs and this is what we learned:
* CFLs do not work well with motion wall-switches
* CFLs take time to warm up. This varies by model and brand. This can be *very* annoying.
* CFLs vary in color, even within the same bin.
* CFLs last longer – but some still fail.
In the end we had to settle on a mix of CFL, incandescent, and halogen.
Whatever you do, don’t convert all at once.
I use a warm white E12 LED bulb on my side porch. I don’t remember the brand but it is one of the higher efficiency (more expensive) longer lasting. I have it on all night long pulling under 4W. It’s just for ambience and to find my way around at night; doesn’t illuminate the porch for reading or playing cards or anything (would probably need about three of them for that). I also added two 10W strings of LED lights around the inside perimeter of the roof with the cool blue look; sometimes I just use the string lights when I’m out there because they look allot like star or moon light when they’re the only lights on.
I also use >1W cheapo Chineese candelabra lights indoors on my Christmas village and window chrismas candles. Not so much for efficiency reasons but I felt they were less of a fire hazard and less likely for my son to burn himself on and less likely for him to break. I use LED’s exclusively for my outdoor Christmas strings for efficiency reasons.
I’ve never had an LED bulb fail yet out of about 30 low efficiency/quality candelabra bulbs, probably 1000 high efficiency/quality strings bulbs, and 1 high quality candelabra bulb. I’ve had a fair amount of CFL’s break to the point that I am getting agitated. At least CFL’s are getting cheap enough that it’s not so big a deal when they fail, but since they were sold to us as “longer lasting than incandescents” I take it personally…
Marko (Comment#22308) October 22nd, 2009 at 7:36 am
* CFLs take time to warm up. This varies by model and brand. This can be *very* annoying.
I especially agree with this one. I use a CFL for my bathroom ceiling flood and, well sometimes you just want to use the bathroom now and not in some sickly colored horror film lighting. I’d switch to an LED but the darn CFL is lasting too long.
We just completed a major remodel. Along the way, we tried to use “green” lights wherever feasible. And, we literally bought examples, installed them, and tried them. E.g. we have a 14’x33′ trey in our ceiling lighted with 40 4′ fluorescent T8 lamps (in pairs in 20 fixtures with 99 percent dimming ballasts) behind the deep crown molding. I doubt we will recover the cost in our lifetime — but, the great room sure is bright when we want it that way and sure is moody when we want it that way. We got a perfect color temperature as T8 lamps have a lot of choice in that regard. We helped “perfect” the color temperature by the paint color used inside the trey.
For real good decorative “spot hi-lighting” (e.g. granite counter tops) we were still forced into dimming halogen incandescent (though we tried dimming CFL and LED) — because the non-incandescent do not provide a strong enough “spot” effect. Moreover, the LED color temperatures did not work with our colors.
Dimmable CFL is not perfected yet. The dimming is mediocre and the wall switches hum.
Our major finding was that “green” lights do not have the same color temperature as traditional incandescent lamps. With fluorescent, there are many color temperature choices. With LED, there were far fewer — and we found no LED that “worked” with our color pallet.
I presume LED will become a standard over time and color temperature will be optimized for decorating. So far, we didn’t see it in our own case.
Merely our experience FWIW.
100% OT lucia
http://www.pnas.org/content/106/37/15555
New article. Anthony posts on it. There is a bunch of interesting charts. couple points.
1. They dont compare to observations. “observations” are reanalysis data ( output from a model of observations– see the SI)
2. It would be cool if you could get there model run data. If you ask nice and promise them brownies.
3. Substitute real observations. Start a good war.
Lucia and all, I had to skip to the end so I may have missed something important, but a number of postings about “white neon” lighting and light efficiency by John de Armond are on Norman Yarvin’s usenet archive, yarchive.net
Happy reading.
It’s a very illuminating thread. I am very impressed with indoor calculations of inefficiency of “green lightening†for Chicago-Minnesota areas, very interesting! For outdoor (and garage) lights, it looks that more light-efficient bulbs are preferred, especially in Texas.
I’d like to add few own observations.
Consider life expectancy of bulbs and associated cost.
1. The modern spiral-like R40 CFL bulbs are advertised as having 7000-10000 hours. Nothing can be further from the truth, especially when they replace incandescent in their usual places. They burn out like hell. The reason is that each bulb has its own “electronic ballastâ€, which is made of 20-22 electronic components, including 4 diodes and two something like triacs/transistors, and a big 22uF/200V wet electrolytic capacitor. All this stuff gets very, very hot, even if the rated wattage is low. The whole design uses a one-sided printed board with hand soldering, I presume done using some Chinese prison labor or something.
I know that most computer-grade capacitors are rated at about 1000-2000 hours at 105C, the best ones are 5000. So I believe I know why compact CFLs burn out so fast. Because of this, better consider an industrial-grade tube CFLs like the 3-ft F40T12 Philips lamps, which are used with external magnetic ballast. The bulb is rated at 40W having 3600 Lumen output (appox as a 300W regilar bulb), and lifetime is 24,000 hours. This is pretty efficient.
2. Where I am using dimmer-base toggles, incandescent bulbs seems to last forever. Yet they are “rated†as 1000 hrs life. I believe I did not change any of eight bulbs in my bathroom for 15 years, and I am sure we are switching these light on-off at least 10 times every day.
3. Most CFLs and all LEDs I tested are not dimmable. With a dimmable switch, they die immediately. The so-called “dimmable CFLs†are a joke. There is no reason for LED to be non-dimmable, except its electronics. I guess a long progress in development of reliable electronics is needed.
… or you could hold on a little longer and wait for ESLs to hit the market. They have all the benefits of LEDs and CFLs with none of the major problems. See here: http://www.youtube.com/watch?v=13O-uAvC3j8 (Sorry about the small piece of AGW propaganda in the intro. 🙂
Here is something for you: When I was nine years old my mother gave me a German made high intensity reading lamp for my birthday. I still have that lamp 45 years later and have replaced the bulbs about 3 times.
The lamp has a 12v transformer on it, and the high intensity bulbs are the same ones you use in your automobile tail lights with a single filament. I buy replacements in pairs at auto parts stores.
The lamp puts out a very bright white light, is very efficient, long-lived, is inexpensive, and contains no mercury. What more could you want?
So, if you really want to change your lighting, go to low voltage, and save yourself some bucks.
http://www.nytimes.com/2009/09/25/technology/25bulb.html
“Build a Better Bulb for a $10 Million Prize”