The recent discussion about lights has pricked my conscience (again!) about the workshop lights, some of which are on several hours per day. I have 2 pairs of modern 6ft lights and one pair of elderly 5ft lights, switchable in pairs. Am I correct that, for equivalent light output, the LEDs consume about half the power of fluorescents? If that's correct, and assuming that all the lights are on for a total of 4 hours per day on average, it seems that I could save roughly
0.6kWh/day by changing to LEDs. If an LED tube is about £5 and electricity is about 50p/kWh then break-even is in less than 20 days. This is all very rough, but is there a glaring error?
"Fluorescent lights produce between 50 and 100 lumens per watt. By comparison, LEDs can produce roughly 130 lumens per watt."
I would say a more conservative estimate, is LEDs are 60 to 130 lumens per watt.
There are still lots of shit 60-80 lumen per watt LEDs in the channel. Quoting "lab numbers" (130 or 200) is cheating, if they aren't really in production, or, are more expensive. Even if the payback period was longer, maybe the expensive LEDs would be better in the long run. The higher output LEDs have fewer crystalline defects (disorder in crystal). Eventually LEDs will reach a "maximum" output some day, but we may not be there yet.
The LED market, is to some extent siloed. You may not find any Nichia LEDs in the lamps you buy (Japan?). Maybe they're Cree. It's possible the efficiency of each of those companies, is different.
The effectiveness of LEDs, depends a lot on the luminaire design. The square LED panels might be better from an absolute point of view, than sticking a LED tube in a florry fixture. (The LED tube should send out light on 180 degrees of the surface, rather than 360 degrees. The LEDs are not inherently 360 degree devices, although you could put them on either side of a PCB if you wanted.)
I learned about LEDs, when I built my own bicycle light. I started with a single "power" LED. I used a commercial grade reflector (polycarbonate, aluminized) and fittings. The light was... terrible. Next, I built an array illuminator,
48 LEDs fitted with 15 degree lenses on each LED by the manufacturer. That produces great light. Part of the reason that works, is there is less "phosphor smothering" when using large numbers of small LEDs. The operating temperature of the LEDs can also be a bit lower that way. (One of the reasons I tried this, is I ordered a "bag of LEDs" at the same time as I ordered the power LEDs, and the bag of LEDs was just going into one of my parts bins for later. Since the power LED was a dismal flop, that bag of LEDs was just "staring me in the face".)
To some extent, you have to try these things out, and see what you think.
Let's say that LEDs come with the following colour temperatures.
2500 degrees -- reading light for living room (least efficient bulb is the result)
3500 \___ -- slightly bluer light for shop
4500 /
9000 -- extremely blue, suited to bicycle lights (most efficient, not pleasant)
The next item is a single point specification for the colour balance. A CRI of 80 is so-so. A CRI of 100 would be an emulation of a black body emitter (like an incandescent).
"Incandescent and halogen light sources have a CRI of 100. Typically, light sources with a CRI of 80 to 90 are regarded as good and those with a CRI of 90+ are excellent! The general rule is: The higher the CRI, the better the color rendering capacity. CRI is independent of color temperature." [ See picture of four apples... ]
(
formatting link
)
Expecting a single point CRI number to reflect colour balance is never going to work. There may be an alternate proposal to CRI, but the thing is, people still make really bad LED light bulbs that will "knock an eye out", whether we have a CRI value or not. I have a LED bulb here that is "mush" and the light, while bright, is awful for any purpose.
To some extent then, you have to get a sample unit of whatever you plan on using, and see for yourself.
In Dubai, they like smaller emitters for their efficiency. But one of the prices for doing this, is the cost of fixturing. The electrician would be there all week, wiring up all those sockets. The stand alone Dubai lamps are low power, like 2W or 3W of mains power. The fake Dubai lamps we can buy here, are SMPS based, and not the (lightly regulated) capacitive dropper of the Dubai lamp. Dubai lamps use twice the filament chains of LEDs that other filament chain lamps use.
formatting link
These bulbs are more efficient than the ones we get, cheaply, at the store. But then you'd have quite a wiring job ahead of you (many sockets).
My 5' LED light for the kitchen arrived this afternoon. I will have to see what the light is like but can anybody tell me how I can balance it on my head while I wire it up please?
I think the idea is you screw it to the ceiling before you try wiring it!
(and if that is the hard bit - then one smallish screw placed through a hole somewhere near the middle - with the end of the driver poked into the head holding the screw in place. Now offer the thing up to the ceiling and drive the screw enough to take the strain. You can now take your time fixing it properly).
Their 5ft and 6ft triphosphor T8 fluorescent tubes are 58W and 70W respectively. The light output is 5200 and 5600 lumens. That's 90 and 80 lumens per watt. The total light output for 2 x 5ft and 2 x 6ft tubes would be 21800 lumens.
For comparison, look at their LEDlite led "fluorescent tubes". The 5ft and 6ft tubes are both rated at 24W and 4000 lumens, so providing 167 lumens per watt. On that basis they are twice as efficient (the real fluorescents would provide light over 360°; the led equivalents are said to provide 330°, so they're essentially equivalent). Strangely, the LEDlite website says the 24W tubes provide only 3230 lumens, or 135 lumens per watt.
To get 21800 lumens from the led tubes, you'd require 5.5 tubes - so 5 tubes in practice. The consumption of 5 tubes is 5 x 24W, or 120W. For 4 hours, that's 480Wh. The fluorescents would use 256W, or 1024Wh over the same period. So, overall, the leds would use 0.5kWh less per day than the fluorescents. That's 20p/day based on 40p/kWh (that might be a bit on the high side, as mine is currently 37p/kWh). At TLC, 3 x 5ft leds and 2 x 6ft leds would cost about £51 + vat; let's say £63. On that basis, it would take £63/£0.20 days, or around 45 weeks to save enough electricity to pay for the tubes. You'd also have to factor in another
5ft tube holder for the extra tube, or put up with 25% less light from 4 tubes.
The wiring is behind it so I have to connect it first. It goes into spring clips so the plan is fix them to the ceiling, balance the light on a step that allows the flex to reach the choc-block (that was used by the spotlights) then clip it up. We'll have to see :-)
If the wiring is in the ceiling void, there is no harm in extending the switched live cable a bit to make it easier. You can push the spare into the ceiling when done. Failing that place a fixing in the ceiling such that it's hole will later be hidden by the light. Attach a strap of some kind to that to hold the fitting while you wire.
to be honest, the new LED battens and plastic casings are much lighter than the old flourries with their heavy iron framed ballasts and metal casings.....
I think the other thing to bear in mind is that the traditional tubes lose efficiency over time, whereas as far as I can tell, LEDs do not. Of cours this depends on the way they make the light, I believe some use UV leds which makes the phosphor light up at visible light in a similar way to traditional tubes using ionised gass, but the driving of the older tubes is less efficient as well, unless its an electronic ballast. I of couse only use light when a sighted person comes around, so there seems little point in changing until it comes time to replace something, when the light panels start to look attractive providing the reliability is there. If you don't want to have to redecorate the sealing however the LED tubes is the only game in town as real tubes are like hens teeth. Brian
Just spotted that I c*cked-up by only costing one tube (and probably under-estimated its cost) :-( Although underlying electricity costs are high at the moment they will (probably?) fall. A bit more poking-around suggests: that T5 tubes give about 100 Lumens/W and T8s around 80 Lumens/W, so size matters! It's hard to find believable accurate info about life and efficiency, but the bottom line seems clear: even for relatively high usage there isn't a strong economic argument for replacing serviceable fluorescents with LEDs. Replacing failed units with LED is, obviously, a no-brainer.
I think that sums it up well. My cost estimate of 37p/kWh should have been 32p/kWh (will it fall more, or is that the final reduction we can expect?), and looking at various sites appears to suggest that the led fluos would be giving nearer to 120 lumen/watt rather than 150. in fact, Philips state about 100l/W - so not that much more than a standard fluo. On that basis, simply replacing working fluos with led equivalents makes little sense.
To a first approximation our electricity prices are driven by gas prices, and they (wholesale) are now below what they were when Vlad started his little exercise ...
though we weren't then paying for Bulb and Ofgem's other lesser disasters, and there is pressure on Ofgem to switch the burden of that and other fixed and mixed costs from standing charges to unit rates
Which would hit families needing to do a lot of washing, drying and heating and elderly people needing extra, often electric, heating in the colder months.
I came to that conclusion as well. In respect of the long tubes anyway. CFLS proved so unreliable, dim , and short lived that they got replaced irrespective of whether they failed or not.
The colour of fluoro light is is not my favourite, but for work areas,m who cares?
LEDS now seem to be getting more and more reliable. I have had a fair few pop, but the remainder are rock solid.
Nuclear power stations sell to the grid currently at around 4-5p. Hinkley point is guaranteed around 9.5p I think. Drax wood is similar to coal at the 4-5p mark I think
Its gas and renewables that together drive the price up.
HomeOwnersHub website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.