Thank you for doing all the work and recommending Auxma strips. I have purchased 2400K 3000K and 5000K rolls. Awesome light quality.
For the 24V 3000K strip, resistors are 56 ohms.
At the beginning of the strip, close to the power supply connector, there is a 3.46V drop on the resistor so LED current is 62mA.
3.46V is 14.4% of 24V so resistors waste 14.4% of total power, and LEDs get 85.6% of total power. Not so bad.
At the end of the 5m strip, 2.7V on the resistor, so only 48mA LED current. Thus 5m is a bit too long for a single power connection.
—
Note efficiency depends on LED color. For 24V having six white LEDs in series is okay, it wastes only a few volts in the resistor. But manufacturers use the same PCBs for all colors! So RED strips, with much lower Vf LEDs like 1.8-2V, will waste like 50% of the total power in the resistors… Not good!
The efficiency loss is a bit of a problem atop of the uneven brightness thing caused by the uneven distance to the power supply which each strip stage has. It is a cheap and simple way to have led lighting, but far from being the best. I'd rather prefer strips in series (without resistors) powered by standard (usually AC powered) led drivers; this requires good isolation and limits concerning minimum and maximum strip lenght, though.
A 10W light (about 1000 lumen) which is on 4 hours a day will use 14 kWh per year or about €2.20 of electricity every year.
Replacing incandescent lights with LEDs really saves energy and money. But optimizing LED efficiency beyond 100 lm/W isn’t really interesting IMO, it gets into diminishing returns. I’m willing to trade some efficiency for better color rendering, better look and aesthetics… but modern LEDs, even the high CRI ones, are already very efficient, so the energy cost isn’t that much anyway.
Bottomline, lowering the heating thermostat by 0.1°C will probably be unnoticeable and save much more energy than optimizing LEDs…
I saw the inclusion of SOME SKV89 data in this spreadsheet:
from this thread:
compiled by fneuf.
But it seemed that fneuf only included measurements that included R12 data (it factors into his “grading” formula) so most of the SKV89’s data is not included. (R12 was measured for only 3 out of 34 products)
CRI_Grade was created by me, I added it to his spreadsheet for him. It uses Ra, R9, and R12. It really targets the weak areas of modern LED technology works stellar. There is a link in my signature about it.
It’s a good question, I just don’t have much time to type. Maybe this along with google will help… CCT is the warm/cool choice you have. DUV is unwanted tint shift, usually green or rosy. CRI is a measure of the how balanced the rainbow is, under 90 is not very good. R9 is one of the rainbow measurements, it’s the hardest with modern tech, it is usually far lower than 90.
The others are a bit harder to explain in a short amount of time. I hope this helps.
Thanks Joshk.
Which parameter is relevant for blue light level? I guess that low CCT and an even spectrum mean low blue light but is there a more direct measure of it? Do you have a definition of the blue peak value and typical values for incandescent bulb, for example. Google didn’t help me with that.
In general, the higher the CRI, the lower the blue wavelength because the phosphor needs to convert more blue wavelength into other wavelengths to achieve the high CRI. Of course lower CCT means lower blue light because more blue light needs to be converted in order to achieve the lower CCT. But you never know unless you test them. You can compare the blue peaks in the ones I tested.
The 460nm “blue peak” of a 2700k LED bulb is about 1000x LESS than the sun at noon today. “blue fear” mostly comes from shady salesmen and click-bait writers.
To better quantify what I am saying, The sun is putting out 2000mw/m2 of 460nm blue light right now. My 2700k LED bulbs put out 2mw/m2. My meter was in relative mode back when I was testing cool-white bulbs, so I don’t know the numbers for them right now. Factory defaults…
Thanks for the answers. Just to be sure - is it true that whenever the blue peak is low and the spectrum is similar to incandescent light there will be not much UV? So there’s no need for getting a separate UV measure?
Yea a quality 2700k LED bulb will look incredibly similar to a 2700k incandescent to your eye. The meter will know the difference though. LEDs get away with the difference so well because our sensitivity to light drops off bad out where the LEDs drop off. Honestly, I think we (humanity) has the tech to fill the graph a lot better, it’s just not worth the extra power and heat to do so.
2660k Incandescent bulb:
2600k LED bulb:
There are some specialty LEDs that use a UV emitter (I think user Sunlike sells one), but I have never seen a second seller of them.
I ordered a new batch of Auxmer (formerly Auxma) LED strips to test. My all time favorite was their 2400k. Seems the quality haven't changed after two or more years. They recently released the 2200k, which is also amazing. The 2000k is like candle light and perfect for the bedroom as it produces negligible amount of blue light, which means it is the best for sleep.
Auxmer 120LEDs/m LED Strip, CRI95 DC12V/24V, 28.8W/m 2000K
