Over on YouTube, [Technology Connections] has a new video: Induction lamps: fluorescent lighting’s final form.
This video is about a wireless fluorescent light which uses induction to transfer power from the electrical system into the lamp. As this lamp doesn’t require wiring it is not prone to “sputtering” as typical fluorescent lights are, thus improving the working life by an order of magnitude. As explained in the video sputtering is the process where the electrodes in a typical fluorescent lamp lose their material over time until they lose their ability to emit electrons at all.
This particular lamp has a power rating of 200 W and light output of 16,000 lumens, which is quite good. But the truly remarkable thing about this type of lighting is its service life. As the lamp is simply a phosphor-coated tube filled with argon gas and a pellet of mercury amalgam it has a theoretically unlimited lifespan. Or let’s call it 23 years.
Given that the service life is so good, why don’t we see induction lamps everywhere? The answer is that the electronics to support them are very expensive, and these days LED lighting has trounced every lighting technology that we’ve ever made in terms of energy efficiency, quality of light, and so on. So induction lamps are obsolete before they ever had their day. Still pretty interesting technology though!
Thanks to [Keith Olson] for writing in about this one.
Which is also at the upper end of the practical luminous efficacy of most consumer LED lamps. 80 lm/W.
A 60 Watt equivalent LED bulb does 600-800 lumens with 9-10 Watts. That’s roughly 60-88 lm/W. You can do better, but only by sacrificing color quality (CRI). Individual diodes can do over 100 lm/W but that’s excluding the power supply and light diffuser, and the operating temperature effect, which reduce the efficiency.
Mind: part of the “efficiency” of LED light is by the way you rate the diodes. The industry standard is specifying luminous output at a junction temperature of 25 C which is never the case in a real operating lamp, especially after years of use. The lifetime is specified down to an arbitrary amount of dimming. One company can say 80%, another 70%, a third can specify the lifetime to 50% brightness before they call the LED “done”. LED’s in recessed cans and under closed lampshades without ventilation get dimmer faster.
What happens in the real world, the brightness of the lamp drops significantly in the first minutes after you turn it on simply because it gets hot, but that’s difficult to notice by bare eye because your eyes adapt to the light level.
The current standard seems to be testing the bulbs at 15 seconds after turning them on with an ambient temperature of 25 C (free ventilation), and if the light output has varied less than half a percent in that time it’s considered “thermally stabilized”. This gives inflated brightness values for lower power bulbs that take longer to heat up.
Fascinating, I did not know these specifics. Good to know when evaluating LED efficiency claims
You’re quite wrong here dude. LED efficacy goes over 200lm/Watt these days, and LED’s with such a high light output also generate much less heat. The biggest problem here is “industry”. The 200lm/W led’s may be exceptional, but I’m guessing around 150lm/Watt is quite normal, but what most often happens, is that LED’s get over driven to get more light out of them. And by over driving the LED’s you do get more light out of them, but efficiency also goes down dramatically.
I am most often “too cheap” to buy quality lighting, but I did some measurements at cheap Aliexpress COB light panels. They were advertised as 10W, and they work fine at 10W, but they do get quite hot. When I run these panels at around 6W you barely see a difference (Eyes are extremely bad ad estimating light output differences), but the panels also barely get warm. There is a very clear “knee” around 5 to 7W, above which the panels suddenly get much hotter. (And that’s much easier to measure, even with bare fingers).
Also, when you look at:
https://www.lighting.philips.com/prof/led-lamps-and-tubes/led-bulbs/master-ultraefficient-led-bulb/LP_CF_8932020_EU/family
Then you see the “LED CLA 4W A60 B22 4000K FR UE SRT6 929003624219” Which (clams to) generate 840lm from 4W of juice, so that’s 210lm/W.
Also, these higher efficiency light bulbs also tend to have a much longer life expectancy. Simply because the LED’s are not overdriven and do not get as hot.
Another factor are the armatures in which the lamps are used. Open armatures that help with keeping temperatures down also improve life expectancy of the LED’s.
That’s fine, but the light bulbs I can find at the market are still only 60-90 lm/W. I checked.
Efficiency isn’t really the selling point of these things anymore, because we’re already in diminishing returns. For a constant amount of light, going from 14 lm/W (halogen) to 90 lm/W (LED) represents an 85% drop in energy consumption. Doubling the LED efficiency to 180 lm/W only reduces the energy consumption further by 7-8%. Considering that lighting was only around 20% of household energy consumption even when we were using incandescent bulbs, the real effect of further improving LED efficiency is on the order of 1-2% of the total. It’s negligible. Fluorescent tubes in the 80 lm/W range, as well as similar LEDs, are already “good enough” and we don’t need to sacrifice any more light quality to gain more efficiency – in fact we’d rather have the opposite.
Speaking of which, you have my point: “You can do better, but only by sacrificing color quality (CRI).”
The MASTER ultra efficient bulb has a CRI of only 80. That’s worse than the sickly colored CFL tubes we had back in the day with a typical CRI of 85. The LED bulbs I buy are CRI 90 minimum as I won’t suffer hospital quality light in my house. CRI 95-97 is my preferred option if available. It’s a bit pricey though.
Down to 70% at 50,000 hours. It’s non-linear: you lose 10-15% in the first few hundred hours and then it stabilizes to a gradual fading. That means you can expect about 160-170 lm/W average luminous efficacy over the whole lifespan.
Doing the same analysis for the common supermarket LED bulb suggests that the practical luminous efficiency, even if it starts initially at 90 lm/W would be around 70 lm/W. That’s also why the “60 Watt equivalent” ratings are total bunk. Halogens don’t dim in use.
“Why high CRI is always less efficient”
https://www.waveformlighting.com/tech/why-high-cri-is-always-less-efficient
The theoretically “ideal” white light source would have a luminous efficacy of 251 lm/W. A completely monochromatic non-white light source can achieve 683 lm/W. This is assuming that the light source itself is 100% efficient.
The highest recorded external quantum efficiency of a white LED (organic WOLED) is 31% so far, which means that for perfect CRI the theoretical efficacy limit of a LED bulb at the current state of art is… drumroll please… 78 lm/W.
“Given that the service life is so good, why don’t we see induction lamps everywhere? The answer is that the electronics to support them are very expensive, and these days LED lighting has trounced every lighting technology that we’ve ever made in terms of energy efficiency, quality of light, and so on.”
Try getting a tan under them.
Also a few caveats under the “quality of light” part. I would think that causing accelerated macular degeneration is a hit to quality
Now do the sulfur lamp !
And also the ESL. These were supposed to be compact fluorescent tubes on steroids.
https://en.wikipedia.org/wiki/Electron-stimulated_luminescence
Back then, in CISPR15, they even agreed on higher EMI limits for conducted and radiated emissions solely for electrodeless lamps. See for example here
https://www.lisungroup.com/wp-content/uploads/2020/01/EMI-9K-CISPR15-2018.pdf
and here
https://www.trilux.com/en/lighting-practice/lighting-equipment/luminaires/electromagnetic-safety/en-55015/
In addition:
I have found a quite valuable website dedicated to these kind of lamps a few years ago. It is still online:
https://edisontechcenter.org/InductionLamps.html
This seems a little surprising to me. I work in LED lighting driver design and our initial sales pitch was yeah our stuff is way more expensive than incandescent, but it also lasts 20x as long and since the price of industrial lighting is almost entirely driven by the labor to replace bulbs, we’re way cheaper in the long term. And people bought stuff, lots of stuff, based on that premise. So I’d think that expensive drivers and lower CRI would be offset by the lifetime of these in industrial and commercial applications, especially now that the market has driven LED lighting lifetimes down into only an order of magnitude better than incandescent.
Not even an order of magnitude. I keep replacing some of my LED bulbs every year. Those that are inside glass covers or in the fume hood in my kitchen all start to blink and turn off randomly after a year.
Consumer grade stuff is worse, yeah. Industrial/commercial baylighting is still pretty good. The stuff that runs off 480V is solid because they’re not on a six month redesign cycle.
Industrial situations where replacing the bulb is especially difficult is exactly where this tech saw the limited success that it did have. This is speculation but based on the timing it was probably just starting to gain momentum when white LEDs became viable and wiped them out.
I don’t know guys, I half get some of the censorship (even when I disagree) but removing that comment that that guy placed just because it criticized the posting of popular youtubes on HaD seems a bit extreme. And petty.