Interesting domestic LED design circuitry

Eventually, jamming enough LED sticks in a bulb, they will heat one another.

The real problem is, that 'bulb' is why your fingers don't touch the AC voltages, and jamming LEDs together, always a bad idea, is necessary for a screw-in lamp replacement that runs off high voltages. Future light fixtures using replaceable LED elements in an interchangeable fashion, with safe power wiring, won't have these problems.

...if engineered well, they'll have other problems entirely.

Not in this case. The whole idea is to use several low-power led "strings" instead of a single high power string. The overall power consumption is the same, but it is being spread across, for example in the 4W bulb, four times as many leds. The lumen output is the same - maybe with a slightly different spectral range - but the leds never get anywhere near as hot as those in a single string.

I assume the designers have looked at the "sweet spot" for cost vs power vs lumens vs life, and they've ended up with those new bulbs.

Actually, the power is not the same because by operating at a lower current and a lower temperature the LEDs are much more efficient. He explains that pretty clearly. In fact, that was the whole point to the bulb. The purpose was to be more efficient.

As far as I can see, those are the regular filment LEDs. One can buy them without any problems in Poland. I bought one, but due to the lack of a capacitor it was flickering like crazy. Not a problem in a static environment, but if there was even a tiny motion, you could see a very disturbing 50Hz stroboscopic effect. I gave mine to somebody for that reason and returned to using a halogene bulb.

Best regards, Piotr

these are special in that they have 4x the number of filaments each run at 1/4 th current, so they are more efficient and last longer

OK, but the amount of flickering should be about the same, which makes its longevity irrelevant.

Best regards, Piotr

I doubt they have any flicker, they have smoothing capacitors and flicker would have been obvious on video

You watch the video? It uses a series capacitor AC voltage-dropper, then into a bridge rectifier and filter so the "filaments" in parallel are almost driven constant-voltage off the bus at the nominal dropped and rectified unregulated mains voltage. but then has a MOSFET/bjt discrete current regulator in series to smooth residual ripple.

Eletrolytic caps in a tiny unventilated enclosure doesn't shout "high reliability" to me but the lighting is pretty low-power so so long as it isn't put in a housing where it's taking the direct Dubai sun 16 hours a day should be alright for quite a while. There's a fair amount of redundancy, four series-dropper caps in parallel, two electrolytic filter in parallel, two zeners in parallel strapped between the low side of the filament and the negative supply so if the current regulator circuit fails it'll still work more-or-less.

By royal decree his Majesty the King Sheik Mohammad bin Rashid blah-di-blah has proclaimed new construction must use this lighting whenever possible and if Phillips is selling these buggers at $5-$8 a pop in small volume as claimed I can see why they signed an exclusive agreement, the margins must be pretty sweet, and this market has decreed, er I mean decided that it's the best

You say 50 Hz, but wouldn't it be 100 Hz? My understanding it that should be very hard to observe. I recall in the days of CRT monitors a refresh rate above 60 Hz was required for everyone to be happy with the images, but not much above 60 Hz.

This one is rectified and filtered and current-regulated so I expect the flicker is minimal at normal operating voltage, but the simple circuitry means it isn't compatible with standard dimmers

ld be very hard to observe. I recall in the days of CRT monitors a refresh rate above 60 Hz was required for everyone to be happy with the images, but not much above 60 Hz.

You are confused. Pitor is talking about an LED bulb he bought, not the on e Big Clive reviewed. I can see where a very inexpensive bulb would omit t he capacitors and simply run at a current controlled by a regulating chip, shutting off every time the voltage drops below some level. In fact, I see m to recall a review by Big Clive on exactly that sort of bulb.

There's a type of "bulb" that has no electros and perhaps no rectifier for the LED strings, it has an IC that monitors the line voltage and dynamically switches in and out various strings of current-regulated series/parallel LEDs in some clever optimized way that keeps light output approximately constant while minimizing flicker. Can't remember the IC offhand but 3M makes one, like a floodlight-type.

In static conditions flicker was barely visible. Moving objects looked as if they were illuminated by stroboscopic light, which indeed was the case. It was easily visible and confusing to the point of getting rid of the bulb by donating it to somebody with a slower vision tract. The person does not see the flicker.

Best regards, Piotr

ld be very hard to observe.

I shouldn't argue that 100 Hz is adequate to not see the flicker. I was al ways sensitive to CRT flicker although I don't see it at 60 Hz on LCDs (per haps it is not as apparent on LCDs). I also see taillight flicker on movin g my eyes. In fact, the first time I saw it I was on a ramp with zero merg e lane, at night with one car approaching in the left lane. After convinci ng myself it was safe to proceed into the right lane I turned my eyes back to looking forward and suddenly I saw a half dozen pair of tail lights in m y peripheral vision!!! I nearly hit the bridge railing. I find when discu ssing this with friends they typically have no idea what I am talking about . One friend said she finally saw it but thought it was literally no big d eal. It bugs me even when it isn't a hazard.

I don't know what rate car tail lights flash at, but I bet it's not much ab ove 100 Hz if at all. Cadillacs are the worst and some have huge vertical tail lights that really stand out.

Yes - my error. Power is the same, but many more lumens/watt.

Can somebody summarize a long winded 32 minute video about s***ty LED bulbs into a few sentences?

Philips designed a light bulb to be properly efficient and as a result runs cooler and lasts longer. This was at the urging of the government of that country and now these light bulbs are mandated but likely are two or three times more expensive than the units we get that have to compete on price r esulting in less efficient lamps that burn out quicker.

Is that short enough? I don't think there is anything special about the bu lb design other than the high degree of redundancy which means even if a co mponent fails the bulb will still operate even if not quite as well. Examp le: they added zeners to the current regulation circuit so a failure would allow the bulb to continue to work at reduced brightness, but then they use d TWO zeners I suppose in case one of them failed. Triple redundancy in ef fect.

I think missing the video misses the point. I find his videos to be quite entertaining. But to each his own.

sounds pointless. Even LED traffic lights don't rely on double zener diode nonsense.

A few of his vidoes are OK, the rest just go on and on and never end. That terrible combo screwdriver he uses bugs me too. Just can't watch them anymore.