A shorter version of a full (40 minute!) video that I had trouble uploading.
These LED "filaments" are now available loose on ebay, so I decided to get some and improvise a simple mains power supply for them and try them in various series and parallel combinations.
Check out Electronupdate's video about a possible glass version here:-
https://www.youtube.com/watch?v=25j2C4jq2HI
And Julian Iletts DC booster experiments here:-
https://www.youtube.com/watch?v=E-e_Wen-Oms
These LED "filaments" are now available loose on ebay, so I decided to get some and improvise a simple mains power supply for them and try them in various series and parallel combinations.
Check out Electronupdate's video about a possible glass version here:-
https://www.youtube.com/watch?v=25j2C4jq2HI
And Julian Iletts DC booster experiments here:-
https://www.youtube.com/watch?v=E-e_Wen-Oms
How is it that these tiny Power L.E.D.'s can be used without the quite extensive Heatsinking we seemed to need in earlier L.E.D. Light Bulbs?
Are they more advanced Power L.E.D.s Clive?
A very clear informative explanation of how filament LEDs are constructed… Thank you.
Need to use battery 🔋 not electronic is dangerous
LED light sucks
Hi Clive, I have a 4W version of this lamp made by Meridian lighting sold at Toolstation. Out of the 8 that I have bought 2 have started flashing. It's not a problem with the fitting just those two particular lamps which I have replaced. I have not dismantled them yet but would appreciate it if you could have any idea what would cause them to flash. The internal circuitry appears very simple so what component would you think might cause this problem?
the holes i think help the phosphor stick on. AND promote the light to go through both sides somewhat more evenly.
I bought 10 of these. They're really nice but man are they delicate. 2/10 came dead on arrival. I broke 2 more soldering them, the pins become detached with very little force rendering them useless. I think you're right about they're supposed to be crimped, but I can't imagine them being very sturdy then either. But the end result was pretty nice, and now I have 4 I can dissect.
As long as he doesnt touch both wires. he'll be fine.
What are the grey clips you're using for the cables?
That was weird we just had an earthquake here in Oklahoma. No idea how strong yet since I am typing this minutes after it happened.
Hey Clive, I didn't have time to watch the whole video so I'm sorry if you talked about this in it. Do you think the filament would be more evenly colored if the LEDs were placed in a spiral?
PS this is really damn cool and I can't wait for my whole house to finally be lit like this. You could maybe even do away with dimmers by just having a variable number of filaments be switched fully on or off using small mosfets or the like instead of having to add resistance or chop the mains wave, which is both inefficient and seemingly pretty unfriendly to anything that isn't tungsten…
or have a bunch of these in a cool spiral pattern within a bespoke light fitting…
or just woven into your ceiling with an even distribution
or.. etc.
OK, I'm not big on the full knowledge of how electronics works at this level – how come you could put one, two, or three of these in series without issues or the voltage seeming to significantly change? Especially a chain supposedly rated for first 80, then 160v, on a 230v supply?
Also if they're nominally 80v (and 0.4w) each, could you just chain three together with a bridge rectifier and wholly optional capacitor for a cheap and dirty 1.2w lamp? Or indeed, make the rectifier out of two (four?) 3-unit chains of them and plug it directly to the mains for a flickery but super simple 2.4w (4.8w?) one? ((or would it only actually be 1.2w still, using twice as many filaments, each only receiving half as much average voltage, but no other components save maybe a couple of smoothing caps?))
FWIW, 80v divided by 28 is a little under 3v … or to be exact, 250 divided by 3, then by 28, is JUST under 3v. So they've been careful to engineer things so they'll stay in spec quite nicely.
You might even be able to make the aforementioned no-other-components light-emitting-rectifier work with a pair of single-filament conductors on a 110-ish volt supply, as the peak will be just on the low side of 170v before RMS is considered… and it only actually gets half the wave…