Great LED bulb ...

holes, or unfold the origami bulb and re-fold it inside-out with all the

I'd also be worried about people who think 133 lm/W is 'about 200%' better than

Rather ugly IMHO. If you think of all the design possibilities, this seems rather pedestrian. All you need are a couple of parallel conductors, with LEDS on at intervals. Then you can bend it into any shape you like. Plus a space for a tiny transformer and rectifier.

Actually do you need a fair amount of passive heatsinking even if the production LEDs used really are in the 140Lm/W class. All the waste heat from LED is in thermal band IR and hot air. They cook the emitter junction very rapidly if operated with inadequate heatsinking.

An incandescent lamp gets away with radiating away a lot of its heat near IR and will happily work at 100-200C envelope surface temperature.

An LED or CFL will cook and ruin the control circuit capacitors in no time flat if its temperature gets anything close to 100C. The junction temperature also has to be kept reasonable or the plastic goes dark.

I don't like the idea of exposed LED dies on this design with a surface brightness that is potentially dangerous to the human eye.

For my money simpler is better and something like a simple cross of aluminium with LEDs on and a smart thin plastic cylindrical diffuser.

load balancing dear boy.

Cant guarantee all the LEDS have EXACTLY the same V-I curves

I've seen interesting prototype OLEDS that emit across a thin film.

The control circuit is a little more complicated that that.

MBQ

On Thursday, January 31, 2013 2:06:08 PM UTC, The Natural Philosopher wrote= :

kids who'd poke knitting needles into the >>> holes, or unfold the origami = bulb and re-fold it inside-out with all the >>> electronics exposed ... >>>= >>> >>> >> >> Rather ugly IMHO. If you think of al= l the design possibilities, this >> seems rather pedestrian. All you need a= re a couple of parallel >> conductors, with LEDS on at intervals. Then you = can bend it into any >> shape you like. Plus a space for a tiny transformer= and rectifier. > > Actually do you need a fair amount of passive heatsinki= ng even if the > production LEDs used really are in the 140Lm/W class. All = the waste heat > from LED is in thermal band IR and hot air. They cook the = emitter > junction very rapidly if operated with inadequate heatsinking. > =

ar IR and will happily work at 100-200C envelope surface temperature. > > A= n LED or CFL will cook and ruin the control circuit capacitors in no > time= flat if its temperature gets anything close to 100C. The junction > temper= ature also has to be kept reasonable or the plastic goes dark. > > I don't = like the idea of exposed LED dies on this design with a surface > brightnes= s that is potentially dangerous to the human eye. > > For my money simpler = is better and something like a simple cross of > aluminium with LEDs on and= a smart thin plastic cylindrical diffuser. > I've seen interesting prototy= pe OLEDS that emit across a thin film. -- Ineptocracy (in-ep-toc=92-ra-cy) = =96 a system of government where the least capable to lead are elected by t= he least capable of producing, and where the members of society least likel= y to sustain themselves or succeed, are rewarded with goods and services pa= id for by the confiscated wealth of a diminishing number of producers.

Cree are well beyond this at 200 lumen per watt

Usually from binning can get them close enough.

But taking an LED sticking it on a bit of metal clad board and calling it twice as efficient as anyone elses is beyond laughable and extends well into knowingly selling snake oil.

Presume this lot are using SSC devices, Cree`s 200 l/W is still a bit of a lab curiosity,

Martin , you must have been reading this page:

Kickstarter is always entertaining reading.

Cheers Adam

with LEDS on at intervals. Then you can bend it into any >> shape you like. Plus a space for a tiny transformer and rectifier. > > Actually do you need a fair amount of passive heatsinking even if the > production LEDs used really are in the 140Lm/W class. All the waste heat > from LED is in thermal band IR and hot air. They cook the emitter > junction very rapidly if operated with inadequate heatsinking. > > An incandescent lam p gets away with radiating away a lot of its heat > near IR and will happily work at 100-200C envelope surface temperature. > > An LED or CFL will cook and ruin the control circuit capacitors in no > time flat if its temperature gets anything close to 100C. The junction > temperature also has to be kept reasonable or the plastic goes dark. > > I don't like the idea of exposed LED dies on this design with a surface > brightness that is potentially dangerous to the human eye. > > For my money simpler is better and something like a simple cross of > aluminium with LEDs on and a smart thin plastic cylindrical diffuser.

Ineptocracy (in-ep-toc?-ra-cy) ? a system of government where the least capable to lead are elected by the least capable of producing, and where the members of society least likely to sustain themselves or succeed, are rewarded with goods and services paid for by the confiscated wealth of a diminishin g number of producers.

In the lab and with a near infinite heatsink. We are talking practical designs here. It is actually quite tough to make an LED bulb that will survive in fittings designed for relatively heat tolerant incandescents.

No. Not seen it before, but the high power parts all come with a disclaimer warning you not to stare at them when lit.

The surface brightness of some high power LEDs are now in the same ballpark as the flux at the surface of the sun. They are pinpricks of light that are borderline for triggering the blink reflex and they can worst case if abused potentially damage the retina.

The famous Belgian Blind Experimentalist was blind because he did an experiment on persistence of vision by staring at the sun for just 25s. Best known now for his work on the laws of bubbles. He may have been unusually susceptible but it really isn't wise to stare at the sun.

If you make a mistake with the sun during a solar eclipse when there is only a sliver of photosphere and the iris is wide open then you can get retinal burns. The nasty thing is that there are no pain receptors in the eye so you only find out about 8 hours later.

The same is almost certainly true of these LEDs being unsafe to look at without a diffuser. I use them and know how bright the 3W and 5W units can be. Think pretty much like striking an arc.

It is debatable if there is enough shortwave or near IR from LEDs to do actual permanent damage but the flash blindness and afterimages from making a mistake with high powered LED lighting is annoying.

I've not looked into it. I assume it needs a bit of smoothing. What else?

In article , GB writes

Some are driven at a high frequency (not sure why, maybe to reduce heat production?), so that would need additional circuitry.

For a mains powered unit, such as those shown, start with turning 240V AC into the much lower DC voltage required by the LEDs. This needs to be done within the space available, which isn't large. The efficiency needed to keep temperatures low enough for the electronics to survive long term mean it needs a switch mode supply.

The design shown would almost certainly not get approval for UK use, as it would be possible in theory, if not in practice, to come into contact with mains voltages, even ignoring the fact that it's a screw fitting.

It could be made to work, and has been made to work, but for large scale production and sale, a lot more work needs doing on the design.

with LEDS on at intervals. Then you can bend it into any >> shape you like. Plus a space for a tiny transformer and rectifier. > > Actually do you need a fair amount of passive heatsinking even if the > production LEDs used really are in the 140Lm/W class. All the waste heat > from LED is in thermal band IR and hot air. They cook the emitter > junction very rapidly if operated with inadequate heatsinking. > > An incandescent lam p gets away with radiating away a lot of its heat > near IR and will happily work at 100-200C envelope surface temperature. > > An LED or CFL will cook and ruin the control circuit capacitors in no > time flat if its temperature gets anything close to 100C. The junction > temperature also has to be kept reasonable or the plastic goes dark. > > I don't like the idea of exposed LED dies on this design with a surface > brightness that is potentially dangerous to the human eye. > > For my money simpler is better and something like a simple cross of > aluminium with LEDs on and a smart thin plastic cylindrical diffuser.

Ineptocracy (in-ep-toc?-ra-cy) ? a system of government where the least capable to lead are elected by the least capable of producing, and where the members of society least likely to sustain themselves or succeed, are rewarded with goods and services paid for by the confiscated wealth of a diminishin g number of producers.

Just wondering what pile of s**te you are using as a newsreader?

For me, it has merged all quoted text into one lump. Similar elsewhere that you have posted.

Well the headers say (in part)

Injection-Info: glegroupsg2000goo.googlegroups.com; posting-host=62.25.109.196; posting-account=SmOwTAkAAABVKZxU4E5lzwejTu4PYYAZ NNTP-Posting-Host: 62.25.109.196

User-Agent: G2/1.0

So no newsreader/client at all!

Avpx

the physics of voltage conversion which nearly always has to involve a chunk of iron or ferrite say that the higher the frequency, the smaller the chunk needs to be for a given power.

Its not more efficient, its cheaper.

Farnell do the Osram ones. I can't find them online, but I've seen Farnell exhibiting them. ISTR one that was something like a couple of sq inches was around £100, but that was a while back now.