LED in hermetic envelope?

My speculation, a hydrogen or helium fill would give improved cooling. I don't know if that's used or not.

Is the pumping stem sealed or just left open?

?t=436 Recently I tried to run white LEDs in pure nitrogen in a glovebox at work. This was very pure nitrogen, with maybe a few ppm or oxygen or water vapour at most. The LEDs degraded quite rapidly during operation in notrogen, and became very dim. Adding a little bit of pure oxygen whilst they were still running restored the brightness to about normal.

I have read about white LEDs degrading in sealed luminaires in the presence of some organic vapours (from elastomeric seals etc.) but at least in my case, it seems that the lack of oxygen was important in causing or enabling the degradation.

The nipple was sealed.

Suggesting that oxygen is needed seems odd.

Oxygen and nitrogen have roughly the same thermal properties, but oxygen seems to be involved in just about every aging process. Helium and hydrogen have about 6x better thermal conductivity.

RL

Wish they'd extended their graph to include 0% He. That might have suggested the practical commercial value of the technique.

My specimen was a smaller bulb of the same ganeral shape, but the envelope was only 2x the height of the edison base.

4W consumption - similar led structure but with a 15degree twist and no spread at the base. Led strips effectively have the same angles w/r to each other as the larger bulb, without increased spacing of ends.

'globe' brand. marked A15.

300 lumens 3000K 15000hrs Not in US commercial website. 06-3051138 bubble pack stock number 667888253029 bar code

'not for use in ovens'.

RL

RL

Helium would be nice, but there is a finite supply and at some point, there will be limitations on how it is used. Maybe at some point in the future we will capture helium resulting from fusion reactions to provide high thermal conductivity in LED bulbs. LOL

Helium-3 does seem to come only from working fission reactors - you expose Li-6 atoms to the neutron flux within the reactor, and collect the tritium that come off, and that tritium decays to He-3 with a half-life of 12,3 years. The tritium is the primary target, but the He-3 is the eventual product.

Dimwit. The joke was in the contrast between getting He-4 out of fusion reactors in some hypothetical future, and getting He-3 out of fission reactors now.

<snip>

Is liquid feasible? I mean, fill the globe with liquid instead of gas. Just don't drop it.

Don't completely fill it either. Differential expansion can be a swine. The heat capacity of a liquid filling would be higher, but it wouldn't convect as fast as a gas, and it's still only moving heat from the light source to the surface of the light bulb, and you are still relying on free air convection to cool the bulb as a whole.

Sure, you can do a lot of cooling with drilling mud, for instance. ;) (As you know)

Cheers

Phil Hobbs

Weren't there some light bulbs filled with krypton or xenon or something? And iodine.

Tungsten-halogen incandescent lamps

Iodine and bromine both work, diluted by an inert gas - like krypton on xenon.

The tungsten that gets evaporated off the filament turns into tungsten iodide or bromide, which circulates in the inert gas until it gets close to the hot filament, where it decomposes, dumping the tungsten back on the filament. You could run the filament a bit hotter, and still get a decent filament life.

There was an inner silica bulb, which ran hot enough that the tungsten halide didn't condense on it's inner surface.

Quartz Halogen.

Joe Gwinn

Usually argon--it's cheaper. Xenon makes a nice bright white arc, of course. Krypton is way too red-looking, iirc.

Cheers

Phil Hobbs

I meant incendescents.

Sure--the ones with thin glass envelopes are all argon-only, and the ones with small, thick quartz envelopes are all argon/iodine, AFAIK.

Krypton and xenon are far more expensive than argon, and no better for incandescent lamps--if it don't react, it don't react.

Cheers

Phil Hobbs

There WERE krypton-fill lamps, for flashlights. The volume is small, but the thermal conduction is small, too: that makes for longer battery life. That fad didn't last, the quartz/halogen lamps were better for my bike light, before LEDs took over all those items.

Yep, not just little flashlight bulbs. In 1950s-70s West Germany Osram made filament bulbs with krypon (probably mixed with N2 or Ar) gas fill boasting longer life. I have some little flashlight bulbs with Xenon fill. The heavier gas pressure helps restore tungsten atoms to the filament. All these have no halogens and are in regular glass envelopes. Halogenated gas in quartz bulb permits running the filaments even hotter.

piglet