Jon Kirwan wrote:
[...]
Like I said, although I can imagine possibilities, my experience flies
in the face of the idea of a mere night light LED using a rare earth
phosphor with 3 eV photons completely failing to provide visible light
levels for human night vision in 3 months being due to the phosphor
itself degrading to that point. I just cannot go there. The stuff
has to have been chemically unstable to begin with, or poisoned in
some way. ...
That's probably the key sentence here. There are threading dislocations,
plain old thermal runaway because of surface inconsistencies,
Thermal runaway in a night light???
contaminations, etc. Considering the price of such lights the
manufacturers would probably hump every opportunity to scrape off half a
penny by going to a different "distributor".
Well, some phosphors are very cheap stuff. I don't know how much they
use in these LEDs, but in my applications a pound of it will last for
tens of thousands of probes (.02 gram each.) Of course, they don't
sell for 10 cents a piece. But cripes, it's not expensive.
Combustion synthesis began to be used in the late 1980's, I think. The
method is cheap (ammonium nitrate oxidizer and urea or glycine fuels)
and produces well-crystallized, very fine particle sizes very rapidly
and often without the necessity of very high temps in a separate step
(meaning above say 1500C) needed for annealing; or other expensive
steps like mechanical separation (grinding, milling, etc., which also
have the downside of often quenching the effect in the product.) It's
a mature, production process producing a cheap, uniform product so far
as I know.
And who knows where the stuff is really coming from.
Well, this is why I started asking in the first place. I'm just
sitting here flummoxed about the idea of a weak night light LED
phosphor getting completely destroyed by a few 3 eV photons over its
occasional use in a 3 month time. Every aspect of that tells me
something is wrong with the claim. The photon energies just don't cut
it. The low intensities just don't cut it. The short calendar time
just doesn't cut it. The complete loss of function doesn't make any
sense, either. And rare-earth phosphors are about as stable as a
piece of ceramic tile. They don't have _any_ H2O in them -- their
formation temperatures make darned sure that isn't the case. And I
have to buy the whole idea when there are so many reasons why it
shouldn't be able to happen?
I'm not saying it can't. Like everyone else, I live a tiny life span
in a small, narrow part of the world and I do not have comprehensive
experience. Something could certainly surprise me. So that's why I
was asking about it. I'd really like to know exactly _what_ the
phosphor is. Then I could go over to the books on my shelves and take
a look, at least, to see if that makes sense in this context.
I guess it's like telling me that sometimes a rock falls upwards. I
might believe it, if I knew what kind of exact rock it was. (Filled
with a hydrogen gas??) But I'd sure have a hard time buying it if
someone didn't spend a little time describing the rock.
I have heard from people who had plain old
jelly-bean opamps and similar chips do weird things. Leakage increases
that couldn't be explained, and so on. In most cases they had been
bought at, ahem, second tier sources.
Hehe. I have had similar, unexplained behaviors in really good
quality Burr-Brown chips. The ACF2101, for example. I would sit and
monitor the integrator output for weeks at a time. It would sit
wonderfully at one charge level for many minutes, then suddenly shift
to a figure 3 times higher and stay there for another 5 or 10 minutes,
then suddenly shift to a third value different from the other two, and
then cycle around these three places. I was using a stock demo board
from Burr-Brown in these tests, which was pretty well designed and
made up in effect a faraday cage for it, too. One of the weird things
was that the stepped levels were at discrete places. No level in
between was every hit. It was as though there were a few stable
places where the bias current could sit and that it would randomly
jump from place to place and just sit there for a while. Then move
again. The time between jumps appeared to be rather Gaussian in
distribution, too. (One of the things I looked at and why I ran this
for a few weeks.)
But that has nothing to do with the phosphors. I completely accept
the idea that some manufacturers don't care at all and will cheat
anywhere they can get away with it, even if only for a few months at a
stretch. That's definitely believable. But phosphors are pretty
stable from my modest experiences and I'm still struggling with all of
the combined factors of the claim on that web site. I'll believe it,
but I'll need to know what the phosphor is. I'm really curious about
that aspect.
Jon
