Want to read Slashdot from your mobile device? Point it at m.slashdot.org and keep reading!


Forgot your password?
Polls on the front page of Slashdot? Is the world coming to an end?! Nope; read more about it. ×

Comment: Re:Cool, but way overstated. (Score 1) 47

by jeffb (2.718) (#49805377) Attached to: First Ultraviolet Quantum Dots Shine In an LED

It almost seems like you're interpreting "UV-C" to include the range from 360-380nm. There are apparently some results indicating that emitters in this range can be germicidal, if you use enough power and enough exposure time; is that where our disconnect is arising?

The Nichia page you linked lists only longwave emitters, with 365nm the shortest wavelength. I'm sure they have shortwave emitters, and maybe even samples for some of them, but if they aren't listed on the Web site, I'm not confident how much of a "product" they are to date.

Cree has never sold UV-C LEDs, as far as I can tell. They sold longwave emitters for a while, but then discontinued them.

I spent some time prowling around HaSun's list of UV LEDs. I haven't waded through every listing, but most of the emitters under 300nm seem to be in the range of 1.5mW or less; I found one ("New Technology!!!") that claimed 0.2-0.3W optical output power in the specs, but in the chart below, it said 0.2-0.3mW. WIth forward current of 20 mA and forward voltage of 7-8.5 V, getting out 200mW of anything would be quite the trick.

Again, we can see that the shortwave emitters exist, but it doesn't look like they're common enough or powerful enough to start appearing in products yet.

Comment: Re:Yes and no (Score 2) 47

by jeffb (2.718) (#49805259) Attached to: First Ultraviolet Quantum Dots Shine In an LED

I'm flattered that you're reposting my links from below, but I think you're missing GP's point. None of those three links appear to describe units that are "commonly available" -- in one case, it's only engineering samples, and in none of the links do they say a word about pricing or actual availability (the last one claims "mass production", but doesn't back it up).

"UV lasers" are mostly 405nm, not really UV, and the quantum dots from TFA are firmly in longwave territory. So, GP's points stand.

Comment: Re:Water sterilization is the big thing here (Score 1) 47

by jeffb (2.718) (#49804679) Attached to: First Ultraviolet Quantum Dots Shine In an LED

Actually, 420 would be a bit too violet for this application. Cree's XR-E emitters seem to use a blue emitter centered around 450nm (pdf), coupled with a yellow phosphor -- blue + yellow = blue + (green + red) = white. That's how most "white" LEDs work. If you used a shorter-wavelength emitter, you'd need to downconvert all its output power, losing efficiency. By using a blue emitter, you pass some of the blue light, and downconvert just enough of it to yield the perceived color temperature you want.

Comment: Re:Cool, but way overstated. (Score 3, Informative) 47

by jeffb (2.718) (#49804653) Attached to: First Ultraviolet Quantum Dots Shine In an LED

Links, please? Last I checked, UV-C emitting diodes were essentially experimental items, with output power below 1mW, and sky-high prices. Feel free to ridicule me mercilessly if I'm wrong, but please post links; I'm a lot more interested in learning about UV-C emitters than in defending my hypothetical reputation.

This page crows about UV-C LEDs, but is conspicuously silent about output power (beyond calling it "stable"), availability, or price.

This page claims 5-10mW per device, adding that "limited release engineering samples are available today." The datasheet is mum on device lifespan.

This looks more promising -- 10mW per device, 10,000 hour life -- but where are the products, and where are the prices?

Obviously, I could be missing some products that are already shipping. But if you actually have a bank of UV-C LEDs that's been putting out enough power to kill algae in your aquarium for the last five years, it looks like a lot of electronics and physics journals would be interested in hearing from you. And so, as I said, would I.

Comment: Cool, but way overstated. (Score 3, Informative) 47

by jeffb (2.718) (#49802933) Attached to: First Ultraviolet Quantum Dots Shine In an LED

We already have robust conventional LEDs that emit high power at comparable wavelengths (considered "longwave UV"). This wavelength is not especially useful for purification and sterilization. For that, you need UVC, in the range of 250nm and below. That's still difficult with anything other than a fluorescent emitter or an arc; solid-state emitters in that range have very low power and short life, at least last time I checked.

The other problem with very short wavelengths is finding packaging materials that will transmit and withstand them over long periods. Even longwave UV will cause materials to deteriorate over the lifespan of a solid-state emitter; UVC is much more harsh.

Comment: Okay, I thought *I* took a long time... (Score 1) 132

by jeffb (2.718) (#49785497) Attached to: A Ph.D Thesis Defense Delayed By Injustice 77 Years

I just really enjoyed grad school too much to want to finish, and they eventually made it clear to me that I'd better defend soon or they'd kick me out. Of course, I also wasn't accomplishing nearly as much in the interim as the article's subject.

It's an inspiring story. All the same, my dissertation defense is something that I'm just as happy I won't have the opportunity to tackle later in life...

Comment: Re: Don't Mess With Taxes (Score 4, Funny) 379

The main one being I pay 4000 a year in taxes to send your to school and then spend another $8000 a year each to send of my 4 kids to a better school than the you get for "free".

If you're spending that much to send them to a "better" school, the least they could do in return is proofread your posts for you.

Comment: Yes! (Score 1) 116

by jeffb (2.718) (#49740889) Attached to: Hydrogen-Powered Drone Can Fly For 4 Hours at a Time

I remember sketching/doodling a hydrogen balloon with a fuel-cell panel stitched into it when I was in 9th grade. Of course, I didn't know much about the practicalities of fuel cells -- just that I probably couldn't afford the platinum it would require. And this was pre-TRS-80, so about the only thing it would have done autonomously was go up until it used up its hydrogen, then come down.

Comment: Re:Bigger scenes were impressive IMO (Score 2) 87

I don't know -- I'd say the character is well up the slope on the "realism" side of the valley.

As for the cost of hardware, well, an ATI video card that I installed about ten years ago came with some canned demos. It casually mentioned that one of the demos implemented a technique that was first displayed at SIGGRAPH in, I believe, 1995 or so -- some ten years earlier -- at which time each frame took several minutes to render on a large server farm. In the space of ten years, hardware advances took us from the render farm to real-time rendering on a sub-$1K card.

Now, maybe we're too close to the far end of the S-curve to see that kind of improvement over the next ten years. But I'm not convinced.

Comment: Re:Worlds Biggest Utility Battery (Score 1) 334

by jeffb (2.718) (#49568313) Attached to: Why Our Antiquated Power Grid Needs Battery Storage

Okay, 26 MW for 15 min is about about 6.5 MWh, ignoring load scaling (total capacity tends to be higher if you don't draw it as quickly). $35 million for 6.5 MWh is about $5.40 per watt-hour.

The Tesla battery is 10KWh, with an initial retail price of $13K. That's $1.30 per watt hour, less than a quarter the price you quote for the ABB battery.

Does this mean the Tesla solution's cost is 1/4 that of the ABB solution? Hard to say. Building something at the ABB scale, with reliability guarantees suitable for a utility, surely adds expense. On the other hand, mass-production leads to economies of scale that ABB probably won't ever enjoy.

Thanks for the links!

When the bosses talk about improving productivity, they are never talking about themselves.