Slashdot Mirror
Toward Micro-Diode Display Panels?
VernonNemitz asks: "Light-Emitting Diodes (LEDs) have been around for decades, and they come in all colors. Here are the basic principles behind their operation, information on what's currently in production , and other practical info. Now, you would think it obvious that video displays should be made from LEDs, to take on the various competitors. Certainly this has been done on a large scale, and I am not the first to be interested in LED display technology for home viewing. The apparent goal of silicon LEDs these days seems to be tied up with optical communications between circuits, have they forgotten the possibility of making high-resolution display panels?"
"Achieving a good image will require smaller pixels, and now I have a better understanding of what has been holding it up, so would like to share what I found out. Basically, unlike a lot of electronic technology, LEDs are mostly -not- made from silicon. So, while silicon diodes are common, and large arrays of them are also made (for sensors), LEDs are not easy to incorporate into standard integrated-circuit manufacturing. Nevertheless, researchers have been pursuing and steadily making progress toward integrating silicon-based LEDs. Even whole arrays, 'suitable for large two-dimensional areas', of silicon laser LEDs have been made, as far back as three years ago."
I can say that this technology is at LEAST 25 years away due to the lack of precision in nanostructure topology lithographic processes. The author seems to be ignorant of a few of the fundamental laws of nanotech, namely that: 1) Maxwell's equations must be obeyed in light of quantum gravity. Yes, quantum gravity is CRUCIAL to the process. This prevents the inadvertent bonding of junction pairs and 2) The superposition principle does NOT apply to these non-inertial frames of reference! Trust me, this technology is pretty much canned until we can solve a few engineering problems in the litho. area.
The largest problem with this technology isn't that it can't be done on a larger scale (it can). The problem is more financial. With all of these companies having geared up for LCD production and building huge plants while downsizing CRTs, it's unlikely that the shift to LED tech is going to happen anytime soon - no matter how cheap it is.
As always, you have to find a way to pay for the R&D that you use - even if it becomes outdated beforehand.
"...Well, there's egg and bacon; egg sausage and bacon; egg and spam; egg bacon and spam; egg bacon sausage and spam..."
I think we're barking up the wrong tree using light-emitting technology for displays. I think that we would be far better off having a very high-res, paper-white display that was 100% reflective. Indeed others have called this "digital paper" and it has become a holy grail of display technologies. However the technology to do this is just about here now. I'd much rather see that kind of display. Give me a reflective, 600 dpi display and then I'll read books on a computer. Light-emitting displays are much harder on my eyes than a magazine or newspaper.
3 years ago you would have spend $2,500 for a LCD of the same quality (if something compareable was even available).
Why do you think that in 3 years people wont look back and say "back than lcds were expensive, a 300$ 20" did cost 1000$ back then...."?
After all, people can make a 19" CRT for 150$, and if you look at the materials and needed manufacturing steps, they arent much simpler than LCDs.
But today, the companys are still paying for their fabs they had to build to supply the increasing demand. Once lcds have replaced crt, demand will stagnate, there wont be any new expensive fabs needed (or at least not as much as now), and prices will be dropping....
HI O WISE PRINCE. WHT TOOK U SO DAM LONG?
Just keep getting OLED displays larger and larger...
http://www.kodak.com/US/en/corp/display/
It uses an thin organic film between conducting layers rather than electronic components, but OLEDs are just a few years away. They're self-luminous and more efficient than a backlit LCD. Lots of /. stories about them too.
Red leds have been around for decades, but the higher you go in frequency (orange, yellow, green, and recently blue), the more expensive they get. Since you need many many triads of them to make a display, that may not be the cheapest one in town.
What's more, more often than not, LED colors aren't pure primitive colors, so it's even harder to get accurate RGB mixing with them.
"A door is what a dog is perpetually on the wrong side of" - Ogden Nash
...But not with Silicon. Silicon LED's may be (relatively) cheap for red and green models, but the blue ones are insanely hard to make, and thus are quite pricy (just look at the DigiKey catalogue). Also, Si LED's are very ineffecient (I don't remember the number, but it's like 20%ish I thought). The next "wave" of things are going to come from OLEDs (Organic Light Emitting Diodes). These are easier and chaper to make for screens and such, and eventually can be made in a "roll to roll" production type deal since these things are "grown." They're already being used in some cell phones, I believe.
A problem with these is that they decay if they're not protected, so currently they use a glass-substrate and a glass coating for some, which can make them bulky. Also, OLEDs have a shorter life span than LCD's, but that's being worked on too. So hark, have patience, OLED screens are coming. Another interesting technology to look into is Light Emitting Quantum Dots... those little suckers are cool, and could also replace LCDs and lights as we know it (from what I remember, they're also quite efficient... I think around 60%).
Of course we'll still be bitching. We're American.
I suspect that the technical limitation is that of delivering enough current across the span of the array. Unlike plasma displays, LEDs are low voltage beasties. And unlike LCDs, LEDs are high current devices. To get say 20 W of brightness, you will need to inject on the order of 10 Amps in (and that 10 Amps only gives you 11 microamps per LED in a 640 x 480 x 3 color display). Routing that much current along the array conductors(which are probably plated or grown on the glass substrate) with acceptably small voltage losses would seem to be very tricky.
That's just my guess as to the technical hurdles for high resolution LED displays. I'm sure some bright young soul will solve the problem, however.
Two wrongs don't make a right, but three lefts do.
some new tech
Reflection based displays, like eInk, are great for displaying text. But when it comes to representing graphical concepts they fall far short, since they're limited to greyscale.
It's fairly simple to adjust the amount of light reflected to make greyscale images. What is very hard to do is adjust the wavelengths reflected to make different colors. Until there is an easy way to do that discovered, light emitting displays is pretty much the only option for computing chores that rely on color.
I have experience in the industry, and can tell you that inorganic LEDs HD displays are not gonna happen. The inorganic LEDs cannot be deposited on silicon wafers. The inability to deposit means no way can it be cheap, or feasible. Not to mention the heat output by that many LEDs. I created a test box to evaluate LED's for lighting systems. Using the latest mass produced ones, I still had a significant heat output with 100 tiny ones. A HD inorganic LED panel would simply melt. Organic LEDs remain viable, yet still not ready. OLEDs can be deposited and can use much of the same wafer design. They have many issues, most dealing with lifetime and color shifting, but it works for HD. Kodak is spending a lot of effort on the technology, and I was allowed to use two of the OLED displays and gotta say they have a good design, but just need more help in the material science.
I'd say more, but my guild is raiding.
Check this out. Highlights (emphasis mine):
The cost of the $13 million, four-year project will be shared among NIST, GE and ECD Ovonics. The program goal is to create a cost-effective system for the mass production of products such as flexible electronic paper displays, portable TV screens the size of posters, embedded sensors, solar powered cells and high-efficiency lighting devices.
It isn't because it can't be done. It is because the Veruca Salt money managers on Wall Street can't tolerate a four year return on investment. GE's stock went down when they announced this.
Life is the leading cause of death in America.
So you need a red, green and blue to make one pixel just like a phospher display, and the circuitry to provide a precise analog current to all three LED chips and every pixel. And you have to avoid crosstalk between pixels.
Still, I use blue LEDs on my front panels a lot as simple indicators. They am kool. :-)
--- Ban humanity.
These devices are a more natural successor to the CRT based systems - excellect contrast and brightness (as good as existing CRTs), flat panel design, lower power consumption (as compared to plasma). The technology is essentially one cathode per pixel.
From IEEE Spectrum, September 2003:
Samsung and Motorola are aggressively pursuing field-emission display technology using nanotubes. Samsung, for example, has already demonstrated a full-color 38-inch field-emission display capable of handling normal video frame rates. What's more, a Japanese government-funded consortium was announced earlier this year to develop similar displays, and Sony Corp. (Tokyo) is developing its own nanotube display technology as well.
Just found something... OSRAM is producing a
LED-Package not to form a display as itself, but
rather a backlight for common displays (presumably
LCD, but they say nothing about it specifically).
The good thing they say is, that the colour
temperature can be adjusted with that technique.
As far as I am informed this is not the case with
current LCDs? Or am I wrong?
Could be somewhat in between the both technologies.
Sort of bringing "the best of both worlds" together.
Link to the Osram Website:
http://www.osram-os.com/news/news_multiled.html
Meme of the day: I browse "Disable Sigs: Checked". So should you.
Back when the MacII first came out, I realized that it could handle 2GB of ram, so I did some paper napkin calculations to figure out how to go about it... The state of the art for RAM were 1megabit chips, with a rated power consumption of 1watt each. I figured you could camolflage the memory unit as a desk... The entire top of the desk would contain cards of ram chips. One pedestal would be a cooling unit. The other would be a 16kilowatt power supply.
Free Software: Like love, it grows best when given away.
First, each led must have a wire on it. This limits the led to the size, that a ball will fit on. This is usually about 25 microns. (approx 1 mill)
;)
So now the led must be bigger than that, so now you would have a die, normally around 10 mills, to be nice and bright. Now space them out with say 7 mills in between, and do an array, with perfect dimensions between die, (remember, you have to have a spacing big enough to place the wire down to the pcb, and you will end up with a spacing of around 17 mill centers or so.
So you see, the pitch is rather large, then you add some optics, so you can make the whole thing look good, and a 1024 x 768 to be pretty clear.
Then throw in some optics to change the view, to a smaller pitch, and walla you have the viewing scope used on so many tanks
This has been built for along time, and is VERY expensive.
Why not use persistance of vision? Make your display out of three "arms" at 0, 120, and 240 degrees with one arm each for red, blue, and green.
Space the LEDs at less than 0.5mm or so to minimize the "empty" space between each circular line. You can offset the leds slightly if more room is needed or to help manage heat dissipation.
The three arms spin at 60RPS or so to get a decent refresh rate.
The downsides:
1. The circular display area will bother some folks.
2. The outer LEDs will need much more power than the inner ones because they will have to display more rixels (radial pixels) in the same amount of time at the same brightness as the inner LEDs..
3. A large moving part. (Don't move the TV while it's running.)
Go here to create your own Slashdot dis
The ROPOD.
"Sic Semper Tyrannosaurus Rex."
But first LEDs will be used as a backlight source. They are vastly more efficient than the most incandescent sources. The best incandescent lights can achieve an output of 40 lumens/watt and these tend to be halogen lamps, too hot for your lcd panel. Discharge lamps run at 50-60 lumens per watt but the colder versions (fluorescent tubes) produce horrible light. The best white LEDs can produce over 120 lumen/watt; vastly more efficient.
Combined with an infinitely variable colour (within the gamut of the LED) they make an exceptionally good light source, not only for LCDs but also in architectural situations. Already a large number mobile phones and some PDAs use a white LEDs as a backlight source. I'm sure laptop LCDs will be next.
OLEDs provide the next evolution in display technology and hopefully the next jump in efficiency. A black pixel is a very inefficient thing on an LCD panel, think of all that wasted light absorbed into the pixel, much better to be able to switch off the light for that pixel.
(Recapping what some other people have said but I hope I've said something new of note)
Thanks for all the responses!
I should mention that it has been my understanding (perhaps mistaken) that OLEDs are not actually diodes, but "devices", based on electroluminescent phenomena.
I confess I posted the Question mostly out of curiosity: the technology SEEMS so obvious....
I personally agree that reflective-light display technology is potentially inherently superior. My personal favorite (so far) can be found here.