And here is another example of transcendental numbers in music. This one was created in the early 1970's by Conlon Nancarrow, a composer who essentially created a parallel form of "electronic music" by using heavily modified player pianos instead of electronics. He was born just a little bit too early, but the concepts are extremely similar to what people do in sequencing.
This one is designed to have two melodies, one in which each measure takes pi inches of player piano roll and the other which takes e inches of player piano roll. The result is two melodies which will never align after the very beginning, using a true e/pi relationship.
Getting a Rhodes scholarship in college does not put you on the list of the smartest 50 people on the planet... if for no other reason than that there are 32 chosen every year. Assuming a modest lifespan for the recipients of 50 years, and assuming a Rhodes scholarship makes you smarter than anyone but another person with a Rhodes scholarship, you're still off by at least one order of magnitude!;-) Maaaybe top.0001%, but not.000001%!
I've met a lot of very smart people. The ones who *I* would classify as true geniuses (and I'm actually probably just barely qualified to judge) never bothered with things like the Rhodes. They had better things to do than do yet more school in England when there's so much exciting science to do!
Quantum tunneling was first theoretically understood in 1927, and since then it's just been a matter of engineering to take advantage of it. I think her point was that if it's taken 80 years to develop discoveries experimentally evaluated using relatively primitive and low-energy techniques, how much longer is it going to take to every apply something which requires the LHC just to observe. I agree with her, both as a physicist, and as an engineer. There are intrinsic difficulties in applying physical principles which require energy densities which approach that found in the Big Bang.
I don't agree that it means we shouldn't do it, because inquiring minds want to know. However, I do agree that duplicating effort in an attempt to discover things a few months sooner is more about scientist/politician pride than about sane expenditures of resources. If the LHC is the better piece of equipment, then mothball the Tevatron since they're nominally collecting similar data, except that the LHC uses better equipment. All that matters, as there are unlikely to be any national security/interest in the results, is that everyone has access to the data.
The reforming process to convert long chain hydrocarbons into methane is counterproductive, loses energy, and costs money. Why would you ever want do to this just for fuel? Everyone is trying so hard to go the other way =)
Methane is extremely chemically inert. The only value it has is as a fuel, so it's super cheap already. Going from a useful hydrocarbon that can be used as a chemical as well to a molecule that can only be burned is never going to be economically favorable.
So I actually am working on the technology to make hydrocarbons from air and water (it's actually what the caltech research was based on, although I'm a latecomer to the lab that developed the technology). I think it's a great idea, although for other reasons I think that it's not great -- after all, CO2 is only present at 300ppm in the air.
But no, I wasn't thinking as long term as fusion. I meant replacing all cars with a combination of electric short range rental vehicles on a model similar to a cell phone combined with electrified rail for transporting long distances, combined with hybrid or electric trucking for transportation from rail nodes to the distribution grid (i.e. the grocery store). That model would work for a very long time, and trucking is easier to upgrade than anything else.
I realize that the compressing process isn't that hard, but you aren't going to build a car that can run on both gasoline *and* CNG. If for no other reason than because the CNG tank is pressurized while the gasoline tank isn't. It's probably not as bad for mixing as ethanol (at least before they upgraded the gaskets, etc), but it probably still requires some rebuilding. I've never seen a car that said "feed me CNG, petrol, or diesel", so I suspect you are oversimplifying it a bit.
In any event, so long as you can't use the same car with either fuel, the chicken and egg problem exists. Until you have alternative fuel stations everywhere, no one will buy the cars, and no one will built those fuel stations until there are customers. The only way around this is massive and painful government subsidies to convince fueling stations to carry the alternative fuel, and it's much better to do that exactly one time, with exactly one fuel. Doing a simultaneous transition to either electric *or* CNG in the US is extremely unwise. It will cost twice as much, and offer only a minor change.
CNG is methane. Am I misunderstanding your comment? The article is commenting that methane is very cheap now, making it no longer economically interesting to replace natural gas power plants with wind. At the 2008 prices, wind was cheaper than natural gas power plants, making it advantageous to replace them and use the CNG you can produce from that natural gas in cars.
Cars are extremely good when it comes to pollution, unless you are referring to CO2 generation. I treat CO2 as a different problem than "air pollution" that resulted in acid rain, smog, and so on. The majority of actual air pollution is from coal power plants. You can track it coming across the Pacific ocean from China even.
But in any case, the drive to wind power isn't entirely political. In my opinion, if we're going to switch to anything other than gasoline, we may as well be switching straight to electric cars for local travel plus hybrids for trucking plus electric rail for long distance travel. It's going to be a *very* hard transition regardless of what we go to, so we may as well go all the way.
That would be part II of his plan -- replace natural gas power generation capacity with wind, and use the saved natural gas to replace oil as a transportation fuel. 1 gallon of saved compressed natural gas is 1 gallon of fuel for a vehicle. It was fairly sane in that respect, I just don't think CNG stands a chance of taking off in the US. It's extremely hard to transition to a new transportation fuel due to the well modeled chicken/egg problems with fueling stations. And if we're going to try to transition to a new fuel, better to pick something more long term than CNG.
Okay. I'd buy 50-60 percent as a maximum theoretical, I guess if you assume a carnot cycle since this is done at 1200-1400C most often.
The fischer-tropsch process was developed in 1926. It is extremely mature, and was used by Germany in WWII to produce nearly all of their gasoline and diesel from syngas. South Africa did similarly. In both cases it was for the same reason; they wanted to take gasified coal and convert it to liquid fuels because trade embargoes prevented importing those liquids.
The reaction is very mature, more mature in fact than alcohol synthesis from syngas. It is costly, but it's very well understood.
I'm not aware of any systems that are robust enough to be used commercially yet, but they aren't terribly far away. I would be surprised if it goes more than 2-3 more years before at least someone is doing it.
I know that this company is doing something related, although non-catalytic. It's some pretty ninja chemistry though.
What on earth are you talking about? The process of thermochemical reduction is going to be:
CO2 -> CO + O* (where * represents O absorbed by the supports change in oxidation state). O* -> O2 (oxygen released during temperature change and accompanying change in oxidation state).
Yeah, you can mix it with water as well, but why not just do them separately and produce CO and H2 in two tanks which can be combined to get whatever carbon number you want on average in your fuel after a one-step fischer-tropsch synthesis?
You have this thing going through methanol? Huh? These are all going to decrease your overall yield.
Next, you claim a 50-60% efficiency in converting to gasoline? Are you just making stuff up as you go along? The thermochemical cycle has a pretty decent amount of loss simply in the requirement that you cool down and heat up the precursors. Hard to move heat around without losses and all. Not to mention the chemical inefficiencies, plus regular thermal losses, plus mirror losses. 20% is an excellent estimate of the maximum photon to chemical conversion efficiency. Now, cost efficiency on the other hand... mirrors are way cheaper than PV panels, so in that regard you're talking about a 20-fold improvement to energy cost efficiency.
Disclaimer: I work on this professionally, so I'm biased.
Disclaimer: I work on this professionally, so I have a vested interest.
There are other well known cycles. Ceria is one of them. So is iron oxide, cobalt oxide, and a few others. Those are solids. I think the solids have a lot more potential than the gases. Ceria is actually what I did my PhD thesis on, and it's my favorite contender. We use it for water splitting and chemical reduction (the same thing they did at caltech), but I'm rather surprised their efficiency is so low. We get quite a lot higher, certainly far higher than solar PV + electrolysis, but the catalysts just don't last very long at high temperatures.
That said, I'm excited that if it's getting in the news, new or not, because it improves my odds of getting funding to use that tech. The more people working on it, the safer of a bet it will look like to funding agencies. It's a robust, efficient, and cheap technology that should be used everywhere. Just a question of who will solve the catalyst stability and reactor material problems first.
NPR? At least so long as you don't mind hearing from frustrated rational conservatives as well. I've found that the guests that I disagree with I find I can at least respect for their thoughtful analysis. Also, not guiding your programming by what will improve your ratings the most helps a lot towards making the discourse more informative and less bombastic.
I know that I, for one, would use something like this for parallel parking as well... this seems like a pretty useful gadget to install right above the bumper so you see where the back of your car actually is.
No offense taken. I'm aware of the competition. No harm in trying though =) Every club I've taken them to so far has been pretty interested in them, but really the target market in terms of clubs is a place without any existing lighting, not a club with $300k+ of lighting already. The advantages of being wifi controlled from a computer, having a flex arm to point it wherever you want, and free software to do music analysis and light control without requiring actually hiring a lighting engineer is significant... if you don't already own lots and lots of DMX based lighting.
When it comes down to it though, I can't sell them at a loss, and "Made in America" carries a high price tag, especially for low quantity goods. If I was making 1000 at a time in Chinese factories, yeah, I could make them $300 each. But I don't want to manufacture in China, I want to manufacture someplace where employees don't commit suicide due to low pay and high stress. I figure it's the least I can do as an American citizen.
In any case, you're not the target market quite yet.
Step 1: Build hobbyist light in small quantities, targeted for people with minimal electronics experience and at clubs without existing light infrastructure. Step 2: Build DMX -> OSC converter and stabilize software so that it is able to control DMX lights as well. Step 3: Build large enough quantities to get the price down to the $400-500 ballpark. Step 4: Market to clubs like the one you presumably work for.
And once I am at that point, I can hopefully just give the lights away to hackers to do cool projects with =)
I actually have lots of recommendations for doing cheap controllers for lower power LEDs =) I helped an extremely good friend of mine design several LED displays and have done a few myself. His design is extremely cool, although it took a lot of labor to do:
Highly, highly, highly recommend the TLC5940 (or the most recent iteration in that line). Does onboard 12-bit PWM, 16 channels, and open drain so you can in principle run an arbitrary number of LEDs on it so long as the voltage difference between the last LED and the chip isn't too high. I don't like it for stuff like this spotlight because it's limited to 100mA per channel, and is a linear regulator. If I used something like that for my lights, they'd burn out really, really fast. However, for a LED strip, it's *perfect*.
You get to offload all the PWM that wastes CPU cycles better spent on harder tasks, operates off of a simple SPI bus, has 12-bit resolution, don't have to worry about aliasing the buck converter frequency with your PWM signal, and you can daisy chain them together to allow you to operate I think as many as 16*16 lights off of one SPI bus. It's crazy.
I have circuit designs for it if you'd like to work from them as an example. Just send me a note either on the saiko-dev@saikoled.com development list, or directly to me at neltnerb@saikoled.com. I also know lots of other people interested in similar projects. Here's one that Xander is doing that looks very similar to what you're describing:
Oh... um, send me a note offline and I can help you design a LM3404HV that will take up to 72VDC and provide a stable 700mA output. National Semiconductor has some nice simulation tools to deal with it. Did you try using those and were still unable to get it to work well? I'd also mention that the current varying between 650 and 750mA depending on input voltage isn't *that* bad of a scenario. Generally an end user can't tell the difference anyway, especially with the dramatically different LED efficiency bins you can buy. I went with the most efficient LEDs because I like nice things =)
Slashdotters are always critical, especially of MIT alums. I'm used to failing at things, whereas most slashdotters don't bother doing something unless they're guaranteed success. *shrug* Guess who's more likely to actually make something useful in the long run? God knows I learn an awful lot more from taking on a lot of very hard challenges and doing my best, succeed or fail, than any armchair engineer. Despite being a physicist and material scientist working on catalysts for renewable energy, I had no problem jumping into and being wildly successful with designing ultra-high frequency digitally controlled PID feedback loops (literally a 5MHz response time), building these and similar LED lights, and so on. I just didn't have the pleasure of taking classes on it, or working at a company doing electronics design so I had to teach it all to myself over the last four years. I think that regardless of what armchair engineers might think, teaching yourself something in your spare time and doing it well enough to produce useful things is not bringing shame on the institvte. The funniest comments are from the people bashing my EE skills. Usually, the solutions they lay out are not workable in a reasonable real world system, once you think it all through. It's sort of like those hilarious comments where someone brings up "what if gravity doesn't work" in a thread about GR. Seriously? Or mentioning basic statistical analysis and sample bias as the "reason" why a 3 year study is wrong. Gee, who do you think knows more about statistical analysis... I can't imagine they tried to account for things that you learn about in high school math...
Anyway, I am serious about collaborating... at the least, my site has some publicity now, would be happy to help sell things, and would be happy to be a sounding board for new ideas as well. I personally checked out the existing hobbyist offerings at sparkfun and also found them to be overpriced and annoying to use. However, there is definitely a place for discrete controllers and chained rebel star boards. It's just not in a quasi-professional lighting fixture that I'm (actually) expecting to make a lot more money from selling to clubs than to hobbyists.
You know what would be awesome -- check out the new 40W RGBA LEDs from ledengin. That's what I plan to use in my next fixture, after I make back enough money (hopefully) from this to design new stuff. I want to put three of them in each light, and use the same power electronics. The hard part is fitting those huge optics and dealing with heat sinking... I might end up having to use active cooling if I want it in the same 5" cylinder form factor (which I do!). But, if I can solve that, which I think I can, I have a 3000 lumen fixture in a tiny size! Awesomeness. Naturally, too expensive for slashdot to approve; surely they could do it for $40 due to my idiocy, but whatever... daddy needs new brake lights for his car...
As the designer of this light fixture, I would love to have you participate in the project... I used the LM3404 for PWM control because of heat sinking issues, but I do prefer using an external chip for PWM signal generation. For this project though it just seemed like overkill since the Arduino has PWM outputs set up already.
I'd be happy to try to work with you to either drop the price of our stuff, make your stuff work more effectively, and so on. It's supposed to be an open project after all =)
Competition in open hacker projects is kind of silly. Pool resources, etc, etc. Right now the reason that board is so expensive is largely because I'm not making quantities large enough to bring down the price.
That is not true. It uses the aubio audio analysis library for beat detection, and uses puredata to make it easily extensible by anyone not scared of graphical programming languages.
This is not a light organ, or a VU meter.
The point is that *anyone* can go in and improve the puredata code to make it more awesome.
Actually, FWIW I'm going to use the 40W version of that in the next version of this light. The cooling is a *serious* problem. Good luck. I doubt you can even turn the 10W one on for more than about 20 seconds before it melts.
Am looking forward to tackling it once I have some funding again though =) It should end up being about a 3000 lumen RGB fixture, but it's going to cost a lot more for me to make it. Have to switch to 1/2" aluminum back plate, put on fans potentially, etc, etc.
Next project is to build a DMX -> OSC converter. It won't be terribly hard, but we're all busy with day jobs.
If you'd like to help, feel free to join the development mailing list. The plan was to do it using a maple board, but FPGA would probably be simpler and cheaper.
The spasticity is entirely controlled by an open source puredata patch. Even a non-programmer could tune it until they like the behavior. I happen to like it as is =)
Feel free to join the music analysis development list at http://saikoled.com/mailinglists/ if you want to make it work better. You'll find that it's designed to make the barrier to entry minimal. I literally wrote the music analysis software in 45 minutes after the python server was done.
As the person who made that design choice, it was because I want to make it accessible to hobbyists with minimal electronics experience. I'm not a moron, I have done work with FPGAs, and plenty of work with integrated atmels and PICs (as you can see from my older projects at http://led-artwork.com/
Don't worry, if I was targeting clubs I'd do it all on an FPGA with a built in network stack and hardware PWM outputs. But that's a lot harder, and no beginners would be able to use it. We probably still *will* do that. But the development time is a lot greater, and I'd rather help beginners make some cool stuff over making money. Development time is a pretty serious constraint given that this is a project done as a hobby... I still have a 40-60 hour a week "real" job, so if I can do it with an arduino in 100 hours versus an FPGA in 200 hours, it's a no brainer at present.
Slashdot Mirror
Oh, we'll just engineer them to require lysine to reproduce. We'll control the supply, so if they get out of control they'll gradually die off.
Easy peasy!
And here is another example of transcendental numbers in music. This one was created in the early 1970's by Conlon Nancarrow, a composer who essentially created a parallel form of "electronic music" by using heavily modified player pianos instead of electronics. He was born just a little bit too early, but the concepts are extremely similar to what people do in sequencing.
http://www.youtube.com/watch?v=BXFiq19-KSE
This one is designed to have two melodies, one in which each measure takes pi inches of player piano roll and the other which takes e inches of player piano roll. The result is two melodies which will never align after the very beginning, using a true e/pi relationship.
I do agree that Clinton is a bright guy... but...
Getting a Rhodes scholarship in college does not put you on the list of the smartest 50 people on the planet... if for no other reason than that there are 32 chosen every year. Assuming a modest lifespan for the recipients of 50 years, and assuming a Rhodes scholarship makes you smarter than anyone but another person with a Rhodes scholarship, you're still off by at least one order of magnitude! ;-) Maaaybe top .0001%, but not .000001%!
I've met a lot of very smart people. The ones who *I* would classify as true geniuses (and I'm actually probably just barely qualified to judge) never bothered with things like the Rhodes. They had better things to do than do yet more school in England when there's so much exciting science to do!
Actually, you're just demonstrating her point.
https://secure.wikimedia.org/wikipedia/en/wiki/Quantum_tunnelling
Quantum tunneling was first theoretically understood in 1927, and since then it's just been a matter of engineering to take advantage of it. I think her point was that if it's taken 80 years to develop discoveries experimentally evaluated using relatively primitive and low-energy techniques, how much longer is it going to take to every apply something which requires the LHC just to observe. I agree with her, both as a physicist, and as an engineer. There are intrinsic difficulties in applying physical principles which require energy densities which approach that found in the Big Bang.
I don't agree that it means we shouldn't do it, because inquiring minds want to know. However, I do agree that duplicating effort in an attempt to discover things a few months sooner is more about scientist/politician pride than about sane expenditures of resources. If the LHC is the better piece of equipment, then mothball the Tevatron since they're nominally collecting similar data, except that the LHC uses better equipment. All that matters, as there are unlikely to be any national security/interest in the results, is that everyone has access to the data.
The reforming process to convert long chain hydrocarbons into methane is counterproductive, loses energy, and costs money. Why would you ever want do to this just for fuel? Everyone is trying so hard to go the other way =)
Methane is extremely chemically inert. The only value it has is as a fuel, so it's super cheap already. Going from a useful hydrocarbon that can be used as a chemical as well to a molecule that can only be burned is never going to be economically favorable.
So I actually am working on the technology to make hydrocarbons from air and water (it's actually what the caltech research was based on, although I'm a latecomer to the lab that developed the technology). I think it's a great idea, although for other reasons I think that it's not great -- after all, CO2 is only present at 300ppm in the air.
But no, I wasn't thinking as long term as fusion. I meant replacing all cars with a combination of electric short range rental vehicles on a model similar to a cell phone combined with electrified rail for transporting long distances, combined with hybrid or electric trucking for transportation from rail nodes to the distribution grid (i.e. the grocery store). That model would work for a very long time, and trucking is easier to upgrade than anything else.
I realize that the compressing process isn't that hard, but you aren't going to build a car that can run on both gasoline *and* CNG. If for no other reason than because the CNG tank is pressurized while the gasoline tank isn't. It's probably not as bad for mixing as ethanol (at least before they upgraded the gaskets, etc), but it probably still requires some rebuilding. I've never seen a car that said "feed me CNG, petrol, or diesel", so I suspect you are oversimplifying it a bit.
In any event, so long as you can't use the same car with either fuel, the chicken and egg problem exists. Until you have alternative fuel stations everywhere, no one will buy the cars, and no one will built those fuel stations until there are customers. The only way around this is massive and painful government subsidies to convince fueling stations to carry the alternative fuel, and it's much better to do that exactly one time, with exactly one fuel. Doing a simultaneous transition to either electric *or* CNG in the US is extremely unwise. It will cost twice as much, and offer only a minor change.
CNG is methane. Am I misunderstanding your comment? The article is commenting that methane is very cheap now, making it no longer economically interesting to replace natural gas power plants with wind. At the 2008 prices, wind was cheaper than natural gas power plants, making it advantageous to replace them and use the CNG you can produce from that natural gas in cars.
Cars are extremely good when it comes to pollution, unless you are referring to CO2 generation. I treat CO2 as a different problem than "air pollution" that resulted in acid rain, smog, and so on. The majority of actual air pollution is from coal power plants. You can track it coming across the Pacific ocean from China even.
But in any case, the drive to wind power isn't entirely political. In my opinion, if we're going to switch to anything other than gasoline, we may as well be switching straight to electric cars for local travel plus hybrids for trucking plus electric rail for long distance travel. It's going to be a *very* hard transition regardless of what we go to, so we may as well go all the way.
That would be part II of his plan -- replace natural gas power generation capacity with wind, and use the saved natural gas to replace oil as a transportation fuel. 1 gallon of saved compressed natural gas is 1 gallon of fuel for a vehicle. It was fairly sane in that respect, I just don't think CNG stands a chance of taking off in the US. It's extremely hard to transition to a new transportation fuel due to the well modeled chicken/egg problems with fueling stations. And if we're going to try to transition to a new fuel, better to pick something more long term than CNG.
Okay. I'd buy 50-60 percent as a maximum theoretical, I guess if you assume a carnot cycle since this is done at 1200-1400C most often.
The fischer-tropsch process was developed in 1926. It is extremely mature, and was used by Germany in WWII to produce nearly all of their gasoline and diesel from syngas. South Africa did similarly. In both cases it was for the same reason; they wanted to take gasified coal and convert it to liquid fuels because trade embargoes prevented importing those liquids.
The reaction is very mature, more mature in fact than alcohol synthesis from syngas. It is costly, but it's very well understood.
I'm not aware of any systems that are robust enough to be used commercially yet, but they aren't terribly far away. I would be surprised if it goes more than 2-3 more years before at least someone is doing it.
I know that this company is doing something related, although non-catalytic. It's some pretty ninja chemistry though.
http://www.sundropfuels.com/
What on earth are you talking about? The process of thermochemical reduction is going to be:
CO2 -> CO + O* (where * represents O absorbed by the supports change in oxidation state).
O* -> O2 (oxygen released during temperature change and accompanying change in oxidation state).
Yeah, you can mix it with water as well, but why not just do them separately and produce CO and H2 in two tanks which can be combined to get whatever carbon number you want on average in your fuel after a one-step fischer-tropsch synthesis?
You have this thing going through methanol? Huh? These are all going to decrease your overall yield.
Next, you claim a 50-60% efficiency in converting to gasoline? Are you just making stuff up as you go along? The thermochemical cycle has a pretty decent amount of loss simply in the requirement that you cool down and heat up the precursors. Hard to move heat around without losses and all. Not to mention the chemical inefficiencies, plus regular thermal losses, plus mirror losses. 20% is an excellent estimate of the maximum photon to chemical conversion efficiency. Now, cost efficiency on the other hand... mirrors are way cheaper than PV panels, so in that regard you're talking about a 20-fold improvement to energy cost efficiency.
Disclaimer: I work on this professionally, so I'm biased.
Disclaimer: I work on this professionally, so I have a vested interest.
There are other well known cycles. Ceria is one of them. So is iron oxide, cobalt oxide, and a few others. Those are solids. I think the solids have a lot more potential than the gases. Ceria is actually what I did my PhD thesis on, and it's my favorite contender. We use it for water splitting and chemical reduction (the same thing they did at caltech), but I'm rather surprised their efficiency is so low. We get quite a lot higher, certainly far higher than solar PV + electrolysis, but the catalysts just don't last very long at high temperatures.
That said, I'm excited that if it's getting in the news, new or not, because it improves my odds of getting funding to use that tech. The more people working on it, the safer of a bet it will look like to funding agencies. It's a robust, efficient, and cheap technology that should be used everywhere. Just a question of who will solve the catalyst stability and reactor material problems first.
NPR? At least so long as you don't mind hearing from frustrated rational conservatives as well. I've found that the guests that I disagree with I find I can at least respect for their thoughtful analysis. Also, not guiding your programming by what will improve your ratings the most helps a lot towards making the discourse more informative and less bombastic.
I know that I, for one, would use something like this for parallel parking as well... this seems like a pretty useful gadget to install right above the bumper so you see where the back of your car actually is.
No offense taken. I'm aware of the competition. No harm in trying though =) Every club I've taken them to so far has been pretty interested in them, but really the target market in terms of clubs is a place without any existing lighting, not a club with $300k+ of lighting already. The advantages of being wifi controlled from a computer, having a flex arm to point it wherever you want, and free software to do music analysis and light control without requiring actually hiring a lighting engineer is significant... if you don't already own lots and lots of DMX based lighting.
When it comes down to it though, I can't sell them at a loss, and "Made in America" carries a high price tag, especially for low quantity goods. If I was making 1000 at a time in Chinese factories, yeah, I could make them $300 each. But I don't want to manufacture in China, I want to manufacture someplace where employees don't commit suicide due to low pay and high stress. I figure it's the least I can do as an American citizen.
In any case, you're not the target market quite yet.
Step 1: Build hobbyist light in small quantities, targeted for people with minimal electronics experience and at clubs without existing light infrastructure.
Step 2: Build DMX -> OSC converter and stabilize software so that it is able to control DMX lights as well.
Step 3: Build large enough quantities to get the price down to the $400-500 ballpark.
Step 4: Market to clubs like the one you presumably work for.
And once I am at that point, I can hopefully just give the lights away to hackers to do cool projects with =)
I actually have lots of recommendations for doing cheap controllers for lower power LEDs =) I helped an extremely good friend of mine design several LED displays and have done a few myself. His design is extremely cool, although it took a lot of labor to do:
http://web.mit.edu/dgrnbrg/www/Eye_Of_Gorlack.html
Highly, highly, highly recommend the TLC5940 (or the most recent iteration in that line). Does onboard 12-bit PWM, 16 channels, and open drain so you can in principle run an arbitrary number of LEDs on it so long as the voltage difference between the last LED and the chip isn't too high. I don't like it for stuff like this spotlight because it's limited to 100mA per channel, and is a linear regulator. If I used something like that for my lights, they'd burn out really, really fast. However, for a LED strip, it's *perfect*.
You get to offload all the PWM that wastes CPU cycles better spent on harder tasks, operates off of a simple SPI bus, has 12-bit resolution, don't have to worry about aliasing the buck converter frequency with your PWM signal, and you can daisy chain them together to allow you to operate I think as many as 16*16 lights off of one SPI bus. It's crazy.
I have circuit designs for it if you'd like to work from them as an example. Just send me a note either on the saiko-dev@saikoled.com development list, or directly to me at neltnerb@saikoled.com. I also know lots of other people interested in similar projects. Here's one that Xander is doing that looks very similar to what you're describing:
http://www.synopticlabs.com/blog/
Anyway, I'm not a big enough fan of slashdot to post much more on here; feel free to send email though.
Oh... um, send me a note offline and I can help you design a LM3404HV that will take up to 72VDC and provide a stable 700mA output. National Semiconductor has some nice simulation tools to deal with it. Did you try using those and were still unable to get it to work well? I'd also mention that the current varying between 650 and 750mA depending on input voltage isn't *that* bad of a scenario. Generally an end user can't tell the difference anyway, especially with the dramatically different LED efficiency bins you can buy. I went with the most efficient LEDs because I like nice things =)
Slashdotters are always critical, especially of MIT alums. I'm used to failing at things, whereas most slashdotters don't bother doing something unless they're guaranteed success. *shrug* Guess who's more likely to actually make something useful in the long run? God knows I learn an awful lot more from taking on a lot of very hard challenges and doing my best, succeed or fail, than any armchair engineer. Despite being a physicist and material scientist working on catalysts for renewable energy, I had no problem jumping into and being wildly successful with designing ultra-high frequency digitally controlled PID feedback loops (literally a 5MHz response time), building these and similar LED lights, and so on. I just didn't have the pleasure of taking classes on it, or working at a company doing electronics design so I had to teach it all to myself over the last four years. I think that regardless of what armchair engineers might think, teaching yourself something in your spare time and doing it well enough to produce useful things is not bringing shame on the institvte. The funniest comments are from the people bashing my EE skills. Usually, the solutions they lay out are not workable in a reasonable real world system, once you think it all through. It's sort of like those hilarious comments where someone brings up "what if gravity doesn't work" in a thread about GR. Seriously? Or mentioning basic statistical analysis and sample bias as the "reason" why a 3 year study is wrong. Gee, who do you think knows more about statistical analysis... I can't imagine they tried to account for things that you learn about in high school math...
Anyway, I am serious about collaborating... at the least, my site has some publicity now, would be happy to help sell things, and would be happy to be a sounding board for new ideas as well. I personally checked out the existing hobbyist offerings at sparkfun and also found them to be overpriced and annoying to use. However, there is definitely a place for discrete controllers and chained rebel star boards. It's just not in a quasi-professional lighting fixture that I'm (actually) expecting to make a lot more money from selling to clubs than to hobbyists.
You know what would be awesome -- check out the new 40W RGBA LEDs from ledengin. That's what I plan to use in my next fixture, after I make back enough money (hopefully) from this to design new stuff. I want to put three of them in each light, and use the same power electronics. The hard part is fitting those huge optics and dealing with heat sinking... I might end up having to use active cooling if I want it in the same 5" cylinder form factor (which I do!). But, if I can solve that, which I think I can, I have a 3000 lumen fixture in a tiny size! Awesomeness. Naturally, too expensive for slashdot to approve; surely they could do it for $40 due to my idiocy, but whatever... daddy needs new brake lights for his car...
Email: neltnerb@saikoled.com
As the designer of this light fixture, I would love to have you participate in the project... I used the LM3404 for PWM control because of heat sinking issues, but I do prefer using an external chip for PWM signal generation. For this project though it just seemed like overkill since the Arduino has PWM outputs set up already.
I'd be happy to try to work with you to either drop the price of our stuff, make your stuff work more effectively, and so on. It's supposed to be an open project after all =)
Competition in open hacker projects is kind of silly. Pool resources, etc, etc. Right now the reason that board is so expensive is largely because I'm not making quantities large enough to bring down the price.
Actually, we posted it on instructables as well, so technically, it is exactly as powerful as the most powerful DIY LED project on there ;-)
http://www.instructables.com/id/Ultra-bright-LED-Color-Changing-Spotlight-using-Op/
Cheers,
That is not true. It uses the aubio audio analysis library for beat detection, and uses puredata to make it easily extensible by anyone not scared of graphical programming languages.
This is not a light organ, or a VU meter.
The point is that *anyone* can go in and improve the puredata code to make it more awesome.
Just needs some cooling and PWM and there you go.
Heh. That's cute.
Actually, FWIW I'm going to use the 40W version of that in the next version of this light. The cooling is a *serious* problem. Good luck. I doubt you can even turn the 10W one on for more than about 20 seconds before it melts.
Am looking forward to tackling it once I have some funding again though =) It should end up being about a 3000 lumen RGB fixture, but it's going to cost a lot more for me to make it. Have to switch to 1/2" aluminum back plate, put on fans potentially, etc, etc.
Next project is to build a DMX -> OSC converter. It won't be terribly hard, but we're all busy with day jobs.
If you'd like to help, feel free to join the development mailing list. The plan was to do it using a maple board, but FPGA would probably be simpler and cheaper.
The spasticity is entirely controlled by an open source puredata patch. Even a non-programmer could tune it until they like the behavior. I happen to like it as is =)
Feel free to join the music analysis development list at http://saikoled.com/mailinglists/ if you want to make it work better. You'll find that it's designed to make the barrier to entry minimal. I literally wrote the music analysis software in 45 minutes after the python server was done.
As the person who made that design choice, it was because I want to make it accessible to hobbyists with minimal electronics experience. I'm not a moron, I have done work with FPGAs, and plenty of work with integrated atmels and PICs (as you can see from my older projects at http://led-artwork.com/
Don't worry, if I was targeting clubs I'd do it all on an FPGA with a built in network stack and hardware PWM outputs. But that's a lot harder, and no beginners would be able to use it. We probably still *will* do that. But the development time is a lot greater, and I'd rather help beginners make some cool stuff over making money. Development time is a pretty serious constraint given that this is a project done as a hobby... I still have a 40-60 hour a week "real" job, so if I can do it with an arduino in 100 hours versus an FPGA in 200 hours, it's a no brainer at present.