(a) It's naturally radioactive. Also, from TFA: "Coauthors of the present paper learned of a new water-filled fracture inside a South African gold mine near the Johannesburg metropolitan area and viewed it as an opportunity to study subsurface rock uncontaminated by human activities."
(b) It's not practical to use its radioactivity as a power source, however, because it's only mildly radioactive in the natural state; said another way, it's not appreciably warm, so the amount of heat given off of natural uranium due to its radioactivity is negligible.
(c) Most (nearly all) human-generated nuclear waste has the same answer as (b); of that that is appreciably warm, there's too little of it to be useful as a power source.
(d) You got it.
Note that the bacteria do not use radioactivity directly, but rather use hydrogen from their environment, made from decomposing water exposed to radioactivity, as an energy source. Again from TFA: "This fracture water contained hydrocarbons and hydrogen not likely to have been created through biological processes, but rather from decomposition of water exposed to radiation from uranium-bearing rocks."
It's always been our position that if enough people go on the air with their stations, the FCC will be overwhelmed and unable to respond.
...and what will this nice gentleman do when a second pirate interferes with his pirate station, due to ideological differences or just to get more advertising revenue? Buy a bigger transmitter? The FCC was created in 1934 specifically to bring sanity to this wild-west, most-powerful-transmitter-wins warfare.
The present-day record for car pacing is held by Fred Rompelberg of Holland, who reached 268.831 km/h (167.043 mph) on a bicycle following a car at the Bonneville Salt Flats, Utah, on October 3, 1995.
Interestingly, he set the record 27 days shy of his 50th birthday, so there does seem to be a trend here....
What you say is true, but HP and Varian were in Silicon Valley as second-generation startups.
The first generation was the Federal Telegraph Company, founded by Cyril Elwell in the winter of 1909-1910 (originally as the Poulsen Wireless Telephone and Telegraph Company). Never heard of it? It was built to commercialize the arc (not spark) transmitter developed by Valdemar Poulsen, and by 1918 had succeeded in building and operating 1-megawatt continuous-wave radio transmitters.
Elwell was not only a Stanford graduate, he got his first financing for the company from Stanford faculty members, including the president of the university. So it can truly be said that Stanford itself acted as the first venture capitalist for the Valley.
Like the startups to follow, Federal people often left to do great things:
--Since it needed receivers to go with its transmitters, Federal hired a man from New York to develop a receiver for it, and set him up in a laboratory in the bay area. There, Lee DeForest would invent the triode vacuum tube (valve).
--To transfer the arc transmitter technology from Denmark to the Valley, Poulsen sent some of his employees with the equipment. One quickly became disillusioned with Federal, but liked the Valley, and started working with speakers. Soon thereafter, Peter Jensen formed his own company, Magnavox. Jensen's name lives on today in several lines of audio products.
--Leonard Fuller, longtime chief engineer of Federal, eventually ended up on the Berkeley faculty. The story goes that one day during the Great Depression, he was sitting in the faculty cafeteria when Ernest O. Lawrence was complaining that his cyclotron research was limited by the size of magnetic pole pieces he could obtain. Fuller realized that the 1-megawatt arc transmitter Federal had designed had very, very large magnetic pole pieces and, as they were too heavy (80 tons) to scrap, several had been sitting unused in a Valley warehouse since the end of World War I. A donation was quickly arranged, and the unused Federal components came to play a significant part in the development of large particle accelerators.
In addition to their work on the microbial fuel cells themselves, they've also made the interesting discovery that the bacteria naturally form nanowires to transfer electrons outside the cell--something potentially [sorry!] useful to connect to an external electrode.
To sum up, the rationale for the Japanese to work on a supersonic transport is based on three assumptions:
1. The scramjet engine will reduce operating (read: fuel) costs per average passenger mile significantly below that of the Concorde (by supporting a larger plane and being more fuel-efficient at cruise),
2. The plane will be capable of nonstop trans-Pacific flight (an ability also largely due to the fuel-efficiency of the scramjet), and
3. The much longer trans-Pacific flights in which the Japanese are interested will more dramatically show the time-of-arrival advantage of the supersonic plane than the shorter trans-Atlantic flights of the Concorde, and make it more appealing to seat-weary passengers.
I suppose there is also a fourth assumption, that cheap, fast, trans-Pacific travel would greatly improve the national economy of Japan in general and the Japanese aircraft industry in particular. This is the reason the Japanese government is expressing interest.
Whether these assumptions turn out to be factual or not requires research, which the Japanese are now doing.
I now return you to your previously-scheduled discussion, already in progress.
The key point not brought out in TFA is that the rainfall prediction scheme is not based on the link from the handset to the cell tower, but on the wireless backhaul links of the cellular system. The backhaul link is the link from the cell tower to the rest of the world (or at least the phone system of the rest of the world)--in many places in the world it is fiber or some other line, but increasingly often it, too, is wireless, using something called digital fixed radio systems (DFRS; check out standard EN 301 751 at ETSI).
The wireless backhaul links are much better for the meterological application than the handset link, because: (a) It's a fixed link; since the cell towers don't move, like the handsets do, the location of the link, and therefore the rain, is known, and (b) It's at a much higher frequency. The DFRS links used in this paper are at 8-23 GHz, much higher than the 0.8-1.9 GHz (depending on your local regulatory environment) of the handset link. This is important because rain attenuation increases as the signal frequency increases; it would be quite difficult to reliably detect rain fades at the handset frequencies (although in a bad enough storm--a cyclone comes to mind--it's probably possible; TFA notes the anecdotal evidence of fading television signals in bad weather).
I note in passing that the web-based supplimental material to the article references a US patent application, # 60/698,491.
You were doing well until you repeated that old hybrid-EMT scare. Any first responders that were afraid to approach a hybrid weren't well informed in their profession. I can't speak for Honda, but not only did Toyota work with national first-responder organizations to get their comments on the design of the US model, it made presentations on its design at their national conventions, made publications about it in the trade press, and distributed literature about the car freely and widely. The locations of the high-voltage elements of the car have been available on the web since time immemorial, and Toyota, at least, spent a lot of time repeating over and over that there's no high voltage in the roof pillars (how do these rumors get started?!?).
Both Toyota and Honda were and are exquisitely well-aware of accident procedures involving their cars; that's why the high-voltage lines in the Prius are armored International Orange cables isolated from the ground of the chassis, surrounded by identified conduit, and centered under the car floor, where the jaws of life and other EMT tools are least likely to be used. The battery itself is placed in the statistically safest place in the car (just over the rear axle), and protects first responders by an accelerometer-based circuit breaker, a Ground Fault Interrupter, and interlocks. Criminy, what do you want?
I think equations like this are how the myth got started in the first place. The equation does not describe what it claims to describe ("free-space loss"), because it makes the (unstated) assumption that constant-gain antennas are used. It then combines the free-space loss (the "proportional to the square of the distance" term, caused by the increasing area of the sphere enclosing the transmitter as one moves away from it) with the antenna term (the "proportional to the square of the frequency" term), to get some sort of combination that in fact can be used for communication link analysis if you're using constant-gain antennas. Defining some terms (trust me, I'll be gentile),
Umax = maximum radiation intensity (i.e., the maximum watts/unit area of a sphere enclosing a (transmitting) antenna in some preferred direction).
Uavg = average radiation intensity (i.e., the average watts/unit area of a sphere enclosing a (transmitting) antenna) = 4pi/(power radiated).
D = directivity = Umax/Uavg.
G = antenna gain = D (for antennas that are 100% efficient, a reasonable approximation for most cases with which we're concerned here).
A = effective area = (power produced at the antenna terminals)/(power density of the incident received wave); e.g., watts/(watts/square meter) = square meters.
wavelength = (speed of light)/(frequency); e.g., wavelength in meters = 300/(frequency in MHz).
You may have notice a gear-shift in moving from the definition of directivity to that of effective area--I defined directivity using a transmitting antenna, but effective area in terms of a receiving antenna. Fortunately, the terms are reciprocal; both terms apply equally well to both transmitting and receiving antennas.
The crux of the matter, and the source of the frequency term in the "free-space" equation, is the next equation. The directivity D (and therefore the gain G, under our assumption of 100% efficiency) of any antenna is related to its effective area A and the wavelength of the incoming wave:
D = G = A*4pi/(wavelength)^2 = A*4pi*(frequency)^2/(speed of light)^2.
If we assume that the antenna gain G is constant then, as the wavelength (i.e., frequency) changes, its effective area must, too. The gain of a dipole is constant at 2.14 dB, relative to an isotropic source; its effective area, however, is (30/73pi)*(wavelength)^2, or about 0.13(wavelength)^2. However, if we assume that the antenna is a different type, with constant area A, then as the wavelength changes its gain must change. Either assumption is valid, depending on the type of antenna we are employing. The confusion arises when people use the Wikipedia "free-space" equation to model path loss, then want to experiment with different antenna types. They don't realize that they're modeling their antennas as part of the "free-space" path loss. Wild errors result, for exampe, if you use Wikipedia's equation with parabolic dish antennas, like you often do in microwave point-to-point systems.
I guess a more subtle and pervasive problem arises when the equation gives people the idea that there's something inherently "bad" about propagation at higher frequencies, and that we should therefore all fight for operation at lower frequencies. Which, I guess, is what so motivates my rant.
In summary, you're right--frequency shouldn't come into this at all, and the second term of the equation in Wikipedia should be removed.
p.s.: And I thought my starlight/sunlight explanation was bulletproof. Maybe I should use flashlights? Lighning bugs?
Pager batteries last a long time because, while both pagers and cell phones spend most of their time asleep (neither transmitting nor receiving), as you suggest the pager can tolerate higher latency than a cell phone. People tolerate page latency of 30 seconds or more; it's a store-and-forward system. Trying to keep a caller on the line for 30 seconds while a cell phone call is set up (in addition to other inherent network delays) is another thing entirely. Having the luxury of long latency enables low pager system duty cycles, and therefore low average power consumption and long battery life.
It is also true that cellular protocols are far more complex than paging protocols (cf. the GSM spec with, say, POCSAG or FLEX), and the hardware to implement them requires more power consumption when active. Not only is the physical layer more complex (and often at a higher frequency), but there is additional decryption, data deinterleaving, etc. required, too. Phone manufacturers have been largely (if not totally) unable to justify the additional cost of a separate, simple, paging-like receiver in phones to improve their standby time, although a search of the USPTO records will indicate that lots of people have considered the idea.
Finally, it is also true that the need to include a relatively high-powered transmitter in the phone requires some compromises in standby power consumption. For example, to get high efficiency from a transmitter power amplifier one likes to have a high supply voltage, fed from a source of low output impedance (internal resistance) and high capacity. Ergo, lithium batteries for maximum talk time. However, the receiver has no such requirement; in fact, its power consumption drops with the supply voltage. (This is another reason why pagers ran from a single 1.5V AAA cell.) The cell phone, however, can't operate from separate transmit and receive batteries--I, at least, wouldn't buy one--so the compromise is that the receiver either runs from the high voltage of the lithium battery directly (an unappetizing choice), runs from the low-voltage output of a linear voltage regulator supplied by the battery (only slightly more appetizing, since the system efficiency is still poor), or runs from the low-voltage output of a switching voltage regulator (possibly the best choice, if the electrical noise can be tolerated, even though the regulator still adds power consumption). Or some combination of the above, any of which is still less power-efficient than running from a low-voltage source in the first place.
The RF signal is typically a sine wave generated by a VCO (voltage-controlled oscillator) that drives the first stage of the divider (prescaler). At the high frequencies used by RF synthesizers, the output pulses of the first stage aren't square, either. They're a sort of a mush; a compromise between the power consumption of your concern and the desire to use CMOS logic for its other low-power features in a highly-integrated system. The first stage of the divider typically draws half or more of the total power, but it is only driven hard enough to enable reliable CMOS operation, not produce square waves. As you go down the divider, the signals gradually move from the sine wave of the input to asymptotically approach a square wave at the output.
I'm considering the devotion of the rest of my professional career to the eradication of the "propagation loss increases with frequency" myth.
Repeat after me:
Propagation loss does not increase with frequency! Propagation loss does not increase with frequency! Propagation loss does not increase with frequency!
Think about it: If the propagation loss of an electromagnetic wave increased in proportion to its frequency, there would be so much so much attenuation at the THz frequency of light that we'd never see sunlight--or stars. Propagation loss is independent of frequency, except for scattering due to molecular and atomic resonances that are insignificant at the frequencies we're discussing. (There are also changes in scattering behavior that become relevant in indoor applications, like propagation around corners.)
What is dependent on frequency, however, is the performance of the antennas we use to transmit and receive electromagnetic waves. Antennas can be characterized by a parameter called effective area. Returning to the sunlight example, recognize that the output power of a solar panel is proportional to its physical area; the larger this area, the greater the fraction of the incident power transmitted by the sun is received by the solar panel and converted to available output power. Receiving antennas, and antennas in general (even wire antennas), have an effective area; it's the area required to produce the measured output power, based on the density of transmitted power (watts/unit area) at the location of the receiving antenna.
Antennas can also be characterized by their gain, a function of their directivity and efficiency.
Interestingly, based on these two parameters any given antenna can be placed into one of two categories: There are constant-area antennas, the effective area of which is constant with frequency, and constant-gain antennas, the gain of which is constant with frequency. Constant-area antennas have gain that increases with frequency; constant-gain antennas have effective area that decreases with frequency.
The source of the myth is that most portable consumer wireless products use constant-gain antennas, usually some variant of a dipole. While the gain of a resonant dipole is constant with frequency, as the frequency goes up its physical length, and therefore its effective area, goes down. 2.4 GHz dipoles are physically smaller than 900 MHz dipoles. They therefore have less effective area, and recover less power from the incident wave. It seems like the path loss at 2.4 GHz is greater, but it's really just a result of the antenna choice in the product design. If consumer products used constant-area antennas, like a parabolic dish of fixed physical dimensions, exactly the opposite result would be found: Since constant-area antennas have gain that increases with frequency, the recovered power at 2.4 GHz would be greater than that at 900 MHz, and we could start a myth that propagation loss decreases with frequency.
Interestingly enough, if the transmitter has a constant-gain antenna and the receiver has a constant-area antenna (or vice-versa), the recovered power at the receiving antenna terminals would be independent of frequency (i.e., constant), and we could avoid the generation of propagation loss myths entirely.
There are, for example, plenty of scientists who are not programmers, but who learn enough to adequately accomplish their scientific tasks. The results are rarely pretty, but often functional. Is the same not true for patents?
The analogy breaks down. The definitions and requirements of the legal "language" change over time, due to various court rulings and changes in the law, and one has to keep up with such things--and, like programming, subtle differences can have big differences in the outcome. I guess it's analogous to saying that the compiler changes it's interpretations over time:)
Or should small fry not even bother to play this game?
I would say that it depends on how much you depend on the result. You wouldn't have a professional scientist write the code your company depends on for its existence, so don't have an amateur write the patent your company depends on, either. On the other hand, if you're just patenting something that you think is cool, or will pad your resume, go ahead, but don't be surprised if it takes longer and is a bigger pain than you imagined, and realize that your odds of success are lower.
Is the only reason to apply for a patent to sue someone?
To be able to sue someone is the only legal reason. There are lots of personal or business reasons, though--the aforementioned padded resume is one. One can also patent something in a defensive manner--i.e., so no one else can patent it--then open the patent to all, royalty-free, if you're an altruist and you've just invented the water-to-gasoline conversion process or something.
Should patents only be applied for if the immediate stakes are high enough to justify spending obscene amounts of money on gold-plated lawyers, as a means to generating even more obscene amounts of money through legal action?
Let's start off by saying that we are discussing the most advantageous legal behavior for honorable and truthful people to have, not some scam.
Of course both the inventor and his attorney are required to disclose all known prior art when filing a patent. Not only is it a legal requirement, but it's stupid not to do so. The USPTO has its faults, but its patent search system is not one of them. You waste your money (and the attorney wastes his time, for a first-action rejection without a response limits the hours he can bill) filing if you know of a preexisting relevant patent you do not disclose for, in my experience, the USPTO will find it if it exists. And yes, attorneys can and do get disbarred for such things.
I also agree that it is neither the inventor's nor the attorney's responsibility to perform prior art searches, that the inventor, as one of (presumably at least) ordinary skill in the art, is assumed to be able to make a determination of novelty, and that the patent office is the ultimate authority on what is, and is not, prior art. I note, however, that what we are discussing are informal searches, not those leading to a binding statement from the attorney; we are discussing those typically taken to avoid the somewhat awkward situation that arises in the office of the attorney when he presents the client with a first-action rejection with dead-on prior art.*
Finally, I agree that having two attorneys is a superior solution.
However, if you'll read my comment and the GP's comment carefully, you'll note that we are discussing a somewhat different point. In many fields (e.g., recombinant DNA), ordinary skill in the art cannot be obtained without knowledge of existing patents. We are discussing the situation in which you, as the inventor, are presented with prior art relevant to (future and as yet unanticipated) application B when filing for application A. The danger is that one leaves a paper (or electronic) trail showing that one was exposed to relevant art to application B even years earlier, in a different context. While the attorney is not assumed to be a subject matter expert, the inventor is (as you say), and it's a lot harder for the inventor to say that he was unaware of some relevant art in his field of expertise when he is presented with evidence that he has seen it.
I don't know how you feel about your memory, but mine is like a black hole, and while it's not at all impossible for me to see something yet later have no recollection of it, that's not a position I'd be comfortable taking in court. It's much easier to pay an attorney to be subject to that privilege--especially since he's not presumed to be a subject matter expert. ________ *Attorney: Well, your application was rejected because of Jones' patent. While he claims something completely different, he mentions in the specification something very close to your invention. The examiner says it would have been obvious to one of ordinary skill in the art to use Jones' teachings to produce your invention. Client: I never can understand this legal mumbo-jumbo. What? Attorney: Your invention is not novel. It's known in the art. Client: Eh? I know the art. Jones never published anything in a journal. He's never manufactured his invention, and anyway it's completely different from mine. I've never seen this before. Attorney: Apparently your knowledge of the art is not as good as you thought it was. Client: This isn't the art. This is patents. I hired you because you have 20 years' experience in the field. Did you know of this patent before, and just took my money to write and file the application? Attorney: Hey, look, you're the subject matter expert.... [Etc.]
IANAPL, but I have spent several dozen years talking to patent attorneys on professional matters, so I'll answer the question anyway. Patent attorneys and agents are encouraged to correct any errors.
In the US, those that hold patents are entitled to sue for damages (e.g., payment for lost sales) resulting from someone infringing on a patent. However, there is an element in US patent law little-known outside the legal profession: If the infringer can be shown in court to have *knowingly* infringed (i.e., known of the patent yet infringed anyway), the patent holder is entitled to sue for triple damages. People attempting to file a patent are therefore frequently advised not to do patent searches, because if they are later sued over some related patent and evidence of their knowledge of the related patent exists as a result of this search (perhaps obtained via the discovery process prior to a trial), their potential losses are three times as great as they would be otherwise.
The use of an attorney, as the GP suggested, provides an abstraction layer to prevent this potential increased liability. The attorney can take the inventor's invention, do a patent search and advise his client accordingly. Since any professional communication between attorney and client is not subject to the discovery process, the client is protected against the potential triple-damages threat whether or not the client is made aware of the prior patent.
In my experience, attorneys will do one of two things: If the prior art is absolutely dead-on, and the attorney can see no way to obtain a patent in light of the art, he will not tell the inventor of the prior art that he has searched (which frequently includes hundreds of patents), and simply advise the client that his invention is not patentable. More often, however, the attorney will give the client the prior patent(s) and the two will examine the relevant claims together, looking for ways that the client's invention is different from that claimed by the earlier patent. This is a team effort, combining the inventor's technical expertise with the attorney's legal expertise. Identified differences will form the basis of the new patent application.
Note that all this applies to prior patents. The inventor is free to search prior art in the form of technical journal articles, conference proceedings, etc. without penalty, AFAIK, and is frequently advised to do so, since the USPTO certainly will when the inventor's patent application is examined. Nothing is more annoying than going through the hassle of applying for a patent, only to have the examiner reject it based on a passing reference in a Byte magazine article from the 1980s (not that I would know what that was like).
[dtmos pauses to don his aerogel suit] The antagonism between/. denizens and patent attorneys has always puzzled me. Like people writing software, the main function of patent attorneys and agents is to write precisely, using terms and phrases the definition of which have previously been determined. A well-written patent application is not unlike a well-written (albeit uncommented) computer program: It has a specific form and internal structure; everything there has a specific purpose and nothing is there that does not have a purpose. Nothing is duplicated. Variables are given specific names the first time they are used, and the names are consistent throughout; definitions are precise (if general). Like computer programs, they are written in a pre-determined and defined high-level language and, like computing, much of the hassle occurs when two different entities interpret the code differently.
The solution isn't shielding, which is impractical to do effectively after the speaker system is manufactured, but capacitive bypassing. Put a 100 to 1000 pF capacitor (of appropriate voltage rating for the audio and DC voltage present) across the speaker leads as close to the speaker as possible. The capacitor bypasses the RF current of the cell phone away from the speaker, keeping the RF voltage (and its TDMA power variations occuring at an audio rate that are causing the interference) low.
The value range is such that a capacitor in that range provides a low impedance at the RF (800-2000 MHz) while producing a negligable reactance at audio frequencies (below 20 kHz). (Capacitive reactance (in ohms) is equal to 1/(2*pi*f*C), where f is frequency in Hz and C is capacitance in Farads.)
They're closer together for the purposes of making a filter at the RF frequencies. However, filtering in most radio designs is not done at the RF frequency. Rather, in both cases superheterodyne receiver designs are commonly employed, in which the signal is mixed down to a lower frequency, the so-called intermediate frequency or "IF", where filtering is easier to do.
The IFs of "traditional" radio designs are 455 kHz for broadcast AM radios (putting the adjacent channels at 445 and 465 kHz) and 10.7 MHz for FM (putting the adjacent channels at 10.5 and 10.9 MHz). The worst-case fractional frequency difference (you don't really need logs to see this effect) for AM is then (465 - 455)/455 = 0.022, while for FM it is (10.9 - 10.7)/10.7 = 0.019, making filtering in an FM radio slightly harder to do than in an AM radio. (It's done this way because the FM design was done in the 1950s, when filter technology was better understood than when the AM design was developed in the 1930s.)
(Most modern receivers employ a variant of the superheterodyne called a "zero-IF" receiver, in which the signal is mixed down to an IF of, surprisingly enough, 0 Hz, centering the adjacent channels around 10 kHz for AM and 200 kHz for FM, where they are more easily filtered by a (usually digital) lowpass filter.)
I suspect that the real reason FM does not suffer from this problem is the digital signal is modulated with quadrature amplitude modulation (QAM), a sort of hybrid between AM and FM to which most legacy FM receivers will be relatively immune--for the same reason they are relatively immune to other AM noise like the static from thunderstorms.
I don't know if it ever made it to the market, but a wristwatch powered by temperature variations of the air would be really cool. (Its inventor, Steven Phillips, died in March, 2004, and I can no longer find his shop, the Budapest Watch Company of Guilford, Connecticut, on the web.)
Scale, packaging, testing the bigger issues
on
Who Makes Custom Chips?
·
· Score: 2, Informative
You didn't really supply enough information for a definitive reply, so I'll make some assumptions as I go along.
First, I don't understand how a "consumer" product could need only a wafer's worth of chips. In the industry, consumer == high volume. I assume, therefore, that this isn't a commercial venture, but a hobby of some type. (Oh, and a word to the wise: Don't go around anyone in the industry with the line about the fabs for older-generation designs being in Taiwan--you'll be marked immediately as either a newb or an idiot. TSMC and UMC are leaders in the semiconductor foundry business, not also-ran bottom feeders.)
Since you mention a VLSI class I'll assume you want a purely digital chip, and that you have no special needs (ultra-high speed, analog circuits, etc.). As others have suggested, if you're doing this yourself an FPGA or microcomputer is the obvious way to go, but I'll add another reason why: A single individual, working in his garage writing Verilog or VHDL from scratch, cannot conceive and design enough VLSI logic in a year to fill up even the smallest ASIC in any process even remotely modern. (Even a five-generation-old IC process is good for 25k gates/mm^2, with the smallest die typically 5 mm^2 or so; that's a lot of Verilog!) So even if you did an ASIC, the size of the die likely would be determined by the number of pads, not the logic--a so-called "pad limited" design--and so isn't likely to be economical to produce. So, FPGA.
What you're looking for, you say, are "bare chips." Your biggest challenge isn't going to be the logic design of the chip, it's going to be this--finding a vendor that will supply bare die FPGAs that you can flip-chip or wirebond and pot to your substrate (whatever it is). Discuss the issue with your local Xilinx and Altera reps. Packaging is a far bigger problem for you than your logic design, especially if you care--maybe you don't--about nasty environmental conditions like humidity and vibration. Do a google search for "custom IC packaging" and look for a custom manufacturing house that will do this for you. Bring to the first meeting a wirebonding diagram (a drawing showing the locations of all the pads on the die to which you want to connect) of your FPGA, a technical description of the substrate material (manufacturers' trade names often suffice), and clues to your overall plan that you can share with the people making your product. If you're doing flip-chip (a.k.a. C4, or other names) packaging, be advised that the die must be specifically designed for such packaging; your task, should you choose to accept it, is to find a mutually-acceptable packaging method between yourself (who has the end-product vision), the chip vendor (who has to supply the chip), and the contract manufacturing house (who can only do the packaging/mounting techniques for which he has the materials and equipment). Oh--and be sure that the FPGA vendor supplies you TESTED bare die--not just bare die.
You may wish to ask the FPGA vendor about "chip-scale packaging" options for his part. Often these packaging schemes, which can look like bare die to the naked eye, are simpler to use than true bare die.
Finally, don't forget that you'll want the contract manufacturing house to test your product after packaging your chip, to ensure that you get good working product and not just high technology waste. Provide him a written test procedure, which typically exercises each pad he was to connect.
Best of luck, and welcome to engineering. Isn't it fun?
Slashdot Mirror
(a) It's naturally radioactive. Also, from TFA: "Coauthors of the present paper learned of a new water-filled fracture inside a South African gold mine near the Johannesburg metropolitan area and viewed it as an opportunity to study subsurface rock uncontaminated by human activities."
(b) It's not practical to use its radioactivity as a power source, however, because it's only mildly radioactive in the natural state; said another way, it's not appreciably warm, so the amount of heat given off of natural uranium due to its radioactivity is negligible.
(c) Most (nearly all) human-generated nuclear waste has the same answer as (b); of that that is appreciably warm, there's too little of it to be useful as a power source.
(d) You got it.
Note that the bacteria do not use radioactivity directly, but rather use hydrogen from their environment, made from decomposing water exposed to radioactivity, as an energy source. Again from TFA: "This fracture water contained hydrocarbons and hydrogen not likely to have been created through biological processes, but rather from decomposition of water exposed to radiation from uranium-bearing rocks."
The present-day record for car pacing is held by Fred Rompelberg of Holland, who reached 268.831 km/h (167.043 mph) on a bicycle following a car at the Bonneville Salt Flats, Utah, on October 3, 1995.
Interestingly, he set the record 27 days shy of his 50th birthday, so there does seem to be a trend here....
You mean those organic lights, made from grain?
Oh, fluorescents. Sorry.
Still at the 3-inch wafer stage, while the silicon industry has been using 12-inch wafers for some time now.
What you say is true, but HP and Varian were in Silicon Valley as second-generation startups.
The first generation was the Federal Telegraph Company, founded by Cyril Elwell in the winter of 1909-1910 (originally as the Poulsen Wireless Telephone and Telegraph Company). Never heard of it? It was built to commercialize the arc (not spark) transmitter developed by Valdemar Poulsen, and by 1918 had succeeded in building and operating 1-megawatt continuous-wave radio transmitters.
Elwell was not only a Stanford graduate, he got his first financing for the company from Stanford faculty members, including the president of the university. So it can truly be said that Stanford itself acted as the first venture capitalist for the Valley.
Like the startups to follow, Federal people often left to do great things:
--Since it needed receivers to go with its transmitters, Federal hired a man from New York to develop a receiver for it, and set him up in a laboratory in the bay area. There, Lee DeForest would invent the triode vacuum tube (valve).
--To transfer the arc transmitter technology from Denmark to the Valley, Poulsen sent some of his employees with the equipment. One quickly became disillusioned with Federal, but liked the Valley, and started working with speakers. Soon thereafter, Peter Jensen formed his own company, Magnavox. Jensen's name lives on today in several lines of audio products.
--Leonard Fuller, longtime chief engineer of Federal, eventually ended up on the Berkeley faculty. The story goes that one day during the Great Depression, he was sitting in the faculty cafeteria when Ernest O. Lawrence was complaining that his cyclotron research was limited by the size of magnetic pole pieces he could obtain. Fuller realized that the 1-megawatt arc transmitter Federal had designed had very, very large magnetic pole pieces and, as they were too heavy (80 tons) to scrap, several had been sitting unused in a Valley warehouse since the end of World War I. A donation was quickly arranged, and the unused Federal components came to play a significant part in the development of large particle accelerators.
A lot of good microbial fuel cell work, including the discovery of the geobacter genus, has been done by D.R. Lovley and the group at UMass/Amherst.
In addition to their work on the microbial fuel cells themselves, they've also made the interesting discovery that the bacteria naturally form nanowires to transfer electrons outside the cell--something potentially [sorry!] useful to connect to an external electrode.
In August 2005.
To sum up, the rationale for the Japanese to work on a supersonic transport is based on three assumptions:
1. The scramjet engine will reduce operating (read: fuel) costs per average passenger mile significantly below that of the Concorde (by supporting a larger plane and being more fuel-efficient at cruise),
2. The plane will be capable of nonstop trans-Pacific flight (an ability also largely due to the fuel-efficiency of the scramjet), and
3. The much longer trans-Pacific flights in which the Japanese are interested will more dramatically show the time-of-arrival advantage of the supersonic plane than the shorter trans-Atlantic flights of the Concorde, and make it more appealing to seat-weary passengers.
I suppose there is also a fourth assumption, that cheap, fast, trans-Pacific travel would greatly improve the national economy of Japan in general and the Japanese aircraft industry in particular. This is the reason the Japanese government is expressing interest.
Whether these assumptions turn out to be factual or not requires research, which the Japanese are now doing.
I now return you to your previously-scheduled discussion, already in progress.
Who're they kidding? It's got the most brutal weather of the contiguous 48 US states. Is this some sort of gag?
Ah, that would be, "meteorological" and "supplemental."
Sorry, rented fingers.
Geez.
The key point not brought out in TFA is that the rainfall prediction scheme is not based on the link from the handset to the cell tower, but on the wireless backhaul links of the cellular system. The backhaul link is the link from the cell tower to the rest of the world (or at least the phone system of the rest of the world)--in many places in the world it is fiber or some other line, but increasingly often it, too, is wireless, using something called digital fixed radio systems (DFRS; check out standard EN 301 751 at ETSI).
The wireless backhaul links are much better for the meterological application than the handset link, because:
(a) It's a fixed link; since the cell towers don't move, like the handsets do, the location of the link, and therefore the rain, is known, and
(b) It's at a much higher frequency. The DFRS links used in this paper are at 8-23 GHz, much higher than the 0.8-1.9 GHz (depending on your local regulatory environment) of the handset link. This is important because rain attenuation increases as the signal frequency increases; it would be quite difficult to reliably detect rain fades at the handset frequencies (although in a bad enough storm--a cyclone comes to mind--it's probably possible; TFA notes the anecdotal evidence of fading television signals in bad weather).
I note in passing that the web-based supplimental material to the article references a US patent application, # 60/698,491.
You were doing well until you repeated that old hybrid-EMT scare. Any first responders that were afraid to approach a hybrid weren't well informed in their profession. I can't speak for Honda, but not only did Toyota work with national first-responder organizations to get their comments on the design of the US model, it made presentations on its design at their national conventions, made publications about it in the trade press, and distributed literature about the car freely and widely. The locations of the high-voltage elements of the car have been available on the web since time immemorial, and Toyota, at least, spent a lot of time repeating over and over that there's no high voltage in the roof pillars (how do these rumors get started?!?).
Both Toyota and Honda were and are exquisitely well-aware of accident procedures involving their cars; that's why the high-voltage lines in the Prius are armored International Orange cables isolated from the ground of the chassis, surrounded by identified conduit, and centered under the car floor, where the jaws of life and other EMT tools are least likely to be used. The battery itself is placed in the statistically safest place in the car (just over the rear axle), and protects first responders by an accelerometer-based circuit breaker, a Ground Fault Interrupter, and interlocks. Criminy, what do you want?
Umax = maximum radiation intensity (i.e., the maximum watts/unit area of a sphere enclosing a (transmitting) antenna in some preferred direction).
Uavg = average radiation intensity (i.e., the average watts/unit area of a sphere enclosing a (transmitting) antenna) = 4pi/(power radiated).
D = directivity = Umax/Uavg.
G = antenna gain = D (for antennas that are 100% efficient, a reasonable approximation for most cases with which we're concerned here).
A = effective area = (power produced at the antenna terminals)/(power density of the incident received wave); e.g., watts/(watts/square meter) = square meters.
wavelength = (speed of light)/(frequency); e.g., wavelength in meters = 300/(frequency in MHz).
You may have notice a gear-shift in moving from the definition of directivity to that of effective area--I defined directivity using a transmitting antenna, but effective area in terms of a receiving antenna. Fortunately, the terms are reciprocal; both terms apply equally well to both transmitting and receiving antennas.
The crux of the matter, and the source of the frequency term in the "free-space" equation, is the next equation. The directivity D (and therefore the gain G, under our assumption of 100% efficiency) of any antenna is related to its effective area A and the wavelength of the incoming wave:
D = G = A*4pi/(wavelength)^2 = A*4pi*(frequency)^2/(speed of light)^2.
If we assume that the antenna gain G is constant then, as the wavelength (i.e., frequency) changes, its effective area must, too. The gain of a dipole is constant at 2.14 dB, relative to an isotropic source; its effective area, however, is (30/73pi)*(wavelength)^2, or about 0.13(wavelength)^2. However, if we assume that the antenna is a different type, with constant area A, then as the wavelength changes its gain must change. Either assumption is valid, depending on the type of antenna we are employing. The confusion arises when people use the Wikipedia "free-space" equation to model path loss, then want to experiment with different antenna types. They don't realize that they're modeling their antennas as part of the "free-space" path loss. Wild errors result, for exampe, if you use Wikipedia's equation with parabolic dish antennas, like you often do in microwave point-to-point systems.
I guess a more subtle and pervasive problem arises when the equation gives people the idea that there's something inherently "bad" about propagation at higher frequencies, and that we should therefore all fight for operation at lower frequencies. Which, I guess, is what so motivates my rant.
In summary, you're right--frequency shouldn't come into this at all, and the second term of the equation in Wikipedia should be removed.
p.s.: And I thought my starlight/sunlight explanation was bulletproof. Maybe I should use flashlights? Lighning bugs?
Pager batteries last a long time because, while both pagers and cell phones spend most of their time asleep (neither transmitting nor receiving), as you suggest the pager can tolerate higher latency than a cell phone. People tolerate page latency of 30 seconds or more; it's a store-and-forward system. Trying to keep a caller on the line for 30 seconds while a cell phone call is set up (in addition to other inherent network delays) is another thing entirely. Having the luxury of long latency enables low pager system duty cycles, and therefore low average power consumption and long battery life.
It is also true that cellular protocols are far more complex than paging protocols (cf. the GSM spec with, say, POCSAG or FLEX), and the hardware to implement them requires more power consumption when active. Not only is the physical layer more complex (and often at a higher frequency), but there is additional decryption, data deinterleaving, etc. required, too. Phone manufacturers have been largely (if not totally) unable to justify the additional cost of a separate, simple, paging-like receiver in phones to improve their standby time, although a search of the USPTO records will indicate that lots of people have considered the idea.
Finally, it is also true that the need to include a relatively high-powered transmitter in the phone requires some compromises in standby power consumption. For example, to get high efficiency from a transmitter power amplifier one likes to have a high supply voltage, fed from a source of low output impedance (internal resistance) and high capacity. Ergo, lithium batteries for maximum talk time. However, the receiver has no such requirement; in fact, its power consumption drops with the supply voltage. (This is another reason why pagers ran from a single 1.5V AAA cell.) The cell phone, however, can't operate from separate transmit and receive batteries--I, at least, wouldn't buy one--so the compromise is that the receiver either runs from the high voltage of the lithium battery directly (an unappetizing choice), runs from the low-voltage output of a linear voltage regulator supplied by the battery (only slightly more appetizing, since the system efficiency is still poor), or runs from the low-voltage output of a switching voltage regulator (possibly the best choice, if the electrical noise can be tolerated, even though the regulator still adds power consumption). Or some combination of the above, any of which is still less power-efficient than running from a low-voltage source in the first place.
The RF signal is typically a sine wave generated by a VCO (voltage-controlled oscillator) that drives the first stage of the divider (prescaler). At the high frequencies used by RF synthesizers, the output pulses of the first stage aren't square, either. They're a sort of a mush; a compromise between the power consumption of your concern and the desire to use CMOS logic for its other low-power features in a highly-integrated system. The first stage of the divider typically draws half or more of the total power, but it is only driven hard enough to enable reliable CMOS operation, not produce square waves. As you go down the divider, the signals gradually move from the sine wave of the input to asymptotically approach a square wave at the output.
I'm considering the devotion of the rest of my professional career to the eradication of the "propagation loss increases with frequency" myth.
Repeat after me:
Propagation loss does not increase with frequency!
Propagation loss does not increase with frequency!
Propagation loss does not increase with frequency!
Think about it: If the propagation loss of an electromagnetic wave increased in proportion to its frequency, there would be so much so much attenuation at the THz frequency of light that we'd never see sunlight--or stars. Propagation loss is independent of frequency, except for scattering due to molecular and atomic resonances that are insignificant at the frequencies we're discussing. (There are also changes in scattering behavior that become relevant in indoor applications, like propagation around corners.)
What is dependent on frequency, however, is the performance of the antennas we use to transmit and receive electromagnetic waves. Antennas can be characterized by a parameter called effective area. Returning to the sunlight example, recognize that the output power of a solar panel is proportional to its physical area; the larger this area, the greater the fraction of the incident power transmitted by the sun is received by the solar panel and converted to available output power. Receiving antennas, and antennas in general (even wire antennas), have an effective area; it's the area required to produce the measured output power, based on the density of transmitted power (watts/unit area) at the location of the receiving antenna.
Antennas can also be characterized by their gain, a function of their directivity and efficiency.
Interestingly, based on these two parameters any given antenna can be placed into one of two categories: There are constant-area antennas, the effective area of which is constant with frequency, and constant-gain antennas, the gain of which is constant with frequency. Constant-area antennas have gain that increases with frequency; constant-gain antennas have effective area that decreases with frequency.
The source of the myth is that most portable consumer wireless products use constant-gain antennas, usually some variant of a dipole. While the gain of a resonant dipole is constant with frequency, as the frequency goes up its physical length, and therefore its effective area, goes down. 2.4 GHz dipoles are physically smaller than 900 MHz dipoles. They therefore have less effective area, and recover less power from the incident wave. It seems like the path loss at 2.4 GHz is greater, but it's really just a result of the antenna choice in the product design. If consumer products used constant-area antennas, like a parabolic dish of fixed physical dimensions, exactly the opposite result would be found: Since constant-area antennas have gain that increases with frequency, the recovered power at 2.4 GHz would be greater than that at 900 MHz, and we could start a myth that propagation loss decreases with frequency.
Interestingly enough, if the transmitter has a constant-gain antenna and the receiver has a constant-area antenna (or vice-versa), the recovered power at the receiving antenna terminals would be independent of frequency (i.e., constant), and we could avoid the generation of propagation loss myths entirely.
I would say that it depends on how much you depend on the result. You wouldn't have a professional scientist write the code your company depends on for its existence, so don't have an amateur write the patent your company depends on, either. On the other hand, if you're just patenting something that you think is cool, or will pad your resume, go ahead, but don't be surprised if it takes longer and is a bigger pain than you imagined, and realize that your odds of success are lower.
To be able to sue someone is the only legal reason. There are lots of personal or business reasons, though--the aforementioned padded resume is one. One can also patent something in a defensive manner--i.e., so no one else can patent it--then open the patent to all, royalty-free, if you're an altruist and you've just invented the water-to-gasoline conversion process or something.
Um, no.
Let's start off by saying that we are discussing the most advantageous legal behavior for honorable and truthful people to have, not some scam.
Of course both the inventor and his attorney are required to disclose all known prior art when filing a patent. Not only is it a legal requirement, but it's stupid not to do so. The USPTO has its faults, but its patent search system is not one of them. You waste your money (and the attorney wastes his time, for a first-action rejection without a response limits the hours he can bill) filing if you know of a preexisting relevant patent you do not disclose for, in my experience, the USPTO will find it if it exists. And yes, attorneys can and do get disbarred for such things.
I also agree that it is neither the inventor's nor the attorney's responsibility to perform prior art searches, that the inventor, as one of (presumably at least) ordinary skill in the art, is assumed to be able to make a determination of novelty, and that the patent office is the ultimate authority on what is, and is not, prior art. I note, however, that what we are discussing are informal searches, not those leading to a binding statement from the attorney; we are discussing those typically taken to avoid the somewhat awkward situation that arises in the office of the attorney when he presents the client with a first-action rejection with dead-on prior art.*
Finally, I agree that having two attorneys is a superior solution.
However, if you'll read my comment and the GP's comment carefully, you'll note that we are discussing a somewhat different point. In many fields (e.g., recombinant DNA), ordinary skill in the art cannot be obtained without knowledge of existing patents. We are discussing the situation in which you, as the inventor, are presented with prior art relevant to (future and as yet unanticipated) application B when filing for application A. The danger is that one leaves a paper (or electronic) trail showing that one was exposed to relevant art to application B even years earlier, in a different context. While the attorney is not assumed to be a subject matter expert, the inventor is (as you say), and it's a lot harder for the inventor to say that he was unaware of some relevant art in his field of expertise when he is presented with evidence that he has seen it.
I don't know how you feel about your memory, but mine is like a black hole, and while it's not at all impossible for me to see something yet later have no recollection of it, that's not a position I'd be comfortable taking in court. It's much easier to pay an attorney to be subject to that privilege--especially since he's not presumed to be a subject matter expert.
________
*Attorney: Well, your application was rejected because of Jones' patent. While he claims something completely different, he mentions in the specification something very close to your invention. The examiner says it would have been obvious to one of ordinary skill in the art to use Jones' teachings to produce your invention.
Client: I never can understand this legal mumbo-jumbo. What?
Attorney: Your invention is not novel. It's known in the art.
Client: Eh? I know the art. Jones never published anything in a journal. He's never manufactured his invention, and anyway it's completely different from mine. I've never seen this before.
Attorney: Apparently your knowledge of the art is not as good as you thought it was.
Client: This isn't the art. This is patents. I hired you because you have 20 years' experience in the field. Did you know of this patent before, and just took my money to write and file the application?
Attorney: Hey, look, you're the subject matter expert....
[Etc.]
IANAPL, but I have spent several dozen years talking to patent attorneys on professional matters, so I'll answer the question anyway. Patent attorneys and agents are encouraged to correct any errors.
/. denizens and patent attorneys has always puzzled me. Like people writing software, the main function of patent attorneys and agents is to write precisely, using terms and phrases the definition of which have previously been determined. A well-written patent application is not unlike a well-written (albeit uncommented) computer program: It has a specific form and internal structure; everything there has a specific purpose and nothing is there that does not have a purpose. Nothing is duplicated. Variables are given specific names the first time they are used, and the names are consistent throughout; definitions are precise (if general). Like computer programs, they are written in a pre-determined and defined high-level language and, like computing, much of the hassle occurs when two different entities interpret the code differently.
In the US, those that hold patents are entitled to sue for damages (e.g., payment for lost sales) resulting from someone infringing on a patent. However, there is an element in US patent law little-known outside the legal profession: If the infringer can be shown in court to have *knowingly* infringed (i.e., known of the patent yet infringed anyway), the patent holder is entitled to sue for triple damages. People attempting to file a patent are therefore frequently advised not to do patent searches, because if they are later sued over some related patent and evidence of their knowledge of the related patent exists as a result of this search (perhaps obtained via the discovery process prior to a trial), their potential losses are three times as great as they would be otherwise.
The use of an attorney, as the GP suggested, provides an abstraction layer to prevent this potential increased liability. The attorney can take the inventor's invention, do a patent search and advise his client accordingly. Since any professional communication between attorney and client is not subject to the discovery process, the client is protected against the potential triple-damages threat whether or not the client is made aware of the prior patent.
In my experience, attorneys will do one of two things: If the prior art is absolutely dead-on, and the attorney can see no way to obtain a patent in light of the art, he will not tell the inventor of the prior art that he has searched (which frequently includes hundreds of patents), and simply advise the client that his invention is not patentable. More often, however, the attorney will give the client the prior patent(s) and the two will examine the relevant claims together, looking for ways that the client's invention is different from that claimed by the earlier patent. This is a team effort, combining the inventor's technical expertise with the attorney's legal expertise. Identified differences will form the basis of the new patent application.
Note that all this applies to prior patents. The inventor is free to search prior art in the form of technical journal articles, conference proceedings, etc. without penalty, AFAIK, and is frequently advised to do so, since the USPTO certainly will when the inventor's patent application is examined. Nothing is more annoying than going through the hassle of applying for a patent, only to have the examiner reject it based on a passing reference in a Byte magazine article from the 1980s (not that I would know what that was like).
[dtmos pauses to don his aerogel suit] The antagonism between
The solution isn't shielding, which is impractical to do effectively after the speaker system is manufactured, but capacitive bypassing. Put a 100 to 1000 pF capacitor (of appropriate voltage rating for the audio and DC voltage present) across the speaker leads as close to the speaker as possible. The capacitor bypasses the RF current of the cell phone away from the speaker, keeping the RF voltage (and its TDMA power variations occuring at an audio rate that are causing the interference) low.
The value range is such that a capacitor in that range provides a low impedance at the RF (800-2000 MHz) while producing a negligable reactance at audio frequencies (below 20 kHz). (Capacitive reactance (in ohms) is equal to 1/(2*pi*f*C), where f is frequency in Hz and C is capacitance in Farads.)
They're closer together for the purposes of making a filter at the RF frequencies. However, filtering in most radio designs is not done at the RF frequency. Rather, in both cases superheterodyne receiver designs are commonly employed, in which the signal is mixed down to a lower frequency, the so-called intermediate frequency or "IF", where filtering is easier to do.
The IFs of "traditional" radio designs are 455 kHz for broadcast AM radios (putting the adjacent channels at 445 and 465 kHz) and 10.7 MHz for FM (putting the adjacent channels at 10.5 and 10.9 MHz). The worst-case fractional frequency difference (you don't really need logs to see this effect) for AM is then (465 - 455)/455 = 0.022, while for FM it is (10.9 - 10.7)/10.7 = 0.019, making filtering in an FM radio slightly harder to do than in an AM radio. (It's done this way because the FM design was done in the 1950s, when filter technology was better understood than when the AM design was developed in the 1930s.)
(Most modern receivers employ a variant of the superheterodyne called a "zero-IF" receiver, in which the signal is mixed down to an IF of, surprisingly enough, 0 Hz, centering the adjacent channels around 10 kHz for AM and 200 kHz for FM, where they are more easily filtered by a (usually digital) lowpass filter.)
I suspect that the real reason FM does not suffer from this problem is the digital signal is modulated with quadrature amplitude modulation (QAM), a sort of hybrid between AM and FM to which most legacy FM receivers will be relatively immune--for the same reason they are relatively immune to other AM noise like the static from thunderstorms.
Ah yes, but it's electronic. Phillips' design is mechanical.
I don't know if it ever made it to the market, but a wristwatch powered by temperature variations of the air would be really cool. (Its inventor, Steven Phillips, died in March, 2004, and I can no longer find his shop, the Budapest Watch Company of Guilford, Connecticut, on the web.)
You didn't really supply enough information for a definitive reply, so I'll make some assumptions as I go along.
First, I don't understand how a "consumer" product could need only a wafer's worth of chips. In the industry, consumer == high volume. I assume, therefore, that this isn't a commercial venture, but a hobby of some type. (Oh, and a word to the wise: Don't go around anyone in the industry with the line about the fabs for older-generation designs being in Taiwan--you'll be marked immediately as either a newb or an idiot. TSMC and UMC are leaders in the semiconductor foundry business, not also-ran bottom feeders.)
Since you mention a VLSI class I'll assume you want a purely digital chip, and that you have no special needs (ultra-high speed, analog circuits, etc.). As others have suggested, if you're doing this yourself an FPGA or microcomputer is the obvious way to go, but I'll add another reason why: A single individual, working in his garage writing Verilog or VHDL from scratch, cannot conceive and design enough VLSI logic in a year to fill up even the smallest ASIC in any process even remotely modern. (Even a five-generation-old IC process is good for 25k gates/mm^2, with the smallest die typically 5 mm^2 or so; that's a lot of Verilog!) So even if you did an ASIC, the size of the die likely would be determined by the number of pads, not the logic--a so-called "pad limited" design--and so isn't likely to be economical to produce. So, FPGA.
What you're looking for, you say, are "bare chips." Your biggest challenge isn't going to be the logic design of the chip, it's going to be this--finding a vendor that will supply bare die FPGAs that you can flip-chip or wirebond and pot to your substrate (whatever it is). Discuss the issue with your local Xilinx and Altera reps. Packaging is a far bigger problem for you than your logic design, especially if you care--maybe you don't--about nasty environmental conditions like humidity and vibration. Do a google search for "custom IC packaging" and look for a custom manufacturing house that will do this for you. Bring to the first meeting a wirebonding diagram (a drawing showing the locations of all the pads on the die to which you want to connect) of your FPGA, a technical description of the substrate material (manufacturers' trade names often suffice), and clues to your overall plan that you can share with the people making your product. If you're doing flip-chip (a.k.a. C4, or other names) packaging, be advised that the die must be specifically designed for such packaging; your task, should you choose to accept it, is to find a mutually-acceptable packaging method between yourself (who has the end-product vision), the chip vendor (who has to supply the chip), and the contract manufacturing house (who can only do the packaging/mounting techniques for which he has the materials and equipment). Oh--and be sure that the FPGA vendor supplies you TESTED bare die--not just bare die.
You may wish to ask the FPGA vendor about "chip-scale packaging" options for his part. Often these packaging schemes, which can look like bare die to the naked eye, are simpler to use than true bare die.
Finally, don't forget that you'll want the contract manufacturing house to test your product after packaging your chip, to ensure that you get good working product and not just high technology waste. Provide him a written test procedure, which typically exercises each pad he was to connect.
Best of luck, and welcome to engineering. Isn't it fun?