Should I be considering this differently from solar? They're both harvesting ambient EM radiation; and the energy density of solar is much higher. I guess that radio goes through walls and never sleeps, so that's a plus. Anyway, I thought that the mentioning of a solar calculator was quite appropriate and warranted further discussion.
It sounds like the device on the hard hat is more like a passive RFID chip; It's hard to be sure, but the article says all the dangerous equipment has transmitters of their own.
Of course, the effective difference between the passive RFID chip and a device that simply consumes ambient radio-spectrum EM radiation is small, mostly a paradigm change.
Your initial question begets an interesting question- since light is EM radiation, just like radio waves, shouldn't we be able to pick up light with an antenna? If we can figure that out, we can forget about solar cells with their sad efficiency levels.
edit: with some quick research, the answer is obvious; while it seems extremely weird to imagine, light can be absorbed by an antenna- and the first and foremost reason this isn't being done already is because the appropriate antenna would be ~700 nanometers long.
You can absorb EM radiation just fine with an antenna that's "too long". It's when it's too short that you run into problems.
500nm-wavelength light oscillates at about 600 terahertz, with a period of about 1.7 femtoseconds. If you want to rectify that and turn it into DC current so you can run current semiconductor devices, you need a diode that can switch on once and off once in that period of time. So your forward recovery time plus your reverse recovery time needs to total less than 1.7 femtoseconds. Among other things, I think this implies that the depletion region in the diode needs to be less than 0.9 femtoseconds in width --- at the electron drift velocity of the semiconductor, which I think is typically around 12 orders of magnitude less than c, although in silicon it can be as high as only three orders of magnitude less than c. Which means that your depletion region needs to be 3 orders of magnitude smaller than the wavelength. Unfortunately the wavelength we're talking about here, at around 1000nm, is only four orders of magnitude bigger than a smallish atom, at 0.1nm. So you're pushing up against the bounds of possibility here with an insulating depletion region of a few atoms in thickness.
Forward and reverse recovery times for silicon diodes vary widely. Typical values for discrete components are measured in the tens to hundreds of nanoseconds. Schottky diodes bring that down to tenths of nanoseconds. One nanosecond is one million femtoseconds, so that's still five orders of magnitude too slow.
Anyway, I don't know anything about this stuff, really.
Wouldn't 700 nm antennas would be easy to make? We've been fabricating features smaller than that on silicon wafers since 1994 (according to Wikipedia).
important bit: simple processing & sound using ambient radio waves in a construction hat. Which is very cool. Other cited uses cover small temperature sensors and possibly replacing AAA battery powered devices.
Add in a healthy dose of sparkly-futurist-hope (devices that run FOR EVAR), and some actual data on another experimental device:
>The device collects enough power to produce about 50 microwatts of DC power, Dr. Smith said. That is enough for many sensing and computing jobs, said Professor Otis. The power consumption of a typical solar-powered calculator, for example, is only about 5 microwatts, he said, and that of a typical digital thermometer with a liquid crystal display is one microwatt.
Didn't see anything on the size of the "device", but that's still pretty cool.
The technology in the linked article would appear, at least superficially, to be similar to RFID. Although one difference might be that RFID works at a fixed frequency (to match the harmonic frequency of the receiving radio circuit) whereas this apparently works with "ambient" radio, implying a broad spectrum of frequencies, which is quite clever.
Either way, the broadcast signal is losing power at a rate proportional 1/d^2. So these devices necessarily have to work at very low power. What Tesla did - or is supposed to have done - was to figure out how to transmit power without the 1/d^2 loss. I.e, much like a collimated laser beam, he could transmit power over large distances and power high current devices.
He had dramatic ideas for free, wireless power transmission for everyone. His wealthy backer (was it JP Morgan?) said he couldn't see the profit angle in that and wasn't interested, saying something like "Where do you attach the meter?".
This is same technology as crystal radio of first half of 20th century, which was also powered only by absorbed EM energy (which is also same as passive RFID and similar things). Tesla's designs are slightly different (in that they are probably not exactly wireless). See http://amasci.com/tesla/tmistk.html
Readit News
Of course, the effective difference between the passive RFID chip and a device that simply consumes ambient radio-spectrum EM radiation is small, mostly a paradigm change.
Your initial question begets an interesting question- since light is EM radiation, just like radio waves, shouldn't we be able to pick up light with an antenna? If we can figure that out, we can forget about solar cells with their sad efficiency levels.
edit: with some quick research, the answer is obvious; while it seems extremely weird to imagine, light can be absorbed by an antenna- and the first and foremost reason this isn't being done already is because the appropriate antenna would be ~700 nanometers long.
500nm-wavelength light oscillates at about 600 terahertz, with a period of about 1.7 femtoseconds. If you want to rectify that and turn it into DC current so you can run current semiconductor devices, you need a diode that can switch on once and off once in that period of time. So your forward recovery time plus your reverse recovery time needs to total less than 1.7 femtoseconds. Among other things, I think this implies that the depletion region in the diode needs to be less than 0.9 femtoseconds in width --- at the electron drift velocity of the semiconductor, which I think is typically around 12 orders of magnitude less than c, although in silicon it can be as high as only three orders of magnitude less than c. Which means that your depletion region needs to be 3 orders of magnitude smaller than the wavelength. Unfortunately the wavelength we're talking about here, at around 1000nm, is only four orders of magnitude bigger than a smallish atom, at 0.1nm. So you're pushing up against the bounds of possibility here with an insulating depletion region of a few atoms in thickness.
Forward and reverse recovery times for silicon diodes vary widely. Typical values for discrete components are measured in the tens to hundreds of nanoseconds. Schottky diodes bring that down to tenths of nanoseconds. One nanosecond is one million femtoseconds, so that's still five orders of magnitude too slow.
Anyway, I don't know anything about this stuff, really.
Add in a healthy dose of sparkly-futurist-hope (devices that run FOR EVAR), and some actual data on another experimental device:
>The device collects enough power to produce about 50 microwatts of DC power, Dr. Smith said. That is enough for many sensing and computing jobs, said Professor Otis. The power consumption of a typical solar-powered calculator, for example, is only about 5 microwatts, he said, and that of a typical digital thermometer with a liquid crystal display is one microwatt.
Didn't see anything on the size of the "device", but that's still pretty cool.
Either way, the broadcast signal is losing power at a rate proportional 1/d^2. So these devices necessarily have to work at very low power. What Tesla did - or is supposed to have done - was to figure out how to transmit power without the 1/d^2 loss. I.e, much like a collimated laser beam, he could transmit power over large distances and power high current devices.
Not exact quotes, but the source is my print version of : http://books.google.com/books?id=40NzjS5FunkC&printsec=f...;