Not enough power. Using that method is going to throw away 80+% of the energy as heat and radiation.
If you want long term implanted power, this is doable http://www.citylabs.net/ It's tritium based, and uses the electrons from the tritium decay to provide electric current. These devices provide 2v/300nA which is basically nothing. I can't think of a single implantable project that uses that little power. Maybe a pulsed RFID, but even that could be done in a better way.
More importantly, don't implant radioactive stuff into yourself please. This little won't kill you but it will increase your risk for cancer.
Piezo discs are only useful if you can get a flex on them frequently. RF harvesting is just another take on inductive charging, and you won't find the density in normal environments.
You also really need to quantify what you mean by low-drain device as well.
By a low drain device I mean something that is not in constant use but only activates when used or a microcontroller that wakes up to do a task on a set timer.
Something like an alarm clock that directly stimulates nerves for example.
Not worth. There are some glowing glucose hydrogel implants, but they seem like a pain in the ass and the hydrogel is still not tested for long term use. Glows under uv light when your glucose levels are up if memory serves but im not sure.
http://forum.biohack.me/discussion/253/powering-devices-within-the-body/p1 @garethnelsonuk solar power for tickle charge has been close to ruled out. From the numbers you are stuck with tiny amounts of energy output even if you are sunbathing in 1000W/m². The faint glow of a tritium light source would be magnitudes below. Piezo or even mechano-electric generators would be way more efficient.
Glowing glucose based on your levels seems like it would be a good way to monitor it if you can make it with different sensitivity. I am going to need to look at that stuff cause I do want something that will glow on me
Glowing is sort of a pain in the ass to do hard/wetware wise. Either the chemistry is expensive/complex or you use something electronic and as we know, making light uses energy which is a pain. Option 3 is radiation but we just disccused why that's no good. Option 4 is a biological approach which means playing with genetics and that opens a whole new can of worms.
Option 5 is uv tattoos, which are still pretty damn cool.
If you could source the stuff for making a glowing hydrogel implant i'd be game to try and make it but i foresee it being a pain.
just to complete this thread's request for information for future readers. friend of mine recently stumbled across http://www.citylabs.net/ they have specs for tritium powered betavoltaic cells (pretty much nuclear power but without the light-detour). They run at up to 2.4V and generate between 20 and 350nanoAmps. They apparently last more than 20 years and cost you roundabout 3000 bucks.
Cool things, but very limited in use due to the rather low power output.
Comments
Thinking of combining a few of these: http://www.survivaldepot.co.uk/product/tritium_vial
With a solar cell encased inside a sealed unit
If you want long term implanted power, this is doable http://www.citylabs.net/
It's tritium based, and uses the electrons from the tritium decay to provide electric current. These devices provide 2v/300nA which is basically nothing. I can't think of a single implantable project that uses that little power. Maybe a pulsed RFID, but even that could be done in a better way.
More importantly, don't implant radioactive stuff into yourself please.
This little won't kill you but it will increase your risk for cancer.
RF harvesting is just another take on inductive charging, and you won't find the density in normal environments.
You also really need to quantify what you mean by low-drain device as well.
Something like an alarm clock that directly stimulates nerves for example.
I think it'd be pretty stupid to do this without shielding
@garethnelsonuk solar power for tickle charge has been close to ruled out. From the numbers you are stuck with tiny amounts of energy output even if you are sunbathing in 1000W/m². The faint glow of a tritium light source would be magnitudes below. Piezo or even mechano-electric generators would be way more efficient.
friend of mine recently stumbled across http://www.citylabs.net/
they have specs for tritium powered betavoltaic cells (pretty much nuclear power but without the light-detour).
They run at up to 2.4V and generate between 20 and 350nanoAmps. They apparently last more than 20 years and cost you roundabout 3000 bucks.
Cool things, but very limited in use due to the rather low power output.