Energy-Harvesting Electronic Holiday Card 2024
keacher.comBack in the days of 2G the pre-wake-up pulses, aside from causing massive FM/AM interference like demented Morse code, would light up tuned antenna spark gaps on stickers for your Nokia phone. In Japan they sold cute phone tokens which had glowing eyes.
Energy harvesting the same way Theremin did for his passive wall bug.
I remember when you could tell you were about to receive a cellphone call based on radios and other analog electronics behaving weirdly. I got various people thinking I could see a few seconds into the future.
Tuned spark gaps? Or just an antenna hooked up to a pair of back to back LEDs? Just recently I saw something about light-up fake fingernails that would blink when you got a 2G phone call.
Probably the LED.
Reminds me of an old Nokia proof of concept charger for a mobile phone that used energy harvesting. It never was enough to make a call though:
From 2009 https://www.eetimes.com/nokia-working-on-energy-harvesting-h...
But it seems that the idea is still alive, from 2023:
"Relying more on ambient energy sources could prove monumental in automated warehouse inventory tracking, in medical instrument management and for deployment in airports, shopping centers and even individual smart homes. Nokia’s goal is to have energy harvesting technology in cellular networks that can support this massive IoT deployment."
https://www.nokia.com/blog/the-future-belongs-to-zero-energy...
I read the technical description of this and the whole work seems like magic. Antenna design, extremely low-power passive networks, etc. On top of that, it can tap into network signal bursts as a communication medium.
It feels like a project sent from the future.
I've always been curious what energy harvesting systems are capable of.
Also, what is the third type of energy harvesting besides light and 2.4GHz? I couldn't figure out what that might be.
“At its core, the card harvests energy from light, radio, and/or a USB connection to enable the blinking of LEDs.”[0]
So the last is USB, the sneakily obvious one.
Heat, voltage drop, and vibration are some other methods.
This card apparently has three - but none of those additional ones you listed, unfortunately.
For a while I had a very overpriced phone case from moeco that lit up when cell service was being used - https://www.moeco.jp.net/ - it came with a big warning not to use wireless charging with it (due to instead heating the case). Sadly, I wouldn't get one nowadays, I love wireless charging too much.
lol screw wireless charging, if I had the $$ for it and shipping and taxes wouldn’t be so darn expensive it’d be a nobrainer for me
I wouldn’t blame you - it was a sweet case. But wireless charging ensures I don’t lazily decline to charge my phone at night.
It's only for the iPhone and still 17,600円 though. 高いですね！～
This is super neat. By modulating the timing of the data being sent to a websocket, (which is basically a /dev/null data sink) it implements a covert air-gaped side-channel data transmission mechanism.
That's the part that's blowing my mind. Energy harvesting is cool and tricky to get right, but it's not magic.
Sending data by modulating the data flow itself, is spooky. Absolute madness. I love it and I'm a little scared.
> That last one consumed an hour of diagnostic time and involved using time-domain reflectometry (with a 20 ps rise-time pulser and 20 GHz scope) to locate the fault to within a region of a couple millimeters on one trace.
How does one even obtain the skills, much less the equipment to run such precision?!
A good EE degree with some RF specific course parts will teach you the concept. The scope .. well, you kinda have to borrow it from your employer as they're in the $10k range at that frequency.
There are plenty of amateur- to semipro-level youtubers out there in the metal and woodworking spaces with a total amount invested in their workshop that is way larger than that. I wouldn't be surprised if there are EE hobbyists with picosecond level scopes.
They also mention having access to a source-meter, which is not cheap either. I wouldn't mind spending some time in that lab!
Source meters are expensive for normie hardware, but certainly not out of reach. A brand new keithley 2401 is ~$5k, so I imagine a used one could be picked up for a few thousand or perhaps less.
Interesting, thanks for sharing.
One question: what load is the matching network designed for? Did the designer find the equivalent small signal impedance of the diode network via simulation? Is the SPICE model even valid at 2.4 GHz? Is small signal even applicable?
This is the same principle as those mobile phone led flashing stickers from the 90’s/00’s.
Clever idea.
very nice! They mention the use of Barker codes, which I'd not heard of before;
This project is so inspiring. I just spent about a half hour+ oogling over the details. Great documentation and I learned about so many things.
Very cool magic/tech!
Apologies if this goes against the hacker spirit, but do you know where I can buy a similar thing? In Europe?