Booting the Prop from low voltages, 2 of 3: Scavenged Headphone Audio @ 0.42V

Tubular · 2011-07-31 05:36

This time I'm using the LTC3105 to boot the prop from energy scavenged from my laptop's headphone audio output. It should work with most other headphone outputs too (iPods etc) !

The headphone signal is rectified using a humble 1N4148 into a 0.9F supercap, which accumulates over a couple of minutes to a critical voltage level of 0.42 volts. That voltage appears to be the minimum sufficient to get the Prop to boot, read its EEprom, and start flashing its blue led, using RCslow to drop the current requirement right down after boot.

The Prop then runs for about 110 seconds, without any additional audio energy input, before the supercap dips below 0.32V and everything dies (requiring further audio electrical energy to charge the supercap once again). Encouragingly, after boot, the power usage by the prop (even with the LED flashing on a low, aeroplane-wingtip-strobe-like duty cycle), is less than the incoming energy, so the Prop will keep running forever so long as the audio signal is being provided.

To generate the audio signal, I am generating a ~2 kHz sine wave at maximum laptop volume using FreeMat (Matlab clone), which is very easy and takes only 4 or 5 commands. Finer detail in next post.

It should also be possible to scavenge enough energy from an iPod or other media player. My initial tests (using non optimal Australian 80's music) indicate I may need a schottky diode to get past the 0.42V hurdle within a reasonable time. Or a dedicated full volume test tone track.

Using the second (stereo) audio channel for data, and suitable reset sync, perhaps we can even distribute prop applications over iTunes (kind of a modern C64 datasette ) !

Anyone want to hazard a guess what my 3rd energy source might be for part 3 of 3 ?

Tubular · 2011-07-31 05:40

FreeMat details:

Ok I lied its only 3 lines of code.

--> x=linspace(0,20000,20000);
--> y=sin(x);
--> for(i=1:50); wavplay(y); end

Wavplay(y) sends the matrix 'y', which has 20,000 samples ranging from -1 to +1, to the audio output device at a sampling rate of 11025 samples per second. So each wavplay(y) lasts just under 2 seconds. The output frequency is about 11025/2*pi or 1754 Hz, but this isn't critical.

Freemat is case sensitive. Don't forget to turn the value up to full (after the headphone connection is plugged in)

You can find FreeMat for Windows, Mac and Linux here

attachment.php?attachmentid=83569&d=1312116982

attachment.php?attachmentid=83569&d=1312116982

Dr_Acula · 2011-07-31 06:08

This is really clever!

Anyone want to hazard a guess what my 3rd energy source might be for part 3 of 3 ?

Wind power from a CPU cooling fan?
Piezo devices in the heel of your shoes?

mindrobots · 2011-07-31 06:16

Methane from you neighbor's dairy farm?

Keystrokes as you type things into the Parallax Forum?

Sound from your daughter's constant chatter.....oh, wait, that's my renewable energy source!

Heater. · 2011-07-31 06:22

...non-optimal Australian 80's music...

I would have thought AC/DC would work rather well in this test:)

Tubular · 2011-07-31 06:39

Heater. wrote: »

I would have thought AC/DC would work rather well in this test:)

LOL!... wonder if "High Voltage Rock n Roll" does indeed produce more energy than the average track

Good suggestions Drac, Mindrobots, but no banana yet. You know sometimes it really pays to ask an open question - some of these ideas are probably worth trying

Phil Pilgrim (PhiPi) · 2011-07-31 07:10

Tubular,

Very cool! You might be able to harvest more energy by using an audio step-up transformer ahead of your rectifier. That way, the diode's forward voltage will account for a lower percentage of the loss.

-Phil

Tubular · 2011-07-31 14:49

A very good suggestion, Phil. That should help the efficiency for AC based sources, and lower level ones too.

Right now I'm happy that it just works at 0.42v DC. The supercap is bigger than it needs to be, which means I can reduce it and cut the charging time before entering the "boot phase". Harvesting stereo is another possibility.

I am trying to work out a simple scheme to bring the Prop out of reset - am thinking something like having both audio channels locked in phase for generating initial charging power, then after a suitable period switching to antiphase to release the prop from reset.

btw I (rightly or wrongly) attribute this to your comment ages ago about programming the prop over audio. I just thought it would nice to power it too.

Mark_T · 2011-07-31 15:16

Perhaps the next energy source is the ever-popular lemon with two different metal rods stuck in it?

Tubular · 2011-07-31 15:55

You're brilliant Mark_T, but for bonus points and fabulous prizes, what breed of lemon ??

Was it the "close but no banana" that gave it away??

Cluso99 · 2011-07-31 18:02

With inside info, I had thought it was a potatoe.

"A lemon powered propeller" has a bitter sweet taste to it (pun intended).

So, next is... can the lemon be doped to give more power???

Tubular · 2011-07-31 18:06

Lol...

You're right I did start off with a potato, which would work fine too. Its just that our lemon tree is overflowing at the moment...

Phil Pilgrim (PhiPi) · 2011-07-31 18:37

Eureka! Final answer (to a previous question).

-Phil

ejwong · 2011-08-24 03:43

how about harvesting RF noise from those AM/FM stations?

this isnt stealing right? you're just scavenging leftover signals...

Tubular · 2011-08-24 04:51

Yep, it isn't stealing, and just as well because they would have a hard time catching you.

I'm not sure about the 'noise' harvesting, but I've seen a couple of examples where they 'tune in' to a nearby TV or Radio transmission tower as a source of small amounts of power. I gather you need to be within a few km's before other sources have more appeal (solar, kinetic, heat etc). Intel were looking at it if I remember correctly

Phil Pilgrim (PhiPi) · 2011-08-24 09:31

This may just be an urban legend, but I've heard of people burying loops of wire under high-tension lines to harvest power inductively. Apparently, the energy providers do consider that to be theft of services, since it results in an actual load -- albeit minuscule -- on the transmission system.

-Phil

ejwong · 2011-08-24 10:21

^ thats a true story, well not really... but its been done. burying a coil of wire under a powerline(the ones that carries 100K volts) will get you about >100watts. thats enough to power a couple of lightbulbs for free, plus these harvested energy is practically a waste of those transmissions(commonly known as electromagnetic field radiation)

its like fire on cooking, not all the heat goes into the frying pan, 60-80% of the heat goes into the room where the cooking is taking place.

gnuarm · 2013-02-12 16:20

Phil Pilgrim (PhiPi) wrote: »

This may just be an urban legend, but I've heard of people burying loops of wire under high-tension lines to harvest power inductively. Apparently, the energy providers do consider that to be theft of services, since it results in an actual load -- albeit minuscule -- on the transmission system.

-Phil

This is about to not be an urban legend any longer. I am currently working on a device that will scavenge power from the stray magnetic field from power lines. I expect it will require being fairly close to a power line and the amount of power will indeed be miniscule, just enough to power a clock. But scavenge it will!

Tubular · 2013-02-12 21:27

Brilliant, gnuarm! All the best to you...

I remember a lively discussion at uni one time when they had a guest lecturer from a high voltage transmission authority. Some of the other lecturers were trying to calculate the amount of power they could expect to scavenge from a safe distance.

Are you going to use the same LTC chip?

Keep us updated!

cheers
tubular

jfox · 2013-02-13 13:12

Just saw this yesterday http://phys.org/news/2013-02-german-student-electromagnetic-harvester-recharge.html#nwlt

Tubular · 2013-02-13 13:59

Thats a good story jfox.

I live for the day when batteries top themselves up - the longer you leave them alone, the more charged they are.

Perhaps someone could design a C or D sized battery, that has an AA cell inside it, plus harvesting circuitry in the spare space. That would be really something.

jfox · 2013-02-13 14:45

I just need to get my zero point energy harvester going and well be all set. I often wonder if those people that call the talk shows to complain about smartmeters are on their cell phones Jim

Lawson · 2013-02-13 15:08

FYI Tubular, if you can get the scavenged audio to charge the super cap above about 1.1 volts you can run the propeller direct off the super cap without the LTC chip. About 1.5 volts is needed to make the EEPROM happy though. Since you're using a fixed frequency audio signal, have you tried driving a resonant LC circuit to boost the available voltage?

Lawson

Mark_T · 2013-02-13 17:39

gnuarm wrote: »

This is about to not be an urban legend any longer. I am currently working on a device that will scavenge power from the stray magnetic field from power lines. I expect it will require being fairly close to a power line and the amount of power will indeed be miniscule, just enough to power a clock. But scavenge it will!

I smell a dubious business opportunity: aluminium "barbed" wire - sell to farmers to run across fields alongside overhead power lines - perfect disguise

Booting the Prop from low voltages, 2 of 3: Scavenged Headphone Audio @ 0.42V

Comments