EngIP 31 Posted August 17, 2012 Share Posted August 17, 2012 Does anyone know anything about leeching power? I'm contemplating powering a 430 through inductance of an AC power line, but I can't seem to find any decent resources on how best to achieve this. Quote Link to post Share on other sites
rockets4kids 204 Posted August 17, 2012 Share Posted August 17, 2012 Just wrap some wire around a power cord, rectify and filter whatever is induced, and measure voltage drop over a known resistance. From there you can calculate current and power. If you are not getting enough, experiment with more wraps. EngIP 1 Quote Link to post Share on other sites
GeekDoc 226 Posted August 18, 2012 Share Posted August 18, 2012 Just wrap some wire around a power cord, rectify and filter whatever is induced, and measure voltage drop over a known resistance. From there you can calculate current and power. If you are not getting enough, experiment with more wraps. This seems right to me (see sig below), but remember to just wrap one conductor of the AC line, or the two will cancel each other out. It's basically just a transformer. The supply side just happens to be powering something else as well. EngIP 1 Quote Link to post Share on other sites
EngIP 31 Posted August 18, 2012 Author Share Posted August 18, 2012 Thanks for the replies chaps, I was hoping that trial and error wouldn't have to be my first step but who am I to shun the tried and tested technique! If what GeekDoc says is true, I'm not going to be able to do what I wanted. I wanted to develop a contactless, batteryless remote power indicator - a board you can strap to an AC supply cord (say for a lamp) which is powered inductively and provides feedback on whether power is currently applied or not via RF. Looks like a CR3032 battery holder is going on the BOM. Quote Link to post Share on other sites
rockets4kids 204 Posted August 18, 2012 Share Posted August 18, 2012 I can tell you absolutely that you can get some serious voltage by just wrapping about 10 turns around a standard PC power cord. I tend to fidget, and once I wrapped a small piece of scrap wire around a power cord. I managed to get a nice little wake-up shock in the process. It can be done, and you don't need anything fancy. Quote Link to post Share on other sites
GeekDoc 226 Posted August 19, 2012 Share Posted August 19, 2012 Worked wrapping around both conductors? Huh. I learn something new all the time. Couldn't hurt to try (carefully), EngIP. Will only cost you a bit of time. Go for it! Quote Link to post Share on other sites
chibiace 46 Posted August 19, 2012 Share Posted August 19, 2012 voltage detector pens and those multimeters with the loop at the top can get readings but id be real interested to see a microcontroller powered from this. Quote Link to post Share on other sites
rockets4kids 204 Posted August 19, 2012 Share Posted August 19, 2012 it is *amazing* how little current the msp430 draws in sleep mode. Depending on the application, this could be no trouble at all. You can also use the tapped power to charge a supercap if you need lots of power at for brief periods. Quote Link to post Share on other sites
EngIP 31 Posted August 20, 2012 Author Share Posted August 20, 2012 it is *amazing* how little current the msp430 draws in sleep mode. My thoughts exactly! I was astounded when I saw the tables for power consumption for the 430. I've had a bit of a play today, but I don't think I'll be able to get anything meaningful done without building a circuit to charge a cap. 20 turns of .75mm wire round a 50Hz 240 a.c. power cable doesn't give enough juice to light a LED (I suspected it wouldn't but I got hopeful after hearing of rockets4kids experience). So the next step is rectification, smoothing and messing about with wire size and number of turns. Quote Link to post Share on other sites
rockets4kids 204 Posted August 20, 2012 Share Posted August 20, 2012 The amount of power required to light an LED is ginormous compared to what you need to run an MSP430. You are really going to need to put a load (1k - 10k resistor) across that coil and look at the voltage drop across it with a scope. Quote Link to post Share on other sites
EngIP 31 Posted August 20, 2012 Author Share Posted August 20, 2012 " a scope". Bah! Thanks for the advice, but if a scope is needed I'll be waiting a month till the labs re-open at uni. Or I could make a "scope" out of a LP and an LCD (measuring across the load resistor to the ADC's should work?). Quote Link to post Share on other sites
rockets4kids 204 Posted August 20, 2012 Share Posted August 20, 2012 Yes, you could do that. EngIP 1 Quote Link to post Share on other sites
EngIP 31 Posted August 24, 2012 Author Share Posted August 24, 2012 Just a little update on this, feel free to throw in any suggestions/explain the basics to me :oops: ... My intended application isn't going to be feasible I wouldn't have thought (RF TX/RX), but there may be another application (datalogging?) that this could prove useful for. I knocked together a rectifier and put an electrolytic cap on the output (100uF), and attached a 10k load. Pushing the resolution limits of my multimeter, 0.002V drop across the resistor was recorded. (0.002V/10k = 200nA ?) Leaving the load disconnected and allowing the cap to charge shows a charging rate of around 0.02V/min. After ~45 mins V across the cap had built to around 0.7V. Firstly, principles is not my strong point, so have I made some basic errors here, or do my figures make sense? Secondly, If I am correct in my figures, 200nA with a 0.2V drop indicates 0.4nW instantaneous power. As an 430F2xxx device needs 900uW at 1 MIPS, a charge time/powered ratio of 22000000:1 is needed with this setup (or 62 hours charge for 0.01 seconds activity). My thoughts so far are that with an FRAM chip (I think you can write individual registers instead of blocks of flash?) and the right application (sample temp and record once every hour), an inductively powered MCU could be possible. I don't really know what power and time would be used waking the chip, ADC's, etc., but for my application, it's a no go... ...unless I can code these issues out... Quote Link to post Share on other sites
GeekDoc 226 Posted August 24, 2012 Share Posted August 24, 2012 What kind of inductor did you make to leech the power? Can you add a few more coils for more power? Are you losing much through the rectifier? Pics would be cool. Quote Link to post Share on other sites
EngIP 31 Posted August 24, 2012 Author Share Posted August 24, 2012 Pics would mean showing the state of my "workstation" to the world :oops: But there really isn't anything to see. 4 x diodes arranged for the rectifier, a cap across the output. The inductor is literally a wire wound round the power line 20 or so times. Adding a few more coils won't help this setup much. I could improve the power harvested by building a better inductor (an iron core surrounding the power line, coils around the iron core), but that's too involved for my application. Actually, there'll be something I can salvage from my parts bin, I'll have a look tomorrow. I'm reading a 0.5V drop across the diodes of the bridge (lower than I expected tbh), and they're big ugly looking diodes, so I could probably get a more efficient rectifier setup, but from my preliminary examinations I'll need to do more than eek out another 5% from the bridge. I suppose it's not dead in the water yet, it's just going to need more specialist hardware. Quote Link to post Share on other sites
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.