Print PDF
Favorite
Email
Step 5Putting it All Together (And Caveats)
Great, now we have the primary coil and oscillator, a secondary coil, and a CW generator.
Of course the purpose of this instructable is that there isn't any 'hooking up'! If you power the oscillator and have the secondary coil inside/on/around the primary coil, you will notice that a fairly decent voltage is generated on the secondary coil. Although usually just over half the voltage on the primary, the CW generator takes that AC voltage and conveniently produces a clean 5V. Ideally this can be used to charge a large capacitor bank like I did, or a low power 5V circuit. Now for the caveats.
This first one being that you can't really use this for any high current applications, such as driving motors or a bunch of LEDs. Charging is a different matter of course and will work perfectly ok for that. When you try to pull too much current out of this circuit the voltage will start dropping considerably. For instance when you connect a fully depleted capacitor bank to it the voltage across the secondary is considerably reduced. If you take a look at the primary during this time you will also see that the frequency and amplitude of the wave is considerably different. This frequency shifting is what prevents you from using high current loads. I'm sure there are better oscillators and other measures that can be taken to improve the voltage regulation, but this works as a preliminary model.
The second caveat is that you can't put ANY metal in between the primary and secondary, particularly iron based metals (steel, stainless or otherwise). Even placing the oscillating circuit inside the primary effects the performance, creating drop-out zones on the upper surface of the picture frame that prevent charging when the secondary is placed in certain spots.
The third caveat is that the distance between the primary and secondary coil must be kept to a minimum. This isn't WiTricity, it can't power anything over a distance of even 20cm.
Working around these limitations is quite easy though. My method was to use the circuit to charge a large capacitor bank (3F @ 5V) and then use that bank to power a switching regulator (to keep a constant 5V even when the capacitor voltage drops) and LDO so I have both 5V and 3.3V to work with. It takes about a full night to charge the capacitors, and I can get a considerable amount of run time with proper power saving attention to the rest of my circuit.
Up next? Well maybe a larger, more robust version to trickle charge a car battery, or a nicer looking version to charge a cell phone battery. Maybe some experimentation in flat-wrapped coils or other methods. Feel free to expand upon my methods and improve this tech, and heck, take my idea and integrate it into some of your existing projects. Just as long as you promise to make an instructable or something out of it, I'm interested to see what people will do with this! Also make sure to respect my creative-commons by-sa license ;).
Happy Hacking,
-Devin
Of course the purpose of this instructable is that there isn't any 'hooking up'! If you power the oscillator and have the secondary coil inside/on/around the primary coil, you will notice that a fairly decent voltage is generated on the secondary coil. Although usually just over half the voltage on the primary, the CW generator takes that AC voltage and conveniently produces a clean 5V. Ideally this can be used to charge a large capacitor bank like I did, or a low power 5V circuit. Now for the caveats.
This first one being that you can't really use this for any high current applications, such as driving motors or a bunch of LEDs. Charging is a different matter of course and will work perfectly ok for that. When you try to pull too much current out of this circuit the voltage will start dropping considerably. For instance when you connect a fully depleted capacitor bank to it the voltage across the secondary is considerably reduced. If you take a look at the primary during this time you will also see that the frequency and amplitude of the wave is considerably different. This frequency shifting is what prevents you from using high current loads. I'm sure there are better oscillators and other measures that can be taken to improve the voltage regulation, but this works as a preliminary model.
The second caveat is that you can't put ANY metal in between the primary and secondary, particularly iron based metals (steel, stainless or otherwise). Even placing the oscillating circuit inside the primary effects the performance, creating drop-out zones on the upper surface of the picture frame that prevent charging when the secondary is placed in certain spots.
The third caveat is that the distance between the primary and secondary coil must be kept to a minimum. This isn't WiTricity, it can't power anything over a distance of even 20cm.
Working around these limitations is quite easy though. My method was to use the circuit to charge a large capacitor bank (3F @ 5V) and then use that bank to power a switching regulator (to keep a constant 5V even when the capacitor voltage drops) and LDO so I have both 5V and 3.3V to work with. It takes about a full night to charge the capacitors, and I can get a considerable amount of run time with proper power saving attention to the rest of my circuit.
Up next? Well maybe a larger, more robust version to trickle charge a car battery, or a nicer looking version to charge a cell phone battery. Maybe some experimentation in flat-wrapped coils or other methods. Feel free to expand upon my methods and improve this tech, and heck, take my idea and integrate it into some of your existing projects. Just as long as you promise to make an instructable or something out of it, I'm interested to see what people will do with this! Also make sure to respect my creative-commons by-sa license ;).
Happy Hacking,
-Devin
| « Previous Step | Download PDFView All Steps | Next Step » |
If I understand the description you are suggesting two pathways for the secondary
- Just use the two capacitors and the 4-diodes to get V(sec in mV) X (1.41X1.41X1.41X1.41)
in step three you provide an applification with a transitor ( +-5V supply) and you get the voltage from the ground-node between capacitors
The primary resonates at 50 KHz
How do I find the resonance of the secondary?
Send info also, with drawings to hitemag@ims.demokritos.gr
The secondary and primary coils will resonate at the same frequency.
I don't understand what you mean by 'it flashes'. Do you mean it flashes once and then doesn't turn on? Or do you mean that it flashes continuously. If it does either you probably have set up the circuit wrong. Technically when the circuit is working correctly the LED will flash, but at many kHz so you can't see the flashing.
I'm a bit confused at how your LED is doing what it's doing. Are you using the correct capacitors for the CW generator? What happens when you remove the CW generator and simply measure the RMS (root mean squared) AC signal across the secondary coil? Just turn your multimeter to AC and measure the voltage, it should be approximately the same voltage as what's on your primary coil.
It sounds to me like you aren't using an oscilloscope, which would make it exceedingly difficult to match the coils and the two circuits. That may be your main problem.
1. increase oscilation frequency, SEE what MIT did is to use 10MHz, which will induce much more voltage.
2. consider about the Q factor of the oscilator. to increase the Q should give better efficience.
for your second caveat, I think you just mean can't put iron based metal between, but actually to put a soft rion material in the middel of secondary coil should make the circuit work better.
anyway, you really done a good instructables. thanks for share
Problem with that of course is that everything has to be a lot more precise, so you have to be able to make really nice coils. If I took more time or used some kind of... jig or plastic former I might be able to make them nicer and higher precision. A better oscilloscope would help too (luckily I have access to some nice ones at work).
Q factor is actually related to the frequency:
Q = 2*pi*freq*inductance/(coil resistance) //For Colpits osc.
So we see here to increase Q we increase frequency, inductance, or decrease coil resistance. Bandwidth is inversely proportional to Q.
Now about the soft iron, placing it in the coil 'focuses' the magnetic field, making the effective inductance of the coil much higher. If you increase the inductance of the secondary its resonant frequency would no-longer match that of the primary. However, a good way of increasing the Q would be increasing the inductance of the coils, so putting a soft iron core in both coils would give you a much better Q with a slight increase in coil resistance (due to eddy currents perhaps?). That's a good idea though, it might make the effective power transfer way higher!
Increasing the inductance decreases the frequency, so the percieved benefit may not be there, and would require some calculation.... Hmmm...
Q=2*pi*f*L/R
f=1/(2*pi*sqrt(L*C) sub f into Q
Q=L/(sqrt(L*C)*R) multiply both sides by 1/sqrt(LC) and reorganize
Q=sqrt(L*C)/(C*R)
Therefore increasing L with a soft iron core will increase Q and the efficiency, but the practicality of it limits how much (the sqrt hampers any linear increase in efficiency).
Q=sqrt(L/C)/R
obviously, increase L, decrease C or R, could increase Q.
but I think to make such a induction power, the first is to satisify a higher resonant frequency, that is f =1/(2pi*sqrt(L*C)).
why?as I think and consider about the quantum theory, E=hv, h is Planck constant and v is the frequency of a paticle. I just guess higher frequency could concentrate(or transfer energy more efficient) much more energy compared to a lower frequency on a same bandwidth.
after decide which f , then, consider about the Q, I read from wikipedia, it says MIT the make the Q to be around 1000.
I roughly cal your Q=sqrt(53e-6/100e-9)/R
assumed R =1 ohm , then Q= 23,
acctually, there should be many details if we want to make a good and professional resonent power, but your home-made is already very good.
1) You have to make sure that the secondary is exactly tuned to the primary. My circuit has a modest bandwidth so it doesn't have to be perfect. Lowering the bandwidth means you have less room for error and the secondary won't be excited as easily unless it's right on the frequency.
2) Decreased bandwidth means that there is more available energy to be transferred across the link, so the link is more selective.
My Q won't be that high (I haven't bothered to figure out the real-world Q) but it sure is easier to build! Also, the Q for both circuits is different because the R will be different between the two (I'm using a large coil and a small coil). And the R will be <0.5Ohms for small coils, roughly speaking. The coil calculator I linked at the bottom will actually tell you a rough approximation of the R that you will get from your coils.
I'm a bit surprised that you brought out the Planck equation. That describes photons and other quantum particles, and doesn't deal with magnetism. Remember, this is only magnetic, there are no electromagnetic effects happening here (barring EMI). Someone else also brought this up in the comments and it is simply incorrect. The effect is the same though; high frequency=higher Q and lower bandwidth=more available energy to transfer :).
It sounded do-able at first, but what got me thinking was tuning the wavelengths to the resonant frequencies and how hard that would be. I assumed that was the dead end, so i am extremely surprised you have made a tutorial explaining this for me!
Just so you know though, There is a company known as "wi-tricity" who are past the prototype stages of a commercially available product. Even though they could only patent the product and not the concept as it isn't exactly theirs, i thought i would let you know anyway.
Good one! :)
Then again, people have found out numerous ways of gaining light energy - even from sound! I believe one can create light from sound by "imploding bubbles in a liquid" which gives off a short burst of light. I think this is called "sonoluminescence"?? Either way, you never know what someone may discover next. We may be close to discovering something entirely different - a different force maybe? In this age of smashing particles together, you never really know what's around the corner.