Faraday Flashlight Project/Experiment

[schnookumz03]

Hi everyone! I am a high school student and my final project is to create my own Faraday Flashlight. I used the following website to find out what materials I needed:
http://www.shake-flashlights.com/how-they-work.html

After having bought all the supplies, coiling the wire manyyy times around a PVC pipe, sandpaper-ing the ends of my wires, and hooking everything up my flashlight did not work. I tried moving the magnet quickly but it still was not working.
I understand that LED lights only work in one direction and that the circuit would have an alternating current, and we (my teacher and I) have done just about everything to see where I went wrong. The wire works just fine (we tested it using a battery and the light bulb which also works just fine), the capacitor works, and the magnets are relatively strong.

We hooked up the system so that the light bulb and the coils were in parallel (just like in the diagram in http://www.shake-flashlights.com/how-they-work.html).

We even tried making the flashlight with only the light bulb, coils and magnet but we could not get a current out of it. What am I doing wrong? Please help me.

Thank you!
Charlene

 

[lewando]

I've seen these things work well when they use a diode rectifier. Not sure how the circuit in the link you provided would work without one. Something like http://www.electroniccrafts.org/?n=Main.Shakelight should work.

 

[schnookumz03]

Thank you!

 

[schnookumz03]

Hi again.

I did my project with the bridge rectifier using four IN4004 rectifier diodes, but it still would not work. I set up everything in the correct manner, but and I have no problem generating a current with my coils and magnet. I have gotten up to 0.25A and -0.21A with the alternating current. After having plugged in the bridge rectifier, all my current is gone. It's as if it ate up all my current. What should I do? I built the bridge correctly and I plugged my coils into the correct spot.
Please help me.

Thank you!

 

[lewando]

These are some thoughts:
1) Coil may not be putting out enough voltage. Voltage may not be high enough for 1N4004 to conduct. Do you have access to any Schottky diodes? They require less forward voltage to conduct.

2) If 1) is the case, you'll need to change out the diodes or beef up your coil (more turns) to put out more volts. Probably should do both.

3) Before you replace the diodes or change the coil, let's make some measurements. If you only have only the coil and bridge connected, connect your ammeter to the two points where your coil connects to the bridge (these two points are the only points where the cathode of one diode connects to the anode of the other). You should see an AC current (positive and negative current pulses, like you did before) when you shake the magnet through the coil. Can you try this and report your result? While you are still setup for this, put your meter into voltage mode, repeat, and report the voltage.

4) What LED are you using? what is the value of the current limiting resistor?

5) Assuming you can eventually get some voltage out of the rectifier, you should consider putting the capacitor back in the circuit. What size/kind were you using?

6) There probably should be a switch in series with the LED so that when it is turned off, you can shake and charge the capacitor for a while (a couple minutes maybe?). Then turn the switch on.

 

[NascentOxygen]

I can't see any mention that you are using a capacitor. You will need it--your handwound coil and magnet will not generate current enough to flash the LED with every shake. Expect to need hundreds of shakes to generate enough charge to flash the LED just once.

First, shake to charge the capacitor, for this you will need a diode in series with the coil to stop the capacitor immediately discharging back through the coil. With each shake the voltage on the capacitor should step up a bit. Finally, close a switch to allow the capacitor to discharge through the LED. A 100 ohm resistor in series with the LED should do.

If you have a digital voltmeter, connect it across the capacitor as you shake, to monitor the voltage stepping up with each shake.

I think I'd start with a capacitor of a 100 uF and see how that works out.

Don't use a bridge rectifier, there's probably not enough voltage to overcome 2 diode drops. Just use a single diode.

 

[vk6kro]

The magnet should be completely outside the coil for part of each shake, too.

If it stays inside the coil, the two poles produce equal and opposite voltages in the coil and these tend to cancel each other out.

You do need a diode, but it should be a Schottky diode for best results. These have a very low forward voltage drop, so more voltage will get stored in your capacitor.

White LEDs need a voltage of about 3.5 volts before they will show any light at all. If you just want to observe an effect, a red LED will be more likely to light up since these only need about 1.2 volts.

 

[NascentOxygen]

White LEDs need a voltage of about 3.5 volts before they will show any light at all.

The data sheets show a drop of a couple of volts for white LEDs, yes. But then how do those solar garden LED lights work? They operate off a 1.2v cell, and there is no transformer or coil in their circuit board.

Any idea? Maybe that would be a good source for a LED for this project.

 

[vk6kro]

Garden lamps have a small oscillator circuit with a transformer which steps up the voltage. This is then rectified and used to operate the white LED.

You can see many such circuits here:
http://www.google.com.au/search?tbm...830l0l26l24l0l9l9l0l790l5272l0.4.3.2.3.2.1l15

 

[NascentOxygen]

Garden lamps have a small oscillator circuit with a transformer which steps up the voltage. This is then rectified and used to operate the white LED.

I took a closer look at a one-rechargeable-cell garden solar light I have. What I'd taken to be an electrolytic in heat-shrink plastic is in fact a 0.25mH inductor, a 2 terminal device, enclosed in heat-shrink.

The whole circuit comprises: 4 transistors, 4 resistors, one tiny ceramic cap near the 0.25mH inductor, and 2 diodes and a 47uF cap. I swung the thing madly around in the air, but could discern no strobe effects, so conclude the LED is being fed DC, not a chopped waveform. The component list supports this conclusion, with the 47uF storing the flyback kick from the inductor.

 

[lindaopa]

Garden lamps have a small oscillator circuit with a transformer which steps up the voltage. This is then rectified and used to operate the white LED.

 

[NascentOxygen]

Adding to a thread from 12 months back ...

A solar garden light that has given 5 years of sterling service recently died. I took the opportunity to examine its innards. What a surprise. No transistors! Its white LED is soldered to a tiny PCB, along with just two other components: a 220 ohm resistor and a 1N58?? diode. There's nothing else. There are two wires directly from the PV panel, plus the two directly from the NiCd cell holder (though after 2 years I substituted NiMH). That's all the electronics it contains, thus demonstrating this white LED runs at less than 1.2V. (I'll try to borrow a meter and measure its actual voltage; I'm away from home base ATM.)

This debunks the belief that a single-NiCd solar light requires a voltage booster before it can power a white LED.

What further marvels are set to emerge from the China powerhouse?

I took a closer look at a one-rechargeable-cell garden solar light I have. What I'd taken to be an electrolytic in heat-shrink plastic is in fact a 0.25mH inductor, a 2 terminal device, enclosed in heat-shrink.

The whole circuit comprises: 4 transistors, 4 resistors, one tiny ceramic cap near the 0.25mH inductor, and 2 diodes and a 47uF cap. I swung the thing madly around in the air, but could discern no strobe effects, so conclude the LED is being fed DC, not a chopped waveform. The component list supports this conclusion, with the 47uF storing the flyback kick from the inductor.

 

[rude man]

Couldn't the solar cells be wired in series to produce higher voltages?

 

[NascentOxygen]

Couldn't the solar cells be wired in series to produce higher voltages?

Remaining in context, mooonlight is insufficiently bright.

 

[vk6kro]

Couldn't the solar cells be wired in series to produce higher voltages?

Yes, you could wire some solar cells in series, but they would still be unable to charge a single NiCd cell to more than 1.2 volts or so.

This would not normally be enough to light a white LED which needs 3.5 volts to light up.

So, there must be a transistor or an IC in there to switch the voltage from the NiCd to an inductor to step up the voltage.

I suppose if none is visible, it would be possible that the LEDs are special ones with the switching transistor built in. I've never heard of this being done, but it might be possible.

 

[NascentOxygen]

LED package is transparent, so I can see it contains no storage elements.
The LED is not strobed, there's no switching going on in there.

 

[vk6kro]

LED package is transparent, so I can see it contains no storage elements.
The LED is not strobed, there's no switching going on in there.

Can you view the voltage on the LED with an oscilloscope and verify that it is not strobed and that the maximum voltage is that of one NiCd cell? If someone is producing white LEDs that can do this, it would be amazing.

You can't tell if it is strobed by "I swung the thing madly around in the air" because it is probably strobed at 40 KHz or more.

 

[NascentOxygen]

Can you view the voltage on the LED with an oscilloscope and verify that it is not strobed and that the maximum voltage is that of one NiCd cell? If someone is producing white LEDs that can do this, it would be amazing.

The LED voltage is much less than 1.2V because of that 220 ohm series resistor. What's amazing is that they were manufacturing this at least 5 years ago. I'm away from home and don't even have a voltmeter with me. But you could strobe the LED all you like, with no L or C your switching would be futile in pumping up V.

It's the PV panel that is broken, so I'll still be able to measure the working voltages when I get hold of a meter.