# Free Energy | searching for free energy and discussing free energy

## Solid States Devices => Joule Thief => Topic started by: omdano on January 17, 2014, 08:50:19 PM

Title: Capacitors and Resistors ?
Post by: omdano on January 17, 2014, 08:50:19 PM
I told the Electrician in the store about my needs and he asked me about Voltage for my capacitor and Wattage for my Resistor , what should i tell him ?
How can i decide the voltage and wattage ?
Title: Re: Capacitors and Resistors ?
Post by: MarkE on January 17, 2014, 11:06:44 PM
Omdano, for ceramic capacitors you should choose a voltage rating that is at least twice the maximum voltage you expect that will appear across them in a circuit.  For Joule Thief circuits running from one AA cell, 16V will be good enough.  For a tantalum electrolytics the voltage rating should be at least 1.3 times the voltage.  On the LED side you should stay at 6.3V or above.  There is no harm in using a capacitor that has a higher voltage rating than you need.  It will just be more expensive and possibly bigger than you can get away with.  There are different kinds of ceramic capacitors.  You should use either NP0 or X7R for the timing capacitor, with NP0 preferred.  The alternative is a polyester film capacitor, but they are much more expensive.  Whatever you do, do not use a Y5V ceramic capacitor.

You can figure out the resistor wattage from Ohm's Law and a 2X safety factor.  P = 2 * duty_cycle^0.5 * V^2/R .  If you don't know the duty cycle then just assume that it is 1.  Nothing in a typical Joule Thief circuit is more than 10V, so any resistor 1.6K Ohm or bigger can safely be 1/8 Watt or bigger.  Anything over 800 Ohms can safely be 1/4W or bigger.  If you know the voltages more precisely then you can be less conservative.

Perhaps the most important thing in a Joule Thief circuit from a standpoint of making measurements will be what you do for your current sense resistor, or resistors.  A non-inductive current sense resistor with leads soldered right to the body for your oscilloscope connections and the normal circuit connections through the device leads is a good simple approach.  For examples see poynt99's you tube videos.
Title: Re: Capacitors and Resistors ?
Post by: omdano on January 17, 2014, 11:10:36 PM
So i am safe using a 35v Capacitor (cermaic) and a 0.5 watts resistor on a circuit with low voltage input (200mV)
Title: Re: Capacitors and Resistors ?
Post by: MarkE on January 17, 2014, 11:31:09 PM
Omdano, the input to the circuit may be 200mV but you also need to know what the boosted output is.  If it is driving an LED it will be a little bit more than 3V, with switching spikes that might be twice that.  35V is fine for the capacitor, and I would be very surprised if any of the resistors need to be a higher Wattage than 1/4W.  Post your schematic and I will tell you if anything needs to be changed.
Title: Re: Capacitors and Resistors ?
Post by: omdano on January 17, 2014, 11:47:26 PM
Thanks buddy the Schematic is attached
Title: Re: Capacitors and Resistors ?
Post by: MarkE on January 18, 2014, 12:29:57 AM
The 100uF capacitor I recommend that you put use a 6.3V or higher voltage electrolytic.  You want this connected in a very short loop from the top of the 4 turn winding to the source of the JFET.

The 2.2K resistor can be 1/10W such as an SMT 0603 size, or greater Wattage.  You will probably be using parts with leads, so 1/8W or greater is fine.

The circuit will perform best if you keep the loop of the 100 turn winding to the LED to the 1n5711 to the capacitor very short.  I would add a 0.1uF capacitor across all that again keeping the leads short.  Wiring inductance in the two loops makes the circuit noisier and less efficient.
Title: Re: Capacitors and Resistors ?
Post by: Pirate88179 on January 18, 2014, 02:19:44 AM
Wow, this question depends on a lot of things.  All of my Joule Thief circuits run from a single AA battery and output over 400 volts.  Some in the 900 volt range.  The resistors I use are only to establish a good resonance for the circuit.  A lot of times, I use some sort of VR or trimpot so I can continue to tune the circuit and keep it in resonance as the input voltage drops.

I don't know if this helps or not but I suggest looking at the main Joule Thief topic and see what others have done with this circuit.

The only thing I use a cap for in the circuit is in place of the battery.  In that case, I use supercaps.

Bill

***EDIT***

Mark knows his stuff and can best advise you on the circuit that you posted.  (I am not sure that is an actual JT circuit)  The ones I am making can easily light hundreds of leds for a long time on a "dead" battery.  Here is an example of the Lidmotor/Jeanna JT circuit.
Title: Re: Capacitors and Resistors ?
Post by: MarkE on January 18, 2014, 03:30:11 AM
Pirate, the circuit shown is limited to low voltages by the LED in series with the Schottky diode.  Construction quality will determine the extent of voltage spikes.  6.3V is safe for the input capacitor over a good range of construction practice.  The JFET input current is miniscule allowing the resistor Wattage rating to be similarly small.
Title: Re: Capacitors and Resistors ?
Post by: Pirate88179 on January 18, 2014, 03:47:36 AM
I am sure you are correct on all counts.  What I do not understand is why someone would want to build that circuit?  You can run hundreds of leds from other JT circuits with very low mA draw.  What would make that circuit worth building?

PS  I edited my previous post while you responded.

Bill
Title: Re: Capacitors and Resistors ?
Post by: MarkE on January 18, 2014, 07:27:41 AM
Pirate, because a JFET is a depletion mode device circuits using JFETs can be made to start down to very low voltages.  With care down to a mV or less is theoretically possible.  A lot of people don't know this, but  there are some depletion mode MOSFETs out there with decent voltage ratings.  Those allow for harvesting circuits that can produce much higher voltages than JFETs without the losses and low gain of a BJT like the venerable 2N3055.  When you get into high turns ratios it is also a good idea to add a snubber and/or clamp to protect the switching transistor against spikes caused by the transformer leakage inductance.

I think that low voltage circuits like the JT are fun because they afford experimentation where a lot of people don't expect that a circuit can work.  With some creativity and effort people have been able to show impressive results.  If you want to try something a little different you could devise up a two stage device:  A first low power stage that provides 3V of bias to a second stage that uses a power MOSFET that switches very cleanly.