RESONANCE EFFECTS FOR EVERYONE TO SHARE

[gyulasun]

Hi Duff,

It would be better to use full wave rectification so that half of the output power be not lost.

You may wish to use a bridge made from 1N4148 or 1N914 diodes if you do not happen to have low power Schottky types.  (The best would be to have Germanium types  -1N34 or similar- to minimize forward voltage drop.)

Or as an alternative you may consider a not often used voltage doubler  which consists of two half wave rectifiers added together, see title "Rearranging the Doubler Circuit"  here:
http://www.play-hookey.com/ac_theory/ps_v_multipliers.html   Capacitor C1 has its positive leg on the common point of the diodes and C2 has its positive leg on D2 cathode of course and no need for a ground connection as the text suggests, it is a possibility for a reference point.    (Notice: I am aware of a voltage doubler does not double the power of course.)   The point is: this rectifier utilizes both half waves of the output voltage and uses only one diode for each half cycle, cf to a bridge where two diodes conduct in each half cycle.  You may use the full wave voltage doubler too of course, shown above the rearranged one.

rgds, Gyula

EDIT:  Although Luc mentioned his rectifier diode and puffer capacitor,  I cannot recall he used them, he did not use them in his video, so consider this to compare apples to apples...

[gotoluc]

Hi Duff,

it sounds like you're getting a better output.

Did you try your 0.68 ohm series resistor to calculate the current on either side of the Signal Generator to see if the noise level is less on one side then the other. I had the same thing but I used the side that is the cleanest. I also noticed if I unplugged the power to my laptop and use it on batteries to power it and the USB scope I would get a slightly cleaner form. I even connected my Signal generator to a inverter powered by a marine 12vdc battery and that also help clean noise. I can make the circuit work and test it completely off batteries!... which also eliminates noise or possible coupling.

BTW, I don't know at what size your capacitance is at to get your output of over 3 volts but keep in mind that too high a capacitance will reverse the resonant circuit savings. From all the tests I have done to date, there seems to be only one ideal (energy saving) capacitor size at a specific frequency for an inductor within a specific pulse generator circuit.

Thanks for all the time you're taking to research and share your findings.

Luc

[Montec]

Hello Duff
This may or may not apply to your coils but a Tesla Coil (TC) is usually made up of two air coils that have the same resonant frequency. The voltage increase/decrease is not based on the turns ratio but is based on a ratio of the inductance's of the two coils.  See http://www.scribd.com/doc/3876818/Denicolai-Tesla-Transformer-for-Experimentation-and-Research-96pp2001  for more info.

[duff]

@Gyula,

It would be better to use full wave rectification so that half of the output power be not lost.

You may wish to use a bridge made from 1N4148 or 1N914 diodes if you do not happen to have low power Schottky types.  (The best would be to have Germanium types  -1N34 or similar- to minimize forward voltage drop.)

Luc used diodes in some of the previous videos. First I wanted to replicate his results before I started changing the circuit. Adding a FWB was the one of the first changes I was going to do to improve the output.

Luc had determined several weeks ago that SBL3040PT diode got the best results, so I built a FWB using them. I then compared the output with that of a 1n4148 FWB. The SBL3040 did the best job, however strangely the half-wave bridge performed better. Not what I expected...


1n4148 Full-Wave Bridge
3042 KHz - 3.018VDC

SBL3040 - Half-Wave Bridge
306.2 KHz - 3.59VDC

SBL3040 - Full-Wave Bridge
303.2 KHz - 3.285VDC

All were using a 330V 80uF capacitor from a disposable camera


Take a look this video Luc previously posted. It  immediately caught my attention, Why? Because Luc charges 12,000 uF caps almost instantly from the secondary using a 1/2 wave bridge with the SBL3040.


http://www.youtube.com/watch?v=ETZaMrnIwxk


@Luc

BTW, I don't know at what size your capacitance is at to get your output of over 3 volts but keep in mind that too high a capacitance will reverse the resonant circuit savings. From all the tests I have done to date, there seems to be only one ideal (energy saving) capacitor size at a specific frequency for an inductor within a specific pulse generator circuit.

Initially I used a 10uF cap with the SBL3040PT diode and was getting the 3.221V out, however now I have a 330V 80uF cap in the circuit. I'm still getting about the same voltage but I was trying to smooth out some of the ripple.

It seems you have to increase the magnetic flux density within an small area. I think that's why my taller inductors were not working even though they had little higher inductance. If you remember the output was only 0.66V

I know you did some experiments with the flat bifiliar spiral but I might try a few more being I already have a small one wound.

Did you try your 0.68 ohm series resistor to calculate the current on either side of the Signal Generator to see if the noise level is less on one side then the other. I had the same thing but I used the side that is the cleanest.

Moving the current sensing resistor to the other end of the circuit places it at the positive lead of the function generator. Then when you connect your probe and ground lead across the resistor you introduce a ground in the middle of your circuit. That upset things...


Below are a few images. I wanted to show the wave form across the 0.68 current sensing resistor which for me presents a problem of calculating input current.

The voltage is now up to 3.281V. Last night I reported 3.221 but I did not have the function generator amplitude turned up all the way.

As stated above I've got a FWB in place.
Voltage input from SG is 25Vpp / 8.84rms

-Duff


[EDIT]
The text in the scope wave form image came out too small to read but it is the waveform across a 0.63 ohm resistor.
1uS/Div
1V/Div D/C

The one below says: Voltage across a 66.3 ohm resistor: 3.281V

[gotoluc]

Luc had determined several weeks ago that SBL3040PT diode got the best results, so I built a FWB using them. I then compared the output with that of a 1n4148 FWB. The SBL3040 did the best job, however strangely the half-wave bridge performed better. Not what I expected...

1n4148 Full-Wave Bridge
3042 KHz - 3.018VDC

SBL3040 - Half-Wave Bridge
306.2 KHz - 3.59VDC

SBL3040 - Full-Wave Bridge
303.2 KHz - 3.285VDC

All were using a 330V 80uF capacitor from a disposable camera

Great job Duff , I'm happy to see you also confirming my finding on the diode. I tried about 30 different kinds of diodes that I pulled out of old CRT monitors and it is quite surprising to see how well these SBL3040 do.

@Luc

Initially I used a 10uF cap with the SBL3040PT diode and was getting the 3.221V out, however now I have a 330V 80uF cap in the circuit. I'm still getting about the same voltage but I was trying to smooth out some of the ripple.

The information I posted was about the capacitor within the primary LC.

It seems you have to increase the magnetic flux density within an small area. I think that's why my taller inductors were not working even though they had little higher inductance. If you remember the output was only 0.66V

I think that maybe part of it . Also try raising your secondary by hand and see if it gives you a better output at a certain location but also keep an eye on the input so you don't use more power. Once you find the sweet spot place some non metallic supports to keep it there. That is why in my last video I have supports on the bottom of my secondary

I know you did some experiments with the flat bifiliar spiral but I might try a few more being I already have a small one wound.

Sounds good. Please let us know what you find good or not.

Moving the current sensing resistor to the other end of the circuit places it at the positive lead of the function generator. Then when you connect your probe and ground lead across the resistor you introduce a ground in the middle of your circuit. That upset things...

Good point. Maybe I was able to do it when working my scope on laptop battery

Below are a few images. I wanted to show the wave form across the 0.68 current sensing resistor which for me presents a problem of calculating input current.

Yes, I see what you mean .... this is why I got this USB scope... the software does the RMS surface calculations... I don't know how you can calculate that

Excellent post Duff ... thanks for sharing

Luc