Use for the Tesla Switch

[Kepszlok]

In the drawing you post under the 555 timer you have "Driver Transistors". This is where the problem is. You are most likely using an NPN to turn on from high signal, and PNP to turn on the low.
I do not know what power level your using through your 55 but 12 volt is most likely.
So picture what I am telling you.
The NPN is ON to drive one side. You have 12 volt on the base of the NPN to keep it ON and hold the PNP OFF.

Now your signal goes LOW, but it take time even if it seems fast it takes time. The NPN will not shut off until you are around .5 volt on the signal. The PNP turns ON as soon as you .5 volt of difference between the signal and the power on the Emitter. Current can start to flow.
Think about what I am telling you.

During the transition between HIGH and LOW, both transistors are on. So are the transistors between the batteries. You had big heat problems I did not follow that earlier because I did not know how your were driving this.

Those "driver transistors" are small BC549's to drive the optocouplers.
555 -> Dflipflop -> BC549 -> optocoupler
I simply don't want to drive the optocouplers directly from the Dflipflop's outputs. There was an early TS version with 2*3 optocoupluer

So, i shoud use a small delay after every switching to let the conducting transistor goes off before the other one starts conducting right?

I can do it, i have a spare board with an fpga around. I realy want to get rid of the 555 anyway.

[Matthew Jones]

Those "driver transistors" are small BC549's to drive the optocouplers.
555 -> Dflipflop -> BC549 -> optocoupler
I simply don't want to drive the optocouplers directly from the Dflipflop's outputs. There was an early TS version with 2*3 optocoupluer

So, i shoud use a small delay after every switching to let the conducting transistor goes off before the other one starts conducting right?

I can do it, i have a spare board with an fpga around. I realy want to get rid of the 555 anyway.

Just enough off time to make sure things have completely turned off. Its easy with an IC it won't execute the next command until its done with the current command so that give you ample time.

[Kepszlok]

Just enough off time to make sure things have completely turned off. Its easy with an IC it won't execute the next command until its done with the current command so that give you ample time.

Got it, thank you! I will be back when the new system works.

[Kepszlok]

Well, my new clock source with fpga are produces nice and clean square waves but it didn't really improved the situation.

See picture1, that shows the optocoupler's 'output'. Picture2 shows the voltage in the toroid's primers.

That crossing between the switchings are conserns me. It takes 2ms while both switches are on, right? Shoud i add a 3ms "all off" delay between the switchings?
2ms rise / fall time are not a short time, what can cause it? I can only think to the transformer's inductance.


Picture3 is different, i put a 0.22ohm power resistor in series with one of the primers to get some current values to calculations. At the moments of switching on, the current goes up to around 3A (630mV spike) and then it stabilizes very fast at around 0.6A (120mV).

[Matthew Jones]

Well, my new clock source with fpga are produces nice and clean square waves but it didn't really improved the situation.

See picture1, that shows the optocoupler's 'output'. Picture2 shows the voltage in the toroid's primers.

That crossing between the switchings are conserns me. It takes 2ms while both switches are on, right? Shoud i add a 3ms "all off" delay between the switchings?
2ms rise / fall time are not a short time, what can cause it? I can only think to the transformer's inductance.


Picture3 is different, i put a 0.22ohm power resistor in series with one of the primers to get some current values to calculations. At the moments of switching on, the current goes up to around 3A (630mV spike) and then it stabilizes very fast at around 0.6A (120mV).

You should have a clear off time between the two side. Your close but it might still be crossed over a bit.
I am not sure if it will make a difference. You can look to the signal of an inverter both from the low voltage side and then to the inductor. A pure sine inverter has very small pulses but on the inductor they look like one pulse.
So maybe adding a larger off time will help clear the inductor before it switch's directions. If you can, that is.

Matt

[Kepszlok]

You should have a clear off time between the two side. Your close but it might still be crossed over a bit.
I am not sure if it will make a difference. You can look to the signal of an inverter both from the low voltage side and then to the inductor. A pure sine inverter has very small pulses but on the inductor they look like one pulse.
So maybe adding a larger off time will help clear the inductor before it switch's directions. If you can, that is.

Matt

I added 3ms delay between the switchings, where both driver signals are off. This has absolutly no effect on the crossing, the oscillogram are the same. I will try with more delay tomorrow.

[shylo]

Sorry to interrupt, But can a cap that is gaining in charge(as the mag rotor is turning) be used to fire a transistor?
It's just a switch right?
Thanks for any help.
artv

[Matthew Jones]

Sorry to interrupt, But can a cap that is gaining in charge(as the mag rotor is turning) be used to fire a transistor?
It's just a switch right?
Thanks for any help.
artv

Confirm this because it is basic electronics 101. When you serialize 2 capacitors you loose 50% of your energy. If you are strictly using capacitors you are throwing away power. Make sure to confirm this by understanding BASIC electronics.

Confirm this as well...Capacitors do not charge and discharge and the same ratio unless they are high quality capacitors. Which means LOW ESR rating and High frequency resonant behavior with very small capacitance.

Last capacitors are not the answer for every project posted. Please avoid confusion and specifically state your theory as per the SUBJECT of each thread. Along with theory please post relevant examples of your relatively simple answer to everything. Only through example is a case made for change or possible experimentation. Using the same suggestion multiple time over multiple thread may expose ignorance and result in humiliation.

Thanks you for your suggestion, no further suggestions without example are needed. Thank you.

[Kepszlok]

I added 3ms delay between the switchings, where both driver signals are off. This has absolutly no effect on the crossing, the oscillogram are the same. I will try with more delay tomorrow.

Adding much more delay showed an intersting thing. After switching off a side, it's voltage are drops slowley while the other primer's voltage startes to rise up. It's normal, the mutual induction does this, but this can also be the source of the problems because there will be always some crossing while the two primer sits on the same core. - if i'm right.

I'm running out of ideas. I will rebuild the whole circuit tomorrow from wire to wire, hope it helps but i'm sure of it is built properly.

[Matthew Jones]

Must have something to do with the toroid. I have something running right now that is similar with a small ferrite transformer, EE core or something like that, anyway it has clear off time. I have never tried the outlined circuit with a toroid. Always used EI cores.

I wish I could hand you the magic bullet and fix the problem, unfortunately...

If your willing to move on you come to the Basic Free Energy Device and start using boost circiuts instead of transformers. It really does work better.


Matt

[Kepszlok]

Hmm, intresting, but now i got a small EI core to play with, just need some time to do the wireing. Hope that will change something. I put too much hours in this circuit to just give up.


I also thinking about high frequency switching with a ferrite toroid core. I have some dead pc PSU with a couple of them.

[Turion]

Toroid

Glad to see you have moved on from the torrid. As I said in post #4221

Here's one hint I will give you for free. You will spend a lot of time beating your head against a wall if you are trying to use a torrid. But then again, you may discover something no one before you has discovered. These things DO happen.

That I know of, nobody has gotten a Tesla type switch to work with a Torroid.

[Kepszlok]

Glad to see you have moved on from the torrid. As I said in post #4221

That I know of, nobody has gotten a Tesla type switch to work with a Torroid.

I'm verry curious about the reason behind this. I hope at least when the EI works, we can formulate some theory with inductivity, different core materials, etc. From a pure electronic view, a low frequency toroid and an EI core are verry similar, just the toroid has higher efficency in voltage coupling, requires less secunder wireing.

I also like to see a .xls or something about the voltages vs running time from somebody who has a working TS. I'm here for some time, but i don't remember such a document.

[Matthew Jones]

I'm verry curious about the reason behind this. I hope at least when the EI works, we can formulate some theory with inductivity, different core materials, etc. From a pure electronic view, a low frequency toroid and an EI core are verry similar, just the toroid has higher efficency in voltage coupling, requires less secunder wireing.

I also like to see a .xls or something about the voltages vs running time from somebody who has a working TS. I'm here for some time, but i don't remember such a document.

I have always looked at toroid more like a flywheel, you know "Once in Motion..." and all that. You can store quite a bit of energy in certain size core. And that is not necessarily what your looking for.

Matt

[holtage]

Why transformer instead of inverter as the load?

I want to asked John Bedini or Mathew Jones(Others also welcome) why use transformer as load instead of an inverter (12V to 220V)? Is it because the diodes, transistors do consume some voltage which will lead to the fact that the load(inverter) cannot get enough voltage for it to operate?

Also, is an inverter belongs to an inductive load or resistive load?

Also, can I use the sink/source current of output pin of ne555 timer to drive the bases of the transistors(which acts as switches) instead of opto-isolator because the adding of optoisolator to the tesla switch circuit make it more bulky and tedious to handle?