Use for the Tesla Switch

[ldissing]

That is why my transistors get so hot

Dave, Matt and Murray will also be looking over these and adding input, we will be building and adding ASAP. There is also a new Solid state design being added in there.
Ash

Thank you, thank you, thank you. I will give this a try, but the following is also an interesting thing. (When I was using relays and the original diagram per the mueller.pdf, if the capacitor was NOT in parallel with the motor after the bridge, then the relay contacts where sparking. Is this not a radiant event?)


I was thinking exactly the same thing about the solid state version...it has a direct path to ground - but the relay version did not. I kept thinking...how can this work? Unfortunately, I did not suppose that it was drawn incorrectly...(or is it?)

I say. "or is it", because it does indeed "do work" the way it was drawn on the icehouse (JB) site. Battery run time length is the key to "working" though....can we recharge and run a load at the same time. (The devices I was using were very small, 200-500ma motors and lights but they do run and/or light up. My guess was is that the direct shorts or a huge radiant buildup was causing the transistors to get hot and some portion of that was being sent to the "load". But if the capacitors are empty, they will fill up BEFORE trying to "recharge" the battery, no? Maybe I should have been using even bigger motors to use up all the current that was generated/stored in the capacitors. I'm using two 60V 8600 uF caps, because that the only "pair" all I have at the moment.)

Doing it the way presented in the post above by Ash would more than likely solve the hot transistor problem too, but do we lose the radiant? Time will tell and I'm going to be giving this new schematic a shot as soon as possible!

This could be considered just like a patent application. It isn't correct until is you figure out what is missing...or in the wrong place. Then it works just fine, but you really have to work at it, be creative, and have some knowledge. Wish I was just a little more creative.

Keep working....someone will get there. Matt has done a great job and maybe you guys have just solved the problem with solid state too.

Thanks again,

L

[Sephiroth]

Jsut a note on using a four point double throw relay for the switching..

I have been trying to get that kind of switching to work properly for 2 weeks but have found it unreliable. although you would think that mechanical switching would make a sharper on/off time, I have found that at higher frequencies large relays are unreliable in making a solid on/off switching action... mine work ok up to 100hz but that really is the upper limit. I have tried solid state before but ended up frying the entire circuit so gave up on it as I couldn;t face all that soldering again lol

Using Mat's rotary switching is probably going to give the best results, but I am going to try an adaptation using solid state switching but combining what I have learned from previous experiments regarding charge conservation. Not keeping my hopes up as solid state is supposed to be less effective than mechanical switching, but it will be an interesting project.

[ashtweth]

Guys there are new solid state switching schemo's in the doc now

[ashtweth]

Doh forgot to post the docs URL to the PDF

Panacea University
Tesla Switch - updated January 8 2009

[ldissing]

Mueller.pdf - Direct Shorts.....?

Question:

Has anybody looked at Figure T-6 page 14?

Observation which caused the question:

This diagram shows a direct short via a switch between the two lower batteries, like it is intended to be there. Maybe this is an on/off for the load? But in any case, it would appear that it is okay to have the direct short without the load.


Panacea circuit works:

I've hooked up the circuit as suggested by the panacea folks and it definitely runs a motor or a light, but not with the same speed as the original circuit, HOWEVER, the transistors do not get hot at all, which is fantastic! I'm not sure about the charging yet.


My thoughts:

I'm still thinking that the original circuit is probably correct and that it is the trigger that is important, i.e. the gating signal to the transistors. NPN transistors have a Hfe (MJL21194 about 75 at room temperature), so if we can limit the current into the base then we can limit the current through the transistors, i.e. the direct short won't matter...so to speak and the transistor will not get hot. What we want is again....potential without current flow if all these theories are correct. I think that is what you will get if you can limit that direct short current through the transistors. If we can do this, then we won't have hot transistors and the caps will still see the potential and supposedly charge themselves from the "vacuum".


Interesting thing 1:

I had never seen this before, but depending on the tuning and having a motor across the bridge I saw something interesting. The wires going to the AC side of the bridge move when a magnet is placed close to them. I mean it makes sense because it is AC, but it is only at certain points depending on frequency and delay time. The motor will run at various frequencies and various delay times, but the wires only move (with a magnet close to them) over a finite range of those values. Just thought it was interesting. Anyone know what that means?

Interesting thing 2:

With the panacea SS circuit mods, the transistors S3 and S6 are running really cold. I mean it is not warm here in the garage, but the other transistors are warmer than S3 and S6 which I also found to be kind of strange. I'll try to take a relative temperature reading soon.


L

[ldissing]

Charging with SS

Well, I recharged the batteries overnight with the battery minder product which seems to work very well.

Then, I hooked the batteries back into the circuit and nothing. It was pretty strange, so I played around with the pot and eventually, the motor started to run.

One of the batteries was at 13.01 and the other at 13.1. After running for about 2 hours, the new readings are 13.09 and 13.3 respectively. Does it mean that it will continue to run, too soon to tell, but it looks promising.

I'll keep ya'll informed.

I was looking at the AC side of the bridge with a scope and it looks like a compression wave, or a LC tank at the AC side of the bridge when I set the scope to 100ns/div. It is hitting over 16V occasionally and hits 14/15 quite often. I sure wish I knew what I was looking for in regards to the waveform.

L

P.S. I'm hoping that someday, I'll actually say something that at least seems intelligent and JB will call me up and say....you got it! I'm such a dreamer.

[Jetijs]

sounds great, but you are right that it is too soon to tell. You are talking about two batteries, what about other two? Or are you using a different setup?

[ldissing]

sounds great, but you are right that it is too soon to tell. You are talking about two batteries, what about other two? Or are you using a different setup?

No, I'm using 4 batteries with the panacea mods to the circuit for ground on the batteries 1 and 3. I am checking batteries 2 and 4, respectively. I did not check the others, but probably should. I just know that battery 2 was the most prone to decline in all my previous tests, but I'll check the others too. Might as well be thorough.

L

[Matthew Jones]

Using Matt's rotary switching is probably going to give the best results, but I am going to try an adaptation using solid state switching but combining what I have learned from previous experiments regarding charge conservation. Not keeping my hopes up as solid state is supposed to be less effective than mechanical switching, but it will be an interesting project.

You have to buy some real high end relays. Most mechanical relays don't go fast enough.
These are about the best you can get with manageable size but they are ballpark $350+ for anything that will run a high amp motor.
Power Industrial Solid State Relays


I have found some on Ebay but they come from China (the supplier) and I just don't want to buy them. I'm not trusting at all even though they have a good rep.
Solid State Relay SSR 24-220V DC, 40A + Heat Sink - eBay (item 230318896804 end time Jan-12-09 00:31:41 PST)

These are what I intend to buy for a vehicle setup.
High Voltage Relays
They are mechanical but they are fast.

In regards to mechanical switching.

You could setup a small trigger generator to drive the relays. Since its an either or setup timing would be the biggest thing. Both coils would have to be rather flat and the magnets would have to be small. But thats also extra energy into the system because you could Diode the energy into the bridge. It give you some options at least.


Matt

[ldissing]

sounds great, but you are right that it is too soon to tell. You are talking about two batteries, what about other two? Or are you using a different setup?

Jetijs, you would just have to bust my bubble, woulnd't you? I used to check all of the batteries all the time, but there was this one particular battery that would seem to drain more than the others. So, after weeks of testing all the battery voltages, I started mainly looking at this one....bad engineering/testing practice.


The other two batteries are losing in voltage. I believe this to be because the transistor that brings the negatives together is not turning on at all, or only partially. I've a few ways around this, so I'm going to give that a shot today and I'll let you know.

Even so, to see two of that batteries coming up in charge is a plus that was not seen in the past. It is running the motor and charging batteries. I'm guessing it will be much harder to get all 4 to charge at the same time, but time, wonderful time, magical time will tell. I'm getting loopy. Maybe I'm generating x-rays and they've fried my brain.

Thanks for asking the question, so I do not remain too happy for too long .

L

P.S. I think John Bedini is using transformers, because they are basically inductors and become a source of power for a little while after pulsing, which means there is no way that the transistors will not turn on, i.e. they have both voltage and current going through them. NPN transistors need some current going into the base to have any amount of current flowing from collector to emitter. Maybe that is the trick to potential without current. Have a transistor with a base that can provide a voltage to turn it on, but no current can flow through it, then you'd have potential without current at the emitter. I'm such a goofball and will quit rambling now.

[Matthew Jones]

You could generate a trigger that could fire a transistor too. Just like a monopole. You would just have to diode 1 side of the coil with 2 diodes back to back. Use the other side of the coil to power the transistors at the base.
You would only have to adjust the resistance on the base from the coil at that point.

See terrible quik drawing

Matt

[Sephiroth]

You could generate a trigger that could fire a transistor too. Just like a monopole. You would just have to diode 1 side of the coil with 2 diodes back to back. Use the other side of the coil to power the transistors at the base.
You would only have to adjust the resistance on the base from the coil at that point.

See terrible quik drawing

Matt

Hi Matt...

Interstingly this is almost precisely what I have in mind... except I have gone through every adaptation of the circuit possible and can't figure out how to trigger the transistors using generator coils without using optocouplers.... I have a two phase bedini motor built for the porject that is running nicely... I just need those d**n optocouplers! The only componants missing!



This is basically what I had in mind...



Not a great schematic either, sorry It's just to give me and idea how to solder it...

[Matthew Jones]

Let me know how it works out. I don't have anything to play with thats why I am pushing my idea a bit. Hoping somebody will try it for me.

Movings Murder, but I can't wait to get there.


Matt

[ldissing]

Most mechanical relays don't go fast enough.
Matt

I was able to get my relays to open for 8ms, closed for 8ms reliably. That is about as fast as they could handle, but they did a pretty good job. Just may need some faster ones.

I could not get the transistor to turn on properly, or at least I could not make it run without the primary running down. I think the transistors have a 0.65 V BE(on), which is difficult to achieve with the floating ground. You should not use the + sides of the battery, though, or you are "using" the battery. You aren't supposed to "use" the battery while you are "charging" it, so this leaves me in a conundrum.

The above is probably the reason for the "direct" short in the original circuit, because then you are using the negative sides of the batteries and "negative potential", so to speak.


L

[ldissing]

Opto-transistor Setup

I've utilized a program to analyze this setup, and it seems to work pretty well.

The voltage generator going to the base of the first transistor is simulating the opto-isolator (program doesn't seem to have these).

If the emitter of this is hooked to the base of the transistor we are trying to control, we can adjust R8 to allow for different base currents into that transistor. The higher the base resistor, the lower the current into the base. So, if that is utilized on just the negative "short" side (actually both sides) of the original circuit, we should be able to limit the current through the transistor, because NPN transistors allow current based on the hFE from 25-100 or so. So the current through the transistor is base current * hFE. If this is true, then it shouldn't matter what is going through the other transistors, because there is a limit to the amount of total current flowing in the circuit. The other transistors should not be at a higher current the the lowest current allowed between either side of the circuit. At least that is my guess at the moment.

If R8 is 330 ohms as in the attached jpg, then the current flowing through transistor T1 should be about 322 mA if the hFE is 25. If hFE is 100, then the current through the transistor should be approximately 1.2A. This should probably be a variable resistor (pot) to tune to the application.

Of course, you must not allow more current through the opto than the opto is capable of handling, which in my case is 100ma.

I think that Jetijs opto (JB) circuit with this modification may work better, so I'll give it a shot and post my results as soon as I am able to implement this.

L

P.S. I'll be happy for anyone to correct me on the workings of a transistor or this particular theory.