Use for the Tesla Switch

[JohnStone]

@ Stefan2
It is not the current. The "non repetive" current expects you to not repeat this shock within some seconds in order to give the semiconductor time to cool and recover from the shock.
Due to the fact that these are diodes of different types the internal properties differ. If the switching speed is different or the reverse current or the capacity you get any effect.
These are quite normal diodes with basic properties and wide tolerances and the manufactures do not give any data regarding switching speed. I found some data regarding reverse capacitance between 30 and 50 pF but this data will vary between manufacturers and their production charge.
Due to the wide tolerated data your measured voltage is not necessarly surprising.

[nvisser]

Sg3524n

This weekend I build myself a nice little pc board with a proper sg3524 flip flop using a 1uf cap and higher resistances.
It also has its own FET, zener power supply and the supply is draw with diodes from all 4 batteries.
At last I can report that it works properly with proper duty cycle adjustment AND it is stable and doesn’t die after awhile when running at low freq. of 0.5Hz.
Now I can also start to experiment!!
Seems like the picaxe will never arrive

[Stefan2]

Dankeschön JohnStone,

"wide tolerated data" sounds like human. So for now i will stay with the 1N5408 then.
Sounds like i need, to get to know all of the used parts habits.
This is quit challenging as nothing stays still and may be "wide tolerated".
As one battery charges the others discharge which causes always changes in Frequency, Voltages, Currents, Induction, others and the so far unknown parts.

best regards,
Stefan

[ldissing]

Hi Lero,

I am the guy that posted one of the circuits that has no diodes, so maybe I can answer your question. In the circuit without diodes the parallel transistors will each be carrying half the current of the serial transistor. They have to carry that current or there is no charging of the parallel batteries. As John K. said the idea was to have a circuit that has as few voltage drops as possible. I believe this circuit is as close as you can come to a mechanical TS and still be solid state.


(QUOTE) I do not think you understand what I'm TRYING to say. In your circuit, what is the load on the transistor from the high positive to the low positive? What is the load on the transistor from the high negative to the low negative? What was the load on the transistors (high and low trannies, not the serial tranny) in the original circuit? (QUOTE)

So how is your own testing coming? Are you ready to drink the wine yet? It sounds like John K. is getting close. I have been side tracked with other issues lately but am planning to try John K's program later today. (I hope)


I hope this answers your questions, Carroll

I was asking the questions for you all to think about, I've thought about it all ready. We may not come to the same conclusions, but just something to keep us busy...

Okay, the testing is going pretty well. I still do not agree with the assessment that Bit's gave on the his original PNP circuit, but he was on the the right track, I think. His assessment was a relief valve, if I remember correctly, and JB liked it for a different reason, which is what I've been working on.

Feel free to disagree with me, but I may not listen. I'm like that.

The goal, is to charge the batteries (and get power, but we don't have to worry about that just yet). So, one side is powering (call it left side) and the other side charging (call it right side). We want to use the ion flow on the right side, once it is moving, which is what Bits was doing but I don't know if he knew that was what he was doing, doesn't matter anyway, it isn't a contest.

Once we get the right side ions moving we are in recharge mode on the right side, and we want to use that movement because the ions can't reverse immediately. So, we hit the right side and get them moving, and immediately after we shut the left side off, we pulse the right side. How long to pulse is still up for grabs. If devices were working perfectly and the timing was just perfect, we would get this working without a uController, but a uController is pretty handy. On my system, I can vary the pulse width with a ADC input, which is really nice too.

So, it goes something like this...
1. Turn on left side for a time T1 (sort of long time)
2. Turn off left AND Turn on right side for a time T2 (short time).

3. Rest.

4. Turn on right side for a time T1
5. Turn off right AND Turn on left side for a time T2 (short time).

6. Rest

At step 1, we get the right side into recharge mode.
At step 2, we use some of the ion movement to give a reverse kick to the left side. It may not stop the movement totally, but it is slowing it down, it is like a HV pulse to the left side. We don't really want any current flow, so this needs to be a sharp, short pulse.
At step 3, we give the left side a rest.
At step 4, we get the left side into recharge mode.
At step 5, we use the ion movement on the left side to give a reverse kick to the right side.
At step 6, we let the right side have a rest.

Goto step 1.

The idea is that once the ions are moving, we can use that movement for free, just like in JBs 1987 booklet with the motor. If the Kick is like the HV pulse (maybe that is why Bits called it a relief valve), the left battery will go into recharge mode easier, thus stopping ion flow on the left side that was powering the circuit and we got that for free from the right side. We didn't stop the right side from being in recharge mode, hopefully, if the time T2 is small enough, or at least, we didn't start discharging it.

That is my theory, and I'm sticking with it for now.

My system is pretty much like this, although, I intersperse a couple of right side small pulses and continue with the left side power for a bit. Finally, ending with a right side small pulse. Then rest.

So, I get the right side into recharge mode, then pulse it (back to the left side), keep it in recharge mode, then pulse it again, keep it in recharge mode, then pulse it again, then go to the other side.

So, left side on for 100ms, right side on for 1 us, left side on for 100ms, right side on for 1us, etc. If course, I can vary both times, and I'm getting very very close to working the way I would like. My batteries are old, and I need to buy some new ones, but would like to wait until after the move is finally over so I don't have to move them.

Leroy

[Bit's-n-Bytes]

I was asking the questions for you all to think about, I've thought about it all ready. We may not come to the same conclusions, but just something to keep us busy...

Okay, the testing is going pretty well. I still do not agree with the assessment that Bit's gave on the his original PNP circuit, but he was on the the right track, I think. His assessment was a relief valve, if I remember correctly, and JB liked it for a different reason, which is what I've been working on.

Feel free to disagree with me, but I may not listen. I'm like that.

The goal, is to charge the batteries (and get power, but we don't have to worry about that just yet). So, one side is powering (call it left side) and the other side charging (call it right side). We want to use the ion flow on the right side, once it is moving, which is what Bits was doing but I don't know if he knew that was what he was doing, doesn't matter anyway, it isn't a contest.

Once we get the right side ions moving we are in recharge mode on the right side, and we want to use that movement because the ions can't reverse immediately. So, we hit the right side and get them moving, and immediately after we shut the left side off, we pulse the right side. How long to pulse is still up for grabs. If devices were working perfectly and the timing was just perfect, we would get this working without a uController, but a uController is pretty handy. On my system, I can vary the pulse width with a ADC input, which is really nice too.

So, it goes something like this...
1. Turn on left side for a time T1 (sort of long time)
2. Turn off left AND Turn on right side for a time T2 (short time).

3. Rest.

4. Turn on right side for a time T1
5. Turn off right AND Turn on left side for a time T2 (short time).

6. Rest

At step 1, we get the right side into recharge mode.
At step 2, we use some of the ion movement to give a reverse kick to the left side. It may not stop the movement totally, but it is slowing it down, it is like a HV pulse to the left side. We don't really want any current flow, so this needs to be a sharp, short pulse.
At step 3, we give the left side a rest.
At step 4, we get the left side into recharge mode.
At step 5, we use the ion movement on the left side to give a reverse kick to the right side.
At step 6, we let the right side have a rest.

Goto step 1.

The idea is that once the ions are moving, we can use that movement for free, just like in JBs 1987 booklet with the motor. If the Kick is like the HV pulse (maybe that is why Bits called it a relief valve), the left battery will go into recharge mode easier, thus stopping ion flow on the left side that was powering the circuit and we got that for free from the right side. We didn't stop the right side from being in recharge mode, hopefully, if the time T2 is small enough, or at least, we didn't start discharging it.

That is my theory, and I'm sticking with it for now.

My system is pretty much like this, although, I intersperse a couple of right side small pulses and continue with the left side power for a bit. Finally, ending with a right side small pulse. Then rest.

So, I get the right side into recharge mode, then pulse it (back to the left side), keep it in recharge mode, then pulse it again, keep it in recharge mode, then pulse it again, then go to the other side.

So, left side on for 100ms, right side on for 1 us, left side on for 100ms, right side on for 1us, etc. If course, I can vary both times, and I'm getting very very close to working the way I would like. My batteries are old, and I need to buy some new ones, but would like to wait until after the move is finally over so I don't have to move them.

Leroy

Leroy, if I were you, I would be drinking the wine. You have said the magic word "Ion's moving". That is where the "magic is".

Bit's

[stonewater]

ions moving

all the way back to JB's 1984 free energy generator booklet, which was republished in the FEG circuits and schematics book. all about the ions

[John_K]

Onya Lero!

I was asking the questions for you all to think about, I've thought about it all ready. We may not come to the same conclusions, but just something to keep us busy...

Okay, the testing is going pretty well. I still do not agree with the assessment that Bit's gave on the his original PNP circuit, but he was on the the right track, I think. His assessment was a relief valve, if I remember correctly, and JB liked it for a different reason, which is what I've been working on.

Feel free to disagree with me, but I may not listen. I'm like that.

The goal, is to charge the batteries (and get power, but we don't have to worry about that just yet). So, one side is powering (call it left side) and the other side charging (call it right side). We want to use the ion flow on the right side, once it is moving, which is what Bits was doing but I don't know if he knew that was what he was doing, doesn't matter anyway, it isn't a contest.

Once we get the right side ions moving we are in recharge mode on the right side, and we want to use that movement because the ions can't reverse immediately. So, we hit the right side and get them moving, and immediately after we shut the left side off, we pulse the right side. How long to pulse is still up for grabs. If devices were working perfectly and the timing was just perfect, we would get this working without a uController, but a uController is pretty handy. On my system, I can vary the pulse width with a ADC input, which is really nice too.

So, it goes something like this...
1. Turn on left side for a time T1 (sort of long time)
2. Turn off left AND Turn on right side for a time T2 (short time).

3. Rest.

4. Turn on right side for a time T1
5. Turn off right AND Turn on left side for a time T2 (short time).

6. Rest

At step 1, we get the right side into recharge mode.
At step 2, we use some of the ion movement to give a reverse kick to the left side. It may not stop the movement totally, but it is slowing it down, it is like a HV pulse to the left side. We don't really want any current flow, so this needs to be a sharp, short pulse.
At step 3, we give the left side a rest.
At step 4, we get the left side into recharge mode.
At step 5, we use the ion movement on the left side to give a reverse kick to the right side.
At step 6, we let the right side have a rest.

Goto step 1.

The idea is that once the ions are moving, we can use that movement for free, just like in JBs 1987 booklet with the motor. If the Kick is like the HV pulse (maybe that is why Bits called it a relief valve), the left battery will go into recharge mode easier, thus stopping ion flow on the left side that was powering the circuit and we got that for free from the right side. We didn't stop the right side from being in recharge mode, hopefully, if the time T2 is small enough, or at least, we didn't start discharging it.

That is my theory, and I'm sticking with it for now.

My system is pretty much like this, although, I intersperse a couple of right side small pulses and continue with the left side power for a bit. Finally, ending with a right side small pulse. Then rest.

So, I get the right side into recharge mode, then pulse it (back to the left side), keep it in recharge mode, then pulse it again, keep it in recharge mode, then pulse it again, then go to the other side.

So, left side on for 100ms, right side on for 1 us, left side on for 100ms, right side on for 1us, etc. If course, I can vary both times, and I'm getting very very close to working the way I would like. My batteries are old, and I need to buy some new ones, but would like to wait until after the move is finally over so I don't have to move them.

Leroy

Hi Lero,

I like your thinking

A couple of questions-

1. How long do you rest in steps 3 and 6?
2. How often do you switch sides?

I'll replicate whatever you suggest, as I've run out of ideas for now


John K.

[Matthew Jones]

John k @ Is your new circuit top out on the charging.

I have finally found the negative resistance of an MJL the transistor. But so far I can only get one at a time to show the property as described by this article.
Negative Resistance Oscillators

If you run the transistor at the Serial position at 82 ms you should get a hint of it. I have tested 12 transistors total and 82 ms always shows signs.
In a Tesla switch it will give you a AC peak to Peak of about 20 volt across the Collector to the Emitter, and the same with a volt across the load.

So far though it does not seem to add to a system, but I have only tested the basics. I am going to try to further the expansion of peak to peak potential by using more transistors and switching progressively through them.

Maybe give it a try and see it for yourself. You must run 100 mhz scope at 1 sec time divisions to see the effect. I'll post a wave in a bit when I get setup again.

Matt

[citfta]

Original thinking

Hi Leroy,

I agree with John K. That is some very original thinking to charge and then reverse pulse the source batteries.

Besides the questions John asked I have a couple. What size batteries are you using? I am thinking the size of the batteries might have something to do with how often we switch and how long the charging pulses should be. I am probably not right but it just seems to me a smaller battery is not going to respond the same as a larger one. And what size load are you using? If you have already answered these I apologize for being too lazy to look for the answers.

I am also like John in that I am about out of ideas to try. Thanks for the inspiration.

Carroll

[Matthew Jones]

ScopeShot of negative resistance

Likewise if you drive it at 75ms or so, you'll see positive resistance. I have not found No resistance though.

Matt

[ldissing]

ScopeShot of negative resistance

Likewise if you drive it at 75ms or so, you'll see positive resistance. I have not found No resistance though.

Matt

@Matt,
Looks pretty much like mine Matt (400 Mhz scope). Now, another interesting thing is if you go down to the us range. Put it on AC, trigger it on the negative slope and down below the 0 line, so you can catch what is going on (one side only). You can see the current pulses switching in and out and looks just like JBs drawing, except that it isn't 500 ms, it happens when the timing is higher, like 82ms or higher (I used your number Matt) switching, but there are multiple pulses down there. It looks like the current pulses shown on JBs solar video, where he has meters all over the place and he shows the current, except they are much faster, and continue for the duration of the pulse.

Leroy

[nvisser]

Question to you guys using the picaxe 18x
I am busy laying out a pc board while I wait for the chips to arrive.
I see that there are 8 outputs.
Can I use any of them?
Because one is labeled sda , the other scl and one pwm out.
Do I leave unused inputs open or are they tied to ground?

[ldissing]

Hi Leroy,

I agree with John K. That is some very original thinking to charge and then reverse pulse the source batteries.

Besides the questions John asked I have a couple. What size batteries are you using? I am thinking the size of the batteries might have something to do with how often we switch and how long the charging pulses should be. I am probably not right but it just seems to me a smaller battery is not going to respond the same as a larger one. And what size load are you using? If you have already answered these I apologize for being too lazy to look for the answers.

I am also like John in that I am about out of ideas to try. Thanks for the inspiration.

Carroll

Carroll,

I'm using 17ah batteries that have been around the block a few times. 12V 20W load, can you figure out why I chose that load? What is the biggest load that these batteries should support on a TS? Sounds like something that JB might say, no? (Matt, I know you disagree with my reasoning on choice for a load, and I may be incorrect (again), but I'm sticking with this for now until it is working perfectly every time, then I will try even bigger loads.)

See my post to Matt above about the pulses, I'm not sure you really need the reverse pulse(s), as the device already sends pulses out by the setup.

I am not using your deviation to the circuit, I am using the 6 transistor (original) switch, FYI.

I just wanted to help it along with the reverse pulses. I'm going to bite the bullet and build a board, I'm so cheap, that I wanted to see something working before I spent another 100 to have it manufactured (packed away my etching equipment (didn't label the box) like an idiot or it would have been built already), and I don't have any doubts that it works. I ran the system for 48 hours on crap batteries, with bailing wire and twine and could have continued running for a while longer without recharging.

@John K., I do not know the exact timing of the pulses, as I use an ADC input that gives me from 4ms-1 sec of pulse width. The on and off times are the same, with small reverse pulses of about 4-50 us every 1 ms during the "off" time. I can also pulse one side for any number of the above ADC input, so I can get over 1 sec in pulse time per side.

When you look at Matts scope shot, you notice that the waveform is pretty wild, and that is hard to trigger on, so it is hard to measure the precise time...guess I could move it to the output and look at the square wave down there for the timing. Haven't done that though.

Leroy

[Matthew Jones]

When you look at Matts scope shot, you notice that the waveform is pretty wild, and that is hard to trigger on, so it is hard to measure the precise time...guess I could move it to the output and look at the square wave down there for the timing. Haven't done that though.

82 milliseconds on then 82 off.

Every mjl I have tried shows the same results at 82ms. If thats what your having trouble with.
I came to that watching the turn on pulse. My scope is PC based so it can measure. Then I just ran 82 on 82 off and the oscillation started.

Matt

[ldissing]

82 milliseconds on then 82 off.

Every mjl I have tried shows the same results at 82ms. If thats what your having trouble with.
I came to that watching the turn on pulse. My scope is PC based so it can measure. Then I just ran 82 on 82 off and the oscillation started.

Matt

@Matt,

I know, I was answering Carroll's question about my timing. I'm sure you are right on the numbers, because if I run to fast, the thing doesn't like to charge at all, of course, my contention is that you need to charge for 125ms minimum, and have up to 125ms before the ions will stop moving, but that is theory right now. Theories are good, but practical experience is better, I'm still learning.

Leroy