Use for the Tesla Switch

[Inquorate]

This is my research [ATTACH]5305[/ATTACH]

When you close the switch the spike is developed, that is why some people uses resistors to charge the cap, but the resistor transfer the energy into heat, now with a coil you can transduce it into real power.

High inductance, very low impedance coil.

How do we make a high inductance low impedance coil with a backyard budget?

[scratchrobot]

Charging dead battery

Charging an old battery that was completly dead, wouldn't take nothing from
a conventional charger. Now it's boiling away

Charging dead battery

@StevanC, what do you think of my timing?

scratchrobot

[ldissing]

@StevanC.

Let us asume the 1st Que is the PV "sucking" one.
Let us Just concentrate on the middle series BJT ("middleBJT"), okay?

1. The "top" bank is "grounded" ("middleBJT's emitter") with a "top-bank grounding"BJT Qued
2. The "lower" bank's positive is "pulled" high by the "low-bank hi-pullingBJT. The node is on "middleBJT's collector"

Those two bank X-BJTs are the ones "heating up" the "middleBJT" if replaced with diodes (or shunted short).
Those two we need to Que to keep good efficiency.

I'm not sure what you mean by "heating up" the middleBJT?

Look up the middle BJT while this "quartet" is in "QUE"?
It is "normal"
It's collector is positive
It's emitter is tied to ground.

I assume you've switched over to the right side middle BJT,

Yes, I agree, it is "normal".

This is exactly the moment stop the QUE, and current stops (dries out) of our "quartet"

Yet, still potential (uAmps) traces bias out middleBJT "right" so current can start flow (it can begin "que")

As it goes to saturation, ti does not filp! ever ( ?! )

I would never "expect" it to "flip" when queued, only when NOT queued, so if you are delivering a higher potential, when the "right" Middle BJT is "de-queued", I'd expect it to flip, putting the caps in series for a short bit...until the potential decreases to X volts (the voltage at which the MJL21194 stops the reverse current). Perhaps your potential is not high enough for it to "flip" when dequeued?

It adds up _negative_ to the series chain:

Battery sees=-BJThi+topCAP-BJTmid+lowCAP-BJTlow

= -4V+17.5V-4V+17.5V-4V = 35V-12V =23V
But the FVD is actually up to 8V at times
= -8V+17.5V-8V+17.5V-8V = 35V-24V =11V dumping to a depleted battery.

How many amperes needs a MJL21194 to develop 8V FVD?

If base current is 1A it has 25A on collector, and if base is 2,5A (which BD140 could provide) it is just over 30A


So just build and slow it down and watch the polarity in slow motion (I was all "wow- what could You know??" )

So, In my (newbie) oppinion, the middle BJT is upside down, not knowing it and does Que just right,
or in other terms,
we draw it "wrong way", it should have it's emitter "downwards".

Does this add up in Your opinion?
Stevan C.

Sorry for taking soooo long to get back to these questions and your response.

Yes, I think it adds up. Well, you didn't draw it the "wrong way", now did you? It is still "reversed from normal" as JB would say, most people would put the collector on the negative of the upper cap bank and the emitter on the positive of the lower cap bank. An SCR should work fine here too. Did you "hook" it up in what most people would do, the "right way"? I think it would still work, but not give you the same results...?

Lero

P.S. I screwed up the quote stuff...not you!

[ldissing]

@StevanC.,

So, how could you develop even more potential to make the middle BJT "go negative"? Of course, with your setup and the 4 transistors on the left side, this isn't possible, but if you didn't have all those transistors, then would it be possible?

What do you think the scalar charger is all about? This is a modified (solar) scalar charger on steroids, because you drive the BJTs harder than is possible with the little optos, but how do you think the (plain jane) scaler charger works.

Lero

[StevanC]

Charging an old battery that was completly dead, wouldn't take nothing from
a conventional charger. Now it's boiling away

Charging dead battery

@StevanC, what do you think of my timing?

scratchrobot

@scratchrobot
Congratulations

re: timing
no pause? You edited it out?

re setup:
I like it , need a favor for me?
Could You shunt the "trio" diodes (LEDs) with a "333" (33nF) capaitors, preferably foil, optionally tnatalum ("pill" like)?


Stevan C.

[StevanC]

in detail

@StevanC.
(snip)
I'm not sure what you mean by "heating up" the middleBJT?

Well, if the two transistors do not separate the power buss off the "trio's" middle BJT, it will burn up buss's current over it self.

I assume you've switched over to the right side middle BJT,

You seem to follow me here

Yes, I agree, it is "normal".

The middle actually does not flip really IMHO, it merely brings "close" (voltage wise) the banks "middle" terminals.
It's FVD does not add up to the banks, it is "resistance" and subtracts off the sum, together with the lateral BJTs of the "trio"
I personally imagine it allows some of the banks voltage to "overlap" across it ?

I would never "expect" it to "flip" when queued, only when NOT queued, so if you are delivering a higher potential, when the "right" Middle BJT is "de-queued", I'd expect it to flip, putting the caps in series for a short bit...until the potential decreases to X volts (the voltage at which the MJL21194 stops the reverse current). Perhaps your potential is not high enough for it to "flip" when dequeued?

I would not think it ever flips, we just pictured it "upside down"? On the other hand, i noticed that "wrongly" connected optos (have BJT on output) seem conduct despite "flipped".
Also a friend of mine, explained this is a "dreatful negative region" we protect our beloved MJLs of, by using a "shunt diode" emitter to base in the "vanilla SSG".
When lead to the "deadful negative region" the BJT conducts in "opposite way" just for the collector is following the base too close (negative!).
it is _known_ to happen

Sorry for taking soooo long to get back to these questions and your response.

You have been excused (this time )

Yes, I think it adds up. Well, you didn't draw it the "wrong way", now did you?

I wonder if I did (not?)

It is still "reversed from normal" as JB would say, most people would put the collector on the negative of the upper cap bank and the emitter on the positive of the lower cap bank.

Maybe this was the "wrong way" JB was reffering to?
or there are more dark secrets in the Dralington vs Sziklai topology (one allows negative region operation, other not?)

An SCR should work fine here too. Did you "hook" it up in what most people would do, the "right way"?

I would have to be a educated EE to answer that, i just fiddle my SCR until it works

I think it would still work, but not give you the same results...?

Lero

Well, why don't You slam up a rig and find out?

P.S. I screwed up the quote stuff...not you!

Yes i know, I was just pulling a prank ? sorry

[StevanC]

@StevanC.,

So, how could you develop even more potential to make the middle BJT "go negative"? Of course, with your setup and the 4 transistors on the left side, this isn't possible, but if you didn't have all those transistors, then would it be possible?

What do you think the scalar charger is all about? This is a modified (solar) scalar charger on steroids, because you drive the BJTs harder than is possible with the little optos, but how do you think the (plain jane) scaler charger works.

Lero

Lero,
I think I already answered the first question above ?

TS>SC:
I think my schematic might contain the config in it self already?
Had You built this unit i posted the HOWT about, You migh just explore it for Your self:
I had shorthed two BJTs with cables and still one with a diode and got operation off 12V


Stevan C.

QOTD:
"Run; You fools!"
. . . . . . . Gandalf the gray

[StevanC]

@Stevan. Hi,

Thanks for your forwarded PM. I also check it everyday.
Get back to the Topic.

I think you already got something. But I am not yet.

What I am chasing is the SPIKE.
Spike is magic to the DIRAC SEA where the energy we're chasing comes from.
NEGATIVE ENERGY.
But the properties of that SPIKE should be currentLESS.
I am not saying the system current.
NO current at the transient moment of swithing at least.
We know the current is the LOSS in the form of HEAT,
but I believe the current at the transient moment is more than just LOSS.

So I am trying to get the currentLESS SPIKE.
I am trying to seperate the Potential from the current which is my enemy.

JB's the Simplified Circuit Can be the solution, I believe.
That simplied circuit is not meaning just SIMPLIFIED saving the cost of parts.
I am checking whether the simplified circuit is seperating the Potential from
the current, then make the PURE Potential go to the battery banks of charging side(12V)
and make the current turn back to the battery banks of discharing side(24V).
If I can make the circuit with the CORRECT SWITCHING and PROPER IMPEDANCE,
I believe I can get there, CURRENTLESS POTENTIAL. I hope.
Regards,

JANG.

@JangYD
, "tune" to BM2, i have a project going on there soon, regarding the leading edge.

Stevan C.

[ldissing]

Lero,
I think I already answered the first question above ?

I believe you mentioned a way to get another few volts out of it, but that was not of what I was speaking. I was thinking of a tesla "style" coil in a couple of different places in your particular circuit, but since I have not built it, I should have not asked the question, you have done a marvelous job both in terms of design and communication, I'm not sure how you have the time for all of that!

TS>SC:
I think my schematic might contain the config in it self already?
Had You built this unit i posted the HOWT about, You migh just explore it for Your self:
I had shorthed two BJTs with cables and still one with a diode and got operation off 12V


Stevan C.

StevanC, your circuit is dandy and the different configurations you can make out if it is nice also. It makes for a great solution when we do not know of what JBs solar charger is "really" made, but I think there is even more to it than what you have shown here in this circuit. We do what we can, right?

I do not have a big interest in the solar charger at the moment, which is why I have not built it. I will build it over the next couple of months more than likely, but I'm doing some other things at the moment.

I probably a more in line with JangYD than what you are doing here. You do get a hefty current push with your device and you do boost the voltage over the battery, my goal was a little different. It is the problem with all forums, our minds have not melded to a single purpose, and we all build things slightly differently and possibly for different purposes. Probably also the reason we do not find the answers faster.

Lero

[scratchrobot]

@scratchrobot
Congratulations

re: timing
no pause? You edited it out?

No pause, maybe I should? Just trying different settings to see how it reacts.

@scratchrobot
re setup:
I like it , need a favor for me?
Could You shunt the "trio" diodes (LEDs) with a "333" (33nF) capaitors, preferably foil, optionally tnatalum ("pill" like)?


Stevan C.

Thank U, I will give it a try

My battery like me really likes those quick pulses, I'm charging/discharging it to see if it holds it's charge better over time.

scratchrobot

[ldissing]

from post #1028

MY first guide is no.4) can be curtailed to
1)HOW MUCH(before the current builds up in the circuit)
2)CURRENTLESS POTENTIAL(before the current builds up in the circuit)
3)move in A MICRO second(before the current builds up in the circuit)

I'm starting to see the spike in my every circuit. Negistor is beyond me.
1)Spike is pulsing in a microsecond(may be 200~500 nano second)
2)Before main current flows on steady state(need 10~30us to reach max)
3)So this spike is not related to how many current flows in the circuit.
4)Impedance needs to be checked very carefully.
CURRENT is not the solution.
The only soultion is the swithing and impedance.

Onemore thing.
I saw something on the simplified circuit with 2Trs and 10 diodes.
With the correct pulsing and proper impedance
this siplified circuit CAN BE WORKED properly like othe complicated circuit.
Anyway let's go ahead and further.

Regards,

JangYD.

Jang,

I still believe that these statements are correct and you (and I) are on the right track. I've mentioned several times that my transistors switch on there own. No one has cared to comment on my transistors switching at very fast rates which would be impossible to happen under "normal" circumstances. I have not been able to see this with the PIC oriented setup, but I also packed up my scope, so I can not say positively that it does not happen with the PIC.

When I was using the SG3524, I could get the transistors to switch very quickly, looked like a train of pulses when looking over the collector/emitter of the middle BJT. It only happens after the signal to the BJT is taken away (de-queud...to use a StevanC term). It does not happen at all pulse widths. To little of a pulse and it goes away, to much of a pulse and the length of time for this pulse train is infinitesimal compared to the "power" drawn from the system. There is a region, and it isn't all that long, when the transistors just decide to switch on/off (with no signal) and this would be allowing very little "current" through the system, but providing potential which could be "harvested". I believe this to be the "negative" switching region.

The problem with the above 1us (or faster), is that it flies in the face of JB saying that a PIC could be used with a pot to allow switching from 1-5000hz. 5000Hz is the fastest frequency (I think) of which he spoke.

How are you attempting to match the impedance in your testing? I was using a capacitor with a small bulb across the capacitor and a bridge, of course.

Leroy

[StevanC]

Jang,

I still believe that these statements are correct and you (and I) are on the right track. I've mentioned several times that my transistors switch on there own. No one has cared to comment on my transistors switching at very fast rates which would be impossible to happen under "normal" circumstances. I have not been able to see this with the PIC oriented setup, but I also packed up my scope, so I can not say positively that it does not happen with the PIC.

When I was using the SG3524, I could get the transistors to switch very quickly, looked like a train of pulses when looking over the collector/emitter of the middle BJT. It only happens after the signal to the BJT is taken away (de-queud...to use a StevanC term). It does not happen at all pulse widths. To little of a pulse and it goes away, to much of a pulse and the length of time for this pulse train is infinitesimal compared to the "power" drawn from the system. There is a region, and it isn't all that long, when the transistors just decide to switch on/off (with no signal) and this would be allowing very little "current" through the system, but providing potential which could be "harvested". I believe this to be the "negative" switching region.

The problem with the above 1us (or faster), is that it flies in the face of JB saying that a PIC could be used with a pot to allow switching from 1-5000hz. 5000Hz is the fastest frequency (I think) of which he spoke.

How are you attempting to match the impedance in your testing? I was using a capacitor with a small bulb across the capacitor and a bridge, of course.

Leroy

Now now,
"dandy dude"
You seem to have whats thermed "floating input"?
It has nothing to do with energy excep cosmic radiation.

If Your transistor is not "ground tied" by a 1k ... 10K resistor, "random currents" develop across it's junction leading eventuall to failure or breakdown in high volt applications.

Leading edge is something different:
A BJT is able to copy a signal on it's input onlty to a certain degree, and we seek way to make the MJL21194 "Que" out 30A / 10uSec or twice so, while keeping it sound and safe to be able to do so many millions of times.

We want *cntrolled* surplus energy inflow into the battery.

And guess what I found?
I can get higher battery voltage swing, by changing just the slope of the leading edge.

But i have to recheck this (JangYD, could You possibly verify?)

Stevan C.
QOTD:
Fortune's Real-Life Courtroom Quote #3:

Q: When he went, had you gone and had she, if she wanted to and were
able, for the time being excluding all the restraints on her not to
go, gone also, would he have brought you, meaning you and she, with
him to the station?
MR. BROOKS: Objection. That question should be taken out and shot.

[ldissing]

Now now,
"dandy dude"
You seem to have whats thermed "floating input"?
It has nothing to do with energy excep cosmic radiation.

If Your transistor is not "ground tied" by a 1k ... 10K resistor, "random currents" develop across it's junction leading eventuall to failure or breakdown in high volt applications.

Leading edge is something different:
A BJT is able to copy a signal on it's input onlty to a certain degree, and we seek way to make the MJL21194 "Que" out 30A / 10uSec or twice so, while keeping it sound and safe to be able to do so many millions of times.

We want *cntrolled* surplus energy inflow into the battery.

And guess what I found?
I can get higher battery voltage swing, by changing just the slope of the leading edge.
Stevan C.

What is HV in your opinion? 12V, 24V, 100V, 400V, something else?

What does a negistor do? Naudin wasn't doing anything "wrong", but he did not figure out how to use it, I believe JB said he had not figured out how to use it. Where did Naudin go "wrong", what had he not figured out?

My comments are not about the scalar charger or your version of the solar charger. Again, kudos on your design and communication, but we are working on two different animals.

I'm not saying that the PNP/NPN combo could not be applied to the 4 battery TS, it could. You could get the same 30A/10us in this application and maybe that is what is required.

What is your take on the following: JB says, "you do not want current" in the system, you "should design for the least amount of current"?

Now, for a particular application, you might need butt loads of current, and that is fine if it is "that for which the design calls". We know that current can charge batteries and believe that pulsed current/voltage is even better for the battery, moving ions and keeping them moving.

BTW, I'm fine with your leading edge theory and application of it, not that you care whether I'm fine with it or not!

This has nothing to do with your leading edge theory, but it was in my head, so:
In the SSG, it is the leading edge quick HV rise that charges batteries. That sharp leading edge is not related exactly to the transistor, but the coil and what it is trying to do...the transistor just stops the coil from doing what it wants to do, so it puts up a HV spike because there is no place to push current. Bearden talks about dv/dt not di/dt, I believe, but I could be wrong about that. He, Bearden, continually talks about doing things without current and so does JB. Bearden talks about pulse charging a capacitor even so as not to use much current, keeping an open looped system.

I guess it is good that we are not all the same. Our ideas are different and come from different perspectives and change all the time based on our limited knowledge and the things we "discover" or "think we discover". Time will tell.

Thank you for your time to respond even though not directed @ you. I don't know, what I don't know.

Lero

[scratchrobot]

@scratchrobot
Could You shunt the "trio" diodes (LEDs) with a "333" (33nF) capaitors, preferably foil, optionally tnatalum ("pill" like)?

Stevan C.

I only have some 10nF capacitors, should I try those?

I was also thinking of using the Sziklai setup in the 4 battery TS, but I think
it would also need a capacitor bank in the middle to get the fast current pulse?
Also tried 3 battery's in series (36V) to charge 1 battery but I didn't like the sound the battery was making
Just found out the Picaxe has a Pulse and PWM command, I wil play with them too
scratchrobot

[StevanC]

I only have some 10nF capacitors, should I try those?

I was also thinking of using the Sziklai setup in the 4 battery TS, but I think
it would also need a capacitor bank in the middle to get the fast current pulse?
Also tried 3 battery's in series (36V) to charge 1 battery but I didn't like the sound the battery was making
Just found out the Picaxe has a Pulse and PWM command, I wil play with them too
scratchrobot

Yes try "103" buy seek "473" (47n) or at least "333"?

The cap gives huge advantage, JB used even "104" or "105"

Also I found out the PC817 is better than PC816 opto?

Stevan C.