Use for the Tesla Switch

[Bit's-n-Bytes]

@ Bring your crayons!

Welcome back Leroy. LOL

Jeff

[SeaMonkey]

Can't get it much simpler than in the sch I posted just LOOK at any one of the 6 fet and drivers

2 diodes, 1 driver one fet .
2 caps one for noise one to de couple
1 resistor from driver to gate and 1 diode for faster turn off.

Each driver fet combo only SEEs the battery it is switching.

The Boost for the High side drive is from the SERIES Connected bank on the other side so it is ALWAYS available without all the BS involved with bootstrap ccts that don't work worth a crap in this reversing type cct.

Pretty simple actually if you just look at it Just one simple design applied 6 times cept for the little tweek for high side. that also will FOLLOW the voltage being switched. In other words there will ALWAYS be the required 18volt above for the gate regardless of the voltage being switched.

Power for the Control cct is Floating from ALL batteries too so there is no reference to any single battery and it's associated noise etc either.


OK Your Turn !!
Now lets SEE YOUR DESIGN to do the same.
Curious to see how to do it with Transformers.

For those of us who've spent years poring over
schematics it is not beyond our ability to track
and digest. For those less able and experienced
it is an imposing diagram.

Could you possibly think of a way to simplify the
visualization by means of an altered layout to
better emphasize what your strategy is? To
assist all who have interest in 'getting it?'

Since you've mentioned that your circuit is indeed
functional, will you enlighten the readership by
describing its operational parameters:

(a) What switching frequency have you found to
be most effective?

(b) What sorts of loads have you powered thus far?

(c) Have you observed any 'benefits' to battery
longevity? Re-charging?

(d) What sort of procedure have you developed to
'tune' your circuit for best operation?

(e) In your considered opinion; do you believe that
the so-called Electrodyne Report has credibility?

(f) Has your experience with your circuit confirmed
that Bedini's observations are correct?


My own inclination is assuredly towards transformers. Think
'isolation' and 'point of load.'

[SeaMonkey]

Here might be a way to amplify a small signal into workable energy. Solar that is!

If you're willing to tolerate the switching losses
of bipolar transistors and the cumulative losses
of the several diodes, your circuit should do
something. There aren't any obvious errors.

It would be helpful for ALL who are interested if
you'd provide a written paragraph or two to
summarize what your circuit is intended to do;
its several different functional regions; with
some idea what typical voltages might be across
the various energy storage capacitors.

Others may desire to know more about your use
of the Micro-Controller for monitor and control
and how you've chosen to interface it to your PC.

Any operational data you may wish to provide
will be very helpful. What sized solar array are
you using; how many batteries are you charging;
etc...

You may yet earn a "Well Done!" Candu?

- - - - -
Yeah, about the sorry state of the World and
especially the good ol' U.S.A.

I spent my time 'in country' and saw more blood,
death, destruction and sorrow than I thought
possible. I've since come to agree with Smedley
Butler: "War is a Racket!"

Dystopia is putting it mildly.

[DoubleD]

For those of us who've spent years poring over
schematics it is not beyond our ability to track
and digest. For those less able and experienced
it is an imposing diagram. '

Well I usually don't do schematics for MY stuff as it is an extra un-necessary step for me most of the time.

Could you possibly think of a way to simplify the
visualization by means of an altered layout to
better emphasize what your strategy is? '

NO
If you can't figure it out with this then Sorry, .... I have lost my crayons.

Since you've mentioned that your circuit is indeed
functional, will you enlighten the readership by
describing its operational parameters:
'

None of it is relevant to the cct construction at this point !!

It will take about anything your Control cct can throw at it within the range of MosFet choice!

My own inclination is assuredly towards transformers. Think
'isolation' and 'point of load.'

GREAT
By all means Post us YOUR cct so we can all learn how to do it easier .

[SeaMonkey]

Hi All,

My goal has been to get the batteries to charge themselves or at least charge other batteries without losing their own charge. So far my best effort has been to get the TS to run for several days while powering itself without losing any charge. The only load was a few LEDs. So I think I have a long way to go to get any decent load to run or to get some charging done without losing charge on my TS batteries. I agree with Leroy if we can get some decent charging from a TS then we could just use it to keep our batteries up and not have to have two large banks of batteries. This seems like the easier way to go if it can be done.

@ Seamonkey, several of us have tried relay switching at one time or another. Matt and some of us have determined from testing that we don't need really high frequency of switching to get pretty good results if we are just running a load. I have used relays controlled by a PIC controller and gotten some ok results. Do you think the Mosfets will switch faster than relay contacts? I am asking because if relays are as fast as Mosfets then we are still missing something to get the results we are looking for. John B has suggested we only need switching speeds of around .5 Hz or so. Relays can easily handle these speeds yet we still don't seem to get any charging and our batteries eventually run down if we load them very much. I think Matt has been the only one to be able to run a reasonable load for an extended time. Thanks for the suggestions you have made and the resources you provided.

Carroll

Yes, MosFets will switch faster and more reliably
than relay contacts. The high current units that
are now available have very little "ON" resistance
and so are as efficient as metal-to-metal switching.

Brandt did say that his circuit, even when adjusted
to attain 'resonance' for best operation, did not run
indefinitely. He was able to get approximately 4 to 6
times normal battery use per charge with his and eventually
the batteries would need re-charging with an external
charger.

The stories about 'instant re-charging' and '36 Volts
across each battery without heat' may be nothing
more than fantasy dis-info. Bedini is the only 'credible'
source of info about his own experiences with the
circuit and he 'mysteriously' chooses not to re-investigate
its 'wonders' after it was strangely 'destroyed.'

Relay switching has merit for first attempts with the
circuit to get a feel for what it does. As you've
discovered, the contacts have limited lifetimes when
switching useful current levels.

The MosFet will 'git 'er done!' Whether any miracles
materialize remains to be seen...

[SeaMonkey]

A Blast From the Past

The circuit which Eric posted some time
ago is conceptually very sound.

Apart from the polarity of the circuit at
the "Load" terminals with respect to the
battery banks, it is so very close to a working
implementation.

The notation regarding the method of supplying
power to the various drivers is pertinent.

Whatever has happened to Eric?

[nvisser]

Seamonkey

I’m willing to give Seamonkey a change. I learned a lot about mosfet switching since he came around
We were stuck anyway with the TS.
A new approach could not be bad. If we try what he suggest and in the end still can’t do it I will finally give up on it.
From the Benitez patents you will know that he used spark gaps to get that fast switching and with mosfets we can get very near that speeds. With fast, I mean the fast rising and falling edges of the square waves in the nanosec. ranges.
I already got the ssg to run and produce like hell with mosfets and is currently working on a new solar charger using fast mosfet switching. The will be a combination of the ss ssg (desulfater,buck boost) circuit and a series to par cap pulsar.(charge pump)
On the coils we do need fast cut off pulses to enhance the flyback pulse and on the cap pulsar a fast rising time is important to get that high instantaneous current pulse.
I think the reason that I did not get good results with Dave’s fet TS was because I used opto mosfet driver I could get local and their rise and fall times were in the usec region instead of nsec like the ones Dave used. That caused my 2 high side switches to run hot. While the switch was still cold the results looked good but that did not last long. I only understand that now.
So, yes I will do the test Seamonkey requested because it could tell us something we need to know about the batteries and how it should be switched.
What is there to lose? We are stuck anyway as JB talked in a language only he understood.
And in the process I learn a lot about switching and mosfets that I otherwise never would know.
The mosfet circuits you get all around on this forums are so wrong.
Yes he does step on toes, but then the guys do give him a lot of crap.
He is not a youngster anymore and I don’t know if he still builds stuff. Maybe he was an ingeneur and they normally has technicians to test the stuff.
But I do think he has a lot to offer.

[StevanC]

@SeaMonkey,
Okay, what would be the preffered devices?
A. IRL2204N?
B. IRF3205?
C. Or IRFP260 (55mOhm) but capable of blocking 200V?


1. This looks like one winding primary (10..30 turns as in the article) and 8-filar secondary?
2. How do we test our (different) cores to the right wire gauges?
3. Shall the Primery to secondary copper masses be in balance?
4. How much of a "umph" should the xfmr drive have V? @ A? for what MOSFET set?
5. This looks like a 3 stage project to me: "Quer-"; "driver-"; "power-" - whit BIG troublle in the middle
6. Do we feed the MOSFETs with coaxial cables or twisted pairs their respective "Que" singnals? Do we match trace lengths and to what exact level?
7. Do we size the driver stage to quad "Que" power devices (four devices in parallel) od we make that in a second (scaled up) attempt? How do we control drive level?
8. this is it at the moment

Stevan C.

[StevanC]

meantime (sideshow progress report) news flash

The progress of the TS PVAmp with MOSFET and nsec ragne "Que":
I have soldered the power-stage (IRF3205 55V failover, 53A DC, 150A surge) and the Schottky diodes (MBRS2060 60V, 20Amp DC, 150A max surge) and started to layout cables.

The sketch is roughly this (there are errors and inconsistencies yet, but it is "very about it"):

(soon to be continued...)

[DoubleD]

I’m willing to give Seamonkey a change. I learned a lot about mosfet switching since he came around
We were stuck anyway with the TS.

A new approach could not be bad. If we try what he suggest and in the end still can’t do it I will finally give up on it.

But I do think he has a lot to offer.

Vissie ..
If the guy would ACTUALLY SHOW Something instead of reposting a bunch of preschool crap we went through long time ago would be nice.

What new approach, he wont even Draw up his own design to show WTF he is touting as a BETTER way.

Just posts a bunch of ancient crap that if anyone has followed this thread to this point should know well by now.

WE ALL KNOW THE BASIC FREAKIN CIRCUIT

Show us the Improved way to do it that he so proudly claims to KNOW how to do. !!



Sorry man but he is just wasting every bodys time here.

If he has something to input to IMPROVE the cct then spit it out.
If he wants to teach an electronics cource then he needs to start a new thread just for that.

[nvisser]

I really do not want to jeopardize the friendship's I got here because of choices I made.
SeaMonkey. Maybe it will be a good thing if you start a new mosfet Tesla switch forum .
Then those interested can join it and all this fighting can stop.

[nvisser]

I drew up a diagram as described above and will try and do it in the coming week.
Bedini said:
"one way was the SG machine in constant high voltage pulses. The other is abrupt capacitor discharges in one second pulses, this causes the battery sustained for long periods of time, or time charging.
The Tesla switch is something much different in the way charging is performed. The switch must be set to perform the same function as a capacitor discharge system allowing the differential across the load"
He also mentioned:"The first part of this is how much potential can you move in a micro second before the current builds up in the circuit"

I guess that is what we are looking for.or are we trying to find the batteries resonant frequency?
The picaxe 18x code could be like this:

setfreq m8
Main:
PULSOUT 3,1 ................5usec (10 =50usec,20=100usec ,100=500usec)
pause 20 ................ 100hz - all pause commands are half the length at 8MHz
goto Main

I did the battery spike test today.
The scope shot is probed over the battery terminals, switching at 100Hz with 400usec pulses
The load was a 0.6 Ohm resistor (2 x 1.2E 7W in parallel)
The negative spike stays about the same from 5usec to 500usec pulses.
The spike is in the order of -1.7V
There seems to be a huge un measurable spike after switch off that looks bigger at low pulse widths but it is so narrow that you can hardly see it in x10 time mode.
So I am sad to report that I did not learn much out of this experiment

[DoubleD]

I really do not want to jeopardize the friendship's I got here because of choices I made.
SeaMonkey. Maybe it will be a good thing if you start a new mosfet Tesla switch forum .
Then those interested can join it and all this fighting can stop.

Hey Vissie..
There should be no Friendship problems here with anyone, this is not about that,
it is not a pissing match of who is write or anything of that nature.

It is real simple...

If Seamonkey has all this knowledge on how to make this better then spit it out or buzz off.
there has been enough of this crap here already.

I do not need to explain to him the details of my cct for him to give US ALL HIS Super Simple better way to do it.

The info he has given on the proper requirements to turn a fet on and off efficiently is right and he knows how to look up good info.

For one, I don't see how a transformer as a gate driver will give the required parameters for switching the gate on a Fet as he says it will. Fine for a transistor, but this is a different beast.
I would really like to see how this is implemented so I learn and grow. Would be usefull in a lot of other applications as well.

He keeps talking about Floating this and that. etc.

Quit talking about it and SHOW us a realworld schematic of all this that actually applies to the cct we are working with.

There is a lot more to the way this works (as you have found through actually building it) than normal text book stuff.

So this whole group upset is on his shoulders.

I showed mine now he needs to show His (not some one else's) complete design.

THEN maybe we all can discuss it and move on !!

[citfta]

Hi Vissie, I don't have a problem with anyone who wants to work with Seamonkey. I just can't seem to do it. I ask a question I get back insults and nothing useful. If you can work with him then go for it and I will just read the posts and not comment.

Good luck, Carroll

[SeaMonkey]

I did the battery spike test today.
The scope shot is probed over the battery terminals, switching at 100Hz with 400usec pulses
The load was a 0.6 Ohm resistor (2 x 1.2E 7W in parallel)
The negative spike stays about the same from 5usec to 500usec pulses.
The spike is in the order of -1.7V
There seems to be a huge un measurable spike after switch off that looks bigger at low pulse widths but it is so narrow that you can hardly see it in x10 time mode.
So I am sad to report that I did not learn much out of this experiment

Your willingness to devote time and effort to
the procedure is appreciated. Your feedback,
even though you believe you may not have seen
anything significant, is valuable.


While performing this procedure to evaluate the
battery pulse discharge characteristics there are
some precautions which must be observed:

In order to rule out the development of a 'spike'
due to distributed inductance in the wiring, all
connections should be as short as possible.
Standard RF wiring practices.

Different battery chemistries will produce
different characteristics. Even individual lead
acid batteries can produce different characteristics.

You're looking to verify what Bearden has written
about Bedini's research on battery capabilities;
the response of both electron movement and
ion movement to sharp impulse demands; and
whether you see any evidence of what Bedini
referred to as 'battery popping.'

The very sharp spike which you've observed may
be an inductive response.

You're looking for evidence of a 'signature' that
Bedini found to be useful for adjusting the Tesla
Switch to achieve what appeared to be overunity
performance.

Bedini himself spent many, many hours in search
of the 'signature' which he was looking for; evidence
of the 'negative resistance' phenomenon as a
consequence of the phase differences between the
electron response and the ionic response.

Brandt also stated that modern battery construction
has all but eliminated the tendency of batteries to
'resonate.'

Without this 'resonance' the Tesla Switch may be a
huge disappointment...