Use for the Tesla Switch

[citfta]

Quick turn off time?

@Seamonkey, I am curious where you got the idea we need quick turn off time for our circuits to work. That indicates to me you think we are trying to make better use of the back emf. John B has said several times back emf is not what we are concerned with. He also said we need the quickest possible turn on times. I think you circuits will help with that. Tesla also said he got the strongest radiant events when he had the shortest turn on times. Since we are trying to create radiant events to charge the battery I don't see how a quick turn off time will help with that. Maybe you have done some research into producing radiant events and have learned something we have missed. If that is the case will you share that with us?

Carroll

p.s. By back emf I am referring to the inductive kick from a collapsing field around an inductor when the voltage is turned off.

[SeaMonkey]

Tesla also said he got the strongest radiant events when he had the shortest

turn on times. Since we are trying to create radiant events to charge the battery I don't see how a quick turn off time will help with that. Maybe you have done some research into producing radiant events and have learned something we have missed. If that is the case will you share that with us?

Carroll

Bipolar transistors have a fast turn-on time provided the base drive is sufficient to take the transistor quickly into at least shallow saturation.

Turn off is much slower due to charge storage within the base region so some provision must be provided to quickly remove the stored charge with a reverse bias or 'short circuit.'

Bedini has stated it correctly; the pulses should be as sharp as possible (minimal rise and fall times) and the pulse duration (pulse width) must not be excessively long. Very short pulses are most beneficial.

Tesla used what he had available at his time to produce the necessary pulses; high voltage capacitive disruptive discharge across a spark gap with magnetic quenching to force rapid turn-off. Now we have the ability to simulate those same conditions with low voltage (less than 400 Volts) and high speed switching with modern semiconductor devices.

Therefore, we must use every trick at our disposal to encourage the semiconductor devices to turn on and off with the requisite speed.

[SeaMonkey]

Vissie,

Regarding your diagram with the opto-isolator driven 'Discharge' transistor switch at the mid-point of the two capacitors:

Using a bipolar transistor in this location is difficult to deal with from the standpoint of speeding up its switching since it is a 'high side' application.

By far the best solution is to replace the bipolar transistor with any N-Channel power MosFet (with suitable voltage/current ratings) and use a small High Side Driver Chip such as the LTC4440-5 to drive it. The chip has the needed built in 'level shifter' and 'boost' function to drive the MosFet properly; it is a small chip; and it accepts TTL signal input for control.

By using an N-Channel MosFet for the 'Discharge' switch you'll attain the best possible switching speed with the least amount of power loss. And, the cost will be very reasonable.

It would also be possible (although not necessary) to replace both the 'Charge' transistors with N-channel MosFets if desired, with appropriate driver chips.

The advent of the High Side Gate Driver chip greatly simplifies circuit construction for any number of otherwise difficult circuits by enabling the easy design replacement of bipolar transistors with commonly available N-channel MosFets which are becoming very inexpensive.

I would strongly encourage all experimenters to gravitate towards the superior switching characteristics and very low conduction losses of the MosFet. In most applications they'll not even need heatsinks.

Once you've made the 'transition' you'll rarely have a 'need' for a bipolar transistor (with their anomalies) in any power switching application.

[nvisser]

I will definitely look at using mosfets if I can find that mosfet drivers somewhere.
One more question. JB introduces a SCR as series switch between the two capacitors . It did work and does switch off as soon as the voltages are balanced out. Do you think it give a sharp cut off.
Or do we want that switch to switch of much sooner?
See diagram posted by JB

[Matthew Jones]

Vissie,....
I would strongly encourage all experimenters to gravitate towards the superior switching characteristics and very low conduction losses of the MosFet. In most applications they'll not even need heatsinks.

Once you've made the 'transition' you'll rarely have a 'need' for a bipolar transistor (with their anomalies) in any power switching application.

I am not doubting you hear but we have heard other things. The Anomalies, as you call them, are some of the characteristics we are looking for according to other people.

Matthew, "I'm Not yelling Here"
Look it is real simple if two batteries go into series minus the junction drop most of the heat will be in the transistor. This circuit works like a current charge pump. This is all about switching and how the switch works. The Tesla Switch is nothing more then a dual charger of which you can take energy from. You all must understand I do not like to use the term current because this is just the dissipation of the potential in the system which is the voltage. You want to use the least amount of current to charge the batteries, the batteries understand what the signal is since your scope can not see it. What you will notice is a sharp funny looking square wave, not what I have seen so far, I will try to take pictures for you all. The minimum scope must be at least 100Mhz to see anything. I was showing Peter yesterday when you get it switching right you will see the scope go negative 1 1/2 times the input, This would give you a COP of 2 anymore then this, John would not be around to talk about anything. I will try to draw it for you. The 50% duty cycle will do this, and you can run this down to 10% if you want. Try just building this simple at first so you can see what is going on. Timing is not the key the switch is and the device used. Right Leroy Good Job.
JB The attached image for this post

We can emulate the turn off with the cap and diode setup on the base.

Yes,
The power factor is very important, and it does effect coil/cap pairs.
The important thing is you must capture all the wasted energy in a circuit.
Switching power supplies must be corrected and the cost is high.

It takes a different person to build switching supplies, since I'm at slow speeds I do not worry too much. The first thing, the layout is very important as traces act like inductors and currents show up where you do not want them.
Fet's are the next problem as the gate cap becomes very large and you can not turn them off. You can do simple experiments with Fet's and you will see what I said about gate discharge, that is why I use the diode and cap in parallel in positive circuits as the load wants to be in the drain and not the source in cap discharge circuits.
I always try to run them inverted, Example if you put the Drain on the negative of the battery all you need is a potential over the battery and it will switch correctly.
You want the battery as the load and not the other way around. When I say Inverted I want to use the differentials of potential in the circuits. The biggest thing about semiconductors is the second emitter break down, you must make 1 amp at DC, remember what I said here, 1 amp at DC.JB

Mosfets have problems in these circuits. I know I have watched a few burn. Not at first though. They react like any other switch. In fact better. And they induce the battery to performs expected. The potentials in the battery rise high on charge and drop low discharge. This closes the potential gap between the 2 banks of batteries yet the system starts performing better. A larger current and an unmeasurable potential show up and drive the loads harder and harder.
At that point though the mosfet stops performing. They cook rather quik.
I know what I am saying probably doesn't make sense. Thats because you have to see it to believe it, but I can tell ya first hand mosfets don't hold up.

The relays I use are mosfet based and they cook quicker than egg's. The reason I have been using them is because they are easy to drive and I can see the effect quickly. Giving me a chance to start to investigate the actions while its happening so I know what to look for.

So maybe we should use a mosfet to see the performance and see if the effect can be induced. In my opinion thats worth the loss, cause the things will not hold if it starts working. So we shouldn't just expect that the mosfet will solve our problems and walk away from anything else.

Change of subject...
@SeaMonkey
I guess I owe you a little apology. Since you got references, and you seem to want to help I apologize for what I wrote.

Matt

[nvisser]

Vissie,
By far the best solution is to replace the bipolar transistor with any N-Channel power MosFet (with suitable voltage/current ratings) and use a small High Side Driver Chip such as the LTC4440-5 to drive it. The chip has the needed built in 'level shifter' and 'boost' function to drive the MosFet properly; it is a small chip; and it accepts TTL signal input for control.

By using an N-Channel MosFet for the 'Discharge' switch you'll attain the best possible switching speed with the least amount of power loss. And, the cost will be very reasonable.

Do I still have to use a opto coupler between the pic and mosfet drivers

[SeaMonkey]

Do I still have to use a opto coupler between the pic and mosfet drivers?

No, there is no need for the opto-coupler or opto-islolator.

The necessary level shifting is an integral part of the High
Side Gate Driver chip. It greatly simplifies the hookup
requirements from the Pulse Source to the MosFet being
controlled.

Puts the 'fun' back into making stuff!

[SeaMonkey]

Mosfets have problems in these circuits. I know I have watched a few burn. Not at first though. They react like any other switch. In fact better. And they induce the battery to performs expected. The potentials in the battery rise high on charge and drop low discharge. This closes the potential gap between the 2 banks of batteries yet the system starts performing better. A larger current and an unmeasurable potential show up and drive the loads harder and harder.
At that point though the mosfet stops performing. They cook rather quik.
I know what I am saying probably doesn't make sense. Thats because you have to see it to believe it, but I can tell ya first hand mosfets don't hold up.

The relays I use are mosfet based and they cook quicker than egg's. The reason I have been using them is because they are easy to drive and I can see the effect quickly. Giving me a chance to start to investigate the actions while its happening so I know what to look for.

Once experimenters understand what the MosFet needs to be
driven into 'full conduction' and how to switch between full off
and full on with the greatest possible speed, the 'Burning MosFet'
problem goes away.

MosFets that run 'hot' aren't being driven with enough 'oomph'
and that is where the driver chip will make all the difference in
the world. The gate drive pulse must be a minimum of 10 Volts
up to about 12 Volts max. And the Gate Capacitance must be
charged as speedily as possible for fast turn-on, the discharged
as speedily as possible for a fast turn-off. That is what the
driver chip is able to do.

I agree with what Bedini has said in your quotes. He is sometimes
difficult to 'interpret' but what he says about switching and
the pulse characteristics is true. The length of the pulse is
critical and the efficiency of the switch is very important.

[SeaMonkey]

A larger current and an unmeasurable potential show up and drive the loads harder and harder.
At that point though the mosfet stops performing. They cook rather quik.

Could it be that this 'unmeasurable potential' is exceeding
the rating of the MosFet?

If flyback pulses occur that exceed the voltage rating of the
MosFet then it will go into 'avalanche' mode and become very
hot. If caught soon enough and shut down the MosFet will usually
not be damaged.

When this condition arises in a circuit then some provision
must be made to 'absorb' the flyback energy before it jeopardizes
the health of the MosFet.

The energy can in most cases be 'captured' with a diode/
capacitor 'plug' and applied to some useful work or to a
battery for charging/rejuvenating.

Yes, some circuit operations do produce surprising results!

[Hoxan]

Did you know you can build your own solar panels, Get your video instructions here: Earth4Energy.com - make solar panel and make wind turbine

link for the pdf file: RapidShare: 1-CLICK Web hosting - Easy Filehosting

[Joit]

Did you know you can build your own solar panels, Get your video instructions here: Earth4Energy.com - make solar panel and make wind turbine

100% Scam, all 21 Posts from this Guy

[Hoxan]

100% Scam, all 21 Posts from this Guy

I don't think its all scam because PDF is explaining how to make your own solar cells from damaged ones. If you take the good ones from damaged solar panel and combine them with other working cells there is no reason why it shouldn't be working. Its like connecting a lot of batteries in series / parallel to get desired voltage / amperage.

Back on topic: I've replicated solid state tesla switch with IRFZ44 mosfets and another with BD135-BD712 darlington. Driver circuit is implemented with ATtiny2313 and 4N35 optocouplers. Diodes are BYV42.
There were no unusual results in testing....no overunity. I tested with 4 accus and 2 of them weren't in good shape, maybe that effected testing (or problems with mosfet switching ?!?). Sry don't have pics because my camera is broken. I can post schematics of driver circuit.

Is it possible to make tesla switch with 3 supercaps and 1 accumulator instead of 4 accus?

[Matthew Jones]

What Kind of loads are running on it?

Matt

[Joit]

Hoxan, the Idea is not new to buy damaged Cells at Ebay,
isolate the bad one and replace them, and make a complete Unit again.
There are certain Posts and Pages for this at the Net, and for free.

But this Guys collect all this Informations, make Sites from it,
and try to sell a Pack from it, and even dont care,
when something is protected with a Patent.
Never waste your Money for this, they are not serious
and only out for her Business.

[StevanC]

Could it be that this 'unmeasurable potential' is exceeding
the rating of the MosFet?

If flyback pulses occur that exceed the voltage rating of the
MosFet then it will go into 'avalanche' mode and become very
hot. If caught soon enough and shut down the MosFet will usually
not be damaged.

When this condition arises in a circuit then some provision
must be made to 'absorb' the flyback energy before it jeopardizes
the health of the MosFet.

The energy can in most cases be 'captured' with a diode/
capacitor 'plug' and applied to some useful work or to a
battery for charging/rejuvenating.

Yes, some circuit operations do produce surprising results!

Hi SeaMonkey,
Do You know about the way to turn a To-3 packed BJT into a tiny solar cell?
Is this possible for a MOSFET too?

Wonder why I ask?

Stevan C.