Use for the Tesla Switch

[sucahyo]

Question, is there any way to switch using solid state that allow current goes to both direction ?

Maybe mechanical switch is more efficient because it allow current goes to both direction? Transistor and diode become a load since it prevent the current goes to both direction and reduce efficiency?

Edit:
found it, bilateral switch optocoupler.

[Xenomorph]

@sucahyo : Well technically a so-called TRIAC would do exactly that with two lines however.

[sucahyo]

@sucahyo : Well technically a so-called TRIAC would do exactly that with two lines however.

I see, thanks.

[nilrehob]

@nilrehob
Great experiment!

Thanks.

Would you mind sharing a bit more in detail how you are switching the 4-pole relay from the timer and how the batteries were connected to the poles as your circuit diagram was simplified?

The circuit, especially variant 1, is simplified compared to the "original" circuit, as there are no diods, capacitors or anything extra and there are no crossing lines in the schema. I think its simpler to follow. I havent simplified any circuit any further in my setups except for omitting the 555 and the relay.

Did you have to amplify the timer output with a transistor?

I am using a "Smart Kit" "LED flasher 1089" and replaced of one of the leds/resistors with my relay.
I also replaced R1 (top-left) with a variable resistor so that I could trim it for a 50/50-wave at the relay-terminals.

Thanks and good luck with further experiments

I hope this helps, good luck to You!

[sucahyo]

Do anyone ever do a performance test for tesla switch?

Do anyone ever compare the performace of transistor and mechanical switch? Maybe I do it wrong, my transistor version is a lot more inefficient than relay version that it can't even light up a neon bulb at the secondary. It charge the battery but only slightly.

[Xenomorph]

@sucahyo
Well you obviously seem to have both variants built right there having found out efficiency differences, why dont you do the tests? Do you use a comparable coil at all, i think thats where the power for the neons comes from?
nilrehob will probably do longterm tests soon too.
I am very curious exactly about the long term part myself.
Relay is having a slight spark gap effect that might boost the circuit, but without data that is mere speculation.
I will build a variant too to find that out, but out of money at the moment for the missing parts. The 2-relay young-type switch works quite good already for me.

[sucahyo]

@Xenomorph, as I said, the transistor version of mine charge at slower rate compare to relay version. I use computer power supply as source for charging 12V battery. I know this is not ideal since there is only very small voltage difference between source and charging. But using relay this configuration is enough to light up neon bulb at secondary using 1Amp transformer.

Using relay and car coil, it can also do wireless or one wire neon lighting. Still don't try transistor and car coil combination.

Edit:
Just did transistor and car coil. It can light up the neon bulb but nothing special. No wireless, no one wire, no 4mm spark, etc. It even doesn't sting hand.

I attach my circuit. I wonder if I do it right.


For durability test, using relay here is what I have:
Source and charging part is 12Volts battery. Relay is powered by charging part (which has lower voltage). Both is resurrected gel battery.
initial voltage:
source = 10.28V
charging = 9.90V

after two hour:
source = 10.00V
charging = 9.88V

[nilrehob]

@sucahyo
Its a tricky business, takes a lot of time and thought (and money) just as ashtweth pointed out, so You have to be patient

In Your transistor/555 version, do You not have any spikes from the coil at all?
Do You have an oscilloscope?

edit: BTW, the voltage on the batteries seems *very* low, not sure if the source battery will do much good?

[sucahyo]

Ok .

I don't have a scope but there seems to be a spike in coil enough to light up a LED only.

About the battery test, I think it would not make any difference if the battery is not in perfect condition?

[nilrehob]

I'm not sure about Your particular batteries of course, but my impression is that most 12V-batteries are quite lame when under 11V, no punch. Please anyone correct me if I'm wrong.

Then we have the transistor. I don't know much about them at all, but I've read on several places that the 2N3055 doesn't work as well as other transistors in these circuits. Not sure why, probably the sharpness of the switching, don't know how to determine that from a transistor-spec though.

Anyone please fill in here?
What transistor to use and why?
What specs to look for in a transistor used for these kind of setups?

[Xenomorph]

@nilrehob: I agree that 12V batteries under 11.50 V even lack the punch.
I was once charging a 2nd battery pack and the voltage wouldnt want to go any higher on them. Then i connected a 12.75 V charged up source battery and the charging battery climbed as expected.

[Sephiroth]

I'm not sure about Your particular batteries of course, but my impression is that most 12V-batteries are quite lame when under 11V, no punch. Please anyone correct me if I'm wrong.

Then we have the transistor. I don't know much about them at all, but I've read on several places that the 2N3055 doesn't work as well as other transistors in these circuits. Not sure why, probably the sharpness of the switching, don't know how to determine that from a transistor-spec though.

Anyone please fill in here?
What transistor to use and why?
What specs to look for in a transistor used for these kind of setups?

The main problem with 2N3055 s is they would rather kill themselves than deal with radiant energy they fry REAL easy... though considering their cost, they are very good transistors with a pretty fast switching time...

But I love MJL21194 s they can handle alot of abuse so for high voltage applications or in places where the transistor may be exposed to flyback (without a neon) they are ideal... they will fry if you give them some serious abuse but they can hold up pretty well on a standard SSG circuit without a neon (though I wouldn't try that for an extended period of time)

and the MJL21194 s also have a noticably faster switching time... I'll upload some scope shots when I have time... though I haven't a clue what all the spec means on their datasheets

EDIT : forgot to mention that the MJL's casing is much easier to solder as well

[sucahyo]

I manage to damage a couple of MJ2955, while the 2N3055 seems more robust. I damage 2N3055 once when I forgot to reduce the transistor gain and also damage a 4.5V light bulb by light it up very bright using BEMF induction, the car coil sing at around 10000 hertz. MJ2955 break everytime I achieve around 8volt of BEMF.

The biggest problem I have is heat. Would that transistor can hold decent heat without TO-3 casing?


For transistor circuit, what if we do the switching in high frequency but rotate the battery in low frequency?

Which is better for efficiency, high freq or low freq?

[nilrehob]

and the MJL21194 s also have a noticably faster switching time... I'll upload some scope shots when I have time... though I haven't a clue what all the spec means on their datasheets

scope shots? Yes please!

[nilrehob]

Which is better for efficiency, high freq or low freq?

Except for Bedini saying he did 20Hz (not sure if it's true though), it seems like the higher frequency the better.

I think the spikes is one big component, even in the TS where it is not highlighted at all, and if so to maximize the number of spikes and minimize the cost for producing them. Therefore I'm skeptic about the focus on a 50/50 wave for the TS. Why keep on draining the batteries between the switching when the mag-field is already built up in the coil?

So You either shorten the +/- part of the wave so it just "fills" the coil (and lengthen the 0 part to keep freq), or You inc freq.