Use for the Tesla Switch

[Kepszlok]

Tell ya the truth I don't know how his worked out, not sure if anybody had luck with it.

I put this paper out some time ago for beginners:

https://www.matthewcjones.com/TSGuid...ide_part_1.pdf

You might get something out of it. Most of the projects are affordable and you use what you need to use or buy whats suggested.

To date 14 people have had success with them, but I couldn't tell ya where they are now.

Matt

Thank you, will read it today!

[Kepszlok]

Its not the 2n3055 its the resistors. The turn off time is too slow.

Personally I use fets and IGBT's now with drivers.

The pulse width is also important, as some of what you are seeing is the impedance in the batteries kicking in. A decline in current and in a OPEN situation also a decline in current flow. Batteries can usually handle up 30 MS pulse then the impedance kicks in so the potential drops at that point.

The batteries will actually tell you when you have hit that point but it is algorithmic in nature and depends on the situation you are using to charge it.

The current Tesla switch (Benitez Switch) I run is heavily dependent on filters to decipher the signals coming back from the battery to determine the pulse width and shape. All that can make a difference.

This makes things a bit more clear. If i do an upgrade on the driveing part of the schematic, i will get more speed and less voltage loss on transistors. I will rebuild my circuit in that way.
Even tina simulation shows great decrease in voltage loss.


Also i had little time to read your whole pdf (hard to do anything if you got a job and a family ), but your desing are very simple and asome. Sooner or later i will also build that circuit too.

[Matthew Jones]

Most people have gotten 8x the runtime out of the load than they would if they just were running the batteries. So it makes it worth it.

After you make some runs I can show you how to improve upon that very dramatically. But you have to understand how to tune the things and simple ones are great learning tools for that.

Cheers
Matt

[Kepszlok]

That's sounds good!


For now, I do modified my original circuit with the new base driveing. It shows some results, some of them may be promeising.

2 of the 4 batterys drained 0.2V while the other 2 drained just 0.01V after 3.5 hour of running (same 1.4Ah batts, equaly charged at the start). The load was 24ohm resistor, the output voltage was 6.7 - 7.3Vdc (with puffer capacitor)

I think some of my optocouplers are needed to be replaced, because of what you can see in B-E driveing2 and graetz voltage pictures.

[Matthew Jones]

That's sounds good!


For now, I do modified my original circuit with the new base driveing. It shows some results, some of them may be promeising.

2 of the 4 batterys drained 0.2V while the other 2 drained just 0.01V after 3.5 hour of running (same 1.4Ah batts, equaly charged at the start). The load was 24ohm resistor, the output voltage was 6.7 - 7.3Vdc (with puffer capacitor)

I think some of my optocouplers are needed to be replaced, because of what you can see in B-E driveing2 and graetz voltage pictures.

You may try getting rid of the caps along side the bridge. I never had need for them. Even in a case that was inductive I always used a parallel cap with the load, transient suppression... You won't have that with a resistor.
Thats another thing as well the resistor is just making heat so there is no recovery from a resistive load. Where as an inductive load is easier to recover power from. But for measurements it makes it harder.

One thing you may try is to lower the resistance on the opto coupler input, but only if your voltage is OK at that point. Those LED's lag a bit if under powered.

Cheers
Matt

[Kepszlok]

You may try getting rid of the caps along side the bridge. I never had need for them. Even in a case that was inductive I always used a parallel cap with the load, transient suppression... You won't have that with a resistor.
Thats another thing as well the resistor is just making heat so there is no recovery from a resistive load. Where as an inductive load is easier to recover power from. But for measurements it makes it harder.

One thing you may try is to lower the resistance on the opto coupler input, but only if your voltage is OK at that point. Those LED's lag a bit if under powered.

Cheers
Matt

Yes, i tought about droping the couplers from the circuit, this will be my next move along with the modification of the opto's.

Currently i do not have a proper motor, but i can put a coil of big toroid transformer as load. Or even directly withouth a graetz, hmm.


But for now, i'm duing these thest on work, and i will be in a festival for a week. So i will continue testing the circuit after.

[Donald Haas]

Hi Amp DC switching CCT, very low heat

Hi Matthew,
Glad to see that your are continuing to post here.
I've had your manual and will build.
Spent some time with the TS in the past but did not see my desire like you.
have.
I want to mention that an engineer on the UFO politics threads has developed a cct for UFO and he specializes in hi pwr mosfet switching.

I built his circuit and was able to switch 6, 300w light bulbs from 108v of LA Batts. The mosfets did not heat up. The frequency was around 4khz.

I redrew the circuit in target 3001 to make it easier to build and then etched it for my tests. I can send it to you (pdf) if you'd like.
PM me anytime.

bro d

[Matthew Jones]

I do not know what a CCT is... Let me know when you hit 50 KW we'll compare.

Cheers
Matt

[gyula]

I do not know what a CCT is... Let me know when you hit 50 KW we'll compare.

Cheers
Matt

Hi Matt,

cct is short for the word circuit if you meant to ask that.

Gyula

[Kepszlok]

Hi Matt,

cct is short for the word circuit if you meant to ask that.

Gyula

Lol, i woud never figure it out.

Köszi, hogy felvilágosítottál minket Gyula!

[Matthew Jones]

Hi Matt,

cct is short for the word circuit if you meant to ask that.

Gyula

Thanks. I'm terrible with acronyms.

Matt

[Kepszlok]

You may try getting rid of the caps along side the bridge. I never had need for them. Even in a case that was inductive I always used a parallel cap with the load, transient suppression... You won't have that with a resistor.
Thats another thing as well the resistor is just making heat so there is no recovery from a resistive load. Where as an inductive load is easier to recover power from. But for measurements it makes it harder.

One thing you may try is to lower the resistance on the opto coupler input, but only if your voltage is OK at that point. Those LED's lag a bit if under powered.

Cheers
Matt

Well, i think the opto coupled tesla switch is a no-go. I do made a lot of modification but the optocouplers are performing very poorly and different to each other.

*see circuit diagram

One thing i'm sure is serial driveing of the optocouplers are a bed thing, because of the second and third one gets a not perfect square wave (falling edge will be a slope), so i connected them paralel, with a proper input resistor.
I do also put depleter resistor to the driveing transistors, so now all of the opto's are got the same and perfect square wave, but the C-E voltages are still shows erratic breakdowns around the falling edge of the squars.


Currently i'm thingik about changing the opto's to small transformers (along with the cap's and diodes, and connecting the secunder between the B-E pins). We got a bunch of 600ohm telephone line transformer with a ratio of 1:1, they are ideal for the job. If it's not working, than i will move to plan B and build your circuit.

[Matthew Jones]

Not sure what opto's your using but you may try bypassing the optical part all together and use the base if its available. The H11d1 has a base pin at pin6. It can take 5 volt pretty easy. This would give you Darlington pair, between the opto unit and the transistor.

Just another last chance hope.

Cheers
Matt

[Kepszlok]

Not sure what opto's your using but you may try bypassing the optical part all together and use the base if its available. The H11d1 has a base pin at pin6. It can take 5 volt pretty easy. This would give you Darlington pair, between the opto unit and the transistor.

Just another last chance hope.

Cheers
Matt

I'm using PC817 opto's, they are quite same but the base are not avaible.
I also forget about my transformer coupled idea, it's not good for square waves. I think it's time to move to plan B.

[Donald Haas]

Newbie setup

Hi All,
Put together a TS setup and started with toroid transformer that had a useable config.
It did about 4VAC and had it run over night.
Arduino and SSR's at .049sec on times and no programed time between pulses.
Led's for a load.
All the batts went down a little over night accept #3 which went up .01v.
#4 went down the most. About .7V

Then, finished the small transformer as per Matt's Simple Switch manual.
#18 wire, 12ah Garden Tractor batts and stepper motor for load.

Tuned it to .017sec on times. Scope shows good switching.
Strange mm readings.
Batt #4 reads -.48v under load. (minus .48v) Negative lead on negative terminal. Fluke 87V meter.
Batt #3 reads 16.8V under load.
Batt #2 reads 9.4v under load.
Batt # 1 reads 12.03v under load

Turn off and rest it and #4 slowly returns to about 12v.
#3 comes down like it has been charged.

Looks like I could add batts in parallel at #4 for charging.

I'll run this a while and then try tranny's or mosfets.

I have a hefty APC transformer that looks good to go bigger.

Used ones are about 40-50 bucks.

Any insight and tips would be appreciated.

Thanks,

bro d