Use for the Tesla Switch

[John_Bedini]

Tesla Switch

ldissing,
Where is the Jetijs circuit posted, But Jetijs comes with a great recommendation from Peter L so where do I get it? I want to see why it did not work.
John B

Jetijs,

Can I post your schematic on the BM2 group? You probably lost all rights to it when you posted it here anyway, but oscillator part works fine...not the opto stuff though.

Leroy

[blackchisel97]

blackchise,
Those parts are fine to use, Thanks and no I will not answer any questions on the SG here as this is about the Tesla Switch at this time. Progress is made buy focusing on one thing at a time.
John B

You're absolutely right, I'll keep my mouth shot about other circuits but this question was about SG3524, my apology for using shortcuts

Thank you

[John_K]

theremart,

I haven't used a "conventional" charger for 2+ years now. But gels are crap, good for testing but then I move onto wet LABs. I don't like to hurt my "babies"...

Depending on "how crap", they rest between 12.8 and 13.8 overnight. The RC-1AU will not cook them. This is not just a shameless plug 'cos I'm a dealer either.

John K.

I am encouraged to see 2 success reports here using the switch. So these batteries have been radiantly charged then. If I may ask what is the resting voltage when these batteries are fully charged after y ou use the RC-1AU and let them sit for a few hours ?

I ask that because I have found when my 12 amp hour batteries seem to be the best charged they have a resting voltage of 13.20.

Thanks for the info!

[Jetijs]

Hi John
I am happy to see you here I played around with the Tesla switch some years ago, but did not get it to work properly, the batteries did go down eventually. But back then I did not know much about electronics, that was before I even made my first simple SSG. I had a help from a fellow guy who was smarter than me in electronics. I just told him what I wanted and he drew a circuit which I had to build an test. This circuit was designed so that I could vary the frequency by adjusting the capacitor value of the 555 timer circuit part. The 555 timer gives a frequency with a very small duty cycle, probably 3-5%, this goes through a flip flop to get exact 50% duty cycle, then the output of the flip flop along with the 555 timer output goes through the 4001 chip that makes the small delay between each cycle to insure that the transistors have sufficient time to close. Here is the circuit:

http://www.emuprim.lv/bildez/images/.../sleedzis2.GIF

And here is the waveform of how the switching looks like:


This was the logic circuit part without the transistors and diodes




I am also intending to build the Switch now again because I see success from others. I will be giving the two capacitor circuit a try and see how it works.

Thank you
Jetijs

[Bit's-n-Bytes]

ldissing,
Where is the Jetijs circuit posted, But Jetijs comes with a great recommendation from Peter L so where do I get it? I want to see why it did not work.
John B

John, I think the Optos will work in the Jetijs circuit if they are configured like this.

Thanks
Bit's

[ldissing]

Jetijs Circuit for Oscillator

ldissing,
Where is the Jetijs circuit posted, But Jetijs comes with a great recommendation from Peter L so where do I get it? I want to see why it did not work.
John B

Mr. Bedini,

This is the post that contains the circuit.

http://www.energeticforum.com/renewa...html#post38353

Of course, you know how to use the optos, there need not be resistors to the emitter or from the collector. In fact, you can just use a diode from the opto emitter to the base of the MJL21194. You are using a diode and a 1uf cap in parallel, maybe that is better.

Just run the collector into the collector of the opto, out through the opto emitter to a diode (and maybe the 1uf cap) to the base of the big MJL21194.

I'm driving 3 of the optos off pin 3 of the CD4001 with a 220 ohm resistor to ground. I'm driving the other 3 optos off pin 4 off the CD4001 with a 220 ohm to ground. Collectors to collectors of the optos, emitter of the optos to a diode to base of the MJLs. Works like a charm.

Timing is adjusted, of course, by the capacitor and resistors in the 555 circuit. Easy to build, simple, not costly and you can get the parts at Radio Shack, except for the CD4001. You get a small delay off of the CD4001.

I've also run it off pins 1 and 2 of the flip flop. That seems to work fine too, but I like the delay.

If you want to run off the capacitor, then the timing needs to be rather quick, big capacitors take to long to recharge the cap, but small ones will power the circuit. No external power supply.

Hope this is what you are looking for.

Leroy

[Bit's-n-Bytes]

Mr. Bedini,

This is the post that contains the circuit.

http://www.energeticforum.com/renewa...html#post38353

Of course, you know how to use the optos, there need not be resistors to the emitter or from the collector. In fact, you can just use a diode from the opto emitter to the base of the MJL21194. You are using a diode and a 1uf cap in parallel, maybe that is better.

Just run the collector into the collector of the opto, out through the opto emitter to a diode (and maybe the 1uf cap) to the base of the big MJL21194.

I'm driving 3 of the optos off pin 3 of the CD4001 with a 220 ohm resistor to ground. I'm driving the other 3 optos off pin 4 off the CD4001 with a 220 ohm to ground. Collectors to collectors of the optos, emitter of the optos to a diode to base of the MJLs. Works like a charm.

Timing is adjusted, of course, by the capacitor and resistors in the 555 circuit. Easy to build, simple, not costly and you can get the parts at Radio Shack, except for the CD4001. You get a small delay off of the CD4001.

I've also run it off pins 1 and 2 of the flip flop. That seems to work fine too, but I like the delay.

If you want to run off the capacitor, then the timing needs to be rather quick, big capacitors take to long to recharge the cap, but small ones will power the circuit. No external power supply.

Hope this is what you are looking for.

Leroy

Leroy, since you have a working TS can you post a Vid or annotate the voltages at D9, D10 and D1, D5 with three types of load? Perhaps a standard automotive light bulb, then a DC motor, and lastly a load that has more capacitance. If you are using the Jetijs oscillator I assume that it is oscillating at a fixed value?

Thanks so much.

Bit's

[ldissing]

Question for JB

Mr. Bedini,

Is the right thing to do...to get the average voltage (B1 + B2) * 3/4 on each battery bank to 21.75 volts, i.e. 14.5+14.5 = 29 * 3/4 = 21.5V? At first, it will be around 18 volts but eventually, it should creep up to 21.75 volts. This would be FULLY charged batteries able to supply an easy 12V to a load.

If you have time, in a previous post, I asked a few questions. I do not know what kind of load can be supported on this, but it must be below the differential as you said. If, the batteries were at 14.5, then the differential should be around 29/2 = 14.5 - 4.5 or -6 volts. This would give a load of 10 or 8.5 volts. I'm just unsure about what kind of load I can run on this for extended periods of time. Bench testing will tell me, but you can tell me off the top of your head. I'd write your software (if you decide to go with a micro controller) for your advise. I will have to buy a kit or board to get it going, and maybe a compiler to get it to a code the micro can understand, but you've been more than forth right with your circuits.

Thanks,

Leroy

[ldissing]

Vid...don't know - D9,D10, D1,D5

Leroy, since you have a working TS can you post a Vid or annotate the voltages at D9, D10 and D1, D5 with three types of load? Perhaps a standard automotive light bulb, then a DC motor, and lastly a load that has more capacitance. If you are using the Jetijs oscillator I assume that it is oscillating at a fixed value?

Thanks so much.

Bit's

I'll post my schematic for you all. It is kind of a mess so don't hit me too hard on it.

I took JBs three battery switch and duplicated that. Then I said to myself, how did that work? The 12V battery negative was connected to the collector of the transistor. The emitter went to a diode and then to the 24V side.

Now, how do we do that in a 4 battery version. I'm not sure if this is the most efficient way, but I used 2 bridges and removed D9 and D10. The bridge is D9 and the other is D10. One side of the circuit goes to one bridge, and the other side of the circuit goes to the other bridge. It could probably be done with just one bridge, but mine works and I could never get it to work before, so before I break it....that is how it works.

I can't tell you what the voltage is. It fluctuates, it is AC. You are creating AC with batteries and the bridge is the only place I can tell you what the voltage is. That also varies based on my resistor/capacitor settings. I think we really want a differential voltage of 12V, but right now mine can be varied from 6-12V. When my batteries are more charged, it will be higher, I started with fresh batteries, but they are years old. Right now, with my current settings and a 6V 25ma bulb, it is approximately 8.2V at the bridge (even with a 6V bulb on the bridge). I'd like to see it closer to 12V, but all those diodes do cause a drop in voltage.

I have run a small motor (don't know the amperage) that runs on 3V to 18V on it as well and it seems to work fine too. Don't know what the voltage was across the bridge, but I can check that.

I don't know if my schematic will show up on this post. I used the Jetijs circuit posted a little earlier for the oscillator. Refer to my post to JB on how the optos are connected.

It works! Never thought I'd see it working, because I'm such a bonehead.

Leroy

[John_K]

3 Battery Test - Supersized

Well, after a successful test of charging up all 3 12v 7Ah batteries it was time to supersize.

Believe it or not, it works better with bigger batteries!

I swapped out the 7Ah's for 12v 130Ah's (Trojan SCS225's)
Start of run: (batteries are numbered)
#12 -12.53
#11 -12.63
#3 - 12.66
Total - 37.82

After ~1 hour running:
#12 - 12.21
#11 - 12.36
#3 - 14.01

After ~1 hour rest:
#12 - 12.50
#11 - 12.61
#3 - 13.10
Total - 38.21

Still using the 12v 55w quartz halogen bulb. Yes Leroy, the trannys are heatsinked

I'll let them rest overnight and rotate in the morning.

John K.

[nilrehob]

I find many questions and will get to them all so do not think I'm ignoring you all, it's just I'm going though all the analog circuits to see if I have missed something I did not say to the group. I want you all to be successful with this, no secrets here.

John,

Thank You for your kind answer,
I think language and other differences is a problem in forums like this,
we all mean well but we behave and express ourselves differently,
so please bear with me

If it's difference in potential and not current that charges a battery, then the more batteries in parallel the better, but always two and no more in series, and then circulate them, would be beneficial?

/Hob

[ldissing]

"My" circuit vs. JB RC RB BE Tesla Circuit

Mr. Bedini,

Can I call you John?

BTW, my name is Leroy John and my friends call me John....

In my circuit, I'm always putting the load between the collector and the negative of the battery. In either direction, the load is between the transistor collector and the negative.

In your (JB RC RB BE) circuit, it seems to me the load is kind of charge/potential siphon and it WILL run a load. I hooked it up and it does work.

Q: Is this what you see the load as in that circuit?

The problem I'm having, is that when you connect the 24V bank to the 12V bank through the transistor, you are basically creating a direct link through the transistor and the transistors get hot. I've tried all kinds of frequencies from really slow to really fast and eventually (not a long time), those transistors get hot. And, it doesn't matter if the load is big or small, they still get hot. P.S. I'm using a small delay between transitions to make sure the transistors are off before turning the other side on (because they aren't all matched, I'm sure).

If the transistors were biased properly, then they should not get hot, because the transresistance would not come into play. It seems that the transistors are becoming the load, which is not what we want.

Q: Is that the reason for the 1uF cap in parallel with the diode? It keeps the bias there?

Q: Is the reason the transistors are getting hot because they are not biased properly?

Q: Is the transistor, in my situation, without the 1uF cap in parallel with the diode causing the transistor to become the load?

I can go and buy a bunch of 1uF caps to hook it up, but I'm sure you already know the answer. Time is money and all that.

Thank you for your time and allowing us access to your knowledge. I do not believe any of the others are still alive, so you (and maybe Peter L.) are the sole knowledge bearers.

Leroy (John)

[Bit's-n-Bytes]

Mr. Bedini,

Can I call you John?

BTW, my name is Leroy John and my friends call me John....

In my circuit, I'm always putting the load between the collector and the negative of the battery. In either direction, the load is between the transistor collector and the negative.

In your (JB RC RB BE) circuit, it seems to me the load is kind of charge/potential siphon and it WILL run a load. I hooked it up and it does work.

Q: Is this what you see the load as in that circuit?

The problem I'm having, is that when you connect the 24V bank to the 12V bank through the transistor, you are basically creating a direct link through the transistor and the transistors get hot. I've tried all kinds of frequencies from really slow to really fast and eventually (not a long time), those transistors get hot. And, it doesn't matter if the load is big or small, they still get hot. P.S. I'm using a small delay between transitions to make sure the transistors are off before turning the other side on (because they aren't all matched, I'm sure).

If the transistors were biased properly, then they should not get hot, because the transresistance would not come into play. It seems that the transistors are becoming the load, which is not what we want.

Q: Is that the reason for the 1uF cap in parallel with the diode? It keeps the bias there?

Q: Is the reason the transistors are getting hot because they are not biased properly?

Q: Is the transistor, in my situation, without the 1uF cap in parallel with the diode causing the transistor to become the load?

I can go and buy a bunch of 1uF caps to hook it up, but I'm sure you already know the answer. Time is money and all that.

Thank you for your time and allowing us access to your knowledge. I do not believe any of the others are still alive, so you (and maybe Peter L.) are the sole knowledge bearers.

Leroy (John)

Leroy, this is just a hunch; You may want to try Schottky diodes in place of the 1n4007's. The 4007's maybe slow in recovery compaired to the frequency of oscillation and are being subjected to recirculative current through the transistors, in turn more heat in the transistors.

Thanks

Bit's

[blackchisel97]

I got most of it assembled but SG (I'm talking about SG3524, as I did before) is only pulsing of emitter #14. When I connect PS to it #11 pulses once and after that only #14 works. Freq and DC adjusts nice. I tried only 2 I have. I have 560 Ohms instead of 680 in their collectors but this shouldn't be a problem Don't know what is wrong at the moment.
From diodes which I have I'm thinking of using MUR860, maybe better than 4007. Not a good start

Vtech - BTW, this is my name (Witek), only changed spelling for easy English pronunciation, not a nickname.

[nilrehob]

old tests

Some time ago i did some long-run tests on different TS-setups as shown in the pictures below.
Y is average voltage on batteries and X is time in hours.



T1.1, T1.2 and T1.3 are no TS but all fours batteries in parallel as a comparison.

All other tests are some sort of switching and always mechanical with a relay going as fast as it could while maintaining 50-50, i think it was around 20Hz max.

The best run was T4.1, with a lamp and a coil as load plus diodes to catch the spikes into the batteries.

The reason I posted this is picture 2, witch is a zoom in on the start. All test started high on voltage after the batteries had been charged and then drops rapidly, but on some runs the voltage went up again before going down like the others. While exiting when it happened I figured it was just a kind of bounce in the batteries, a normal thing, nothing to do with the TS and didn't make much out of it.



The best bounce was T4.2, same test as T4.1 but a later run, and T6 same as T4 but with rectifier plus caps around load, and T7 which was a similar test.

Maybe I was wrong?
Maybe that was a success and not just a normal bounce?
But why did the voltage then drop?

I just didn't know what to look for when the switching was on, how to tune it, and i still don't know actually.

Maybe someone knows what a scope on a battery should look like or something? I don't know.

I must admit it was some time ago, and I've learned a lot since, but I'm still in the dark on how to tune it.
I don't know how batteries behave.

/Hob