Use for the Tesla Switch

[lamare]

Bit's
Do you mean this?
John B

"I guess I did not explain myself with this circuit. If the load is in the collector it's no problem turning the transistor on, But what if the load is in the Emitter?, since the switch uses different potentials at different points I had to calculate the difference voltage between collector and emitter. when using the device as a cathode follower you will not get the voltage you need, so I use this method to force the transistor to discharge the most voltage between collector and emitter, a quick discharge without base to emitter resistor. The FET is the real problem with it's gate capacity. With this method you do not need a driver. However you do it anyway you want, I will stay with what I have . This switch uses inverted potentials in some of the devices and can be a real problem."


The way I understand this, is that the capacitor charges up to the diode's bias voltage (about 0.6V) when switching the power tor on. Then, when you want to switch it off, you intend to push the base of the power tor down to -0.6V in order to quickly switch it off, right?

However, if there is no pull-down resistor somewhere, who's pulling the base of the power tor down?

This reminds me of a problem we found some 10 years ago in PLC backplanes I was developing units for. There was one reset signal on that backplane that had an open collector driver without a pull-up resistor. In practice, this was no problem. It worked just fine, ..., most of the time.
However, it was a hell of a problem once you started ESD testing, i.e. firing sparks onto the external connectors of the modules you built for the backplane in order to simulate static discharging. Then, this signal would be raised by the ESD, causing your modules to reset when they were not supposed to. The obvious solution would be to add a pull-up resistor to the backplanes, where they should be, but these things had been in production for several years. So, that was not an option. So, we eventually solved the problem by adding metal shields inside the modules.

So, it appears to me that without a pull-down resistor, the circuit might be pretty sensitive to disturbances....

Another point is that with the introduction of a diode in your base line, you create an additional voltage drop of 0.6V, which will also appear over your power tor, resulting in unneccesary power loss and unneccessary heating of your power tor.

However, with this, I'm not taking any reverse potentials into account here, so I might be totally wrong for at least some of the switches.... OTOH, these might be some points that can be taken into consideration at some point.

Anyway, I'm definately going to order some stuff to experiment myself. I'm thinking about using 9V 150 mAH NiMh batteries, and TIP41 transistors. These are cheap and then I should at least be able to make a proof of concept lighting a small light-bulb or something.

[John_K]

John K,
Are you still up or just getting up?
John B

John B,

Still up, will be for a while.

John K.

[SkyWatcher]

Hi folks, John K, are you still cycling your batteries with the tesla switch setup and how is it going and just wondered if you've tried it without all the transistors in there and just gone direct. I will be testing it soon, so I was just curious how it was going and if you tried it without the transistors since your not pulsing it seems they would not be needed. Thanks.
peace love light
Tyson

[John_K]

Hi folks, John K, are you still cycling your batteries with the tesla switch setup and how is it going and just wondered if you've tried it without all the transistors in there and just gone direct. I will be testing it soon, so I was just curious how it was going and if you tried it without the transistors since your not pulsing it seems they would not be needed. Thanks.
peace love light
Tyson

Tyson,

I'm not sure what you mean by "all the transistors in there and gone direct".

I think you mean have I just taken the transistors out and hooked the batteries together. Please tell me if I'm wrong here.
The transistors are there for a very good reason, even when the circuit is not being pulsed.

I believe this is what John B wants us to learn and understand. I think I know why they are there and what they do, because I just calculated it out on paper. John B gave me the answer I was looking for in the first question I posted on this thread.
But you'll need to see it in front of your own eyes to believe it.

When the transistors are pulsed there is another effect that is also observed, but you won't know what it is until you understand what this very basic 3 battery test is for.

I'm still testing with different things to prove what's happening. More along the lines if knowing what to expect when I chnage something, like rotating the batteries and changing the load.

John K.

[Bit's-n-Bytes]

Bit's
What if !!!!!!! We do not need all these parts to do the job and only need two devices to do the whole thing? Has anybody given any thought to this? What if the SG3524 could be controlled using the error amplifier at the duty cycle pin that senses all the voltages at each battery. What other device could we use to switch the batteries since we can see that the potentials change quickly? I guess as Peter says I have been known for tricky circuits, just a little quiz. After all do we really need all the devices in the circuit to switch four batteries? By the way excellent job on the circuit diagram and the code, I can get this to run in the pic I have just a few changes.
John B

John, thanks. I am all for simplicity. One thought however, how vulnerable is the SG3524 and error amplifier to "noise"? I am thinking this may be a challange to controll "drift". You had also metioned the SG3524 does not like the cold so metal film resistors and proper packaging are essiential. (ok, that was 2 thoughts).

[Inquorate]

Reliable

John, thanks. I am all for simplicity. One thought however, how vulnerable is the SG3524 and error amplifier to "noise"? I am thinking this may be a challange to controll "drift". You had also metioned the SG3524 does not like the cold so metal film resistors and proper packaging are essiential. (ok, that was 2 thoughts).

Vacuum tubes might work..

Tesla electric car

[Bit's-n-Bytes]

Vacuum tubes might work..

Tesla electric car

@Inquorate

Repaired our first TV with tubes. In fact you could go down to any local grocery store and test them yourself. Those stores are no longer, same with the tubes , but technology wins out with better switching devices and availabilty.

Keep the ideas flowing.

Bit's

[Helijason]

Bits, it will take me some time to go over this (holidays) but at a first glance I made a small change on a comment below. I also will need to review the newer pics. I programmed them with assembly back in the day and it was pretty easy to understand code execution time. Our code should not change our pwm.

Start:
SYMBOL D1Sense = W3 '########## Reserves buffer for D1 Value
SYMBOL D5Sense = W4 '########## Reserves buffer for D5 Value
SYMBOL D9Sense = W5 '########## Reserves buffer for D9 Value
SYMBOL D10Sense = W2 '########## Reserves buffer for D10 Value
SYMBOL TotalCompare = W1 '########## Reserves buffer D1 + D5

Bit's
What if !!!!!!! We do not need all these parts to do the job and only need two devices to do the whole thing? Has anybody given any thought to this? What if the SG3524 could be controlled using the error amplifier at the duty cycle pin that senses all the voltages at each battery. What other device could we use to switch the batteries since we can see that the potentials change quickly? I guess as Peter says I have been known for tricky circuits, just a little quiz. After all do we really need all the devices in the circuit to switch four batteries? By the way excellent job on the circuit diagram and the code, I can get this to run in the pic I have just a few changes.
John B

I like the simplicity of using analog to solve this problem. The challenge as I see it is to understand how the pwm is working with different loads. I don't have an oscilloscope and can not measure frequencies once I hit a sweat spot.

Innitially I was thinking we should develop a system that lets us know within some certainty what our pwm is.

Jason

[Bit's-n-Bytes]

Bits, it will take me some time to go over this (holidays) but at a first glance I made a small change on a comment below. I also will need to review the newer pics. I programmed them with assembly back in the day and it was pretty easy to understand code execution time. Our code should not change our pwm.

Start:
SYMBOL D1Sense = W3 '########## Reserves buffer for D1 Value
SYMBOL D5Sense = W4 '########## Reserves buffer for D5 Value
SYMBOL D9Sense = W5 '########## Reserves buffer for D9 Value
SYMBOL D10Sense = W2 '########## Reserves buffer for D10 Value
SYMBOL TotalCompare = W1 '########## Reserves buffer D1 + D5




I like the simplicity of using analog to solve this problem. The challenge as I see it is to understand how the pwm is working with different loads. I don't have an oscilloscope and can not measure frequencies once I hit a sweat spot.

Innitially I was thinking we should develop a system that lets us know within some certainty what our pwm is.

Jason

Thank you Jason, I'll make the update.
Bit's

[Helijason]

Batteries

Something that I was thinking about is the nature of batteries. I always found it interesting that when pulling a load that voltage goes down. We used the load and the voltage to determine the state of the batteries. After removing the load the batteries would float or the voltage would be near nominal (increase, not sure the proper terminology here).

I think that in part that this is helping the Tesla switch. I looked at the ideal model of a diode, and it seems that if we get the appropriate voltages across our diodes then the system is decoupled (not current flowing) and it is purely potential that is charging the batteries. Does that mean that we should switch when we no longer have a balanced system?

I have too many questions, need to build (need a job first ).

Jason

[stonewater]

test1.jpg running

running the 3 bat test1 circuit. as of now 2 of my 3 batts have climbed above 13 volts. soon all three



Tom C

Tyson,

I'm not sure what you mean by "all the transistors in there and gone direct".

I think you mean have I just taken the transistors out and hooked the batteries together. Please tell me if I'm wrong here.
The transistors are there for a very good reason, even when the circuit is not being pulsed.

I believe this is what John B wants us to learn and understand. I think I know why they are there and what they do, because I just calculated it out on paper. John B gave me the answer I was looking for in the first question I posted on this thread.
But you'll need to see it in front of your own eyes to believe it.

When the transistors are pulsed there is another effect that is also observed, but you won't know what it is until you understand what this very basic 3 battery test is for.

I'm still testing with different things to prove what's happening. More along the lines if knowing what to expect when I chnage something, like rotating the batteries and changing the load.

John K.

[Bit's-n-Bytes]

What other device could we use to switch the batteries since we can see that the potentials change quickly? I guess as Peter says I have been known for tricky circuits, just a little quiz. John B

John, are you thinking Triacs?

Bit's

[stonewater]

3 batt switch

ok here is the right pics, not the ones where I put the circuit together wrong

[Dave Michael Rogers]

Basic Tesla Switch

Good Evening John and group,

I have attached my Tesla switch design based on your three battery setup which I tested a couple of years ago, posted the results, but then put it to the back of the bench.

The attached shows two designs of the same idea for a mechanical rotary Tesla switch. The torch was to continue the basic testing of a manual switch, but the second part show the design that will be used to power the Kromrey converter replication I have built.

John, you know me and that I do follow instructions. I will try anything you suggest in order to get these working as claimed, so I hope these basic designs have some merit and that I am not just wasting my own and now everyone elses time, especially yours.

Regards
Dave

[Mtech]

Pic controlled TS v1

Team, here is a first cut of the proposed schematic for the "Digital Controlled TS". Please let me know if you see mistakes. John, can you let me know if I put the sensing points in the right areas?

Thanks

Jeff

Hi Bit's

Just had a peek at your drawing...great job. Noticed that Q2 emit/coll polarity appears to be swapped ??

@John B

Just wanted to echo sentiments of others who have already expressed their gratitude for your mentorship. You continue to inspire us in pioneering the future of alternative energy. I've built a few versions of the TS but haven't seen the magic yet! I have no doubt though that it won't be long now and that there is radiant light at the end of this tunnel!


Kimesh