Use for the Tesla Switch

[Redisnoc]

Glad you're back John,
R

[John_K]

Back to Basics

Hi Group,

After having no real joy with my Tesla Switch and Scalar Charger, I decided it was time to go back to basics. (Yes, I had limited success with both)

Today I replicated the circuit in the Eike Mueller document, called "TEST # 11A" or Figure T-2 on page 8.

I could not get an inverter (rated at 120W) to work because I think my batteries were too small, so I used 2 x #47 bulbs in series. This was a load of 85mA.

The 400Hz switch as per the diagram was made up of a 555 timer with 50% duty cycle. I got it to run at 383Hz. The output of the 555 was driving an opto solid-state relay, which switched an MJL21194. This is a similar setup to "FIGURE T-3"

Using 7Ah gel-cells as batteries I was able to get all 3 batteries charged up over a few hours whilst rotating the batteries when the 12V battery was at 14.5V.

At first I had too much of a load (4 x #47 in series/parallel), which made the 24V bank die pretty quickly, but with 2 x #47 bulbs in series the 24V bank dropped a little when the load was applied, but then only dropped about 0.01V every 30 minutes or so. It was pretty clear that the 3rd battery charges faster when the load is higher (i.e. 8 x #47 bulbs), but IMO this is a superficial charge which WILL kill the 24V bank.

After rotating the batteries several times, I found that the batteries charged slower each time and took longer to get to 14.5V . IMO, the batteries started to show their true impedance, which was evident when the switch was turned off - the batteries only dropped about half a volt. (Down to 13.8V)

So it appears that choosing the load is very important - you need a balance of good battery charging with minimal load on the 24V bank. If the load is chosen well, the 24V bank hardly sees the load at all and will find a place where the 24V bank will not drop much but the 12V charging battery will slowly increase over time. It's a balancing act. My 24V bank seems to like sitting at 25.84V for a hours whilst the 12V battery increases 0.01V every couple of minutes.

I think the trick is to find a load that will allow the 24V bank to stabilize at a steady voltage while the 12V battery charges slowly - in line with nature's natural charging curve. I also think that if you try and charge the 12V battery too quick, by using a lower resistance load, you'll pay for it because the 24V bank can't cope.

Now that I have had success with this replication I need to apply the same theory to the full TS. Slower is better it seems, if you want the batteries to charge.

I was listening to the Bill Jenkins "Open Mind" interview again where John Bedini talks about the Tesla Swicth that he presented at the Tesla Symposium in 1984. There was a caller (Greg) that had some questions about the schematic he handed out at the time. I believe it's the same schematic that is printed on page 7 of the Eike Mueller document. In that interview John explained that the circuit has two modes. The first mode is for charging, where the oscillator is set for around 5Hz. The second mode is for running a load, where the oscillator frequency is around 60Hz to run DC motors, after being rectified.

If you study that schematic, you can see what John means. There is a "S1" switch that sets the SG3524 to "low" mode for charging. I think this is for the 5Hz mode (remember he is using 5V NiCd's in that schematic). This is also evident by the LED labelled "CH" (for charging). When S1 is open the SG3524 probably runs at a much higher frequency to run the "LOAD MOTOR".

John also went on to say in that interview that he was planning on releasing an amplifier based on this technology, that would run for days and only draw 0.5W from the wall (for switching??), then only take 15 minutes to recharge the batteries when they went flat. That would have been nice to see, but I'm guessing that's about the time John was told to shut up.

If anyone is interested in a copy of the interview I have it in mp3 format. PM me if you want me to email it to you, it's only 2.5MB . Don't forget to give me your email address.

I also have a copy of the interview on "Open Mind" where Bill and Ray call in and discuss the Scalar Battery Charger and how they got it to work, with comments by Ron Cole and John Bedini. I haven't figured out how to get the SC to work yet, but it's still a work in progress. I had it running so the 12V battery did not discharge, but would still run the load.

Anyhow, sorry for the long post. JB, I'll give you a call on your Monday afternoon to talk about this and catch up about some other things. I've seen enough to know that the TS works - even for a "555 for dummies" based setup.

For everyone else (without trying to sound too cocky), if you are frustrated like I was I urge you to replicate the Eike Mueller/John Bedini experiments that are written in the Mueller document. It basically covers the 3 battery experiment that shows how to use differnetial potentials. With respect, I feel that many of us are going too fast and need to keep the basic concepts in mind when trying to get a TS to work. Yes, a digital controlled setup will help control the timing but you must know what you are looking for before you dive in. It should not need fancy stuff to get it working. Too many people go for the gold too early - batteries like to charge on nature's time scale, not our's.

I'm thinking of reducing the timing on the full TS to minutes, if not hours until one bank is charged before switching banks. We need to be patient. I wonder how many of us have had a working TS in front of us that just needed more patience, not changes.

John K.

[Bit's-n-Bytes]

Yes, a digital controlled setup will help control the timing but you must know what you are looking for before you dive in. It should not need fancy stuff to get it working. Too many people go for the gold too early - batteries like to charge on nature's time scale, not our's.

I'm thinking of reducing the timing on the full TS to minutes, if not hours until one bank is charged before switching banks. We need to be patient. I wonder how many of us have had a working TS in front of us that just needed more patience, not changes.

John K.

John K., I agree with you on the reducing of the timing. I now have almost a full second in between pulses, (approx. 60Hz) and have gained about 2 volts over night. In addition, the load balancing you mention is so very critical.

Bit's

[Mark]

Hello John K,

Where can I find this Eike Mueller document, called "TEST # 11A". Can you post a link?

Thanks and congrats on your success!

Mark

[billwarner9997]

Eike Mueller Report

Mark:

Check post #766 for link to full report.

Edit: Its on page 26.

[ldissing]

Going Digital

Guys,

I received my little board (PICKIT 2 Debug kit) with a PIC16F887 on it along with a potentiometer and switch.

I programmed it so that the potentiometer (analog input) is used to derive the frequency and the switch is used to decease the duty cycle. Each press of that button decreases the duty cycle by 10%. It works like a charm from very fast to very slow and I don't have to even change the code.

Batteries seem to like 30 Hz pretty well (one pulse on each side every two seconds). I guess I could take it down some more, but I don't know exactly right now what the slowest it will go it. The batteries do seem to be charging, but slowly. I'm using some 6V 25ma bulbs as the load on each side, and only when there is a delay (decreased duty cycle) do I see the voltage on the series bank hit max voltage. Seems that having the delay before the other side pulses does do something unusual.

Anyway, just wanted to let you all know I am working on this and haven't died or anything. Still need to create the basic circuit on a board. I have many loses with all the wire in my "system"...er....my setup. I'm sure things will get better when this is done.....just need to find the darn etching equipment.

If anybody buys that board, I can send the code to you. It is not too complicated, but not too easy either. Have to do an analog to digital conversion to get the potentiometer value and that is the signal to change states too. The timers are great and used for delays, otherwise, it is just to fast. State machine oriented, so it is pretty fail proof.

Leroy

P.S. Did JB say 2 HZ, or once every two seconds?

[Plazma]

With All Due Respect to Heinrich Hertz . . .

Hi Leroy and Welcome Back . . .

Is it possible that "1 pulse each side every 2 seconds" is
30 Pulses/Minute (0.5Hz) and not 30Hz? Sometimes Hertz can hurt -

Cheers,

Plazma

[ldissing]

Hi Leroy and Welcome Back . . .

Is it possible that "1 pulse each side every 2 seconds" is
30 Pulses/Minute (0.5Hz) and not 30Hz? Sometimes Hertz can hurt -

Cheers,

Plazma

The hertz (symbol: Hz) is a unit of frequency. It is defined as the number of complete cycles per second. (WIKI)

So, I'm way too slow. 60Hz is 60 times a seconds, so I'm at .5Hz, thank you. I will speed it up by a factor of 4. Brain farts I guess. I noticed that I could only program it correctly in a state machine too. Worked all day...and could not get it programmed correctly...been away from it too long. Programmed the state machine in about 5 minutes and it worked perfectly first time.

Leroy

[John_K]

So, I'm way too slow. 60Hz is 60 times a seconds, so I'm at .5Hz, thank you. I will speed it up by a factor of 4.
Leroy

I'm still confused if the frequency should be 0.5Hz or 2Hz. JB posted that he sees the magic between 0.5 and 1 second, depending on the batteries.

JB, can you please clarify?

John K.

[ldissing]

I'm still confused if the frequency should be 0.5Hz or 2Hz. JB posted that he sees the magic between 0.5 and 1 second, depending on the batteries.

JB, can you please clarify?

John K.

And just as important....Is the pulse on a side 250ms, 500ms, or 1 second. There are two pulses, one one each side. So, if one side is pulsed at 250ms, the other is also 250ms, is that 2 hz, or 4 hz?

Leroy

[Bit's-n-Bytes]

It's in the math

And just as important....Is the pulse on a side 250ms, 500ms, or 1 second. There are two pulses, one one each side. So, if one side is pulsed at 250ms, the other is also 250ms, is that 2 hz, or 4 hz?

Leroy

There are only 1000 milli seconds in a second, so it becomes an issue of how many pulse's can you start within that second. If the duration is 500ms then you can only start 2. Now if you can, and with the digital TS absolutly, if you fire both sides together within that same 1 second, you still only have 2, side A for 2 and side B for 2.

Right?

Bit's

[John_K]

More Maths...

And just as important....Is the pulse on a side 250ms, 500ms, or 1 second. There are two pulses, one one each side. So, if one side is pulsed at 250ms, the other is also 250ms, is that 2 hz, or 4 hz?

Leroy

There are only 1000 milli seconds in a second, so it becomes an issue of how many pulse's can you start within that second. If the duration is 500ms then you can only start 2. Now if you can, and with the digital TS absolutly, if you fire both sides together within that same 1 second, you still only have 2, side A for 2 and side B for 2.

Right?

Bit's

More maths....

a. 250ms side A, 250mS side B = 2Hz
b. 500mS side A, 500mS side B = 1Hz
c. 1000ms (1 sec) side A, 1000mS side B = 0.5Hz

My guess is "c." will yield the best results, but just a guess. My tests with the 3 battery test switch however showed that the longer you let the charging bank go, the better. Need to prove it on the bench with the full TS though.

John K.

[nvisser]

Charge pulsing

The way I understand it is a pulse every 0.5 to 1 second.
That is 2 to 1 cycles/second or Hz.
Then of coarse the pulse width or duty cycle can vary giving you different dead times which also sound like it is important.

[blackchisel97]

Mueller Report

@John K. It was pretty good idea to re read this again
BTW. I left you PM.

I took My SC apart, frustrated enough with SG having mind on its own. I'll go back to it when I'll get working reliable switch to begin with.
Meanwhile, I built 555 oscillator with adjustable d.c and freq. I set @ 50% and near 400Hz driving TIP41 via H11D3, just like I did in SC. In the negative branch I used 5W automotive bulb in series with my 300T/30T toroid. I got quite a kick from 300T while running through small coil) Charging battery gains steady.
I'll try different frequencies.


Vtech

[ldissing]

...

if you fire both sides together within that same 1 second, you still only have 2, side A for 2 and side B for 2.

Right?

Bit's

And that would be 2hz, or 4 hz? I'm saying I think that the frequency is on a side basis, so that would be 2 hz, because side A fired 2 times in that one second. Pulse width = 250 ms on, 250 ms off (ignoring duty cycle). Have we reached consensus?

Leroy