Bedini Tesla Switch DvD Questions.

[Tom C]

@ Gary

if there is interest, we can probably do component kits minus the circuit board for now, it will be expensive, but you will get matched transistors which is important for the TS. I don't think as a small company its worth time to pick individual components, but a complete component one would be worth the time I think. I am going to talk to JB about diode substitutions, if there is a way to use the SB 560 instead or another les expensive diode.

@ Don,

[brodonh]

Hi all,
I flipped the LEDs now and switched back to 5V. First run under load with a small motor looks very promissing. Bigest voltage drop at one Bat (Pos 2) was from 5.40 to 5.36 after 15min runtime. 10min after disconneting the motor the Bat is back up to 5.38 still climbing.
The 12V Bats are charging on the SG so I can start the next test fully charged.
Time to sleep now...

Thorsten,
Do you have a part number(s) for your LED's?
Thanks,
bro d

[bruno]

Hi all,
When I whas studying the pulses out of the sg to the opto's I noticed that the puls for stage 2 is in between the puls off stage 1. So if the target is 50/50 duty, it would mean that we have to influence both sides (edges) of puls 2 or am I wrong?



I'm going to build a measurement setup to see what the switching delays look like but with one source (one ground) so I can measure it better. The 20 micro seconds are now measurements without grounding on one channel so a lot of noise is influencing the measurements.

[bruno]

Hello all,
Just did some measurements on the pulses with no load attached to circuit and interruped the pulsing of stage 2 by removing an opto out of that stage.
The batteries used are 4 x (8 x 1.2V Ni-mh)
The measurements or done as follow:



The results for the on delay part:





The results for the off delay part:



Then I pulsed the stage with maybe what would be the resonance frequency of 22Hz:



Could this way of measuring be a possibility to find the resonance frequency of the batteries in this stage off the switching?
I'm going to do some measurements on stage 2 and see how many difference there is.

[bruno]

The measurements off stage 2:

On delay stage 2 part 1



On delay stage 2 part 2



Off delay stage 2:



Total overview off puls stage 2:



Now running the circuit at 22.5Hz with no load and see what happens

[Thorsten]

@ bro:
My LEDs are from this package:
http://www.conrad.de/ce/de/product/1...f=searchDetail
red one is a 2V, 30mA type
green is 2.2V, 30mA
blue load Diode has 3.2V, 20mA

@bruno:
I think you have to look at the resistors you've connected to the SG3524 if you want to set the duty cycle to 50:50. Try to use exactly the values from the DVD. For me this works best, even for the dead zone.

[bruno]

@ Thorsten,
The values of the resistors 2 x 330 ohm on circuit drawing are changed to the 680 ohm like shown on pcb layout on the dvd.
All other resistors and components have the values as on dvd.

I did the same measurements on my setup with 12V 7Ah batteries to see what differences came up.

On delay Stage 1:



Off delay stage 1:



Stage 2 On delay:

[bruno]

Stage 2 off delay:



I also measured the influence of a load to the switching times, but this has no influence, only the voltage has a small drop.
The load whas a led with diode in series attached to - bat 2 and - bat 4 (the normal TS load points)



At the moment I still have no charging but I know something more about the delay times.

Summerised:

_____________Stage1 on part 1___Stage1 on part 2___ Stage1 off_____Stage2 on part 1___Stage2 on part 2___Stage 2 off delay
10.3V setup___6.3 micro sec _____23.9 micro sec______20.55 milliSec___4.150 micro sec ___19.7 micro sec_____19.2 milli sec
12V setup ____5.6 micro sec _____34 micro sec________45.2 milli sec___4.95 micro sec_____21 micro sec______45.2 milli sec

[bruno]

I did some extra measurements to see the influence of the duty cycle on the voltage and saw that there wasn't any difference but the led on output whas dimmer.
The grey graph is the one with no influence on the duty cycle (with pot meter 100K from pin 5 to ground as Bro D has found)


I hope that when I study the paper of Tom Bearden on the negative resistors inside batteries that I can make a link to what is not correct on my circuit to charge the batteries. If anyone see a difference with the signals on a working circuit (charging the batteries I mean, AC output on my circuit is completly ok) please let me know.

[brodonh]

[QUOTE=Thorsten;18860]@ bro:
My LEDs are from this package:
http://www.conrad.de/ce/de/product/1...f=searchDetail
red one is a 2V, 30mA type
green is 2.2V, 30mA
blue load Diode has 3.2V, 20mA

Thank you Thorsten,
Have to get to the bottom of why my LED's wouldn't light according to JB's video board, as some others have reported.
Going to maintain spread sheets for data.
bro d

[bruno]

Hello all,
I just found another page handling tom beardens paper and there is stated:

Let me warn you that you must use microwave switching techniques, and you must switch in 5 nanoseconds or less; one nanosecond is better. The entire overpotential is likely to be over in about 20 to 40 nanoseconds, depending upon the specific battery, load, and other circuit conditions. Capacitance effects may extend this in some cases up to a microsecond. So if all you know is ordinary motor switching, go get the services of a microwave switching engineer first. The average motor switching fellow will be amazed at the notion of switching so suddenly. The microwave switching engineer will simply shrug his shoulders and say, "Piece of cake!" He does that every day without a second's hesitation.

I hope this is not true.
The page is: http://www.eternaltruth.net/Science/...;20Battery.htm

Every internetpage has another resonant frequency for the lead acid batteries:

An ordinary battery filled with electrolyte fits the bill nicely. While it's not commonly known, ordinary lead-acid storage batteries have a resonant ionic frequency, usually in the range of from 1-6 MHz.

Taken from: http://readious.blogspot.be/2012/12/...uits_6484.html
But no page is telling how they measured that frequency, it's like a fairy tale told thousand times and changed along the way

[Thorsten]

@bruno: by changing the 330 ohm resistors to 680 you are only changing the current limit for the led side in the optos. Each resistor of the two is for one row of the optos. The duty cycle will be changed by the 4k7s, think you have to look for those.

@ bro d: I watched the LEDs again in the DVD. the two leds from the DVD circuit don't light up. So they are connected like JB draws it to the board otherwise they would light up. In addition JB has soldered two more leds two each side, both blinking. For me it looks like they are connected from the anode of the unlit LED to the anodes of the two SB570s near the - contacts of the bottom bats.
when one of these 4 LEDs is not blinking, something in the circuit is running wrong. Thats my understanding.
Thorsten

[Dave Wing]

Hi Guys,

I am not sure what you guys are doing in your switching but I wanted to try and clear things up a little. In the image below there are two different on and off times, one for the motor and one for the energizer. These two seperate DC pulse waveforms can be explained as both providing a positve 50/50 duty cycle... That is exactly what we are looking for in the Tesla Switch, what you take from side "A" you must put back into side "A" from side"B" and vice versa... it has to be 50/50 or it will not work properly. Now you have only to choose the length of time you want any side to be on for. Each side could be on for 1 minute, 10 minutes, 60 times per second or whatever you choose, it is still a 50/50 duty cycle as long as you do it equally to each side. This is called retroflection.
Image taken from.. http://www.icehouse.net/john1/index11.html


Any comments welcome.

-Dave Wing

[bruno]

Hi Dave,
I studied the pulses to see where differences where with the 50/50 duty.
Because it's not clear how great the working zone is in milli or micro seconds between charging and no charging, I wanted to find out how the switching looks like after the optos and transistors because off the delays caused by all the components. Now I know where and how much the delays are, I can try to make them more accurate to the 50/50 duty on the final switching and not direct on the outputs of the SG3524. Because if we talk about resonance in- or outside the battery I think the working zone is very small as is always the case with resonance so far I know (I'm not an expert). It could be that it's totally unnecessary to go into the micro second range for a working TS, but I have no charging with the delays I've measured, so I thaught that the only way is more accurate switching.

[bruno]

An extra clarification:
I measured for example that when a puls is given it takes 5 micro seconds before the transistor output starts to react. Therefore I whas thinking to have a puls on the SG3524 not exactly 50/50 accounting for the 5 micro seconds but at the end after the transistors end with an exact 50/50 duty.
And this also for the off delay's.
I hope you understand what I mean.