Andy's thread- Don Smith Device

[andy]

Don Smith's Resistor Divider - not two resistors but coil (rheostat) ?

Don Smith shows voltage divider #11 as two resistors. But!
(in red circle)
   


Voltage divider can be wired coil (rheostat)
https://en.wikipedia.org/wiki/Resistor#Wire_wound

or wired potentiometer, as show in this video:
https://www.youtube.com/watch?v=Ca2e3evwVK0&t=62s

I can replace both resistors #11 with wired coil - rheostat.

   

Can the coil from rheostat function as LC circuit with capacitor #12 ?

Probably if I use this correctly, then I get zero resistance in this serial LC circuit.

So maybe interesting redraw of Don Smith's image can be this:

  voltage divider with coil.jpg (Size: 6.31 KB / Downloads: 126)

[Shylo]

If you use DC it must be pulsed
AC is already fluxuating
It needs to be passed through a coil to slow it down
Load as many caps as you want or need
Use the caps for your output

[andy]

Don Smith about MHz Frequency of the Coil


"The highest frequency that you have represented in any of the components is going to be the dominant frequency. That‘s going to be set by the length of the wire."

... said Don Smith during this video

https://youtu.be/BHr3eDELyHk?si=b_9Mdno961iDsHoH&t=2767

   


If I understand corretly, it means that if I have any NST at 30kHz and I connect it to the coil which has selfresonance for example 220MHz, than that NST will make that coil to oscillate at 220MHz regardless of its own frequency is  30kHz.
Am i right?

[unimmortal]

Great thread! Watching and learning with interest.

[lfarrand]

[quote pid="1069" dateline="1706639090"]
If I understand corretly, it means that if I have any NST at 30kHz and I connect it to the coil which has selfresonance for example 220MHz, than that NST will make that coil to oscillate at 220MHz regardless of its own frequency is  30kHz.
Am i right?
[/quote]

Yes, that's what I understand to be the case.

Here is a scope shot showing the effect of triggering two separate coils (pink & yellow traces) using one input coil (blue trace).

The input coil is being triggered using a 10% duty cycle 24V DC pulse running at 44kHz. The other two coils are then resonating at their natural frequencies of 450kHz.

Guests cannot see images in the messages. Please register at the forum by clicking here to see images.

[unimmortal]

So thinking logically, if you were to ascertain coil resonance and use that figure as a harmonic for a calculated base pulse frequency to enhance the effect? Don Smith called out 35.1khz on his NST. Is that because his coil naturally resonated with the frequency of his NST? Given we have to build to constraints, I'm guessing the NST was used as a base.

Simply put, you can hit a bell with a brick, but a hammer works better.

[lfarrand]

Yeah that sounds plausible, based on all the research I've done about Don Smith (a lot!).

I wouldn't get too caught up on the NST frequency, because he's simply charging a capacitor using the NST and then discharging that into his L1 coil. The capacitor discharge into the coil is the hammer that rings the bell. The coil then resonates at it's natural resonant frequency. The frequency of the NST is irrelevant in my opinion. Some of his devices use an older style iron-core NST that operates at 60Hz.

I don't think the newer electronic switching NSTs have a fixed output frequency.

[andy]

Thank you @lfarrand and @unimmortal for your posts.

I finally constructed my driver (on breadboard) as I posted some time ago. I used P mosfet IRF9540NPBF instead of IRF9130
   

I want to make frequency tests, but first I have run basic test for me.

Input is 12V battery, output (50% duty cycle, positive) is connected into the primary coil which is on top of the secondary coil.
Primary coil is multistranded wire (20 strands?). I had only short piece of this wire, so I connected two pieces together - first half is orange and second half is white .
Secondary coil is enameled copper wire (cca 30 AWG).
Coil is air cored.
I measured secondary coil voltage and frequency.

1) Test #1
   
in first case I connected only to the half of primary - to the orange coil - 12 turns
Secondary coil 1000 turns
Measured voltage 515 V peak to peak on output coil and adjusted frequency to 125 kHz (which gives max voltage)
I think this is correct, because for normal iron core transformer is turns ratio formula N1:N2 = V1:V2
In my case 12turns:1000turns = 12V:1000V so for input 12V it means 12V per turn => output should be 1000V. But because I use only 50% duty cycle positive, it is only 500V (one half)


2) Test #2
   
I connected to the full length of primary - to the orange coil and white coil - 24 turns
Secondary coil 1000 turns
Measured voltage 480 V peak to peak on output coil and slightly adjusted frequency again to 125 kHz.
24turns:1000turns => 0.5V per turn => output 500V, but for me for 50% duty cycle should be 250V. Why is 480V ?

And here are my questions:
1) How is possible, that output is not smaller in #2 example? Should be cca 250V and is 480V
I guess I have error somewhere, but I cannot find it.

2) Is 125 kHz resonant frequency of the secondary coil?
(I measure the frequency by oscilloscope probe positive on one end of output coil, negative on second end of this coil)

[unimmortal]

Great thread Andy. I've wanted to build this for a while now.

I've got 3 NST's and a variac just sitting here waiting a for a project to replicate. Any chance you can put up a full end to end schematic?

Sorry I can't offer any input other than to say I'm using the IRF540N and IRF9540N FETs for my pulse motor and they perform well.

[andy]

I am progressing very slowly. So this was only my exciter connected to step up coils to create my own "NST replacement".

   

Now my goal is to achieve larger output voltage of this circuit. I know, I could use ferrite core, but I want to try aircore first for this purspose.  Maybe bad idea :-)