Tesla coil tuner design / General Science and Electronics / Forums

Forums

4hv.org :: Forums :: General Science and Electronics

« Previous topic | Next topic »

Tesla coil tuner design

1 2 next

Move Thread

LAN_403

Njubster

Tue Nov 25 2014, 02:37PM

Registered Member #6857 Joined: Sat Sept 22 2012, 08:25AM
Location: Srbija, Novi Sad
Posts: 40

I came upon THIS TC tuner while googling for a way to find Fres of my TC, but the given design didn't quite suite my needs because most of the coils I play with have a much higher Fres (mostly because of the cost of the driving circuitry). Also, I don't have the proper measuring equipment for calibrating the scale for this analog circuitry.

So, I had a thought, but I am sure that it has many flaws, cause well, I'm just a hobbyist. Anyway, could THIS thing work?

The idea is, telling the possible frequency range (around the modeled Fres) to the microprocessor, telling him the resolution (or the step in increasing the frequency, bigger resolution = more time to compute), and then the microprocessor would feed the frequency to the transistor base, and the frequency would increase over time. Once it reaches the next biggest amplitude in the given range, the microprocessor would remember it's value and the frequency it occurred at. That how, in the end, it would be able to output the last remembered frequency to LCD screen/7 segment display. Or, a more advanced solution would be to send each measured point to the computer, where it could show the whole bode plot, with all the harmonics peaks and Fres peak.

First thing I'm not sure about is whether I understood correctly how the whole setup is used in the first link, areTHESE the correct equivalent circuits?

Next thing, would my arrangement with 2u2 cap and 1k resistor cause any blockage to some frequencies? If observed as a high pass filter, it should be able to let all the frequencies above ~72Hz, but I'm not sure how it will react when additional LC circuitry is added.

If I was to set my limits from 150kHz - 4MHz, would 2u2 cap be functional for the whole range, or would it stop being useful after the LC impedance has exceeded some value?

Thanks in advance :)

Shrad

Tue Nov 25 2014, 02:59PM

Registered Member #3215 Joined: Sun Sept 19 2010, 08:42PM
Location:
Posts: 780

it this can help, here are my suggestions

I'd use a DDS module for this approach in parallel mode so you just need to move your watchdog or standard counter register value to a port in order to be able to sweep frequencies

with this setup I think you'll want to measure the drop of voltage at the coil to ground at a given frequency so that it's easier to measure

also, you'll have to use a really fast diode and a small cap as integrator as well as a way to discharge the cap between each frequency step

I think there can be other approaches like with a dipmeter or something alike

Njubster

Tue Nov 25 2014, 06:02PM

Registered Member #6857 Joined: Sat Sept 22 2012, 08:25AM
Location: Srbija, Novi Sad
Posts: 40

Shrad wrote ...
I'd use a DDS module for this approach in parallel mode so you just need to move your watchdog or standard counter register value to a port in order to be able to sweep frequencies

Woah woah woah, I'm not that advanced in MCUs :) Though, I think I'd be able to achieve what I described above on some primitive way without using fancy functions I know nothing about :)

Shrad wrote ...
with this setup I think you'll want to measure the drop of voltage at the coil to ground at a given frequency so that it's easier to measure

You mean like THIS ? EDIT: I forgot to drive the peak detector / rectifier part.

Shrad wrote ...
also, you'll have to use a really fast diode and a small cap as integrator as well as a way to discharge the cap between each frequency step

Well, bat46 or 1n4148 should do the job, right? I also have some germanium ones, AA113, are those suitable too?
And I thought 10n was small enough?

Sulaiman

Tue Nov 25 2014, 07:00PM

Registered Member #162 Joined: Mon Feb 13 2006, 10:25AM
Location: United Kingdom
Posts: 3140

If you use any non-sinusoidal wave drive then you will have harmonics,
for squarewave at 3x, 5x, 7x ... the main frequency.
So if your TC resonates at say 300 kHz
there will be responses when the main frequency is at 100 kHz, 60 kHz, 43.857 kHz ...... etc.

Although manageable it is a potential source of significant error.

AND

once you have your tuner, how many times will you use it?
================================================= =================

A signal generator from eBay (e.g.

)
would be a good buy;
. general purpose signal generator to 500 kHz
. 'strong' output of +/- 10 Vpeak no-load = +/- 5 Vpeak into 50 Ohms
. in a case (less likely to get broken / cause problems)

All you would need to add is a diode and capacitor (e.g. 1N4148 & 10nF)and voltmeter/dmm
and you will easily and quickly be able to find TC resonances

secondary by itself = one resonant frequency,
secondary with primary//capacitor = 2 resonances

and you will have a good general purpose tool.

or for a little bit more;

which can also be a sweep frequency generator, counter, frequency meter .....up to 2 MHz
with a TTL output for sync. or general purpose clock/oscillator etc.

You COULD build something better, but it would cost more (case, connectors .......)
and would end up as a major project in itself.

If you've been good this year maybe Santa will get you one?

P.S. both of the above require a power supply, e.g. battery or plug-pack (wall wart)

Njubster

Tue Nov 25 2014, 07:18PM

Registered Member #6857 Joined: Sat Sept 22 2012, 08:25AM
Location: Srbija, Novi Sad
Posts: 40

EDIT: this post is a mistake, hope that someone will delete it :$

Njubster

Tue Nov 25 2014, 07:19PM

Registered Member #6857 Joined: Sat Sept 22 2012, 08:25AM
Location: Srbija, Novi Sad
Posts: 40

Sulaiman wrote ...
once you have your tuner, how many times will you use it?

From where I'm standing, I see 3 untuned coils :) And I plan on building at least another big one some day when I become a rich engineer (ha ha).

I have several 3 digit 7-segment displays, several bcd decoders, and several atmega chips I have no use for, so I thought I had enough to make this without buying anything.

Unfortunately, PayPal is not working in my country, so I cannot buy stuff from ebay. I could probably get something from AliExpress, but I'd rather build it :) I'm in a process of making a frequency counter, but I'm kinda stuck, cause once I get above ~100kHz it goes loco :/ I left that for some better times.

I read something about that DDS thingie, that seems interesting, I'll try to understand what that is.

Sulaiman wrote ...
If you've been good this year maybe Santa will get you one?

Nope, I was a naughty boy :(

Shrad

Tue Nov 25 2014, 08:10PM

Registered Member #3215 Joined: Sun Sept 19 2010, 08:42PM
Location:
Posts: 780

if you need a multipurpose tool which would enable you to measure components, so that you can calculate resonant frequency of a system, try one of these :

I have bought one like this and it is really useful

maybe that's the best thing to do before, so that you are able to experiment directly and are not discouraged by other complicated projects... one more chance to get some arcs more easily

Sulaiman

Tue Nov 25 2014, 09:02PM

Registered Member #162 Joined: Mon Feb 13 2006, 10:25AM
Location: United Kingdom
Posts: 3140

I bought one of the generic cheap R/L/C esr pnp/pnp/mosfet/diode/triac... testers for about GBP5
... very useful for testing/repair BUT not accurate on L and C ranges, up to 10% error

For 'accurate' measurement of L and C, I use my trusty old Farnell LF1 sig-gen (used, eBay)
(together with a frequency counter and 'scope)
and one of my 'calibrated' capacitors or inductors to resonate with the component to be measured

I had a great big valve-powered precision L/C/R bridge,
incredibly accurate, phenomenally heavy and large,
took up too much space so had to go.

GOOD L/C/R meters are still not cheap. (If anyone knows otherwise please PM me)

Njubster

Tue Nov 25 2014, 11:08PM

Registered Member #6857 Joined: Sat Sept 22 2012, 08:25AM
Location: Srbija, Novi Sad
Posts: 40

I'm in a process of building my own L/C/F meter (

), but I need to resolve few more program issues, because I was learning to program arduino with this project :) ESR is the next one.

Maybe the best course would be to ditch the above project, and just build a nice DDS function generator and frequency counter, because these two are multipurpose, right?

What does "valve powered" mean? :D Btw if it's some vintage piece of equipment, and you threw it away, then you are a horrible person and should be incarcerated :P

And the popular method that uses high precision capacitor for calibration is pretty reliable, and it all comes down to that capacitor :) It's a beautiful design.

johnf

Wed Nov 26 2014, 05:46AM

Registered Member #230 Joined: Tue Feb 21 2006, 08:01PM
Location: Gracefield lower Hutt
Posts: 284

Njubster
"Valve" what the rest of the world except Canada and USA call vacuum tubes --so in your case "tubes"

comes from "voltage controlled thermionic current valve"

just english language idiosyncracies

1 2 next

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob