SSTC secondary value verification / Tesla Coils / Forums

Forums

4hv.org :: Forums :: Tesla Coils

« Previous topic | Next topic »

SSTC secondary value verification

Move Thread

LAN_403

Ken M.

Thu Sept 30 2010, 06:56AM

Registered Member #618 Joined: Sat Mar 31 2007, 04:15AM
Location: Us-Great Lakes
Posts: 628

Its been a while since I've posted on here, and even longer since I've actually posted anything regarding Tesla Coils, lately I've been getting that feeling that I should start working on some again, but time and other things keeps preventing me from actually doing something. I'll be having some time in the near future to start working on some of these projects again and maybe try some new ideas or at least a spin on some old ideas.

The first Idea is one that most of us encounter and that's figuring out the Fres of our secondaries as well as the Fres of the entire circuit not just the secondary, I realize that this value is figured out more or less through experimentation, even though a few have tried to make circuits that aid in this or give a closer approximation. I was thinking however that one could possibly use a micro-controller to output a large frequency range from 1Hz (not really going to be of any use but, rather just for the principle of the matter) up to possibly 1MHz at 50% duty-cycle, and measure the frequency at the break out point as well as measuring the voltage at both ends, and maybe toss in a little math function so the use can enter in basic values such as coil dia. and height, along with turn count and wire gauge then display the calculated Fres, as well as a screen to show progress and the values being entered and such, although this is more of a general science / electronics thing, I just thought I'd talk about the idea since it pertains to tesla coils, anyways.

Regarding the topic, I wanted to see how well a certain micro-controller I have would work to out put a variety of frequency's at a fairly accurate rate into a secondary I had lying around and to use the o-scope method of finding the approximate series Fres, and comparing it to a calculated Fres, although I can't seem to remember nor relocate the formulas for figuring this up, and was wondering if someone could either verify these values or direct me towards the formulas or links to the formulas that I am missing.

Using a MCU as a square wave generator at 5V P-P through a 100ohm resistor ( closest one I had around) attached to the base of a 4" diameter 20" tall secondary using 30AWG with no topload, I was getting 6.064V at the base with a Frequency of 39.408KHz, however if if I attach the second channel of my scope to the breakout point of my secondary and adjust the frequency I can get up to 29.25V with a frequency of 55.944KHz and the voltage at my base is 5.536.

I would presume these to be harmonics and not the actual Fres since the last time I calculated the Fres of this particular coil it was 325KHz.

Dr. Dark Current

Thu Sept 30 2010, 06:32PM

Registered Member #152 Joined: Sun Feb 12 2006, 03:36PM
Location: Czech Rep.
Posts: 3384

If you attach a probe to the secondary output, you are adding a large capacitance (relative to the TC capacitances) and the reading will be off a lot.
Better method is to drive the base through a ~1k resistor, sweep the frequency and find a frequency which produces the biggest dip in secondary base voltage (the impedance at resonance will be the lowest, but you will get dips also at harmonics/subharmonics of the resonant frequency).

Sulaiman

Thu Sept 30 2010, 08:07PM

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

That's a good method,
since the base impedance at resonance is resistive you can also determine the input resistance at resonance by simple calculation.
Knowing the base impedance is just 'nice' to know as it makes you realise that dc resistance and 'skin' effect are less than 'proximity-effect', I usually get about 2x to 3x ac resistance at resonance compared to dc resistance, much more than 'skin effect' alone, people do not talk about proximity effect much because it's so difficult to calculate, unlike 'skin effect' which is talked about all the time.
Also, once corona and/or sparks start the base impedance becomes much higher.

Ken M.

Thu Sept 30 2010, 10:09PM

Registered Member #618 Joined: Sat Mar 31 2007, 04:15AM
Location: Us-Great Lakes
Posts: 628

I realize attaching the scope probe to the breakout point would affect the Fres, but I was more or less just doing it see how much of a "load" it would place on the signal being applied to the base, not to mention seeing how far off it would throw the measurements.

I managed to find the formulas I was looking for and actually got a Fres of 253.531KHz without a topload and 154.253KHz with a topload, dividing the calculated Fres by the Frequencies I measured, I get just the bottom reading is just about 1/4 of the Calculated Fres with a topload, even though I did not have a topload on the coil while testing, and the measurement from the breakout point comes to just under 1/3 of Calc Fres with a topload, and about 1/5 of the calc Fres without a topload.

At one point, I got a calculated Fres of around 219.105KHz for a coil without a topload, if divided by 55944 comes to just under 1/4 of the Fres, so could that mean that the probe didn't not affect the Fres as much, or that the probe had a similar capacitance to a 12x4 toroid?

Not to mention the point of least impedance at the base with no top load gave 6V and was 39.408KHz or 1/4 wave of 157.632Khz and the max impedeance around 230Khz with 4.8V, but with 55.944KHz was roughly half way with 5.56V at the base and max voltage at the top of 29V and is equal to 1/4 wave of 223.776KHz.

Sulaiman

Sat Oct 02 2010, 03:39AM

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

A TC secondary has two basic resonance modes,
one is an LC resonance,
one is a 1/4 wavelength resonance
due to the total length of wire (and a slow-wave correction factor)

If you are aiming for the highest voltage without breakout
like N.Tesla wanted for his 'world power' madcap scheme
then the two resonances should coincide.

For arcs'n'sparks then the 1/4-wave mode is unimportant.
Medhurst & Wheeler calculations will put you in the right ballpark.

See if you can identify the two resonances.

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