Tuning coils and precision / Tesla Coils / Forums

Forums

4hv.org :: Forums :: Tesla Coils

« Previous topic | Next topic »

Tuning coils and precision

Move Thread

LAN_403

Hazmatt_(The Underdog)

Tue Jan 29 2008, 02:56AM

Registered Member #135 Joined: Sat Feb 11 2006, 12:06AM
Location: Anywhere is fine
Posts: 1735

This thread is for those who wish to get more experience in determinations of their coil performance and the characteristics of the coil at working frequency.

I wanted to know Z, Rac, Rdc, Q, as these characteristics are difficult to measure and are important factors in system analysis. Knowing these factors I will compare different geometries and performances of coils.

In my case, LED tuning methods were not able to provide the sharp tuning indication required to find the resonant frequency of my coils. Current measurement using my early probe designs did not provide accurate tuning information either. The early probes were subject to RF bleeding because of their single diode construction. This led me toward professionally made Boonton meters which meter RMS voltage from 10KHz to 1.2GHz. The meters are an invaluable tool, allbiet very expensive. I have also made a stern effort to duplicate the good tracking performance of the Boonton meters for you so that you too can make qualitative determinations of your coils.

The concept is simple, A voltage divider is setup to match the line impedance. This way a direct reading of the reactance can be known, and impedance calculated. Ideally the source would deliver 3V and the divider node 1.5V, meaning the resistance and reactance are equal, then the impedance is simply sqrt(2)*resistance (or reactance). But this is not always the case since there is a 50 ohm line resistance internal to the source, some of the voltage is dropped across this load due to the low Rac of the line. So insted we make careful measurement of the voltage drops and resistance to determine the load current delivered to the coil at resonance.

The first incarnation of this admittance bridge is the older project box with the 0-100% verniers. This was to be the all inclusive bridge, with RF voltmeters and resistance for a portable instrument. However, I found that the verniers were only spaning 60% of the pot travel, and was unacceptable.

For now the admittance bridge is in seperate components, the R box and the external RF voltmeters, and generator. In the future this instrument will be a full package with hopefully a very stable wide range generator.

The R box for the bridge is nothing special. It is a 100R 10 turn pot, a 1KR 10 turn pot, and a 10KR resistor on the switch. There may be no need for the 100R pot since one can only have so much resolution and be precisce about it. I may have to build a partial decade and counting dial vernier for the best possible precision.

The tuning procedure is quite simple. Initially we set the generator to a known voltage, in this case 3V for the Boonton meter on the input side of the R box, then connect the coil to the out side of the R box and use a 10KR line resistance. Fo is then found when Vout is nearly 0V.

Now that we have Fo resolved to a hundred Hertz or so, we then turn off the 10KR switch so that just the verniers are in-circuit. At this point we dial in the line resistance to match the load Rac. A good match will resolve to 1.5V divided from 3V. At this point you would be able to read XL directly off the vernier because they are equal. Z is simply 1.414*XL as it is the resultant vector of R and XL.

We now have a working measure of XL, R, Z and Fo at resonance. We can then determine our Q using Rac, and my figure of that is 257 ohms. Now of course, all of these figures change when top-loaded, and by determining how the system loads we can determine what the optimal topload is.

Early experiments with toploads of various capacitence were interesting because the topload would reduce the coil impedance to a degree with certian sizes, and then later as the toploads increased the impedance would increase. I have to look into how and where phase changes are concerning loading and for that I will build a noise bridge and use a spectrum analyzer.

This method can be used for many other applications where the data is required at working frequencies. It is a multipurpose approach and can be used in almost any transmission line application.

Later I hope to have a completed bench Admittance bridge for you that includes the RF voltmeters like my initial design. That way you will have a useful instrument for your bench that will help you gain insight into the system you are dealing with and get some quantitative results that aid in design, analysis, or simulation.

Coil under test.

Outside of R box with verniers.

Innards of the R box, nothing special here.

Calculations of the coil under test and impedance resultant of 349 ohms.

A view of the first concept bridge that was almost completed, but had vernier issues.
Displaying RF Genie, R box, and the Coil Tuner

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