Original poster: Bert Hickman <bert.hickman@xxxxxxxxxx> Hi Finn, Malcolm Watts and I struggled with this (off list) over 10 years ago... :^)My MicroSim simulations showed similar results - no evidence of frequency splitting when the system was examined in the time domain, but there was obvious splitting when a Fourier analysis of the system was performed (in simulations). And, similar results could be seen if one used a spectrum analyzer on the system. After considerable head scratching by both of us, Malcolm finally figured it out... :^)
It turns out that Antonio's explanation is indeed correct. Although, in the time domain, the frequency of the primary and secondary stay the same, the primary and secondary voltage or current amplitudes are rapidly changing as energy transfers between the coupled LC circuits. In addition, system losses make the overall envelope decay exponentially. The higher the coupling, the quicker energy transfers between primary and secondary (and the faster voltages or currents decline/grow between primary and secondary). It turns out that the coupled system can be described using amplitude modulated RF signals, where the upper and lower sidebands (USB and LSB) are a function of coupling. The greater the couping, the greater the difference between USB and LSB from the fundamental (or "carrier" frequency). Since Tesla coils are usually overcoupled, the effective "modulation index" is 100%.
BTW, frequency "splitting" only occurs while energy is actively being transferred between primary and secondary systems (i.e, when the spark gap is actually conducting in a SGTC). During energy transfer, BOTH the USB and LSB are simultaneously present. In fact, a heavily coupled system may operate with virtually all of the system energy residing in the upper and lower sidebands and none in the fundamental frequency (called "suppressed carrier" mode).
Only when the gap opens (and energy transfer stops) does the secondary ring at the center (fundamental) frequency. The confusing thing is that, if you scope the system, you'll only see the [heavily amplitude modulated] fundamental frequency. It can certainly be confusing... :^)
Some more info can be found here: http://en.wikipedia.org/wiki/Amplitude_modulation Hope this helped a bit... Bert -- *************************************************** We specialize in UNIQUE items! Coins shrunk by huge magnetic fields, Lichtenberg Figures (our "Captured Lightning") and out of print technical Books. Visit Stoneridge Engineering at http://www.teslamania.com *************************************************** Tesla list wrote:
Original poster: Finn Hammer <f-h@xxxx> Gerry, I tried with k=0.1327433628 which should give a split of 94/100/107 And I got this table:(It says pri-current, but it is secondary current really) Numbers from top and down= peak, zero, trough, zero, peak etc.etc. http://home5.inet.tele.dk/f-hammer/k0.1327433628notchtable.jpegLooks like it confirms with the split, and also the "sawtooth" waveshape although I still don`t understand the change in frequency around the notches.I use this eq for split: upper pole-> Fres/sqrt(1+K) lower pole-> Fres/sqrt(1-K) Help! I need to understand this so I can move on.... Cheers, Finn Hammer Tesla list skrev:Original poster: "Gerry Reynolds" <gerryreynolds@xxxxxxxxxxxxx> Hi Finn,With a coupling of 0.0688.., it seems like the frequency splitting would be small. Have you calculated what the split would be?? (I dont have the equation)Gerry R.Original poster: Finn Hammer <f-h@xxxx> All,In an attempt to find an explanation of the subject that I can understand, I modeled a tesla coil in Orcad. It has Cpri=33nF, Lpri=77uH Csec=33pF Lsec=77mH and K12=0.0688836105 So that it should have it`s first notch after 7,5 primary cycles.