Original poster: Terry Fritz <vardin@xxxxxxxxxxxxxxxxxxxxxxx> Hi, At 12:44 AM 12/29/2005, you wrote:
Terry said, that the source end is terminated, so "That damps the system (Q=0.7071..) so it does not ring or distort at the frequencies we are working in.". btw Terry - why Q=0.707, why not Q=1? but if Terry is right, then you can kill Blumlien effect just by connecting 50 ohm in series with output of unballasted pig?Terry's antenna has a 50 ohm resister in series with the coax at the antenna end. This resister connects to the capacitor divider. The termination is only effective above frequencies that the capacitor divider thevenin capacitance interacts with the 50 ohms (really 50 ohm resister is series with 50 ohm coax). If you thevenize the capacitor divider (20 nf in parallel with say 1 pf) you have a pole associated with equivalent RC combo with a frequency around 80KHz. Above this frequency, the coax is effectively terminated. Below this frequency, the termination tends toward open circuit. So in the frequency range of DC to 80KHz (lets say) the termination at both ends is not good. Granted the rise time of a particular frequency in this band is slower than the round trip propagation time. The real issue, I believe, is how long does it take for the multiple reflections from one transition to die out. Given multiple transitions, there could be many reflections continuing from previous transitions and measurement errors could occur. How significant this error is I dont know. Spice shows an effect that could be attributed to this effect that could affect the transient response of the measurement.Gerry R
Page one here shows the spice model:
http://hot-streamer.com/temp/PlaneWave.pdf
Cheers,
Terry