[time-nuts] locking oscillators - an increase in power and/or stability ?
magnus at rubidium.dyndns.org
Wed Oct 8 20:32:44 EDT 2014
On 10/08/2014 09:07 PM, cdelect at juno.com wrote:
> I came across this phenomena when transmitting with two 5KW transmitters
> via separate parabolic antennas to a satellite.
> If the phase of the TXs was correct the received signal at the satellite
> was 6db hotter!
> I thought at the time that it was due to the power adding in the voltage
> For instance if you take a 1volt signal into 1 ohm you get 1 amp and 1
> watt. but if you take two 1volt signals and add them to produce a 2 volt
> signal then you get 2 amps and 4 watts.
> Not sure if my logic is correct but the phenomena is real!
If you have two non-coherent sources, their power will be added and
hence you get a +3.01 dB gain, which is to be expected from 10*log(2).
If you have two coherent sources, their voltage (or current) will add,
but with the factor of phase-angle between the sources, for same
voltage, the result will be 2*cos(phi) of one of the sources, giving the
maximum of 20*log(2)=6.02 dB. This is however just 3 dB more than
Notice now that for the sum to be 6 dB lower in uncorrected phase of two
coherent sources, the phase angle will be 60 degrees. Remember that the
two antenna may see a difference in distance, contributing to this phase
error, so the satellite's view will be the correct vector sum.
More information about the time-nuts