[time-nuts] Firmware and antenna for Stanford Research FS700 Loran C frequency standard
Dr. David Kirkby (Kirkby Microwave Ltd)
drkirkby at kirkbymicrowave.co.uk
Fri Jul 17 08:31:04 EDT 2015
On 16 July 2015 at 23:23, Bob Camp <kb8tq at n1k.org> wrote:
> Hi
>
> Quick and simple:
>
> 1) Signal power is proportional to the area of the loop. Bigger is better.
> 2) Inductance is proportional to the turns squared. Turns do not directly
> affect signal to noise.
> 3) Inductance may be resonated with a capacitor. This gives a bandpass
> function.
> 4) The coil shapes are very common. The many inductance calculators on the
> web will give you an inductance estimate.
> 5) If the inductance is resonated, the system Q (and thus bandwidth) is a
> function of the coil losses and the amplifier’s input impedance.
> 6) More turns gives a power match into a higher impedance ( more voltage).
> 7) *Practical* matching of the amplifier to the antenna will give you an
> reasonable target number of turns.
>
> Bob
>
It's interesting that
http://www.vlf.it/feletti2/idealloop.html
says that sensitivity is set by the mass of copper used. To quote
"A single turn square loop, 1m side, made with 1kg copper has the same
sensitivity of a 1000 turns square loop made with 1kg copper and same
dimensions. In this context, the sensitivity limit is represented only by
loop thermal noise:
noise floor (nV/sqrt(Hz)) = 4 sqrt(R in kOhm)"
It is not immediately obvious where that equation comes from, but
re-arranging the equation for thermal noise power
P=k T B
(P in watts, k= Boltzmann contant, B is bandwidth in Hz)
and assuming a temperature T of 300 Kelvin, k = 1.38 x 10^-23 J/K, one
finds the constant is 4.06, so the 4 in that equation is fairly accurate at
300 Kelvin.
I'd much rather wind a loop with a few turns than a few hundred turns! But
obviously the voltage rises with the number of turns, so requires less
gain.
Dave
More information about the time-nuts
mailing list