Topband: The WD8DSB mini-flag antenna

[Richard (Rick) Karlquist]

On 2/24/2021 12:26 PM, n4is at comcast.net wrote:
> Hi Rich
> 
> " The problem here is that any loop antenna inherently has a noise
> temperature of 300K, because it is lossy, and because it "sees"
> the earth"
> 
> Would you elaborate on that? Small loop 1/10 of wave length works very
> different from a loop or large loop. I experimented with a large number of

> Local noise is the most misunderstood figure. On top-band during a winter
> Sunday morning the noise can be low as 70K. On topband the only source of
> nom mam made  noise atmospheric noise. No atmospheric activity means no
> noise. Yes it is that low.  I am using horizontal phased loops for almost 15

1.  I am specifically talking about "small" loops.

2.  In small loops, the ohmic dissipation of the conductor
vastly overwhelms the tiny radiation resistance.

3.  This is unlike a vertical antenna or a dipole, which can
be considered non-dissipative.

4.  Therefore, if the conductor is at a physical temperature
of 300K, its Johnson noise will correspond to that temperature.
(-173 dBm/Hz).  IOW, the antenna's noise temperature will never
be less than 300K.

5.  If we were to put a loop in "free space", meaning outer space,
where there is no noise besides the 3K cosmic background noise,
it would still have a noise temperature of 300K, assuming that
the physical temperature were somehow maintained at 300K.

6.  NF is source impedance dependent.  Measuring the NF with
a 50 ohm source doesn't tell you anything about the effective
NF when terminated with the loop impedance (which is no where
near 50 ohms resistive.)

> still in use here, using 6 BF981 and  large Q input filter. It measured .7
> db NF including the input tuning filter loss.  It does make a difference on
> signals at noise level. I built one for NX4D and Doug still use it.  Making

7. It is extremely easy to get a 50 ohm NF of 0.5 dB with a BF981
at 88 MHz as specified by its data sheet.  All you have to do
is transform the 50 ohm source so that it loads the FET with
about 1,000 ohms.  I did this 40 years ago.  Unfortunately,
at 1.8 MHz, the flicker noise of the BF981 dominates, so you can't
get the low NF down there.

8. I did this same experiment with the multiple paralleled BF981's over 
15 years ago and was limited by flicker noise, which is
unspecified, YMMV.  Maybe you had better devices than I did.

> pair of phase loops. A/B tests with NORTON preamps show not possible to copy
> the weak signals I did copy with my 6xBF981.

Norton amps are not noted for low NF, so this is a straw man.
Maybe they are picking up BCB QRM (see below).

> Implementation is a different ball game. The .7 db NF preamp needs 3 level
> of shield including a magnetic shield with steel, just aluminum is not
> enough, and no open shield at all, a 1 mm gap will ruin your system. You can
> find information about my preamp on google or WWROF archives.

Considering that the input of the preamp is deliberately connected to 
the antenna, it is hard to imagine what signal the shield is
shielding the amplifier from.  Large BCB signals out of band
coming up the feedline?  Is the shield still needed for QTHs that have
no nearby BCB stations?

>
> 73's
> N4IS
> JC
> 

73
Rick N6RK