The preamp can improve NF if it has enough takeover
gain..........................in most cases, this is not necessary. A much
better way to go is to have a 1st mixer that has a low conversion loss or
good NF in the first place. Any amount of RF amplification that you add
ahead of that mixer decreases the dynamic range by the same
amount.........for instance, if you add 10dB of gain ahead of the mixer,
your 1dB compression is degraded by 10dB, your IP3 is degraded by 10dB,
etc,etc,etc.
As an example, you can use a Mini-Circuits TAK-3H as your 1st mixer. The 1dB
compression point on this device is +14dBm and the IP3 of this mixer is
+28dBm. The tradeoff here is that +17dBm (50mW) of LO power is required to
properly "turn on" the diodes so that good dynamic range and conversion loss
figures are met. The SSB conversion loss on this device is right around
4.7dB. This is roughly, but not exactly equivalent to the NF. Add in 2dB of
loss for the input filter and you have a system NF of approximately 7dB.
This will allow you to hear a signal down to around .04uV. You do need to
have a good post mixer amplifier that follows the mixer, and a diplexer to
follow that mixer. Since the diplexer provides some selectivity, you will
not have noise at the image frequency degrading the system NF. It is not
expensive to employ this scheme; I don't know why manufacturers don't
generally do it. Exceptions are some of the solid state Collins equipment,
the Drake R7, TR7, R8. Some of the manufacturers use a quad JFET 1st mixer.
This mixer can have an IP3 of +30 to +40dBm. Examples that use this design
are the Kenwood TS-50, TR570, Racal 6790/GM, AOR-7030, Yaesu FT1000.
The big thing here is setting up the proper gain distribution in the system.
Whenever possible, keep all of the amplification AFTER the 1st mixer, and
you will be able to take advantage of the full dynamic range of the mixer.
I think that the Drake R8 has a NF of 10 to 12dB, as does the AOR-7030. For
receiving below 30MHz, this is fine, because in most areas, 15dB is the
amount of excess noise that you will encounter.
I hope this helps to explain why you don't need a preamp, unless you are in
an extremely quiet location such as northern Wisconsin, up in the ore
deposit regions, or out in the country where the excess noise is a bit
lower. When I was living out in the outskirst of Cedar Rapids, the area was
extremely quiet RF wise, so a super sensitive receiving system was helpful.
Now that I am back in the Chicago area there is so much noise that this type
of receiver isn't as necessary. The background noise from the thousands of
furnace controllers, factory equipment, etc, bring up the background noise
quite a bit.

Pete

"jimg" wrote in message
...
jimg wrote:

i don't thjink i've ever heard anything goofier. you cannot improve on
the s/s at the receiver input without altering the antenna. given a
reasonable impedance match from the receiver to the antenna, that's
it. and it's not hopeless, it's just what it is. if there were no
atmosphereic noise or propogation effects, swl would be kind boring.

as far as your receiver, most receivers noise floors are so far below
the antenna s/n that there is no improvemnt in your dxing capability
there...and why playing with your antenna can make such a diff.

btw, there are ways to improve the performance dramatically, but
wideband amps are not the answer..whereas very narrow band rf amps
are...especially if they're LN2 cooled along with the antenna
pre-amp...but then you don't have enough money for this scheme.

now the kiwa design like other LNA's is very basic and uses a shunt
feebback two transistor approach. other use cascoded mos devices
(called dual gate by non-engineers)..all are very low noise designs
and give moderate gain/ in the case of kiwa, you get 10dB gain and a
reasonable NF (noise figure) (p.s. your cell phone gets better) ...
but 10dB is practically nothing (you got a 10dB rf attenuator on your
rcvr?) and usually will not change an overall sinpo from 2 to 3.

20dB is better, but if noise at the antenna dominates, the snr remains
the same.

having an rf lab with about $2M of equip to play with, i built a nice
little preamp with built in programmable rlc bandpass filtering....and
even then there is little appreciable improvement (on an old r1000) on
"real" sigs buried in noise...

some mornings though, when the background noise is low, the preamp
boosts weak sigs to a more audible level. non-linear adaptive LMS
noise cancellation does the rest...

if you dont have an rf front end worth a damn, it might help...and a
bcb hp filter might too....but in general, you paid enuf money
for a more than adequate rf front end...the rest is the sunspot cycle,
your propogation conditions, your patience and tenacity...


jimg phdee


I don't think I've *ever* connected an antenna to any receiver without
hearing an increase in noise. Is the goal of improving the S/N ratio
really that hopeless? I've heard great things about the Kiwa
preamp....it that all just hot air or does it really live up to its
claims? Are you guys speaking from experience with the Kiwa unit?

Steve

jimg
Oregon
USA