Receiver front-end filter #rcvr


Roelof Bakker
 

Hello Ton,

Varicaps can be used, but show a much lower Q than a normal capacitor.
In the previous century I have measured the Q of the BB112, a 20 - 500 pF varicap.
It varied from 50 (400 pF) to 70 (30 pF).

I build a pre-selector using a Wes Hayward, W7ZOI double Peak Low Pass Filter topology.
The beauty of this topology is that it offers very little loss variation over a 1 to 4
frequency range, typical 2 dB.

Using BB-112 varicaps to tune a 10 - 30 MHz pre-selector, the loss was:

MHz loss dB
=============
28 3 (3)
25 3 (4)
21 3 (5)
18 4 (6)
14 5 (8)
10 8

The figures between brackets apply to using two BB-112's back to back instead of a single
BB-112.

73,
Roelof, pa0rdt


Albert Tatlock's Greatest Hits - Vol 1
 

Good old fashioned moving caps are things of beauty these days.


Jim Allyn - N7JA
 

If you use varactors to tune your filter, be sure that the varactor reverse bias does not approach zero.  If the signals approach the level of the varactor bias voltage, you will have intermod.  Even better, use two varactors back to back.  If you can't get enough capacitance with two of them in series,  then add another varactor pair in parallel.


Glen Leinweber
 

Ton,
Yes, Hans would be less-aghast at tiny hyper-abrupt varicaps than that monster
dual-gang variable capacitor - as Cristi points out, its size compares with the
whole receiver module!
The physical scale of Hans designs amazes me - his board layout is as crowded
as a rush-hour bus on this receiver module.
In general, electronic design trends toward microscopic size where robots
out-perform shaky human hands. Hans obviously uses CAD to design his boards,
using component placement tools effectively. My industrial PCB layout days were
with tape & vellum on a drafting table, when half-watt resistors were standard, and
generally lay flat: double-sided boards with plate-thru holes were exotic.

I'd want to do some careful IP3 measurements when substituting varicaps for that
very linear mechanical variable capacitor. If any large unwanted signals sneak in,
they'd cause spurious mayhem. The Tayloe mixer in these radios is well-renowned
for its spurious rejection. Why taint it with a potentially weak pre-filter?

Be aware that this tune-able filter does make compromises that the QRP-Labs
fixed-tuned band filter modules avoid. As stated, bandwidth at the high-frequency
end of the tuning range is very wide, with inferior stop band attenuation.


Ton - PA0ARR
 

You can also use some varicap diodes on the standard bpf boards.
73 de Ton


cris blak
 

Nice idea Glen.
I think the variable capacitor is half the size of the QCX HI! but it will do.

Thank for sharing.

73 de YO3FFF
Cristi

Pe vineri, 23 iulie 2021, 21:27:23 EEST, Glen Leinweber <leinwebe@...> a scris:


Completed QRP-Labs receiver module, meant for most any band.
Hans supplies a fixed-tuned bandpass filter module between antenna & input for some RF selectivity.
Modules for most any HF band are available from QRP-Labs.
Why not attempt a similar module with a big dual-gang variable capacitor that tunes over all HF?
Hans conveniently does allow direct inputs right on the PCB from such a filter.

The circuit arrangement is exactly the same as Hans' filter: two toroids (tuned by each gang
of the big variable), coupled together by a fixed, small coupling capacitor. A link winding couples
antenna signals into one toroid, while the other toroid couples another similar link winding for output.

I used T50-2 toroid cores, with a 5-turn link winding on each....30 turn main winding #24 AWG.
Those dual-gang variable capacitors are rare today. You want one from an AM radio with equal
capacitance, of the 365 pf variety. I pulled one from an old transistor desktop AM/FM radio. The
20pf capacitance sections normally used for FM are left unused.
The fixed 6.8 pf coupling capacitor (hidden in the photo) couples signals from one gang to the other.

How wide a tuning range can such a filter accommodate? Those two toroids must be matched
carefully, especially at the high-frequency end. That 6.8pf coupling capacitor is a bit small at the
low frequency end (3.8 MHz) where capacitor plates are fully meshed. Attenuation is 6.5dB there,
but it is a very effective filter.
Filter losses were quite small at higher frequency, but filter bandwidth grew to be quite wide.
Frequencies from 3.5 - 23MHz were accommodated.
With a Si5351 VFO module, this makes a pretty nice general-coverage receiver.


Glen Leinweber
 

Completed QRP-Labs receiver module, meant for most any band.
Hans supplies a fixed-tuned bandpass filter module between antenna & input for some RF selectivity.
Modules for most any HF band are available from QRP-Labs.
Why not attempt a similar module with a big dual-gang variable capacitor that tunes over all HF?
Hans conveniently does allow direct inputs right on the PCB from such a filter.

The circuit arrangement is exactly the same as Hans' filter: two toroids (tuned by each gang
of the big variable), coupled together by a fixed, small coupling capacitor. A link winding couples
antenna signals into one toroid, while the other toroid couples another similar link winding for output.

I used T50-2 toroid cores, with a 5-turn link winding on each....30 turn main winding #24 AWG.
Those dual-gang variable capacitors are rare today. You want one from an AM radio with equal
capacitance, of the 365 pf variety. I pulled one from an old transistor desktop AM/FM radio. The
20pf capacitance sections normally used for FM are left unused.
The fixed 6.8 pf coupling capacitor (hidden in the photo) couples signals from one gang to the other.

How wide a tuning range can such a filter accommodate? Those two toroids must be matched
carefully, especially at the high-frequency end. That 6.8pf coupling capacitor is a bit small at the
low frequency end (3.8 MHz) where capacitor plates are fully meshed. Attenuation is 6.5dB there,
but it is a very effective filter.
Filter losses were quite small at higher frequency, but filter bandwidth grew to be quite wide.
Frequencies from 3.5 - 23MHz were accommodated.
With a Si5351 VFO module, this makes a pretty nice general-coverage receiver.