I need an LF to 30 Khz preselector (passive) for a homebrewed receiver
I'm building. I have most of the details worked out, but I am stuck on
the preselector design.

I want one band to cover 50 Khz to 500 Khz (or 1 Mhz). The second band
would be 1 Mhz to 30 Mhz. 50 ohm input with moderate Q (don't want to
have to retune it every 100 Khz. The LF unit can be very high Q,
peaking it every few Khz is ok as most of the signals in that range
are narrowband signals anyway. Input impedance is standard 50 ohms.

I'd REALLY like to have one assembly do the entire job, but it appears
to be a tall order as that's quite a wide range to cover with one
tuned circuit.

Google returns to many hits even if I narrow down the search
parameters, I tried google.

It appears I need to use some sort of permeability tuned variable
inductor in conjunction with a variable cap. Have no idea how to go
about this or where to buy permeability tuned inductors.

Any suggestions?

T

On Fri, 09 Sep 2005 22:01:27 -0400, TRABEM wrote:

I need an LF to 30 Khz preselector (passive) for a homebrewed receiver
I'm building. I have most of the details worked out, but I am stuck on
the preselector design.

I want one band to cover 50 Khz to 500 Khz (or 1 Mhz).

That would be 1:10 (or 1:20) frequency range.

The second band would be 1 Mhz to 30 Mhz.

That would be 1:30 frequency range and would require a 1:900 reactance
range, if only one element (inductance or capacitance) is tuned, which
would be completely unrealistic. The alternative would be to tune both
the inductance _and_ capacitance in the 1:30 range. However, with a
300 pF variable capacitor, the minimum capacitor would be 10 pF
_including_ all stray capacitances, not very realistic. I have no idea
what a realistic tuning range would be for a permeability tuned
inductor. I have only seen those in FM receivers (87..108 MHz), which
only about 20 % tuning range.

50 ohm input with moderate Q (don't want to
have to retune it every 100 Khz. The LF unit can be very high Q,
peaking it every few Khz is ok as most of the signals in that range
are narrowband signals anyway. Input impedance is standard 50 ohms.

Maintaining the constant impedance if both inductance and capacitance
is tuned would be a challenge.

While maintaining the power match makes some sense in HF receivers
(and also if you are using some small magnetic loops on LF/MF), but
otherwise, I would question the idea to maintain the 50 ohm resistive
input impedance on LF (and possibly also MF).

On LF, any practical random wire antenna would have a high capacitive
reactance. On LF I have simply used a parallel tuned circuit and
connected the antenna to the hot end of the circuit with a small
(10-100 pF) capacitor. Of course, the tuning scale will not be
accurate if you change the antenna, since the antenna stray
capacitances will detune the resonant circuit. Use a secondary winding
or a source/emitter follower to get the impedance down to feed a 50
ohm receiver input or feed a coaxial cable (if the preselector is just
under the antenna).

If you intend to use magnetic loop antennas on LF/MF, which have a low
efficiency, I would suggest using a (remotely)tunable loop as the
preselector, with possibly a preamplifier at the antenna.

I would strongly suggest splitting the HF bands and switching in
various inductors for various bands and use an ordinary variable
capacitor.

The situation is quite problematic on LF, since you would need a
variable capacitor with a tuning range int several nF to keep the
resonant circuit impedances at manageable levels (otherwise, it would
be quite "touchy" due to stray capacitances). With practical variable
capacitors, you would have to add parallel padding capacitors, with
would further reduce the tuning range.

Permeability tuning might be an option. You could try making a coil
and slide a ferrite bar (from a medium wave antenna) into it, however,
I have no experience with this.

Paul OH3LWR

I realise that you are after a passive solution but I wonder if an
op-amp or two in a Q multiplier configuration *for the first band) might
be a better idea? You could then tune the feedback/gain of the device
with a R/C combination rather than large values of L. You could of
course set the gain to 1 if you wanted.

Just an idle thought.

Cheers Bob W5/VK2YQA

TRABEM wrote:

I need an LF to 30 Khz preselector (passive) for a homebrewed receiver
I'm building. I have most of the details worked out, but I am stuck on
the preselector design.

Thanks Paul,

My old Drake C line had a multiple ganged permeability tuned inductor
mechanically coupled to a variable cap.

One of the assemblies tuned the front end, one peaked the driver
stage, etc.

It covered from 1 Mhz to 30 Mhz with a half turn on the front panel
preselector control. Most likely the variable C and L tuned together
maintained a desirable Q across all the bands.

I was hoping for something similar, but have no idea where to get
permeability tuned inductors today.

Regards,

T


On Sat, 10 Sep 2005 12:16:44 +0300, Paul Keinanen
wrote:

On Fri, 09 Sep 2005 22:01:27 -0400, TRABEM wrote:

I need an LF to 30 Khz preselector (passive) for a homebrewed receiver
I'm building. I have most of the details worked out, but I am stuck on
the preselector design.

I want one band to cover 50 Khz to 500 Khz (or 1 Mhz).

That would be 1:10 (or 1:20) frequency range.

The second band would be 1 Mhz to 30 Mhz.

That would be 1:30 frequency range and would require a 1:900 reactance
range, if only one element (inductance or capacitance) is tuned, which
would be completely unrealistic. The alternative would be to tune both
the inductance _and_ capacitance in the 1:30 range. However, with a
300 pF variable capacitor, the minimum capacitor would be 10 pF
_including_ all stray capacitances, not very realistic. I have no idea
what a realistic tuning range would be for a permeability tuned
inductor. I have only seen those in FM receivers (87..108 MHz), which
only about 20 % tuning range.

50 ohm input with moderate Q (don't want to
have to retune it every 100 Khz. The LF unit can be very high Q,
peaking it every few Khz is ok as most of the signals in that range
are narrowband signals anyway. Input impedance is standard 50 ohms.

Maintaining the constant impedance if both inductance and capacitance
is tuned would be a challenge.

While maintaining the power match makes some sense in HF receivers
(and also if you are using some small magnetic loops on LF/MF), but
otherwise, I would question the idea to maintain the 50 ohm resistive
input impedance on LF (and possibly also MF).

On LF, any practical random wire antenna would have a high capacitive
reactance. On LF I have simply used a parallel tuned circuit and
connected the antenna to the hot end of the circuit with a small
(10-100 pF) capacitor. Of course, the tuning scale will not be
accurate if you change the antenna, since the antenna stray
capacitances will detune the resonant circuit. Use a secondary winding
or a source/emitter follower to get the impedance down to feed a 50
ohm receiver input or feed a coaxial cable (if the preselector is just
under the antenna).

If you intend to use magnetic loop antennas on LF/MF, which have a low
efficiency, I would suggest using a (remotely)tunable loop as the
preselector, with possibly a preamplifier at the antenna.

I would strongly suggest splitting the HF bands and switching in
various inductors for various bands and use an ordinary variable
capacitor.

The situation is quite problematic on LF, since you would need a
variable capacitor with a tuning range int several nF to keep the
resonant circuit impedances at manageable levels (otherwise, it would
be quite "touchy" due to stray capacitances). With practical variable
capacitors, you would have to add parallel padding capacitors, with
would further reduce the tuning range.

Permeability tuning might be an option. You could try making a coil
and slide a ferrite bar (from a medium wave antenna) into it, however,
I have no experience with this.

Paul OH3LWR

TRABEM wrote:

Thanks Paul,

My old Drake C line had a multiple ganged permeability tuned inductor
mechanically coupled to a variable cap.

One of the assemblies tuned the front end, one peaked the driver
stage, etc.

It covered from 1 Mhz to 30 Mhz with a half turn on the front panel
preselector control. Most likely the variable C and L tuned together
maintained a desirable Q across all the bands.

I was hoping for something similar, but have no idea where to get
permeability tuned inductors today.

Regards,

T


Fair Radio used to have some in their catalog, and they were used in
all of the old Delco (and some other brands) car radios of the '60s and
'70s.


--
?

Michael A. Terrell
Central Florida

Michael A. Terrell wrote:

Fair Radio used to have some in their catalog, and they were used in
all of the old Delco (and some other brands) car radios of the '60s and
'70s.

Just about all the radios I've seen with mechanical pushbuttons for
station presets used permeability tuned inductors -- the pushbuttons
each just positioned a ferrite core to a preset depth in a coil. It
wouldn't be difficult to construct one, unless you're mechanically
declined like I am.

Roy Lewallen, W7EL

"Michael A. Terrell" ) writes:
TRABEM wrote:

Thanks Paul,

My old Drake C line had a multiple ganged permeability tuned inductor
mechanically coupled to a variable cap.

One of the assemblies tuned the front end, one peaked the driver
stage, etc.

It covered from 1 Mhz to 30 Mhz with a half turn on the front panel
preselector control. Most likely the variable C and L tuned together
maintained a desirable Q across all the bands.

I was hoping for something similar, but have no idea where to get
permeability tuned inductors today.

Regards,

T


Fair Radio used to have some in their catalog, and they were used in
all of the old Delco (and some other brands) car radios of the '60s and
'70s.

I thought it was the norm for car radios to be permeability tuned, right
up to the time synthesizers came along. No manually tuned car radio
that I've looked at had a variable capacitor, with the exception of
a relatively recent one that had a frequency counter for the readout.

Car radios would be the best source of the mechanics, at least for
front end tuning and not oscillator use. Of course, the issue isn't
changing the permeability of a coil, that's easy with a tuning slug,
but making it easy to do that from a front panel control, ie the knob
doesn't move in and out. The pre-synthesized car radios would proved
the mechanism. But you did see the occasional article in the old
days about making up such a mechanism, though those tended to be
for oscillators where there was more fuss because they wanted/hoped
for linear tuning.

Michael VE2BVW

Michael Black wrote:

I thought it was the norm for car radios to be permeability tuned, right
up to the time synthesizers came along. No manually tuned car radio
that I've looked at had a variable capacitor, with the exception of
a relatively recent one that had a frequency counter for the readout.

Car radios would be the best source of the mechanics, at least for
front end tuning and not oscillator use. Of course, the issue isn't
changing the permeability of a coil, that's easy with a tuning slug,
but making it easy to do that from a front panel control, ie the knob
doesn't move in and out. The pre-synthesized car radios would proved
the mechanism. But you did see the occasional article in the old
days about making up such a mechanism, though those tended to be
for oscillators where there was more fuss because they wanted/hoped
for linear tuning.

Michael VE2BVW


Some of the cheap imported car radios used the same plastic cased
variable capacitor with mylar film between the plates that were used in
a six transistor pocket radio. They didn't have a tuned front end and
amp like the US made car radios. They were ok if you only wanted to
listen to a local station but they were very poor quality radios.

I think I would use a variable capacitor and electronically switch
the inductor for different ranges, like the broadband L-C Based VCO in
the telemetry receivers I worked on. Diodes were used to short segments
of the inductor, at the grounded end and the band segment was controlled
by simple logic. Bias on hard to short a tap to ground, and reverse
bias it to prevent it from clipping the RF when its off. They used
discrete +/- 12 or 15 volt switching , but a decent buffer amp would
work and could be driven by CMOS or TTL compatible logic. If the
receiver has any switched DC signals to show which band its on, you
could make it change bands without touching it. You could even sample
the L.O. with a counter circuit and use the output to control the
tuning. If you want to g to extremes, you could add a microprocessor
with lookup tables to drive a stepper motor that can adjust the variable
capacitor(s) for you as you tune the radio.

--
?

Michael A. Terrell
Central Florida

Roy Lewallen wrote:

Michael A. Terrell wrote:

Fair Radio used to have some in their catalog, and they were used in
all of the old Delco (and some other brands) car radios of the '60s and
'70s.

Just about all the radios I've seen with mechanical pushbuttons for
station presets used permeability tuned inductors


All of the american made radios were, but some cheap imports weren't.

-- the pushbuttons
each just positioned a ferrite core to a preset depth in a coil. It
wouldn't be difficult to construct one, unless you're mechanically
declined like I am.

Roy Lewallen, W7EL


I once told one of the Vice president of marketing at United Video
Cablevision that he was mechanically declined and my boss freaked out.
;-) The VP yelled at him and told him to shut up because it was true,
and he was happy that whenever he had to launch a new channel in the
Cincinnati system that everything was set up and ready for him when he
arrived.


Of course, you could make them as individually tuned coils with a
stepper motor or servo per coil.

--
?

Michael A. Terrell
Central Florida

Thanks to all that commented. Although I don't have an answer, the
content of the discussion was helpful.

I'm not exactly fond of electronic switching, but it's probably
adequate since atmospheric noise dominates in the vlf and lf range
anyway.

I'm trying to understand the butterworth filter now, with the hope of
designing a bandpass filter over a wide range of frequencies rather
than a variable unit.

Regards,

T