EMRFD Message Archive 2004

Message Date From Subject
2004 2008-09-05 09:20:36 ha5rxz Debugging The Hilberling PT-8000 Preselector
As there appears to be a staggering amount of RF Design talent in this
group I want to pick your brains. I'm trying to build a copy of the
Hilberling PT-8000 preselector, if you haven't seen it it's got a
massive 2.5 inch type 2 toroid with about 36 turns and to make the
peak sharper it runs at an impedance of 3 ohms. There are 5 coil taps
and 255 capacitor combinations.

The Problems

1) 5 Taps and 255 capacitor combinations gives 1275 combinations. 30
MHz divided by 1275 gives a step size of about 23.5 KHz. Is the
preselector really this wide?

2) I have a circuit that works well up to the 20m band. After this
point I need to use either very low capacitor values or silly coils
such as a 1.5 turn toroid. Which would be best or can you suggest a
better solution for operation above 15 MHz?

3) How do I tune it automatically? The thought of entering well over a
thousand values into some form of lookup table does not fill me with joy.

4) Why doesn't the shield around the module kill the effect of the
toroid because it will act as a shorted turn?

Yes, I know that there are other preselector designs including one in
a recent issue of QEX. I would however prefer to get a design working
that uses a single large toroid so that I can get a decent IP3 figure.

Peter Gant HA5RXZ
2005 2008-09-05 10:36:39 Niels A. Moseley Re: Debugging The Hilberling PT-8000 Preselector
Hello Peter,

(3) You could use Excel or other programs (GNU Octave etc) to generate
the table automatically. A PIC or AVR processor can store such a table
and select the optimum values, given the RX frequency.

(4) I think it's because the EM field is mostly contained within the
toroid. The toroid won't 'see' the enclosure.

73,
Niels PE1OIT.

> As there appears to be a staggering amount of RF Design talent in this
> group I want to pick your brains. I'm trying to build a copy of the
> Hilberling PT-8000 preselector, if you haven't seen it it's got a
> massive 2.5 inch type 2 toroid with about 36 turns and to make the
> peak sharper it runs at an impedance of 3 ohms. There are 5 coil taps
> and 255 capacitor combinations.
>
> The Problems
>
> 1) 5 Taps and 255 capacitor combinations gives 1275 combinations. 30
> MHz divided by 1275 gives a step size of about 23.5 KHz. Is the
> preselector really this wide?
>
> 2) I have a circuit that works well up to the 20m band. After this
> point I need to use either very low capacitor values or silly coils
> such as a 1.5 turn toroid. Which would be best or can you suggest a
> better solution for operation above 15 MHz?
>
> 3) How do I tune it automatically? The thought of entering well over a
> thousand values into some form of lookup table does not fill me with joy.
>
> 4) Why doesn't the shield around the module kill the effect of the
> toroid because it will act as a shorted turn?
>
> Yes, I know that there are other preselector designs including one in
> a recent issue of QEX. I would however prefer to get a design working
> that uses a single large toroid so that I can get a decent IP3 figure.
>
> Peter Gant HA5RXZ
2006 2008-09-05 10:38:00 Graham Re: Debugging The Hilberling PT-8000 Preselector
Peter:
Comments embedded below.
2007 2008-09-05 19:46:52 Allison Parent Re: Debugging The Hilberling PT-8000 Preselector
2011 2008-09-06 09:44:02 ha5rxz Re: Debugging The Hilberling PT-8000 Preselector
OK, thanks for all the replies. As I am now retired I can fiddle
around with this until I get something reasonable. This afternoon I'm
experimenting with a thirty turn coil instead of thirty six turns with
the first five turns wider spaced (about 4mm apart), this might help
at the higher end because of lower inter-turn capacitance.
Preselectors are really needed on 30m and below here in Europe so I'll
need something after the bandpass filters.

It's such a shame that the PT-8000 was cancelled because I'd love to
see a set of schematics.

HA5RXZ

--- In emrfd@yahoogroups.com, "Allis