Duplex on HF

[AF6LJ]

I can't help but wonder if moving the transmitting antennas isn't the better starting idea since this will relax the performance requirements of any notch / pass system.

Transmitter wideband noise, I would think, would be a lesser issue on one sixty and and maybe even on seventy five meters especially if the transmitting antennas are moved further away.

[W8JI]

I can't help but wonder if moving the transmitting antennas isn't the better starting idea since this will relax the performance requirements of any notch / pass system.

Transmitter wideband noise, I would think, would be a lesser issue on one sixty and and maybe even on seventy five meters especially if the transmitting antennas are moved further away.

I now have a spreadsheet of signal levels in various combinations of antennas, Sue. There are 9 transmit directions with about primary 25 receive system selection choices. That's over 200 possible direction combinations. It looks like I only have a dozen bad combinations that will overload receivers by going over ~0 dBm into the preamps.

Since my hair-brained nulling sense antenna idea did not work, I have decided to add new receiving antennas to replace the most problematic receiving antennas. That's a new 1000 foot feeder extension to some property I have not had antennas on yet, and I can run audio band clean sinewave tones through the signal coax to control remote relays.

That will eliminate the worse of the systems, and hopefully, with a little selective notch system at the receivers, be enough. I have some big coils left over from a BCB 50 kW project, 1/2 inch copper tubing about 18 inch diameter and 30 inches tall that might make a receiver notch.

Big project.

[KL7AJ]

You should be able to approach this with a shorted coaxial stub, if it's made of good low-loss coax.

[W8JI]

You should be able to approach this with a shorted coaxial stub, if it's made of good low-loss coax.

Hi Eric,

Unfortunately, it is not even close with the best coax for stubs. The Q of any stub is far too low. A stub 100 feet long is lucky to have a Q of 20 or 30.

I have harmonic filters, for 160 through 40 meters, made with coax stubs. They pass the entire band without tuning!

An L/C network is much sharper.

73 Tom

[K7ELP]

Pherhaps a remote receiving antenna about 20 miles or so distance with a internet connection to control the frequency of the remote receiver.
I suggest this as a long time ago I used to maintain transmitters at a USCG communications station. We were on a small area of a USN transmitter site. This site was on the island of OAHU in the Hawaiian islands. The location was Waianae. The receiver site for both communications sites were about 20 miles away near Wahiwa. This seemed to eliminate the receiver overload caused by the close proxmity of the receiver/transmitters. As I recall at the receiving site we also had some broadband receiver preamps. This was the mid 1960's so all the equipment was tube and I think the receivers were 51J4's(made by Collins).
As a side note the CG used to monitor 500Khz for emergancies and used 440Khz as a working frequency. One time if the operator keyed the 500Khz transmitter he would hear a local radio station(carrier frequency 940Khz) on the 440 receiver. The problem was traced to a corroided connection on the receiving antenna.
I also in the late 60's used to run full duplex on 2M AM with a station in NYC when I was stationed at governer's island, NY. The receiver was a Clegg Intercepter B. I used a 40M dipole for receiving and 11 element beam for xmitting. We could seperate by 100Khz and still operate full duplex.
I hope any of this is a help to you.
Ned

[AF6LJ]

I now have a spreadsheet of signal levels in various combinations of antennas, Sue. There are 9 transmit directions with about primary 25 receive system selection choices. That's over 200 possible direction combinations. It looks like I only have a dozen bad combinations that will overload receivers by going over ~0 dBm into the preamps.

Since my hair-brained nulling sense antenna idea did not work, I have decided to add new receiving antennas to replace the most problematic receiving antennas. That's a new 1000 foot feeder extension to some property I have not had antennas on yet, and I can run audio band clean sinewave tones through the signal coax to control remote relays.

That will eliminate the worse of the systems, and hopefully, with a little selective notch system at the receivers, be enough. I have some big coils left over from a BCB 50 kW project, 1/2 inch copper tubing about 18 inch diameter and 30 inches tall that might make a receiver notch.

Big project.

Big project indeed; using a little finesse as appose to bruit force looks like the best possible answer.
(moving the antennas)

Using tone remote control is really the only sensible choice given the amount of additional copper required to interconnect everything.


Using shorted stubs had occurred to me but the series R would be too high to obtain a usable Q.


Good luck, keep us posted.

[G3UUR]

I'm at wit's end on this. I'm looking for creative ideas and suggestions.

Has anyone built a filter that can give about 20 dB attenuation in 5-10 kHz of spacing on 160 meters?? Input level would be ~25 dBm or less.

I can get a real Q of about 500-800 with some sample inductors I have measured.

73 Tom

I've never built the sort of filter you require, Tom, but I think your inductor Q values are an order of magnitude short of what you require. To meet the specification you describe in one of your later postings, you seem to need a couple of crystal filters, or possibly a couple of LC-crystal combination filters. The one in the fixed frequency receive path could be a simple band-pass arrangement. I'm not sure how much leeway you need around the fixed frequency, but a bandwidth of 4kHz should be easily attainable at 1.85MHz with 20dB attenuation 5kHz to either side of the passband. The search station would need a crystal notch filter in its receive path to take out a band around the fixed station transmit frequency.

Insertion loss may be an issue, of course, and crystals are not free of intermodulation products of their own either. If you haven't had much experience of crystal filter design, you might prefer the phasing route.

Thought I'd mention the crystal filter possibilities, though, in case you like a real challenge.

73, Dave.

[W8JI]

I've never built the sort of filter you require, Tom, but I think your inductor Q values are an order of magnitude short of what you require. To meet the specification you describe in one of your later postings, you seem to need a couple of crystal filters, or possibly a couple of LC-crystal combination filters. The one in the fixed frequency receive path could be a simple band-pass arrangement. I'm not sure how much leeway you need around the fixed frequency, but a bandwidth of 4kHz should be easily attainable at 1.85MHz with 20dB attenuation 5kHz to either side of the passband. The search station would need a crystal notch filter in its receive path to take out a band around the fixed station transmit frequency.

Insertion loss may be an issue, of course, and crystals are not free of intermodulation products of their own either. If you haven't had much experience of crystal filter design, you might prefer the phasing route.

Thought I'd mention the crystal filter possibilities, though, in case you like a real challenge.

73, Dave.

Hi Dave,

Based on past measurements, the highest Q I can get with is around 1000 or so with air-core inductors (1/2 inch copper tubing about a foot or so in diameter).
Highest on toroids is about 500 or more.

I probably don't want to do a pass, I probably want to do a signal suck-out type of notch with L/C circuits.

My second thought was to build a high power local oscillator and use it with a pair of very hard mixers to up and down convert to some fixed crystals in a filter. I can probably get that so IMD in the crystal itself filter limits the dynamic range.

VFO stability would not be a big issue if the same LO is to up and down convert.

Today I'm moving one of the worse receive antenna systems. It looks like I can get about 20 dB less coupling into that antenna with this relocation.

73 Tom

[AF6LJ]

Good luck Tom.

[G3UUR]

My second thought was to build a high power local oscillator and use it with a pair of very hard mixers to up and down convert to some fixed crystals in a filter. I can probably get that so IMD in the crystal itself filter limits the dynamic range.

VFO stability would not be a big issue if the same LO is to up and down convert.

If you decide to go that route, I can recommend the H-mode mixer using an FSAV332.

OTOH, a simple crystal half-lattice filter with 2 transmission zeroes would give you at least 20dB ultimate attenuation and pretty fast cut-off above and below the pass band. However, I can appreciate that a single BPF on the fixed channel station would be very limiting in a contest, and you wouldn't want to have a whole mess of filters to be able to move frequency. From that point of view, the tunable converter with fixed IF filter would be the more practical approach for flexibility.

Dave.