Return to KLUBNL.PL main page

rsgb_lf_group
[Top] [All Lists]

Re: LF: Influence of a TX antenna to a small rx antenna

To: "[email protected]" <[email protected]>
Subject: Re: LF: Influence of a TX antenna to a small rx antenna
From: Rik Strobbe <[email protected]>
Date: Sat, 26 Dec 2009 23:29:10 +0100
In-reply-to: <38A51B74B884D74083D7950AD0DD85E828AD02@File-Server-HST.hst.e-technik.tu-darmstadt.de>
References: <38A51B74B884D74083D7950AD0DD85E828AD02@File-Server-HST.hst.e-technik.tu-darmstadt.de>
Reply-to: [email protected]
Sender: [email protected]
User-agent: Internet Messaging Program (IMP) H3 (4.1.2)

Hello Stefan,

why not use the "big" TX antenna also for RX. With a sufficient atenuator of course. On 137kHz I alway used my TX antenna for RX, but a 40dB attenuator was required to avoid overload at the RX.

73, Rik  ON7YD

Quoting Stefan Schäfer <[email protected]>:

Dear Lowfers,

Yesterday i saw the extreme influence of my 100m vertical tx antenna to the small active antenna. Now I have some ideas how to avoid that and would be thankful for any comments and hints.

During the QSOs there was much QSB in the RX signals, just as if someone is playing with the AF-volume. First I thought it is caused by a loose contact in the Converter or in the fiber optic cable but later I found, that it was caused by the angle of the tx antenna to ground (Wind was good but with some QSB ;-) ).

In my last QSO with IK1HSS I disconnected the TX wire from the loading coil and than, the whole signal was almost gone (when choosing “visual mode LOW” at the argo monitor). Switching to “visual mode HIGH” and increasing the AF a little bit gave almost the same QRM lines by DFC39 (vy strong yesterday). Now, the SNR was much better than before!

In the following pic you can get a impression of the dimension of the influence of the TX ant (The RX ant is isolated to the rest of the rig by an optic fiber cable and battery supply, so “just” coupled by the E field to the TX ant):

[cid:[email protected]]

In the left part the TX wire was disconnected from the coil and mode was HIGH

In the middle part I connected the wire to the loading coil

In the right part I switched to the LOW mode!

I am sure that the TX ant does not only increase the QRM but also the wanted signal. So the benefit will not be so strong than between left and right part in the pic. But the SNR increases significantly, as I saw in the QSO with IK1HSS. So a disconnection of the TX wire makes sense.

What are the alternatives?

Placing the RX ant apart from the TX ant to reduce the influence? I think with a 100m vertical I can forget that unless I do not want to spend 100s of meters of RX cable ;-)

Bringing the ant out of resonance? That is the one and only solution I think. Any better ideas?

I could disconnect the loading coil at the lower end to ground. That could be done with a relay (My RX/TX sequencer provides an output “12V @ RX” that could directly be used for this). Probably there will be a residual coupling capacity of the coil to ground and the relay contacts and so on, say 30pF but since my ant C is 580pF that would be quite enough for deadjusting, I hope so.

As I thought about protecting the contacts and the PA in the case that there will be a mistake (contacts have to be closed without a voltage at the relay) in the circuit, i thought some Varistors or even better some antiparallel diodes could be a good choice. That brought me to the idea that a power diode out of a SMPS would easyly handle the antenna current with fast enough switching times.

So do you think just two antiparallel diodes (or perhaps 2x2, 2 in series, and perhaps with an additional 10…100k Resistor parallel) without a relay could be suitable to solve the RX TX ant influence problem? Could the TX signal distortion be neglected? When the voltage across the diodes (RX case) is less than the forward voltage, do the diodes act as an open contact, just with a few pF reverse capacity? I think so but I don’t know much about the RX voltages that are given by such an antenna. There will be some experiments… :)

Comments welcome ;-)

73, Stefan/DK7FC




<Prev in Thread] Current Thread [Next in Thread>