Return to KLUBNL.PL main page

rsgb_lf_group
[Top] [All Lists]

Re: LF: 136kHz ferrite rod antenna test

To: [email protected]
Subject: Re: LF: 136kHz ferrite rod antenna test
From: Peter Dodd <[email protected]>
Date: Tue, 23 Aug 2011 22:00:26 +0100
In-reply-to: <24DD2264A48C492C89F1A444F5D47A5D@JimPC>
References: <24DD2264A48C492C89F1A444F5D47A5D@JimPC>
Reply-to: [email protected]
Sender: [email protected]
User-agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-GB; rv:1.9.2.20) Gecko/20110804 Thunderbird/3.1.12

In the early 1960s some aircraft in the RAF were fitted with Marconi ADF sets. The loop antenna was a slab of ferrite, which if my memory serves me right, was some 200mm x 120mm and about 30 or 40mm thick. The windings were just a few turns of litz wire. As the equipment had to cover the whole of the NDB beacon band I guess it was a broad band technology.

In the early days of 73kHz I followed the example of Mike, G3XDV, in constructing a loop antenna and FET preamp from a defunct transistor set to make up a mobile receiving station. These were used to check out the relative field strengths of our primitive early transmitters. My transmitter was a audio signal generator, a 10w solid state vehicle sound system amplifier and a big coil

Peter, G3LDO.

Having said quite a lot about ferrite rods on the reflector recently, I thought I better try one as a quick "feasibility study"!

I have a large bar of ferrite I had assembled from surplus "U" cores some years ago as part of a portable SAQ receiving system (it's a long story...). The"rod" is about 470mm long, with rectangular cross section about 30mm x 35mm. This is rather larger than the rods used by DK7FC and DF6NM, but the ferrite is more lossy - probably 3C8 or some similar "power" grade. 30 turns of litz wire gave L about 360uH and Q of 150 (so using litz probably wasn't justified, but I had that length to hand). This resonates with about 3700pF at 137kHz, giving an equivalent parallel resistance of about 45kohm and a bandwidth a bit under 1kHz.

The preamp is a compound JFET/bipolar follower using a J310 biased to 6mA and a 2N3053 biased to 50mA (probably similar performance to a "mini-whip" preamp at 137k). The high Z input is connected directly across the tuned antenna winding. With the high QRN on the band this evening, it is hard to get a good view of the noise floor, so I used a "dummy antenna" consisting of a pot-core inductor wound with the same inductance and Q, substituted for the real rod antenna to simulate a zero signal and external noise condition. The noise output from the resonated dummy antenna was 10dB or more above the RX noise floor (SDR-IQ). The preamp with input shorted gave a noise level below the RX noise floor. So this preamp arrangement gives sensitivity limited by the thermal noise of the antenna, and adequate gain for the "reasonably good sensitivity" SDR-IQ (also for my RA1792).

I made a rough estimate of the overall sensitivity of the complete system by measuring the level of DCF39, which is a fairly stable 800uV/m here during daylight. On an arbitrary scale on the SDR-IQ spectrogram, DCF39 was -19dB, while the noise floor at around 137.7kHz with dummy antenna was -109dB with 0.75Hz FFT noise bandwidth. This works out to an antenna noise floor of about 0.03uV/m per sqrt(Hz), several dB below quiet 136kHz band conditions. So sensitivity of a receive system with this antenna should be limited only by the band noise; it should also be possible to reduce the amount of ferrite somewhat, especially if lower loss ferrite antenna rods are used. I await lower QRN levels for a "live" test !

Cheers, Jim Moritz
73 de M0BMU





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