Month: January 2016

Working Hellschreiber from a cold war bunker

Today I was able to operate the special event call sign GB0SNB from the Kelvedon Hatch Secret Nuclear Bunker. Hellschreiber was my mode of choice, its fun and easy to identify on the air  and also quite rare. Operating was completed from the original home office radio room deep inside the bunker.



Here you can see the door to the “Home Office Radio Room” and a picture of the mast when it was full of microwave links connecting the site to a number of other locations around the country. The radio room is located 3 levels below ground.


The radio room is located directly below the radio mast, and a tunnel connects you to the surface. Feeder cables run towards the surface and connect to the doublet antenna. Here you can see the radio room and both modern equipment (used for GB0SNB activations) and period equipment.



The laptop is running Ham Radio Deluxe and using the digital mode HELL. The Icom 7100 was running 50w RF into the clubs doublet antenna located 26m on the mast. The doublet antenna is 40m per leg. A cup of tea is required to keep the operator warm, its not often above 12c in the shack.

I managed to work several station from Belgium, Netherlands, Germany, England and Sweden on 7 MHz and on 5.3665 MHz I used Olivia MFSK. The bunker is also a museum so often busy with members of the public most days of the week.

The mast is also home for GB7KH the Kelvedon Hatch Dstar repeater.


472 kHz reception on WSPR

Having played around with the Wellbrook loop on 7 MHz the other evening I decided to try some 472 kHz reception on WSPR. I had never tried 472 kHz before, as the combination of local noise and poor antenna has always put me off.

The Elad FDM-DUO seems quite sensitive down to 100 kHz (the specifications says 10 Khz), and coupled with the Welbrook this seems to work really well. The signal to noise ratio of received stations has improved dramatically, and the screen was soon populated with stations all over Europe.

The Elad also has 2 antenna sockets on the rear, one is configured as RX only, and the other TX/RX. Although I didnt transmit this type of set-up works well if you intend to use a separate receive antenna.

Radio user completed a review of the Wellbrook antenna, and you can read that on-line.

The Elad is a QRP radio that works on all HF bands from 1.9 MHz to 50 MHz with power selectable from 0.3w to 5w. RADCOM did publish a review in March 2015, available for members to read. Its a SDR that can be run standalone, and works like a normal radio with a front panel and VFO or connected to a computer.

Dial freq for WSPR is 474.2 USB. Using software available free from Joe Taylor K1JT.

Here is a screen shot showing stations received.


After 24 Hrs I had received the following stations. This data can be represented on a map


Some graphed data, signal v time for PA3ABL/2



Graphed data DK7FC signal v time.






73 Dave M0TAZ

Comparing the Wellbrook Loop to a Doublet Antenna

This evening I decided to complete a test using two different HF setups in an effort to compare the receiver performance of my antennas.

Radio 1, Kenwood TS 990, doublet antenna 10m per leg @ 9m

Billy offers support.

Billy offers support.

Radio 2, Elad SDR, Wellbrook RX loop @ 3m



I decide the best way to test on receive would be to leave both radios receiving on 7076 KHz for 90 minutes, using WSJT-X. Both receivers had been set to 3 KHz bandwidth, and over the course of the next 90 minutes they each collected over 600 measurement reports.

The test started at 1830 Hrs and continued to 2000 Hrs. The doublet collected 633 measurement reports, and the wellbrook 740. This was the first indication that the Elad coupled with the Wellbrook was out performing the Kenwood / doublet setup. The next stage was to look in detail at the reports, WSJT has an option to write all received data into a text file, this includes time, s/n and decoded text. I used this to select at random some calls that represented a selection of EU (and DX if possible) and compare the relative signals received at the very same time on both radio 1 and 2.

The following table shows a selection of data points

wellbrook v doublet

After reviewing the data it was clear the Wellbrook loop improved the quantity of stations decoded, and in many cases also improved the s/n received.

This is by no means a definitive test, but it does indicate the Wellbrook loop (as many others have reported) work well in somewhat noisy urban environments.  The wellbrook loop can be reviewed online at the Wellbrook communications website, the version I used for the tests was ALA1530  its 1m and designed for medium and shortwave 50 KHz to 30 MHz. Steve Nichols G0KYA completed a review in January 2012 Radcom You can see the Eznec plot as photo 3 in the review.

The other factor to consider is the loop antenna has some directivity, I didn’t rotate the antenna at any point during my tests. In many cases rotation will improve / reduce signals as the antenna has directional properties.

The Welbrook performed well on 7 MHz, I suspect it would be even better on 1.9 and 3.5 MHz… to be continued….

Dave M0TAZ