Wartime transmission experiments and two unknown sites in County Down

The 1930s and 1940s saw huge advances in the development of radio and radar technologies, with one of the most sophisticated networks of Radio Direction Finding (RDF) being deployed in the form of Chain Home stations around the coast of Great Britain and Northern Ireland. Running concurrently with the development of what would become RADAR (RAdio Detection And Ranging) as we know it were a series of experiments to study the characteristics of radio wave propagation; essentially looking at how do radio waves behave over large distances, over different terrain and through varying weather patterns. One such series of tests was reports in the 1946 Summary Technical Report of the National Defense Research Committee, Volume 2, Radio Wave Propagation Experiments. While this is an American publication (available here), allied nations worked collaboratively and many British scientists and engineers were involved in this work. This particular report was penned by E.C.S. Megaw, member of the Ultra Short Wave Propagation Panel.

While all of this work was highly secretive at the time, it has been possible to find traces of some of the developmental sites constructed in support of these experiments. The aim of this article is to record the results of this investigation, and examine a possible location on the County Down coast that may have been involved. All of this stemmed from an email I received from fellow historian Martin Briscoe who was trying to establish the background to a Scottish site known locally as Hush-Hush.

Part 1: Transmission Experiments

The series of trials known as the Irish Sea Experiment has been well documented in post-war technical journals, so it is possible to get a large amount of detail from these to assist the investigation. I won’t attempt to explain the technical nuances of radio wave propagation or the electromagnetic spectrum, but will endeavour to provide suitable explanation where necessary. The story of the mid-war radar experiments across the Irish Sea is ultimately the story of the history of weather forecasting radar. The main object of the tests was to “determine the relationship between radio performance and meteorological conditions in the lower atmosphere, with forecasting as the ultimate aim.1Summary Technical Report of the National Defense Research Committee, Volume 2, Radio Wave Propagation Experiments, British Transmission Experiments, Introduction” A series of experiments were conducted concurrently across the Irish Sea (along the Welsh and Scottish coasts) and in the Massachusetts Bay area of the United States.

The work was being primarily conducted in the British Isles under the auspices of the Ultra Short Wave Propagation Panel, a department within the Ministry of Supply; one of many wartime departments whose name and purpose was suitably obscured within the British civil service. A number of other government departments were also involved; including the Naval Meteorological Service who undertook low-level meteorological measurements, the Radio Division of the National Physical Laboratory (NPL) who analysed the results of the trials; the Admiralty Signal Establishment (ASE) who provided equipment through a contract with General Electric2General Electric Company Research Laboratories; and most crucially the stations were operated by personnel from the Signals Research and Development Establishment (SRDE) within the Ministry of Supply (MoS).

What surprised me most about these experiments, was the very low power levels used during the experiments over distances between 57 and 200 miles. Transmitting a series of on/off pulses from the transmitter3Technically this was a square wave, the output of a very simple transmitter circuit containing no data as such, only the radio energy at a rate of 1,000 per second (a frequency of 1 kilohertz or 1 kHz), the power varied only between 0.15 Watts and 0.6 Watts. The maximum output of a home routes can be as high as 4 Watts!45 GHz channel 155 to 171 at 4000 mW The sensitivity of the equipment and the highly focussed parabolic mirror dishes enabled such large distances to be covered.

3 cm Band0.15 W
6 cm Band0.3 W
9 cm Band0.6 W
A table showing the standard power output for the signal on period. Information contained in Reference 1.

A further interesting note about the Irish Sea experiments was that in the summer of 1944 the behaviour of a light beam was studied across the 57 mile link between stations A and C. The purpose of this was to observe any changes in refraction as a result of temperature change, but these tests proved inconclusive at that stage.

When referring to radio signals we often refer to the frequency and wavelength of the signal. The frequency is the number of complete on/off cycles that are completed in one second. Historically the unit of measure was cycles, but we now measure frequency in Hertz (Hz). Think of frequency as being a measure of time. The wavelength is the total length of a complete cycle, and is measured in meters (m). The higher the frequency, or the more cycles per second, the shorter the wavelength, or the more cycles that need to fit into a notional 1m length.

Wavelength can sometimes be used to refer to the radio band in which the transmission is being undertaken. In the Irish Sea experiments, transmissions were planned in three bands, or using three different wavelengths; 3cm (also known as the X-band), 6cm and 9cm (also known as the S-band). The corresponding frequencies are 8-12 GHz, 5GHz and 2-4GHz respectively.

While the experiments themselves were only evaluating the propagation of radio waves through the lower atmosphere, it appears that the links was also used to provide a radiotelephone link initially between the two closest stations 57 miles away. While only temporary connection was achieved for a couple of months at a time, the system was adopted on the 1m (1 meter) wavelength and installed in all of the stations by the Admiralty Signal Establishment. Using two varying frequencies, they achieved duplex operation allowing an operator to transmit and receive at the same time.

Locating the stations

In both detailed reports I have seen on these experiments, the station locations have been described in detail. While no grid references were supplied, from the descriptions and use of modern aerial photography it was possible to locate almost all of the sites involved in the Irish Sea experiments. There were also over-ground experiments conducted between the Admiralty Signal Establishment buildings at Whitewell Hatch and the General Electric Company Research Laboratories in Wembley; but these are out of the scope of this article. At three of these, the original buildings still exist. It appears that when scoping locations for these sites in early 1942 there was some difficulty finding a suitable compromise. After short field trials, the locations below were agreed.

The first links between south and north Wales were between 2-4 GHz, known as the S-band. These were operational in November 1943, although Garn-Fawr is reported as being operational from July 1943. Portpatrick became operational at the end of August 1943, but it took until March 1944 until signals were able to be received over the 200 mile sea path. The last site to have equipment installed and be operational was Knockharnahan in March 1944. The next link to be trialled was the X-band, between 8-12 GHz, on which all stations were functional in July 1944. Transmission tests were undertaken nearly continuously until the end of the war.

Map of Irish Sea locations showing the six locations chosen, letters A to F. Figure 4.16 from Reference 2.
StationLocationHeightPath Used
3cm X-band
7-11 GHz
Path Used
9cm S-band
2-4 GHz
A (TX)Garn-Fawr, South Wales540 ftACAC
B (TX)Strumble Head, South Wales90 ftAD
C (RX)Rhiw, North Wales825 ftAE*AE*
D (RX)Aberdaron, North Wales95 ftBDBE
E (RX)Knockharnahan, Scotland375 ftBF*BF*
F (RX)Portpatrick, Scotland95 ft
Station and path designations, Reference 2. Paths marked * were only operated for part of the test period. TX – Transmitter, RX – Receiver

Identifying the structures

It was the question over an unusual looking technical building raised by Martin that prompted this research in the first place, so confirmation of the structures purpose was needed. However, thanks to one of the comprehensive journal articles, we have an image of inside one of the Receiver stations, as well as descriptions about the equipment used.

One of the images on Flickr of the previously unknown building at Knockharnahan, Scotland. It was noted that there appeared to be openings that had been bricked up to the front, but their purpose was not initially identified. Also on the roof of the building were two possible concrete bases or anchor points © JM Briscoe 2012

Having located the stations and identified likely structures still standing, and while the standard of construction appears to be in keeping with wartime military technical buildings, they did not conform to known plans of buildings at other sites such as wireless stations, radar sites, observation posts, administrative or technical buildings. The sites were isolated and not obviously associated with other military sites, and did not follow the alignment of nearby roads or seem to be orientated for an optimal view. Three stations are extant and can be viewed through Google Streetview, below:

The image below taken in side the Aberdaron receiver station in North Wales (now removed) shows four parabolic antenna sitting on what appear to be concrete plinths by covered apertures in the wall of the building. The roof seems solid, likely constructed from cast or slabs of concrete, and there are steel beams reinforcing the roof. There are few other fixtures other than ducting for power cables, electric lighting and a clock on the gable wall. From the descriptions of these sites, this was one of two rooms at the TX and RX stations, which appear to have been very similar in design.

General view of the equipment in Aberdaron, North Wales. Image is Figure 1 from Reference 1, 1946.

The level of technical detail in the reports gives us much greater idea of the purpose of these features. 48in diameter parabolic mirrors were used for all transmitters and receivers, mounted inside the stations behind canvas windows. This indicates that both TX and RX stations would be very similar in appearance, with possible a greater number of windows in the TX station. In the image above, the X- and S-band receivers are the two mirrors central to the image, with the S-band equipment in the furthest window. The table on the right side has the S-band receiver and monitoring equipment. The bay to the left of the image is empty. The notes refer to meter wave (1 meter) equipment being located in an adjoining room which was used to carry a duplex radiotelephone link to all stations.

Part 2: Irish Coast Sites

So this article now delves into the unknown. Where County Down, or Saint Johns Point to be more precise, is involved is the result of speculation on my part. I will lay out my through process and the evidence on which I based my conclusions, and you can decide if you agree or otherwise. With each of the two the sites I look at, I will lay out my hypothesis and the evidence on which they are based.

It was a coincidence that I had been reading about the Irish Sea experiments only a month or so before I visited St John’s Point on a bright Sunday afternoon. It was only when walking back to my car that I saw this roofless building constructed form brick and concrete that I thought about a note in one of the research articles from the Irish Sea experiments. It made me do some further investigative work to establish the purpose of this structure and also attempt to interpret a site I had visited months earlier at Killough.

An introductory paragraph in the 1946 Summary Technical Report of the National Defense Research Committee combined with an image in a 1951 journal the Propagation of Short Radio Waves that made me interpret this particular site in a more focused way.

Several investigations of more specific propagation problems have been carried out during the summer of 1944 (…). Measurements on a wavelength of about 3 1/2 metres over a 90-mile sea path, with heights such that the path length was about twice optical range, to provide quantitative data on the importance of refraction in this waveband. Radar measurements from Llandudno, North Wales, with the Isle of Man and the Irish Coast as the main targets, on S, X, and K5K band is a portion of the microwave radio spectrum between 18 to 27 GHz bands. The object was to obtain practical data on the relative performance of K band under a variety of meteorological conditions which were studied simultaneously with the radar observations by ship, balloon, and aircraft measurements.

Summary Technical Report of the National Defense Research Committee, Chapter 5, Transmission Experiments in England, E.C.S. Megaw, 1946.
Revisiting the map of Irish Sea locations from Reference 2, a notional continuation of the bearing of the aircraft soundings lands us at St Johns Point.

The key pieces of information that are useful in this investigation are a sea path of 90 miles with measurements conducted from Llandudno to Irish Coast targets. These trials took place in Summer 1944. With St Johns Point being the closest and most suitable point along the (Northern) Irish Coast, and the Admiralty Signal Establishment having a trials station on Great Orme’s Head (also known as Great Orme) above Llandudno this was a great place to start. Also, the measured6Google Earth measuring tool 95.8 miles distance between the two sites is 95 miles, close enough to continue on the line of investigation to St Johns Point.

Using the map of Irish Sea locations from one of the experiment reports, I have extended the bearing line from Morfa Beach7Morfa Beach is the name given to the sandy strand to the west of Great Orme. I am unsure why this reference point was used; perhaps indicating measurements were at sea level and a ground weather station was located on or near the beach. which is labelled “locus of aircraft soundings.” This bearing terminates at St Johns Point. Meteorological soundings were made by ships and aircraft on or near the transmission paths.8Transmission experiments in the British Isles, Donald E. Kerr

St Johns Point (J 527 339)

It was a single but distinctive technical structure at the side of the road at St John’s Point that caught me eye (which is curiously tuned to red brick and concrete). Characteristically military in nature, it appears to have been constructed for a specialist purpose, not recognisable as any previously known building type. It is still not clear what the exact purpose may have been, but I will attempt an interpretation.

Site Description

This building is approximately 11m x 5m in size, constructed of red brick and cast (presumed reinforced) concrete. The walls extend to approximately 2 meters above the concrete floor, with the lower portion constructed of red brick and the upper of concrete. There is evidence that the upper wall cast using wooden plank formwork. Both wall sections are the same thickness, approximately 300mm. There is no evidence of internal walls or fixtures.

To the front and rear of the structure are what appear to be window and door apertures. To the rear is a single full-height opening which may have at one time contained two large doors similar to those found on vehicle garages. The front openings consist of a central, single, narrow full-height opening, similar to a doorway, which is flanked and adjoined to two half-height openings, laid with a concrete sill, possible window frames. This opening has been infilled with rubble, including what may be a concrete lintel that once sat across the front opening and is now in two pieces. At each gable end there is evidence in the concrete that in these places the structure was higher.

While the structure currently has no roof, on the top of the walls is evidence that there may have been wooden or steel rafters and ceiling joists. However these impressions left in the concrete may be as a result of formwork during construction. There is a row of small nails protruding from the concrete along the edges which may have once held wooden batons for fixing a roof structure to.

On the outside of the structure, at the rear right corner, is a secondary wall. It is likely this was constructed at the same time as the main walls, and while it stands at approximately 1.5m high, is terminated with a square post, rendered in concrete. The outside wall appears to be constructed with concrete blocks, and cement render.

The entire structure has been later adapted for agricultural use, including the addition of a number of animal gates.

Flickr Album Gallery Powered By: WP Frank


To place this structure in the context of the surrounding landscape I have provided an annotated aerial image below. There is possible other second world war military activity near the site in the form of a reported anti-aircraft gunnery range, but I am confident this structure is not related to that.

An aerial photograph of the site at Killough. There are a number of features I have highlighted in the table below.
AA concrete hard approximately 50m long x 10m wide. This is typical of hards used to park heavy vehicles. It does not appear to have been part of a structure.
BIt is this roofless structure that is now cause for further research and speculation. It appears not to sit on the alignment of the road and is constructed from red brick and concrete, similar to military structures of the second world war era.
CA series of four tanks constructed from concrete panels. These are marked in some databases as anti-aircraft gun positions as part of a US Army gunnery range, however they are marked as tanks on historic OS maps, and upon closer inspection they can be confirmed as agricultural tanks, with signs of pipework and internal tar sealing for watertightness.
DSt John’s Church, an ancient church and scheduled monument. Included for reference only.
EA corrugated iron structure along side what appears to be a sunken water tank. Given the close proximity to a freshwater well associated with the church these buildings can be eliminated as not being related to the other structures nearby.

The alignment of the building is 157 degrees front-to-back and it does not appear to line up with the existing road, as would often be expected with roadside buildings. When this alignment bearing is plotted across the Irish Sea, its destination is Station C at Rhiw in North Wales, a distance of 106 miles away. This may of course be coincidental. It is the association with the Admiralty Signal Establishment at Llandudno (Great Orme) that I think is the most likely. Also of note is that the site sits within a couple of hundred meters of a (now removed) BT microwave tower. It was likely constructed in the late 1970s carrying a telephone link between Northern Ireland, the Isle of Man and onwards to Manchester.9BT Microwave Sites website – http://www.dgsys.co.uk/btmicrowave/sites/266.php

An overview map of the Irish Sea experiment locations and St Johns Point placed in context. In this map I have included Great Ormes Head near Llandudno in North Wales © Crown Copyright Ordnance Survey Ireland


This building was constructed to support the series of transmission experiments across the Irish Sea.

  1. Looking purely at the fabric of this structure, I can not reach a conclusion about its purpose. Red brick was used extensively (but not exclusively) during the second world war for all manner of structures, including technical buildings and as a method of formwork when filling with concrete for defensive structures. However this is not a true red brick building, having only a handful of lower courses of brick laid, before concrete was poured into wood formwork. I have no other examples of this technique other than here.
  2. With the background knowledge of the Irish Sea experiments, it is possible to believe that this building may have been constructed as a drive-in vehicle mounted radio laboratory from which to operate mobile transmitting or receiver equipment; providing protection from the weather and (as in the case of HF/DF stations) a blast wall against aerial gunfire or bombing.
  3. Even with little physical evidence, I still assume a roof was once fitted to this building, and the apertures at the front of the building also covered. Perhaps with a radio transparent material as was fitted to the stations in Scotland and Wales.
  4. The lack of windows along the sides of the building is also suggestive that this is a technical building, and not an ancillary building to some other site. There are also no ventilation louvres, although the apertures on the front of the building may have had louvres as opposed to windows.
  5. There is no visible sign of internal fixtures or mountings, certainly no plinths that I would have expected to find if this was a generator housing.
  6. With an anti-aircraft range being reported to have also been active on this section of coastline, the building may have had connections with this, however the design or construction is not like supporting buildings at other ranges.
  7. It is possible, but unlikely that this is an agricultural building constructed using surplus materials from a nearby but now removed site.
Perhaps not so far fetched is my idea that this building would have housed a large vehicle. The lorry shown (not to scale – I guessed) is a second world war vintage Leyland Retriever Type H.

Killough (J 540 352)

Often incorrectly reported as a pillbox, a coastal observation post above Killough coastguard station has been modified so that interpretation of the site could be related to radio or radar; however I have not yet found any record of either. I have covered a few locations around the NI coast that still have lookout posts of this design (Guarding the Coast), but this is the only one that has been modified in this way. I have decided to provide some of this interpretation here.

Site Description

Constructed of characteristic red brick, with cast concrete features such as the roof and the three observation loopholes, this observation post has been modified post-construction. The particular features of note are the roughly chiselled channels running across the front and rear faces. At each corner, on the ground, are four concrete foundation blocks, with evidence of a steel frame set into them. The entrance to the post has also been modified so that the concrete angle at the top of the door follows the line of the cross members. A brick blast wall is placed at the rear of the post and its width extends beyond the main post structure.

Flickr Album Gallery Powered By: WP Frank
An impression of what the Killough observation post may have looked like based on the steel remaining in the four anchor points and the channels carved into the brick and concrete walls © Frontline Ulster

Despite the scars of probably having a mast constructed on top of it, there do not appear to be any ancillary buildings nearby as would have been expected in conventional radio or radar stations (such as generator, transmitter or receiver buildings), nor does there appear to be a vehicle hardstanding to park mobile equipment.


In relation to the unknown building at St Johns Point, the observation post is 1.1 miles north east. The distance to the stations on the north coast of Wales is relatively unchanged at 90 and 106 miles, so I would consider this site to be in contention as one of the Irish Sea experiment locations. The post is elevated on top of a small hillock, near a spot height of 91 feet (27.7m); the same height as the low level sites chosen for the transmission experiments. This is the highest point close to the coast for some miles, and even thought the modern telecoms mast is a higher elevation, it was set inland.

St Johns Point, County Down showing the relationship between the unknown structure and a modified observation post at Killough Coastguard station. I have also circled the site of a BT microwave tower which provided a radio link between NI and the Isle of Man (and on to Great Britain) © Crown Copyright Ordnance Survey Ireland


A tower or elevated platform was retrofitted around an existing coastal observation post in support of the Irish Sea transmission experiments.

  1. Built on a hill of height 90 feet would put this at the same elevation as was chosen for the low level sites in Wales and Scotland.
  2. The main reason for constructing a mast over an existing building is generally to house equipment and reduce the loss of signal that would occur over longer cable runs.
  3. Did the mast support a signal light or beacon to assist the aircraft taking measurements to maintain a constant bearing? Or is this structure unrelated to these radio experiments?
  4. Even though a wind gauge is marked on the OS 1:10,000 map, I believe the it was inside the coastguard compound.


From the diligent scientific reports on the series of experiments conducted in the 1940s we know a great deal about the methods and locations used in the early development of meteorological radar. While sites on mainland Great Britain have been recorded, the fleeting reference to sites on the Irish Coast have led to a missing piece of the jigsaw. It it likely to be coincidence that there are two sites in County Down that have no explanation as to their purpose, but their unique features have warranted further investigation. I am sure the conclusions I have drawn will be brought into question, but I would like to open this topic for further discussion. Perhaps I am close to positive identification, perhaps not. But regardless of the purpose of these buildings, the experiments conducted across the Irish Sea were important in laying the foundations for modern radar weather forecasting.

While I may consider myself an amateur historian, I am keen to continue to develop my knowledge and understanding, so if you have any evidence to support or refute any of the assumptions made in this article then please get in touch.


  1. Radio wave propagation experiments
  2. MIT Radiation Laboratory Series 13 Propagation Of Short Radio Waves
  3. National Meteorological Library and Archive Fact sheet 15 — Weather radar
  4. Images and information from Flickr user Doffcocker