History of the Atlantic Cable & Undersea Communications
of the Transatlantic
While it is difficult to be precise, it would appear that the United States Army's interest in a transatlantic cable began in early 1942. This is evidenced by the earliest date on some of the Western Union Company's blueprints. It is obvious from Western Union documents that a concerted and continuous effort began at this time.
The Army Signal Corps collected the necessary personnel during 1943 and at least in some of the cases they were held on "Rations and Quarters" with little or nothing to do. The author had been ill during the summer and the Army sent him home for a ninety day "Recuperative Leave". The group was assembled in New York in January and was assigned to study at the Western Union Company's Research Laboratory at corporate headquarters at 60 Hudson St. The group consisted of eight enlisted men and two Second Lieutenants.
The course work was taught by Mr. W.F. Wilder and Mr. Dickey, both of Western Union's engineering department. The instruction started with Lord Kelvin's definitive treatise on telegraphy. All aspects of ocean cable telegraphy were taught, from the simple devices like the siphon recorder to the most modern equipment used on high speed multiplex cables. The work included operation and maintenance of the various machines which we would use. We were also taught the protocol necessary to set up and operate a modern cable station.
This all ended, complete with a graduation and diplomas at the end of April 1944. We still were not needed so we marked time, but remained at Western Union. They managed for us to spend time at the operating sites actually seeing the systems work. We visited the terminal at Broad St (CD); we visited the cable terminus at Rockaway Beach (Hammel HM). We visited Commercial Cable's New York offices and their terminal at Far Rockaway. We even spent time at RCA's telegraphic terminal in New York. About June 4 we were told that we were leaving immediately and on June 6, 1944 we were on board the Queen Mary bound for Europe, but we still did not know anything about our organization. The names of the people follow:
We arrived in Cheltenham, England and found that we were part of the 3104th Signal Service Battalion, which was to supply communications for SHAEF (Supreme Headquarters Allied Expeditionary Force) and for COM Z (Communication Zone).
The TO (Table of Organization) said that we were two cable teams of one officer and four enlisted men each. In actuality we were a single team and had ten people because we had to operate a cable station twenty-four hours a day, seven days a week. About the fifth of July we were moved to Southampton and on the sixth we arrived on Omaha Beach. We were stationed at Valognes along with COM Z.
Again we waited because the army had not yet broken out of Normandy. Meanwhile the transport people were moving a truly immense amount of supplies and storing it in many depots, which consisted of fields with very large piles of crates. It is hard to describe the magnitude of these supplies. We marked time helping out where we could but we were sleeping on the ground and for three or four weeks ate K rations.
About mid-August we started to look for our supplies, which consisted of 139 large wooden crates randomly stored in the aforementioned depots. All of the crates had a code marking on them and you found them by walking down aisles looking at each crate. It took about five weeks and we found 138 crates. Fortunately the 139th crate had paper supplies which could be found elsewhere.
I should mention that during this period the one officer (name unknown) had caused a great deal of trouble. He then decided that he wanted to go to the front lines so he could be part of the action. His request for transfer was honored almost instantly and when he left the enlisted men (and some superior officers) stood and waved goodbye. This left the team with one officer and eight men.
About mid September the cable landing site was identified as the beach at Urville Hague, a small town about eight miles west of Cherbourg. The lieutenant and the author of this paper surveyed the site for suitability. The area was a gigantic mine field and had to be cleared. An Army mine clearing company came in and removed the following from the area that we would occupy:
35 Mustard Pots
They missed two mines, one of which blew up one of our trucks and badly injured the truck driver who was evacuated to England and then to the United States. The other mine that they missed, one of our fellows stepped on, but fortunately it failed to detonate. Needless to say you could not go outside of the limits of our site. Later many people were killed in this area.
Our time was occupied in moving our equipment, arranging for buildings to be built, obtaining diesel generators, and two diesel mechanics etc. The English Army had established a cable landing about one half-mile west of us on the same beach and we began to cooperate with each other.
About this time the army got around to ranking the various personnel. Lt. Williams was promoted to 1st Lieutenant, Addison Sheckler and Walter Herman were master sergeants, Jerome Saffer and one other (unknown, but probably Gino ("Rick") Riciardelli; see note above) were Technical sergeants and the remaining four were staff sergeants.
On October 20, 1944 the cable ship John W Mackay arrived at Cherbourg and Lt. Williams and Sergeants Sheckler and Herman went aboard to discuss the cable landing. The cable was the loaded cable 1 HO from New York to Horta, Azores, and then on to Emden, Germany. The cable had been cut in the English Channel at the beginning of the war and the intent was to pick this up and bring it to shore for our use. The cable was landed on October 23 and the splicing at sea was completed on October 27. We established communications with Horta, Azores, by key and siphon recorder.
The operations building was an army pre-Fab and was complete enough for us to go ahead by November 5. The author was the only person on the team with engineering experience, and Lt. Williams and he were the only ones with electrical experience. Lt. Williams had been a telephone installer for Illinois Bell. Western Union's training had been excellent and each person was given his assignment for installation and wiring and in one week the station was complete. However no planning had been done for the power wiring other than to supply the three Hill diesel generators. These were installed at the same time as the other work but there was no transformers or power cable available. We found the transformers someplace and installed them. The British Cable Station down the beach from us had salvaged about 1,000 feet of heavy power cable from a German command center also on the beach. We "liberated" that cable and completed the station. For a short while there were hard feelings between us and the British soldiers but we all got over it.
Now came the hard part of getting the station working! We had been trained well but something had not been taken into consideration. The Horta - Emden cable was special in that it was inductively loaded. There was no loaded cable available to splice on and bring ashore so they used a fourteen-mile piece of unloaded cable. This unfortunately created an impedance mis-match at the splice and the middle frequencies of the cable signal were reflected back. This in turn effectively created a hole in the cable's frequency band-pass response. We had been trained in cable signal correction but this only applied to linear corrections. The author was the only one who truly realized what the problem was and consequently was the one who had to solve it. We got the system up and sort of operating by November 30, but the error rate was intolerable. It took two more weeks to find a compromise solution to the problem but finally on December 13 we were successful. We had lowered the error rate to about 1 in 10,000 and while we could still see this the user could not and the station continued at this rate from that time on. At this time the station was handling military traffic on A and E channels. C channel was reserved for cable management. Later D channel was extended to Cable & Wireless in London and B channel was devoted to the British Foreign Offices Delegation at the organization of the United Nations. The cable had a capacity of about 350,000 five-letter groups
A description of the equipment is in order. Western Union had three special (i.e. loaded) cables 1 HO (Horta), 4 PZ (Penzance), and 2 HO. I HO was linearly loaded and operated at 32.5 Hertz. 4Pz was linearly loaded but shorter and operated at 55 Hertz. 2 Horta was taper loaded and operated duplex at 30 Hertz. It should be noted that modern cables all operate with single fill-in i.e. the data rate is twice the bandpass frequency. For more information the reader is referred to Lord Kelvin's treatise on Telegraphy.
These special cables all had special terminal equipment, which was built for the cable when it was installed. There was very little extra equipment and in our case there was one duplicate distributor and one duplicate tuning fork (clock) in New York and these were the units which we used. This left no spares anywhere. There were no spare amplifiers and the ones in use had been built by Western Electric in the 1920's. Consequently Western Union refurbished the duplicate equipment and designed and built new power supplies and a new amplifier (which became their new standard). The station consisted of a 561A amplifier (with cable shaping) and power supplies. This fed into a five-channel distributor, which was capable of reversing direction automatically. The distributor was driven by a temperature compensated tuning fork identical to those in Horta and New York. The time standard was set by New York; every five letters a time signal was sent and Horta and our station corrected our time to be correct with New York. The signal left the distributor or arrived at the distributor through peripheral signal processing equipment. Signals coming from New York went to five tape perforators which also printed on the tape. This tape then went to a tape reader, which sent the signal out on land lines. Incoming land line signals (which were in Baudot code) went to an electrical/mechanical translator which converted the Baudot code to cable code which then went to the distributor and out to New York.
The station's code designation of RM came about in an interesting manner. When we were having trouble shaping the amplifier we were being sent signals from Horta (HO) and lots of communications. They addressed us as ARMY. The error rate was so high that this often came out RM so we adopted that call.
Maintenance procedures were set up and by the end of December 1944 the station operated without any difficulty from that time on.
A tape relay station, JEAC, was appended to the facility and additional personnel were brought in to operate it. This and subsequent land line operations are covered in Lt. Williams' report (appended). The overall manpower eventually rose to about thirty military personnel, about twenty French civilians, and forty German prisoners. Interestingly the German prisoners were not guarded and we had no fence but we had no trouble of any kind. Apparently this was a safer, more comfortable place than any other they could imagine. They supplied the facility labor and they made the place into a very good looking military camp. The author had many conversations with individual Germans and none of those with whom he spoke wanted to go back to Germany until the war was over.
Lt. Williams eventually got bored since there was very little to do and he asked for a transfer. After him we had a progression of officers who had no knowledge of the facility and who spent their time making the camp more elaborate. Some of this is shown in the appended photos.
Unfortunately the original crew were witnesses to the sinking of a troop ship December 24(?) just off Cherbourg Harbour. A very large number of people died, perhaps as many as 1,000.
The station operated with no trouble from that time on. After VE Day (8 May 1945) the 3104th Signal Service Battalion was sent to the Pacific, but we were left behind because the station continued. The battalion was on board ship entering the Suez Canal on VJ Day and they received orders to turn around and go to New York. They were some of the first people discharged after the war. We continued on and the army in November 1945 asked the author to stay on as civilian station chief at an amazingly high salary. The author refused the position and another man, Daniel Crean, was recruited for the position. Crean arrived about November 15 and had about two weeks to get acquainted with the station.
A final anecdote is in order. Mr. HF Wilder of Western Union, Mr. Breyfogle of London office Cable & Wireless, and Lt. Col. Allen Wharton of War Department all were present at the landing of the cable. Mr. Wilder took home a detonator from a Bouncing Betty mine - a spring snap action device - and when he got home he got in the habit of snapping this device while reading at his desk. Suddenly after thousands of snaps the percussion cap exploded and blew a small hole in his desk top. Fortunately no one was hurt but he had a permanent memento of the cable station.
During 1945 and 1946 Western Union brought together all of the technical documentation and built new equipment to replace that which we used. The author used this documentation for his senior thesis at Drexel University. The documentation still exists.
Technical history of transatlantic cable head by Orville H. Williams, 1st Lt. Sig C.
The cable was landed at Urville-Hague about seven miles west of Cherbourg on the morning of October 23, 1944. The landing was made from a harbor barge under the direction of the crew of the cable ship John W. Mackay and Lt. Col. Allen Wharton of War Department. After the shore end of the cable was landed the cable ship left to complete the off-shore splices.
On October 27th the splicing was completed and contact was made with Horta, Azores by siphon recorder and cable key.
The operations building was sufficiently completed on November 5th to start the installation of equipment and by November 12th the equipment was installed and wired. The rest of the installation was spent in adjustment of equipment and shaping of the amplifier. The adjustments of all the equipment was completed on November 25th; however the amplifier was not yet shaped.
The amplifier shaping started on November 14th and due to an irregularity in the cable characteristics was not finished until November 30th.
On November 30th the final tests were made with Horta, and at 2130 the circuit was extended to CD in New York. Due to some engineering discrepancies the next few days were spent aligning the circuit between New York and Urville-Hague. On December 4th the circuit was connected through to JEJE and JEAR on A and E cable channels respectively. After a day's tests the circuit was put up for traffic and traffic began between WAR and JEJE and WAR and JEAR on December 5th.
In the period from December 5th through January 4th 1945, operations were normal with two exceptions. One of the troubles experienced was with the amplifier shape, however this was cleared completely on December 13th. The other trouble in this period was with landlines and this persisted until about February, when the wire conditions were much improved.
On January 4th a temporary installation of the tape relay station JEAC was completed after about two days work. This installation consisted of typing reperforators and transmitter-distributors terminating the land lines and the same type of equipment terminating the cable channels. This station, along with WFQ at New York, allowed full usage of the cable channels and also served to absorb the time lag between the cable and the landlines.
An alternate traffic route, terminating at JEAC and JETA, was installed on January 8th.
Work was being done on a permanent installation of JEAC during this period and on February 2nd the work was completed, and JEAC went into permanent operation.
Another alternate traffic route was installed on February 28th, terminating in JEAC and JEVL.
On March 6th, a new circuit was put into operation on the cable between Cable and Wireless London and Cable and Wireless Fayal, Azores on the D cable channel.
On March 8th the alternate route between JEAC and JETA was discontinued.
On March 15th a circuit between JBJB and JEAC was installed for the purpose of giving JBJB an alternate route to WAR.
After several tests a circuit was started up between the British Foreign Office in London and the British delegation at the San Francisco conference. The circuit was put through the "B" cable channel for traffic on April 19th, and continued in service until June 2, 1945.
The full capabilities at this cable have never been utilized and are approximately as follows; 70,000 groups total for one channel in a twenty four hour period and 350,000 groups total for all five channels in a twenty four hour period.
Text and images copyright © 2004 Addison C Sheckler. Used by permission.
Editor's notes: The German cable diverted to Normandy in 1944 was the Azores - Emden connection to Western Union's Azores - Newfoundland cable of 1926. This 1926 cable, identical in construction to the 1924 cable, was linearly loaded.
According to an article in Underseas Cable World, Vol 1, No. 10, April-May 1968, the 1926 cable was once again diverted in 1960, when it was restored to its original path. The German Atlantic Telegraph Company relocated, re-connected, and re-established operation of this cable, which then once again carried telegraph signals from Borkum to Horta, Azores.
For a technical description of the 1928 cable on this same route, which used taper loading and was the first high-speed duplex long distance cable, see the 1931 article: The Newfoundland-Azores High-Speed Duplex Cable.
Postscript: An article in the New York Times, Sunday March 31 1957, is evidence that Cal Sheckler's painstaking work paid off.
Note, however, that the cable was not used for trans-Atlantic telephone calls in 1944, nor subsequently; this capability was not available in trans-Atlantic cables until TAT-1 in 1956.
Don Morton adds this further note on the later history of station RM:
I was stationed at RM from mid 1951 through December 1952. I was employed as an engineer at Motorola Government Electronics Division in Phoenix, Arizona, when I was drafted in Jan 1951. I had my First Class FCC Phone license and had worked my way through college working in broadcast stations. I had my amateur radio license (W7LBN) which I still have, as well as my degree from Arizona State.
I was sent to basic training for 14 weeks of infantry training and then to Ft. Monmouth New Jersey for assignment in the Signal Corps. I remember the interviewer saying, "You have a ham license so you can copy code?" to which I agreed. His next question was "How would you like to go to France?". I know I happily replied yes, as the alternative was probably Korea and combat areas.
A number of us were sent to school at Western Union's offices at 60 Hudson Street in New York City for several weeks, then to Rockaway Beach to work on the U.S. end of the cable with Western Union technicians. Then we were sent to France. I think there were six of us—two of us were draftee engineers; another had been working for RCA and I knew him from college and ham radio contacts.
When we arrived at RM, or Cherbourg, the cable had just been cut at sea and the techs on hand were trying to make measurements and determine how far out the cut/break was located. When everyone else went to eat the other engineer and I took a look at the test equipment, made some measurements, calculated a bit and figured out about where the break was located. When the regular personnel came back from eating we gave them our data, they contacted the cable repair ship and they proceed to locate it, pick it up and repair it. Instead of using Morse code to communicate with the ship we used GI field telephones since the break was only 8 or 10 miles off shore.
For routine operation we also put code oscillators in parallel with the telegraph clackers because we were much better at reading Morse from tones than clackers. We later had a break off the Azores that took a while to fix as the cable ship reported it was probably 21,000 feet deep at that point.
We understood at the time that the U.S. had no cable ships and the British did the work at sea for us.
When I was there we were part of 7774 Signal Battalion, headquartered in Germany. Our on-site commanding officer was Maj. Allison D. Melvin, now retired and living in Sierra Vista, Arizona, the last I knew.I arrived there as a private and left as a Staff Sergeant and went back to work as an Engineer at Motorola, and as a civilian. I learned something about multiplex, saw a lot of equipment that seemed like it should be in a museum but it got the job done and we kept it working.
Last revised: 6 June, 2018