Today’s date, 6 August 2020 marks the 75th Anniversary of the atomic bombing of the Japanese city of Hiroshima. Some where in the neighborhood of 70,000–80,000 people in Hiroshima were killed by the blast and resultant firestorm that reached it’s peak three hours after the detonation. Japanese military personnel made up 20,000 of the 70,000–80,000 immediate deaths. This bombing set in motion a train of events including the subsequent atomic bombing of Nagasaki, the Soviet Union’s accelerated invasion of Japanese occupied Manchuria on 9 August 1945 and Emperor Hirohito’s 15 August 1945 broadcast of Japan’s surrender under the terms laid out by the Potsdam Declaration.
Much has been written on these events and I’ve revisited them here on Chicagoboyz annually from 2011 to 2018. This year, 2020, I’m going to address a different part of the Atomic attacks. Namely, how the American military electronically communicated about the Atomic bomb. How the secrecy and limitations of that communications system meant Admiral Nimitz knew about the Atomic bomb long before General MacArthur. And how General MacArthur was working to change that for the proposed and cancelled by A-Bomb invasion of Southern Japan
AMERICA’S SECRET TALKER
In World War 2 many of the major powers developed strategic level code & cypher radio electronic communications systems between it’s top level political & military leaders and the various theater commanders. The German Geheimschreiber (secret writer) is the best known of these systems because British crypt-analysts at Government Code and Cypher School (GC&CS) at Bletchley Park with the the aid of eventually ten Colossus computers.
Much less well know is the Anglo-American equivalent of the German Geheimschreiber, The US Army Signal Corps and Bell Telephone Laboratories SIGSALY. This system was the only form of secret broadcast radio-electronic communications the American and British government trusted to transmit information on the Atomic bomb in the World War II. It was due in large part to that level of communications security that Admiral Nimitz was informed of the atomic bomb before General MacArthur. Admiral Nimitz in Hawaii and later Guam was reachable by SIGSALY after his initial courier briefing. General MacArthur between October 1944 and May 1945 was not, for a number of reasons I’ll get into a little later.
First, a quick introduction: SIGSALY was a highly secret WW2 digital voice communications system that used a special one-time pad encryption. There were only 12 station made in all of WW2 and MacArthur’s had two. The first in Brisbane was sent to Manila. The 2nd SIGSALY meant for Hollandia was instead placed in a Australian built barge barge in the SWPA “Signal Corps Grand fleet,” a motley collection of small ships and barges with powerful Signal Corps radios. The barge mounted SIGSALY was intended for quick sea movement and it was key for MacArthur’s communications at Okinawa and Kyushu during the planned invasion of Japan.
Figure 2 – This is a SIGSALY digital radio-telephone system screen captured from the Crypto Museum web site.
This is how the Crypto Museum describes SIGSALY.
SIGSALY Ciphony 1
Digital voice encryption with OTP
SIGSALY was a digital speech encryption system, developed by Bell Telephone Laboratories (BTL) 1 in the US in 1941/1942, and built by Western Electric in New York (US) in 1943. The system went into service in April 1943, just two months before the invasion of Italy, and was used until at least 1946.
SIGSALY was used heavily during WWII, in particular for confidential talks between British Prime Minister Winston Churchill and US President Roosevelt. The system used the highly-secure One-Time Pad (OTP) encryption and is known under various names, including The Green Hornet.
SIGSALY featured a number of innovative digital communications concepts, including the first transmission of pulse-code modulation (PCM). 2
SIGSALY was completely built with vacuum tubes (valves). A single system consisted of more than 30 full-height 19″ racks, plus 4 synchronisable turntables. It weighted 50,000 kg, consumed 30 kW of power, and had special air conditioning requirements. A single SIGSALY terminal had a price tag of US$ 1 million in 1943. In total, 12 SIGSALY terminals were set up around the world, the first of which was installed at the Pentagon.
WHY SIGSALY WAS CALLED “THE GREEN HORNET”
As a vacuum tube technology digital electronic system, SIGSALY was in some ways both far ahead of its time and in others crippled by it. For example, to create it’s “one-time pad” encryption, recordings were made of the heat decay of a vacuum tube — it’s buzz — and pairs of these audio recordings were sent out between various stations. When a radio-teleconference one station started playing it’s record and the other played their until the signals synchronized on an oscilloscope. Once a match was locked in the teleconference started. After it ended, both records were destroyed. This made the system invulnerable to code cracking with the tools of that time and it would be nearly as hard to break in the 21st century.
On the other hand, there were some really large limitations which can be sussed out from the following passage from Crypto Museum article.
SIGSALY provided a full-duplex voice link via narrow-band HF radio channels in the Short Wave (SW) radio bands. Each half of the link used 12 individual data channels, or carriers, through which the data was sent by means of digital 6-level Multiple Frequency-Shift Keying (MFSK).
The human speech was analysed just 50 times per second (at 20 ms intervals), broken down into its characteristic parts, and then coded and sent across the Atlantic. At the receiving end, the data was decoded and used to reconstruct or synthesize the original human speech again.
This resulted in a low data rate (comparable to 1500 baud today) but made it very difficult to recognize the person at the other end.
SIGSALY used short wave/high frequency radio bands to push a low bandwidth digital signal through the Ionosphere. You had to compute the correct direction, radio antenna critical angle and frequencies at the correct part of the day or night to get the proper distance “skips” for a radio connection. Since SIGSALY used 12 data channels with different H/F carrier frequencies. This required as many antenna’s critical angles to be adjusted for their different channels. See the figure immediately below –
There were two other concerns affecting the quality of digital data packets. H/F bands are naturally static filled and some places on earth are more naturally static filled than others. Between the data distortion of normal static and the added one time pad static from vacuum tubes, the voice that came out of SIGSALY had the buzz of a hornet. Given the green of may Army phone handset, the name “Green Hornet” was a natural name given the annoying sound.
On the upside, this natural static is affected by sunspot activity. The more sunspots there are, the more H/F transmissibility was increased in daytime because the solar wind compresses Earth’s magnetic field thus lowering H/F static.
The issue for the SIGSALY machines generally was that 1943-1945 were in the low sunspot part of an 11-year solar cycle. The added static from the low sunspot activity degraded digital data packets. This affect was intensified in the two SIGSALY serving in the South West Pacific Area (SWPA) because they were in one of the worse areas for H/F because of the local magnetic field effects.
From a lot of the SIGSALY data at Archives II:The ALGIERS installation was begun on 12 August 1943 and placed into service in October 1943. If we assume it went online on 1 October, that’s 50 days to set up a SIGSALY terminal that was already nicely packed for transport.Another example was Terminal #9; the OL-31 installation.May 1944 Arrives in SWPA intended for installation at Hollandia, New Guinea.7 JUL 1944 – Equipment and personnel arrive at Brisbane, Australia1 NOV 1944 – Installation onto OL-31 begins.1 FEB 1945 – OL-31 installation completed. (92 days to complete)17 FEB 1945 – OL-31 leaves under tow. Hits a coral reef under tow, shaking things up and causing a stop to inspect for damage.12 APR 1945: Arrives at Hollandia17 APR 1945: Leaves Hollandia for Manila10 MAY 1945: Arrives Manila. Due to harbor difficulties it takes 11 days to find a suitable docking site for OL-31.21 MAY 1945: During the night of 21 May 1945 at the assigned site, the power/Air conditioning system went out of service due to the low tide causing OL31 to settle on the bottom of the river, clogging the circulating system’s intake. OL31 had to be towed to deeper water and the recirculators cleaned out.23 MAY 1945: OL-31 Terminal #9 turned on for first time in 10 months since it’s arrival in SWPA.23 JUNE 1945: OL-31 Terminal #9 removed from service.Meanwhile in Manila…10 MAY 1945: Construction of quarters at CITY HALL for Terminal 4 is completed.19 MAY 1945: SIGSALY Terminal #4 equipment arrives.20 June 1945: SIGSALY Terminal #4 enters service. (32 Days to complete)So we have the following times of setup for SIGSALY terminals:50~ days for ALGIERS92~ days for OL-3132~ days for MANILA
LIVING WITH THE OL-31
Several members of the 805th Signal Service Company — established by General Marshall to run all 12 SIGSALY world wide — have documented their experience aboard the OL-31 ship and provided an insight into what service aboard this floating terminal of SIGSALY was like. The saga of the OL-31 and this communications flotilla can best be told by two of the men who were there in their own words. They are 805 Signal Service Company Sergeants Bud Brown and Curtis Martin. While overseas experience was a memorable time in any serviceman’s life, their’s had more “SNAFU” (Situation Normal, All Fouled Up) than most with the snake bitten OL-31.
Following is a report that Sergeant Bud Brown wrote about his experiences on OL-31.
“I traveled on Ocean Lighter 31 with several of the Detachment 9 personnel from Manila to Tokyo during August and September 1945. Because the OL’s were not self-propelled, they were towed by three ocean-going tugs, each with a string of two or three OLs. The flotilla was escorted by two destroyer escorts. When we got to the open sea there were always three or four dolphins leading us at the bow of the ship and since the deck was only three or four feet from the water, flying fish easily landed on the deck. Our first stop was at Okinawa where we were moved into Katina Cove to ride out a typhoon that was in the area. On leaving Katina Cove, the Manila rope that connected us to the tug snagged on the bottom and had to be cut off and replaced by a smaller steel cable. From there the ride was rough because steel cable does not give like Manila rope and we felt every wave that hit the tug. As we proceeded, one of the OL’s in another string broke loose. A destroyer escort had to go back and retrieve it and was gone a couple of days before it appeared again towing the OL. Doing a tugs job seemed beneath the dignity of a destroyer. We arrived in Tokyo Bay opposite Yokohama on September 10, 1945, and were overwhelmed by the number and sizes of the U. S. Navy ships in the bay. We didn’t stay long, for a picketboat with a Japanese pilot was added to guide us through the minefields to Tokyo harbor where our trip ended alongside the dock. The Japanese indicated that the OL’s were the first U. S. ships in Tokyo Harbor.”
Sergeant Curtis Martin describes his experiences onboard OL-31.
“Although referred to as a barge, OL 31 had the configuration similar to an oil tanker with superstructure aft. We were towed by a large tug with approximately a mile-long cable. There were three tugs in all with two tugs pulling two OL’s and the other pulling one. Travel was very peaceful with dolphins, flying fish, barracuda, etc. during the day and the phosphorescence moon and stars at night. It was hard to believe that a war was going on. If my memory is correct, there were four other ships in the Seaborne Communications Fleet. These included the transmitter and receiver ships. We had very little contact with the others on these ships. They were very curious about the “Green Hornet” as our signal sounded to them.
“Installation of our equipment on OL 31 was a challenge. Sergeants Durkin, Fruman, and myself, none with any experience in air-conditioning, built all of the ductwork and made the installation. We did have precise installation instructions from the York Corporation and it did work! We probably had one of the very few air-conditioned workplaces and sometimes living quarters in the South Pacific! Our power was furnished by three 50 kW diesel generators. We all had to learn to phase in the engine when switching from one to another. Our destination on leaving Okinawa was Tokyo. Fortunately we arrived after the surrender. The gathering of warships in Tokyo Bay was quite a sight.
“On July 7, 1944, the equipment and operating personnel arrived at Brisbane, Australia. The installation, which was in a vessel designated the OL-31, started on November 1, 1944, and was completed on February 1st. It left Brisbane on February 17 under orders to be towed to Manila. Somewhere off the northeast coast of Australia in the great Barrier area, the OL-31 was towed across a submerged corral reef causing some damage to the hull and shaking up the equipment considerably, necessitating readjustment and repairs. The vessel was delayed at Port Moresby and Ora Bay, New Guinea, for examination of the damage to the hull. It did not prove to be serious enough to warrant repairs, therefore, it continued north. The OL-31 arrived at Hollandia on April 12, 1944 and left on April 17, continuing to Manila. On May 10, the terminal arrived at Manila. Difficulties were experienced with the harbor authorities and it required 11 days to obtain a suitable site for docking the OL-31 vessel so that it could be placed into service. The site was assigned along the sea wall on the south side of the Pasig River just west of the Jones Bridge. On the night of May 21, the power and air-conditioning system went out of service due to the receding tide that permitted the OL-31 to settle on the bottom of the river. This permitted the circulating system’s intake to take in mud and sludge from the riverbed and clogged up the entire circulating system. This necessitated moving the vessel to deeper water and rehabilitating the circulating system.
“On May 23, 1945, Terminal 9 on OL-31 was turned up for service for the first time 10 months after its arrival in the theater. It was removed from service at Manila on June 23rd.
“Terminal 9 is the unit that was sent to the SWPA (Southwest Pacific Area) in May 1944, originally intended for installation at Hollandia, New Guinea. However, it was included in the sea-borne communications plan for SWPA and installed on a small barge referred to as an ocean lighter, which is a small barge type of vessel not capable of self-propulsion. The terminal installation was confined to the hold of the vessel occupying the forward two-thirds with the balance of the space aft devoted to power and air-conditioning equipment. Because of the limited space and particularly head-room, it was necessary to construct a false deck in order to provide sufficient clearance for the equipment bays and this only was possible by stripping the equipment of most of its cabinets and revising such other items so as to reduce the overall dimensions. Considerable modification was necessary in the air-conditioning system, outstanding of which was the Freon condensing and cooling unit which was a constant source of trouble since the installation was completed. It was learned from the Maritime people at Brisbane that the regular condensing and economizer unit normally used with the York air-conditioning system could not be used because of limited space. It would have been necessary to make this installation on deck which would have made the vessel top-heavy and rendered it unseaworthy.
“As a substitute for the cooling and economizer unit, a special heat exchanger was built in Sidney, Australia for this installation that was capable of circulating salt water for condensation purposes. This was the basis of the air-conditioning system’s trouble because the Freon circulating coils were not designed for exposure to salt water. This resulted in electrolytic action to the coils and resulted in decomposition of the lines and leaks. The air-conditioning system capacity was 190 pounds. In four months over 1100 pounds of Freon were required to keep the system in operation despite the fact that the modified system only required 60 percent of the Freon as other terminal systems and the space to be air-conditioned was only 50 percent of that of other systems. The conference room for this terminal was only approximately 5ft. by 8 ft. with a 6 ft. ceiling. It was located on the main deck, was not air-conditioned, and was very poorly ventilated.”
Sources and Notes:
AMERICA’S WORLD WAR II UNBREAKABLE CODE FOR SECRET TELEPHONY
THE U. S. SIGNAL CORPS’ SIGSALY SYSTEM
BRIGADIER GENERAL SPENCER BALL AKIN CHIEF SIGNAL OFFICER, GHQ, SWPA IN AUSTRALIA DURING WWII
J. H. DELLINGER, FELLOW, I.R.E., AND NEWBERN SMITH, SENIOR MEMBER, I.R.E “Developments in Radio Sky-Wave Propagation Research and Applications During the War, ” PROCEEDINGS OF THE I.R.E.-Waves and Electrons Section, Feb. 1948, Pages 258 – 266
SEABORNE COMMUNICATION UNIT AND MOBILE COMMUNICATION UNIT IN THE SOUTHWEST PACIFIC AREA DURING WWII
SIGSALY Ciphony 1
Digital voice encryption with OTP
SPSCC-11 17 December 1945
MEMORANDUM for the Chief, Communications Engineering Branch
Subject: Records for the Post-War Organization – Development of Special Conference Systems.
Found at National Archives II, College Park, MD; RG 111: Records of the Office of the Chief Signal Officer, 1860 –1985;
Series: SIGSALY Project Files, 1942 – 1948
ARC ID: 6949922
HMS/MLR: UD 1029
Appendix 1 – HISTORY OF THE SIGSALY TERMINALS IN WW2
|TERMINAL NUMBER||LOCATION||IN SERVICE||REMARKS|
|1.||London*||July 1943||Removed from service 31 Oct 45 and returned to U.S.|
|3.||Algiers||Oct 1943||Removed from service April 1944 and returned to U.S.|
|Removed from Brisbane Jan 1944.|
|6.||Frankfurt||June 1945||This terminal first sent to Caserta, Italy, but not installed. Reorganization in theater. Moved it to London where it was held pending invasion of the continent.|
|7.||Washington||Mar 1944||Due to increased load and time differentials a second terminal for Washington was required.|
|9.||Barge OL-31||This terminal installed on a ship by SWPA at Brisbane, Australia to facilitate mobility in following the movement of GHQ’s progress north. This terminal was ready for service and first used at Manila, May 1945 and remained in service until Terminal No. 4 was brought up from Brisbane and permanently installed at Manila. From there it was taken to Tokyo and placed in service in Oct 1945 pending the installation of a permanent station there, which is presently in progress. (This permanent station will be Terminal #11, removed from Oakland, Calif).|
|10.||Guam||Mar 45||Installed for the Navy.|
|11.||Oakland, Calif||Oct 44||Removed from service 31 Oct 1945 and shipped to Tokyo for permanent installation there (now in progress)|
|* These terminals have extension units, making it possible to render SIGSALY service from a designated remote point. The locations are as follows:
No. 1 War Cabinet Office London, for the British-Great George Street
Nos. 2 and 7
No. 5 ” ” CINCPAC,Pearl Harbor –for the Navy
No. 10 USSTAF HDQ Guam –for the Air Forces