PDA

View Full Version : Naval Codes & Codebreaking: Enigma, Room 40 & Other.


Harley
25-06-2008, 00:04
So that the Dardanelles thread doesn't go any further OT due to my incessant questioning, I propose this thread be the main one on naval codebreaking in the future, covering everything from Room 40 to Enigma.

One note for interested persons. Contrary to very popular history, SMS "Magdeburg", the German cruiser from which Britain ended up with the "Signalbuch der Kaiserlichen Marine" (SKM), actually yielded three copies of the SKM and not the two which is commonly imagined, or thanks to Churchill the one waterlogged copy of yore.

It is usually written that "Magdeburg" was carrying three or four SKMs. It was four. It is agreed that one was destroyed. One was flung overboard - also accepted, this is the waterlogged copy the Russians ended up with. Another was discovered in a perfectly dry safe on the sea-bed where it had been thrown overboard by the Russians. This is the copy which wound up in London, the Russians correctly ascertaining that it would be of greater use to the British. So much for Churchill waxing lyrical about a book found clutched in the arms, "rigid with death" of a German. The fourth SKM was supposedly taken overboard by a radio operator who swam off and most accounts say was never heard from again, and neither was the book. I have no idea where some of this tripe comes from. To set the matter straight, the sailor in question was named Neuhaus. He survived the jump from "Magdeburg" and swam to Odensholm, dropping the SKM "en route", and became a Prisoner of War. This "lost" SKM was recovered by the Russians at the same time as the other two, in a large diving operation in which powerful electric lights were used to illuminate the seabed - the Russians knew what they were looking for.

Harley

historydavid
25-06-2008, 00:28
Churchill was always full or overful, some might say, of hyperbolae.

It's a good thing the code book wasn't recovered during Uncle Joe's time or we would never have heard of it. Mind you, we didn't let on that we had cracked the German Enigma codes.

Pity really, because he might have listened when we told him Russia was going to be invaded, and that would have saved many lives and ships that were lost on the Arctic convoys.

Harley
25-06-2008, 00:54
Speaking of code-breaking, this article makes interesting - and slightly disconcerting - reading;

http://findarticles.com/p/articles/mi_qa3926/is_200501/ai_n13244764/pg_1?tag=artBody;col1

A very brief overview of the German army's code-breaking effort in WWI - the Bavarian Army was rather successful in cracking certain Royal Navy codes it would seem.

Harley

iantait
09-05-2011, 15:17
Today is the anniversary of the capture of the Enigma coding machine from the German U-Boat.Ceremonies are being held in Bletchly Park

emason
17-05-2011, 18:40
Prologue

This introductory posting is to be the first in a series about the German Enigma cypher machine of WW2, its development and usage; the cryptanalytical effort at breaking it, the consequences and the usage to which the decyphered messages were put. It will not be a detailed description of the machine itself, nor of the methods used to attack it, except where necessary to understand the problems encountered and how they were overcome.

This first article is about the pre-war development of electro-mechanical cypher machines in general, with emphasis, naturally, on the Enigma machine. Later articles will be primarily about naval Enigma messages and the efforts made to decrypt them.


Pre-WW2 Cypher Machines


Codes and cyphers have been used for hundreds, possibly thousands, of years to pass messages secretly, and attempts to intercept and decode these have been going on for just as long. Hand ciphers can be secure, especially the one-time pad which is unbreakeable. But the more secure a hand cypher was, the longer it took to encypher and decypher. With large amounts of messages, this hand method would become impractical. The answer was a cypher machine. These started to appear in the late 1800s; the earliest patent for one was filed for in 1890, but they were not very secure and were unreliable.

With the advent of wireless communication, which can easily be intercepted by anyone, with the resulting necessity for security, it became essential to develop secure communications for both military as civilian use that was both quick and easy to use, and above all secure. Coincidently four such machines were patented in 1918 and 1919 by different persons in different countries in the space of about 18 months.


Early Cypher Machine Inventors

Edward Hugh Hebern (American), developed a cypher machine with rotating disks, each disk performing a substitution cypher. He filed a patent in 1919 for his Electric Code Machine and started a company to manufacture and market them, but didn’t sell a machine until 1926. The US Navy bought several systems in 1931.

Hugo Alexander Koch (Dutch), filed for his rotor machine patent on 7 October 1919, and was granted a patent, but built no machines. Koch sold his patent to Arthur Scherbius in 1927.

Arvid Gerhard Damm (Swedish) filed for a patent on a rotor machine on 10 October 1919. After his death in 1925, his company, AB Cryptograph, was taken over by Boris Hagelin, who moved the company to Switzerland at the start of WW2 and is still in existence, as Crypto AG. Their C-38 model was adopted by the US Army in 1938, who renamed it the M-209, and was made under license by Smith and Corona Typewriter Co. who, by the 1960’s had produced about 125,000 of them, costing only $64 each. It wasn’t very secure (the Germans were able to decipher it quite easily), and so was only used for low level field traffic.

Arthur Scherbius (German) filed for a patent on 23 February 1918 for a cypher machine based on rotating wired wheels which he called ‘Enigma’. Scherbius, with Richard Ritter, started the company Scherbius and Ritter, but in 1923 assigned the patent rights to Chiffriermaschinen-AG, and he and Ritter became directors. But Scherbius didn’t live to realise the fruits of his invention; he died in an accident in 1929.

The problem that all these machines had initially, (except Koch who didn’t make any) was that no-one wanted to buy them. It seemed that a solution had been found to a problem that no-one realised they had, until the late 1920’s.


Principles of Early Cypher Machines

The simplest cypher is a merely a letter substitution method (for example, ‘A’ becomes ‘T’; ‘B’ becomes ‘G’, etc) and can be easily implemented in an electrical device. But it can be quickly broken by statistical analysis, ‘E’ being the most common letter in English.

With variations, the machines used a rotor with 26 electrical contacts on each side, each one representing a letter of the alphabet. The rotor’s internal wiring was such that the input contacts were not connected to their corresponding output contact, but to a different one. For example, the ‘A’ input contact could be connected to the ‘P’ output contact, and so on for all the others, thus implementing a substitution alphabet electrically. If that was all there was to it, there would be no advantage.

But, if the rotor is made to rotate one position after a letter is entered, the rotor’s internal wiring could now connect the ‘A’ input to the ‘X’ output for example. In other words a different substitute alphabet is used. In this way, 26 different substitution alphabets are used sequentially for each letter input, until the rotor has completed a full revolution and has returned to its initial position, and will now start to re-use the same 26 alphabets. With a sufficiently long message, statistical analysis can again be applied to the 1st, 27th, 53rd etc. letters of the message as these use the same substitution alphabet.

If the output from the rotor is passed on to a second rotor wired differently, and it is made to step once for every revolution of the first rotor, the number of alphabets used is now 26 x 26 = 676 without repetition. With three rotors the number is 26 x 26 x 26 = 17,576 alphabets, and statistical analysis can no longer be applied. Following on from this, the more rotors a machine has, the more secure it becomes.

This simplified explanation is the basis of most, if not all electro-mechanical cypher machines used in WW2.


The Enigma machine development

Scherbius had tried selling his design to the German Navy and Foreign Office, but neither were interested. In 1923, he assigned the patent rights to Chiffriermaschinen-AG, and he and Ritter became directors. The Company manufactured, advertised and exhibited the first Enigma Model A, a huge machine with a typewriter, weighing about a hundredweight (50 kilos). In 1925 a similar Model B was introduced.

The typewriter output persisted until the Model C was introduced in 1926, which used lamps instead to indicate the encyphered text. This model was very much smaller than previous models and had many technical improvements, including a reflector that took the output from the last rotor and reflected it back into the rotor and so through the other rotors. This neat addition meant that the Enigma machine could now be used to decode messages generated by other Enigma machines.

In 1927, Scherbius purchased the rights to Hugo Koch’s 1919 patent in order to secure his own patent. That same year, the Enigma D was introduced and commercialized in several versions with different rotor wirings, and sold across Europe to military and diplomatic services. Examples were bought by Sweden, Netherlands, Great Britain, Japan, Italy, Spain, United States and Poland. Also just about every branch of the German military, including the Abwehr, used the machine.

The German Navy had purchased a number of machines in 1926. The system was adopted by the German Army in 1929 and the Air Force in 1934. When the German forces decided to standardise on the Enigma for their communications, the machine was taken off the open market after which, a series of modifications were made which would greatly improve the machine’s security. The main improvements were:

1. A plug board was situated between the keyboard and the rotors and enabled from one to thirteen pairs of letters to be interchanged. The effect was similar to an extra static rotor being introduced, but much more flexible and with many more combinations.

2. Rotors with different wiring to the commercial machine were used.

3. Movable rotor rings allowed the ring around the rotor, on which the alphabet was inscribed, to be rotated with respect to its internal wiring. This meant that even if the initial position of the rotors had been found out, the information would be useless without the ring settings, as there were 17,575 (26x26x26) different combinations of these.

4. Two extra rotors. The normal Enigma had three rotors numbered I to III, each of which had different internal wiring. These rotors were interchangeable within the machine, giving only six (3x2x1) ways they could be positioned in the machine. Now with five rotors, there are 60 (5x4x3) ways in which three rotors can be selected from the five available. The naval Enigma used three more rotors, making eight from which three were selected, giving 336 (8x7x6) ways in which this could be done.

By the outbreak of war, as a result of these modifications, the Germans were confident that they had rendered it totally secure. Even if a machine were captured, the security arrangements and complicated procedure of setting it up initially would defeat any attempt to make use of it. Instructions for arranging and setting the rotors could be changed as frequently as every 24 hours. Anyone not knowing the setting was faced with the problem of choosing from 150,000,000,000,000,000,000 solutions.


The Typex Machine

In 1926 the British government purchased an Enigma machine (this was probably a Model B) after the machine was demonstrated at the Foreign Office. The War Office felt, however, that it would be too ungainly for use in the field. In 1928 two of the commercial Enigmas (probably the commercial Model D) were bought for examination by the Government Code and Cypher School.

In 1934, Wing Commander Lywood, of Air Ministry Signals, became interested in the possibilities of Enigma and asked if he might borrow one of the machines for study by his engineers. GCCS loaned him one with the warning that they were probably covered by patents if he was thinking of using them as a basis for a new cypher machine. They were in fact covered by three British patents, and Lywood was advised by the Air Ministry Contracts:

"In view of Section 29 of the Patents and Design Acts, the Air Ministry may by themselves, or by their contractors authorised in writing by them, make or use any patented invention for the service of the Crown, without any licence from the patentee. The department do not require the permission of the inventor to make his invention. The terms for such making and using may be considered after the making or using of the invention. It is suggested that the question of infringement be referred to the branch in order that any payment due to the patentee may be considered, and put on record. Apparently no action to settle the matter could be taken while the apparatus is secret."

Wing Commander Lywood and his engineers started work, and by 1936 had developed what became known as the RAF Enigma. In March, that year the War office, displayed an interest and raised the patents question again. Air Ministry Signals, informed Air Ministry Contracts: "The machine made for us was copied from the German ‘Enigma’ with additions and alterations suggested by the Government Code and Cypher School."

The full name for this machine was "The RAF Enigma with Type X attachments", but by 1937 it became officially abbreviated to "Typex" for convenience, and to disguise its Enigma origins as by now, with deteriorating German relations, to admit any patent infringement would reveal its existence and subject it to examination. From this point on, the question of patent infringement is left in abeyance, with the matter to be deferred, "until such time as it is perishable to negotiate with them".

From 1939, Typex was in use by the RAF, Army and government departments. With later models the operating was completely automatic. The message could be typed, up to about 60 words per minute, encoded and transmitted in one operation. The same is also true for the reception of encyphered messages.

[Note: The Royal Navy continued to use hand ciphers for operational messages up until 1943, when they were obliged to use the Typex machine in the form of a Combined Cypher Machine, for inter Allied communications.]

About 12,000 Typex machines of all models were produced by the end of WW2. Bletchley Park routinely used Typex machines, reconfigured as Enigmas, to decypher messages whose keys had been broken by the cryptanalysts.

A Typex machine without rotors was captured by the Germans during the Battle of France prior to Dunkirk. A senior German cryptanalyst Dr. Erich Huettenhain wrote "As we believe that the Enigma can not be solved, no great effort was made to solve Typex. Typex has seven wheels (rotors) and we therefore believe it to be more secure than our Enigma". So far as it is known, Typex messages were never read by the enemy.
 

 
 

Sean Roberts
17-05-2011, 18:55
Hi Bill.Your depth of research,and your willingness to share with us never fails to astound me.I know my late father (Steve) would always get me to read your posts,and busy as I am at the moment.This "Pongo" is in awe of your endless depth of research and sharing.Long may it continue.

Dreadnought
17-05-2011, 20:43
Brilliant Bill ... as ever. More please ..

I can see this developing into an excellent thread. There were so many unsung, and brilliant, contributors to the British codebreaking effort .... not just at Bletchley either.

Thanks Harley for starting it.

harry.gibbon
17-05-2011, 21:27
Might as well have this link to an earlier posting in the Comms thread, link to a Bletchley Park site to be found within in the post #1217:-

http://www.worldnavalships.com/forums/showpost.php?p=142934&postcount=1217

and a link to the index page of the link contained in the post included above:-

http://www.colindaylinks.com/bletchley/index.html

Little h

Dreadnought
18-05-2011, 06:06
And I'll add these threads ....

Admiral Sir Reginald Hall and Room 40 ....
http://www.worldnavalships.com/forums/showthread.php?t=1594&highlight=room+40

Operation Lucid and the Syko machine ....
http://www.worldnavalships.com/forums/showthread.php?t=6960&highlight=enigma

Cqpture of U110 and the Enigma machine .....
http://www.worldnavalships.com/forums/showthread.php?t=2782&highlight=enigma

Ednamay
18-05-2011, 11:02
This is fascinating; cryptology is beyond me and I have read with amazement the work of the staff at Bletchley. Nowadays I can just about manage the crossword! Thank you for the challenge

Edna

Teuchter
18-05-2011, 14:01
Thank you very much yet again Bill for your diligent and interesting posts

emason
18-05-2011, 17:14
Thank you all for your interest. There is a more to come when I have got things straight in my mind, as it is all too easy to become bogged down in detail.

John Odom
18-05-2011, 21:45
A most interesting post, we are waiting for more!

emason
24-05-2011, 18:36
Terminology

Just a note on the usage of the words "code" and "cypher".

A code is the substitution of a word or phrase by another word, phrase or number, e.g. "Uncle Fred has broken his leg" could mean "Returning to port with casualties", or more commonly in naval codes, a five figure number for a word or phrase, e.g. "62940" to mean "Enemy fleet at sea". In both these cases, the messages can only be decoded with reference to a decoding manual – a security risk if one of these is captured.
[Note: When the five figure code is used, another five figure number is usually subtracted from it before transmission, using a book of subtraction tables, to increase security.]

A cypher, on the other hand, is the procedure by which the manipulation (usually mathematical) of each character in the message, produces another character apparently at random; the whole message appearing as a meaningless jumble of characters, which can only be decyphered by applying the reverse procedure. So the Enigma is a cypher machine.

 
Decyphering Enigma – Early Days

The Polish connection

In late January 1929, an alert Warsaw customs officer’s attention was drawn to a heavy wooden box by a request from the German Embassy to have the box returned to Germany immediately as it had been sent to Poland by mistake. The curious officer opened the box and found an Enigma machine inside, and reported it. The Polish Cypher Bureau called in two engineers from a communications company with whom they had close links. For the next two days, they thoroughly examined it, then repacked it and returned it to Germany. The machine was the commercial version of Enigma.

With their interest aroused, the Cypher Bureau, who had been intercepting unintelligible radio signals from Germany, began to link these with the Enigma machine, and an unsuccessful attempt was made to decypher them.

Coincidently, also in January 1929, the head of the Cypher Bureau’s German section was looking for mathematics students to attend a course on cryptology he was setting up at Poznan University. Amongst those who attended were Marian Rejewski, Henryk Zygalski and Jerzy Rozycki, who were later recruited by the Cypher Bureau to work on the Enigma cypher.


The French connection

In June 1931, Hans Thilo Schmidt, an employee of the German Cypher Office, visited the French Embassy in Berlin to enquire about selling secret documents to the French Government. He had no political motives; he was driven entirely by money, or rather his perceived lack of it. He subsequently wrote a letter to the French Secret Service, the Deuxieme Bureau in Paris, and a meeting was arranged.

On 1 November 1931, at the Belgian border town of Verviers he had his first meeting with Rodolphe Lemoine, an agent of the French Deuxieme Bureau. One week later they met again at the same place, this time with Gustave Bertrand, the head of the Deuxieme Bureau’s cypher section, in attendance. Schmidt had brought with him the secret manuals on how to set up and operate the Army Enigma machine. Bertrand photographed the manuals and returned them several hours later. Schmidt was generously paid 10,000 Reich Marks.

Back in Paris, Bertrand showed the photographs to his cryptographers who said that, although they showed how to encrypt a message using the machine, they would not enable anyone to decrypt one. It was agreed with his superiors that a second opinion should be obtained from Britain.

The photographs were shown to British officials, but not to the cryptographers. Unfortunately, Bertrand was asking for a huge amount of money for them, which the British were not prepared to pay, especially as they considered the manuals on their own were not much help. This rebuttal meant that Britain received no further Enigma intelligence from France for several more years.

So finally the Poles were consulted who said they would be interested, and a meeting with Major Gwido Langer, the head of the Cypher Bureau, took place in Warsaw on 10 December 1931, where they were shown the photographs, which the Poles confirmed would be very useful. Although again, they would not be much help with decyphering messages, they did show that the commercial Enigma that they had examined in 1929 had been modified to produce the Army version whose signals the Poles were trying to decypher. But what was really needed was the current Army Enigma settings.

Schmidt dutifully provided these and by late December they were in the Poles hands. But even with these documents, the cryptographers still could not read the Enigma messages. Further settings were delivered in May and September 1932. In all, between 1931 and 1939 Schmidt handed over 303 documents.


Unwelcome developments

After their first meeting on 8 November 1931, Bertram was accompanied at further meetings by Andre Perruche who worked in a department of the Deuxieme Bureau that dealt with the wider aspect of Franco German relations. Perruche was not so much interested in cyphers as in other aspects of Schmidt’s work in the German Cypher Office, especially when he discovered that Schmidt’s elder brother Rudolf was a senior Army officer with whom Schmidt had frequent conversations.

What Perruche and his seniors were interested in was German rearmament plans about which Rudolf knew a great deal, so Schmidt started to provide this intelligence. This had the consequence that the handling of Schmidt was taken over by Perruche, and Enigma gradually ceased to be the priority, although Enigma information continued to be supplied for several more years.

To protect Schmidt, and their source, Perruche wanted to end the cross border meetings, because if Schmidt were stopped with incriminating evidence on him, the golden eggs would cease being laid, and the efforts against Enigma would be revealed. In 1936, after the German occupation of the Rhineland, Perruche insisted that Schmidt should not meet French agents outside Germany except in an emergency. Instead, when had information to send, he now had to pass it secretly to a French agent in Berlin who passed it on to the French Ambassador for despatch via the diplomatic bag to Paris.

This system worked well until 6 November 1937 when, instead of using the diplomatic bag, the Ambassador inexplicably sent it in a telegram using the diplomatic code. This telegram was intercepted and decoded by the Forschungsamt who were responsible for monitoring radio and telephone communications. The telegram mentioned a top secret meeting of Hitler with his most senior generals, the information having been supplied unwittingly by Schmidt’s elder brother Rudolf.

[Note: Before these events in 1937, Rudolf Schmidt had been promoted to General in command of the 1st Panzer Division, and later took part in the invasion of Poland as part of the 10th Army under General Walter von Reichenau.]

Fortunately, Hans Thilo Schmidt had heard from his Forschungsamt connections what had happened, and at an emergency meeting with the French, refused to continue working for them unless he had assurances that his information would never again be sent via the Ambassador. In future Schmidt was to communicate by letters written in invisible ink. With this assurance, Schmidt gave them a photograph of the minutes of the meeting in which the annexing of Austria and other ambitions were discussed.

But the damage had been done; German security now realised they had a high level spy in their midst and would spare nothing to find him.


Enigma variations

The German Army, Navy and Air Force all used their own versions of the Enigma machine. The Army had in 1929, the military version with the plugboard and three rotors. The Air Force started to use the Army model in 1934, but with different rotors. The Navy in 1926 had bought the commercial machine (no plugboard), albeit with different rotors, but changed to the army version in 1934, but using five rotors from which to select the three.

[Note: Each of the three services had their own way of setting up and operating their machines. Because the Poles were given the Army manuals and Army Enigma settings, they concentrated on this version mainly. From now on, any reference to Enigma implies the Army machine and settings, unless stated otherwise.]

Throughout the 1930s, the Germans were constantly improving the Enigma, mainly with more stringent operating procedures. For example, initially the rotor starting positions for every message, e.g.’ABC’ (called the ‘grundstellung’) were unchanged for a couple of months at a time, but changes became more frequent, until they were changed every day.

In addition to every Enigma having the same basic key (grundstellung) during the same period, each operator would choose another different one for each message, the problem being was how to tell the receiving station what it was. This was done with a message ‘indicator’ sent at the start of each message, which involved sending the encyphered message setting twice. This was a mistake, to be explained later.

The number of plugboard connections was gradually increased until, in 1938 between five and eight pairs were connected; and in 1939, between seven and ten pairs were connected. If they had plugged all thirteen pairs, there were eight trillion (8,000,000,000,000) ways of doing this.

On 15 September 1938, the Germans altered the message indicator system. On 15 December 1938, another two rotors were introduced, making five from which to select the three.


The British Progress

When the Foreign Office purchased one of the early Enigma machines in 1926, the Government Code and Cypher School of (GC&CS) would have been asked to examine it in order to assess its security, as one of its main remit was to advise on codes and cyphers.

When the British patents for the commercial Enigma were applied for in 1927, GC&CS studied them, and in 1928 they purchased two of the machines for themselves and analysed it fully to see if the claims that it was unbreakable were true. The patents described fully how the Enigma worked, and accompanying photographs showed how to set up the machine, the interchange ability of the rotors, and the movable rotor rings.

In the early 1930s (exact date unclear) Dillwyn Knox, (who had been instrumental in decoding the Zimmerman telegram while in Room 40), devised methods for decrypting messages encrypted on this commercial machine, namely "buttoning up" (for determining the rotor’s internal wiring) and "rodding" (for discovering the rotor order and starting position).
Impetus was given to this work when Italy invaded Abyssinia in 1935.

Consultation with France resulted in their cooperation, with a jointly run wireless station set up in southern France to intercept Italian naval transmissions.

Both Italy and Spain had bought the commercial Enigma (with changed rotors and other differences) in 1936, so that they could communicate with each other during the Spanish Civil War. This machine was "broken" by GC&CS in 1937. Because the "Condor Legion" of the Luftwaffe used the same machine, it gave a useful insight into their general operating procedures and tactics.


The Polish progress (1)

[Note: the dates and chronology in this section are approximate only, as there is considerable variation in the many accounts of this period.]

After handing over the third set of Enigma setting in September 1932, Bertrand expressed anxiety that little progress had been made. The Poles asked him to be patient until they could construct an Enigma replica machine to help them, promising to tell him when it had been completed.

Bertrand wondered if the lack of progress was due to Schmidt passing him off with old settings, or those for a different Enigma model, and confronted Schmidt with this. A complete rift was only avoided by Lemoine handing over a lot more money.

But progress had been made. Marian Rejewski the mathematician cryptanalyst had deduced an important difference between the commercial and Army versions of the machine. Both machines had a disk which provided the connection between the keyboard and the first rotor, called the entry disk. In the commercial version, the inputs to the disk were in the order of the German keyboard layout (QWERTZU etc.). Whereas in the Army version, the connections were in alphabetic order (‘A’ went to position 1 etc.); with the plugboard acting like a programmable fixed rotor between the keyboard and the entry disk.

With this discovery, and the three sets of Enigma settings given to the Poles, Marian Rejewski was able, in December 1932, to work out the wiring inside the rotors. This provided the first break-through, a mathematical one, which resulted in the theoretical breaking of the Enigma machine at the beginning of 1933, but they kept the news from Bertrand. In fact they told him nothing more for the next six years, despite being frequently supplied with current Enigma settings.

Once the wiring inside the three rotors currently in use was known, the Poles could now build an Enigma replica to assist with their work. The first one was completed some time in 1934, after which they read their first Enigma message; an old one, using the settings supplied for the period. Progress was being made, and again the French were not told.


Polish progress (2)

From 1934, the Poles had varying success, often being able to read Enigma messages. One of the difficulties they had was that the Germans, throughout this period up to 1939 were constantly improving their operating procedures, and the Poles had to develop new techniques to combat them.

In 1936 the Polish cryptographers had discovered a method of exploiting the message key indicator. The indicator was the rotor starting position for the message (the message key) and was entered twice. Being three characters long (one for each rotor), the 1st and 4th characters were produced by the same keyboard letter; the same applies to the 2nd and 5th, 3rd and 6th characters. This method became known as the ‘characteristic method’, and was used to help determine the rotor order and their starting positions.

In 1938 the ‘Bomba’ was introduced. A Bomba was a series of six inter-connected replica Enigmas, designed to test the ability of the rotors to produce the given message indicator. A Bomba could test only one of the six possible rotor orders at a time, so six Bomba were needed, running together, to test all six rotor orders. It worked rapidly by stepping the rotors and testing each position for a match to the message indicator.

In 1938, a mathematical method, for quickly eliminating impossible rotor orders, was devised by Henryk Zygalski. The mathematics involved was represented on a series of perforated sheets called Zygalski Sheets. There were 26 sheets for each possible rotor order. Each sheet consisted of a 26x26 matrix of the possible starting positions of two rotors. By positioning the sheets on top of each other and observing where the holes coincided, certain rotor orders could be eliminated.

In 1938, the Germans introduced two more rotors, so instead of there being six possible rotor orders, there were now sixty (5x4x3); a tenfold increase. This meant that ten times the number of Bomba, and ten times as many Zygalski sheets were needed to test for all rotor orders. This virtually brought their progress to a halt.


Tripartite conferences

In 1938, after the annexing of Austria, Bertrand, despairing of the Poles ever reading Enigma messages, asked for a conference with GC&CS, at which he produced the encyphered and plain texts of four Enigma messages and information about the wireless networks of the German Army. It was agreed to pool resources to combat the German Enigma. Bertrand also suggested that the Polish Cypher Bureau might be a useful ally in a triple alliance, but the British were not keen on the idea.

The First Paris Conference 9 January 1939
In December 1938, Bertrand took the initiative in calling a Franco-British-Polish conference of cryptanalysts to discuss Enigma. On 9 January 1939, the meeting took place in Paris between all three, at which they discussed the progress they had made so far. The British revealed that in April 1937 they had broken the Enigma used by the Italians and Franco’s forces during the Spanish Civil War, and demonstrated ‘Rodding’. But it had made little progress with the different model that the German Navy was using. The French admitted that they could make no progress on their own. But the Poles, probably mindful of the fact that Britain and France had recently signed the Munich Agreement with Germany, sent no cryptographers to the meeting and concealed the real progress they had made. But it was agreed that each participant would concentrate on a particular aspect of the Enigma machine, and to exchange information at further meetings.

The Second Warsaw Conference 26-27 July 1939
Soon after the Munich Agreement, Hitler started making territorial demands on Poland, and in March 1939 Britain and France gave Poland a guarantee that Polish territorial integrity would be defended with their support. With the rapidly deteriorating political situation with Germany, together with the 1938 changes to the Enigma, and the realisation that they did not have the resources to increase the manufacture of their Bomba and perforated Zygalski Sheets by a factor of ten, finally persuaded the Poles to call another meeting.

This took place on 26-27 July 1939 in Warsaw. On the first day the Poles finally revealed that they have reconstructed the Enigma machine and have been reading the German Army Enigma messages from 1933 to 1938. Although the Poles admit they can no longer read the Wehrmacht messages, even with the help from their Bomba, they say that the messages can be broken again if the perforated sheets are produced, which GC&CS agreed to manufacture. The Poles also said they would give an Enigma replica to both France and Britain.

These revelations about the Pole’s progress infuriated Bertrand who now knew, that for six years, he had been deceived, and that Schmidt had been putting his life at risk needlessly for much of this time, and he never fully trusted them again. However, the Poles lived up to their promise and sent two Enigma replicas to Bertrand who delivered one of them to GC&CS on 16 August 1939.

On the first day of the conference, there were no Polish cryptographers present much to Dillwyn Knox’s displeasure. But the second day was different, when he and the French cryptographer Henri Braquenie were taken to meet Marian Rejewski, Henryk Zygalski and Jerzy Rozycki, the young cryptographer mathematicians who showed them everything. Much was discussed and information was freely offered. Knox was given the solution to a problem he had been confounded by; the wiring order of the keyboard to the entry disk.


Other developments

In March 1938, Rodolphe Lemoine (Schmidt’s French contact) was arrested during a trip to Cologne by the Gestapo. Fortunately this was before the Poles revealed that they had been reading Enigma messages. He was released and returned to France where he underwent an interrogation by the Deuxieme Bureau, who concluded that he had not been turned. Nevertheless, he was forbidden ever to meet Schmidt again.

On 1 September 1939, Germany invaded Poland. The Polish Cypher Bureau quickly packed everything they couldn’t burn, including the replica machines and Bomba. They were taken eastwards to Brest Litovsk on a special train with members of the General Staff. Their journey was disrupted many times by air raids, taking three days to reach Brest Litovsk where the train was partially destroyed by bombing. They were then transferred to cars and buses and evacuated to Rumania, arriving in Bucharest on 18 September. The trucks had run out of fuel, so the Bomba etc. was buried near the Russian border.

Contacting the French Embassy in Bucharest, they were issued with all the necessary papers and tickets to travel to Paris, finally arriving there in early October.
 
 

Dreadnought
24-05-2011, 20:20
More good stuff Bill ... a huge subject ... you could be here some time ...!!

It is interesting to note that Scherbius originally marketed his enigma machine to the German industriaal and banking sectors to prevent their professional secrets being compromised in the noraml public telegraph system, but the machine was not a commercial success because the internal set-up of the machine was compliacted to the degree that it was considere impossible for an outsider to unravel the coded messages that it produced. And yet, its operation was so simple that virtually anyone was capable of putting messages into code.

However, the German armed forces were becoming increasingly aware of the value of radio communication in warfare. Then, in 1925, the Reichswehr (German Army) tested some of the machines and eventually bought over 30,000. The machines used by the military and other government organisations (such as the railways) had different internal wiring from that of the few commercial machines that had been sold, and the German Navy made even more changes which further increased the security of their machines. Adapted for military use, the machine proved to be of great value particularly because of its ease of operation even under severe battle conditions.

Sean Roberts
24-05-2011, 20:28
Hi Bill. Working in that field myself,I know the compexity of cypher.The Germans thought they had an unbreakable cypher machine, but these days with computer assistance we can break a simple cypher in seconds.The compexity occurrs when you intercept an STU signal.That can take weeks for a number cruncining machine like a Cray to brake down.
As my Dad would have said B.Z Billl

emason
25-05-2011, 19:08
Clive, you are right, it is a huge and absorbing subject. The biggest difficulty is deciding what to leave out!

Sean, thanks for your reply, but can you explain what an STU signal is? That is if you are allowed to tell us mere mortals.

Sean Roberts
25-05-2011, 19:34
Hi Bill.Glad to be of help(As far as I know it's not a being sent to the Tower job!!!:D) STU is simply a secure radio telephone link.It is a pain in the A**E to use when setting up your link.The system changes frequency about every 5 seconds or so.When your entry code and who you want to contact is recognised,you are left holding the unit,listning to all sorts of wierd noises until you are fianally conected.Sounds worse that I make it out,but if you are trying to call in an air strike or counter artillary barrage,it seems as if you are waiting for ever.Though it's rarely more than 30 seconds for patch in!

John Odom
25-05-2011, 20:48
More great stuff!

emason
27-05-2011, 20:55
To demonstrate the operation of the Enigma machine, I have written a simple program that simulates a single rotor machine.

If you would like a copy to play with please P.M. me with your email address and I will send it as an attachment.

emason
30-05-2011, 18:45
From BBC News:

The National Museum of Computing has finished restoring a Tunny machine - a key part of Allied code-cracking during World War II.

Tunny machines helped to unscramble Allied interceptions of the encrypted orders Hitler sent to his generals. Restoration work on Tunny at the museum in Bletchley was re-started in 2005 by a team led by computer conservationists John Pether and John Whetter.

Mr Pether said the lack of source material made the rebuild challenging. "As far as I know there were no original circuit diagrams left," he said. "All we had was a few circuit elements drawn up from memory by engineers who worked on the original." The trickiest part of the rebuild, he said, was getting the six timing circuits of the machine working in unison. The Tunny machines, like the Colossus computers they worked alongside, were dismantled and recycled for spare parts after World War II.

The first Tunny machine was built in 1942 by mathematician Bill Tutte. He drew up plans for it after analysing intercepted encrypted radio signals Hitler was sending to the Nazi high command. These orders were encrypted before being transmitted by a machine known as a Lorenz SZ42 enciphering machine.

Bill Tutte's work effectively reverse-engineered the workings of the SZ42 - even though he had never seen it.

Tunny worked alongside the early Colossus computer, which calculated the settings of an SZ42 used to scramble a particular message. These settings were reproduced on Tunny, the enciphered message was fed in, and the decrypted text was printed out.

By the end of WWII there were 12-15 Tunny machines in use and the information they revealed about Nazi battle plans aided the Russians during the battle of Kursk and helped to ensure the success of D-Day.

"We have a great deal of admiration for Bill Tutte and those original engineers," said John Whetter. "There were no standard drawings they could put together," he said. "It was all original thought and it was incredible what they achieved."

One reason the restoration project has succeeded, said Mr Whetter, was that the machines were built by the Post Office's research lab at Dollis Hill. All the parts were typically used to build telephone exchanges, he said. "Those parts were in use from the 1920s to the 1980s when they were replaced by computer-controlled exchanges," he said.

Former BT engineers and workers involved with The National Museum of Computing have managed to secure lots of these spare parts to help with restoration projects, said Mr Whetter.

The next restoration project being contemplated is that of the Heath Robinson machines, which were used to find SZ42 settings before the creation of Colossus.That, said Mr Whetter, might be even more of a challenge. "We have even less information about that than we had on Tunny," he said.


You can read the original and view a video of it here:

http://www.bbc.co.uk/news/technology-13566878

 

John Odom
30-05-2011, 23:00
The difficulty of that restoration is beyond my comprehension!

emason
31-05-2011, 18:46
The difficulty of that restoration is beyond my comprehension!

Perhaps only exceeded by the feat of the original machine being built from analysed intercepted signals alone. Which is also beyond my comprehension.

Sean Roberts
31-05-2011, 19:08
Hi Bill.I must arrange a visit to Bletchly park.I had heard of the restoration,but heard some years ago that they were having difficulties(Valves and other types of equipment from the era).We all now rely on modern technology and do not realise how difficult the task was that faced the code breakers in those far off days.We now only have to hear a Camel F**t in the desert and we will have it's exact position in seconds!!!
God what must it have been like in the days before electronics,when most of it was done by mathamatical deduction?
Thank God I work in ops nowadays not then. Once again Bill well done on your research and reporting.

harry.gibbon
31-05-2011, 19:30
Interesting thread.

All of our branch members were required to complete a 6 months residential course at 'the Park'. The course, as well as giving us a comprehensive 'cross' training, also formed part of a series of courses we were required to successfully complete before being considered qualifiied.

I attended in the mid 60's and the place had many interesting pieces of kit, some of which we got an 'introduction' to (mainly for background information), whilst intro's to other pieces of kit, well that had to wait until operating in the field so to speak.

One of the things that sticks with me were the simple huts within which all of the activity was going on.

Little h

emason
13-06-2011, 18:15
The German Naval Enigma (Part 1)
 
Basics

The Machine
The Enigma machine used by the Kriegsmarine was basically the same one as used by the Wehrmacht and Luftwaffe, the three rotor machine with a plugboard. But instead of having five rotors from which to select the three, the Kriegsmarine had eight, which gave 336 (8x7x6) possibilities. Their operating procedures were more stringent, and their operators more skilful and careful, generally being able to transmit and receive Morse quicker and more accurately than their Wehrmacht and Luftwaffe counterparts.

The Networks
Each branch of the German armed forces operated a number of networks, each with their own daily settings, commonly called the daily key, which were identified by name. The Kriegsmarine had over a dozen of these, the main ones being: (B.P. name)

‘Heimisch’, later ‘Hydra’ (Dolphin) - the home waters key, used for surface ships and U-boats in the Baltic and coastal waters, and for operational U-boats in the Atlantic until 5 October 1941.

‘Triton’ (Shark) - for operational U-boats in the North Atlantic from 5 October 1941.

‘Hermes’ (Porpoise) – for naval units in the Mediterranean and Black Sea.

‘Medusa’ (Turtle) – for U-boats in the Mediterranean Sea, from June 1943.

The daily settings were different for each network, and each specified the rotor order, the ring settings, plugboard connections and basic rotor setting. It was this daily key which Bletchley Park sought to discover, which once known, most messages for the same network could be read for the day.
 

The naval indicator system

The indicator was the method of incorporating the message key into the message itself, so that the receiving operator would know the key to use to decypher the message. The message key is the starting position of the rotors, (different for each message) consisting of three letters selected by the operator, and were encyphered twice at the start of every message.
This was the weakness that was first exploited by the Polish cryptographers in 1936, upon which their Bomba and Zygalski sheets depended. In May 1937 the Kriegsmarine abandoned this method in favour of a more secure indicator, and for the next four years neither the Poles nor any one else could read these signals.

The purpose of the naval indicator was to permit the receiving station to recover the message key, while preventing unauthorised discovery. Instead of allowing the operator to choose the key, Naval operators were issued with sets of trigram tables - randomly groups of three letters, from which they selected two trigrams at random from different tables, the second being the basis for the message key. These were then manipulated using a set of bigram substitution tables, as follows.

Suppose the first trigram selected is ‘ABC’and the second is ‘DEF’. The operator would add any letter of his choosing to the start of the first (e.g. XABC), and another letter to the end of the second (e.g. DEFZ), then write them down, one under the other thus:

X A B C
D E F Z

Reading vertically, this made four bigrams, namely XD AE BF CZ. Then using the sheets of bigram substitution tables, the equivalent bigrams were substituted, e.g. FG NS OT RM. This was the indicator that was transmitted unencyphered at the beginning and end of every naval message.

The operator then set his Enigma to the daily setting and typed in the second trigram selected ‘DEF’. This gave him three enciphered letters, e.g. ‘KGB’ which became the message key. He then reset his rotors to ‘KGB’ and entered the plain text message and obtained the encyphered text.

The indicator was transmitted unencyphered, followed by the encyphered message, and then the indicator again at the end. The receiving operator reversed the bigram procedure to obtain ‘DEF’, set his rotors to the daily setting and typed in ‘DEF’, which gave him ‘KGB’ – the message key. He reset the rotors to ‘KGB’ and typed in the enciphered text and obtained the original plaintext message.

[Note: This detailed explanation is to illustrate why it was so difficult for the cryptographers to understand the indicator, and why the bigram tables were so important.]
 

The insoluble problem

Hut 8 at Bletchley Park was where the cryptographers attacked the naval Enigma cyphers. In the first 12 months of WW2 they had made little progress. The Wehrmacht and Luftwaffe cyphers were easier and considered more relevant to the threat against Britain (Battle of France and Battle of Britain) than the apparently insoluble naval cyphers, and so received the attention of the majority of the cryptographers available. The other main reasons were:

1. The indicator
The problem the cryptographers had was that there was now no logical, mechanical or mathematical connection between the message key and the indicator. That link had been severed by the use of the bigram substitution tables. The method of producing the indicator was not understood, although bigram tables were suspected of playing a part. In short, the key could not be deduced from the message without possession of the current bigram tables. To make matters worse, nine different bigram tables were used each month and were changed from time to time.

2. The rotors
The Kriegsmarine had introduced three extra rotors, making a total of eight from which to select the three to use, and at least one the three new ones was always used. The Polish cryptographers had determined the wiring of the first five rotors, but nothing was known of these three new ones.

3. The short signal books
There were two of these, and were widely used for naval signals. The main one in general use was the Kurtzsignalheft, literally the short signal book, and the Wetterkurtzschlussel – the short weather signal book used by weather reporting stations and ships. These were code books to convert standard messages into short coded messages in order to minimise transmission time, to make direction finding by the enemy more difficult. The problem from BP’s perspective, was that this code was then encyphered with the Enigma machine, so double encyphering (super encyphering) the message. Even if the message could be decrypted, its meaning could not be understood without the short signal book.

The problem impeding complete decryption, was that to solve any one of the above, required knowledge or possession of the other two – a Catch 22 situation. This is what a young mathematics graduate of Cambridge University, Alan Turing decided to tackle at the end of 1939, as he said. "because no one else was doing it", and because he preferred to work on his own.
 

The first lights in the darkness

On 4 February 1940, Karl Doenitz was at the quayside in Wilhelmshaven to see off U-33 on its operation to lay mines in the Firth of Clyde, considered a dangerous mission as they would be operating in enemy shallow coastal waters. Normally on operations of this type they would carry no secret equipment with them that could be recovered by divers, in the event of being attacked and possibly sunk in shallow waters. This included their Enigma machine, but for some reason this was overlooked on this occasion.

On the early hours of 12 February, while operating in the Clyde estuary, HMS Gleaner, a survey ship converted to anti-submarine duties, picked up the sound of diesel engines on its hydrophones. Turning towards the sound, the ASDIC soon acquired a contact and a depth charge attack was initiated. U-33 dived deeper but hit the sea bed at only thirty six metres. A further two attacks damaged U-33, and her captain Hans Wilhelm von Dresky, decided to move her, but U-33 was stuck on the sea bed. The only solution was to blow the main ballast tanks and hope to sneak away unseen in the darkness.

When U-33 reached the surface HMS Gleaner was waiting and fired a few rounds, whereupon Dresky gave the order to abandon ship and for the prepared demolition charge fuses to be lit. The Enigma rotors were distributed amongst a few of the crew who were told to drop them in the water as soon as they left the boat.

The surviving U-boat crew were picked up by HMS Gleaner, and the Bohemian Queen who transferred them to the Gleaner. In the pockets of one survivor, three Enigma rotors were found, although no one knew what they were. These eventually found their way to Bletchley Park where two of them were found to be rotors VI and VII, two of the three unknown extra ones. Although the rotors would assist the cryptographers, they did not provide an immediate breakthrough.

[Note: The eighth rotor was eventually "recovered at sea" in August 1940; no details are available, but is presumed to be from a sunken vessel and recovered by divers.]

On 26 April 1940 off the coast of Norway, a German trawler disguised as a Dutch fishing vessel was sunk by the cruiser HMS Birmingham after ramming HMS Arrow. So when another apparently Dutch trawler the Polares was seen, she was ordered to stop by HMS Griffin, and was found to be the armed German patrol boat V2623.

After stopping, a Polares crewmember was seen to throw two canvass bags overboard, one of which sank immediately, the other floated. A gunner named Foord, from the Griffin’sboarding party led by Lieutenant Alec Dennis, jumped into the sea and grabbed the canvas bag. After several abortive attempts to pull him from the rough sea, he was eventually hauled back on board.

[Note: The Polares was sailed to Scapa Flow with a prize crew, entering harbour with the white ensign and Kriegsmarine swastika flying, much to the horror of John Godfrey, Head of Naval Intelligence. Even worse, the whole entrance was filmed by a Universal News film crew, when there should have been as little publicity as possible to preserve the ‘secret’ of her capture – the film was confiscated. Also, Lieutenant Dennis had permitted the crew to take away souvenirs, and their searches for these gave the appearance of the ship having been looted, with papers, documents and cypher forms scattered everywhere, instead of leaving it intact so that a systematic search for material could be done. Lieutenent Dennis was reprimanded and ordered to return a pair of binoculars he had claimed for himself.]

The contents of the recovered canvass bag were the Enigma operators’ logs, some Enigma settings and plaintext messages, plus some other non-Enigma codes. These enabled the Hut 8 cryptographers to read some April messages in May; the first naval Enigma messages read in WW2 by BP. Later in June and July, messages for six days in April were deciphered with the aid of the first Bombe. All these messages were too old to be of operational use, but they gave Hut 8 an insight into the naval Enigma.

Perhaps the most valuable insight gained was from the Polares cypher operator’s notes, when Alan Turing finally worked out how the indicator system worked. From this he was able to start reconstructing the bigram tables and short signal book.
 
Attachment: The capture of Polares.

100076


The Bombe

Dillwyn Knox, from his meetings with the Poles had recognised that ultimately, the only way to crack a machine cypher quickly was with another machine, as the Poles had done with their Bomba. But he was thinking more generally than the single purpose Bomba which relied on the original indicator method of double encypherment of the message key. Knox was a Classics scholar, methodical and adept at piecing together fragments of obscure ancient Greek papyrus, but lacking any mechanical aptitude.

Alan Turing, who joined B.P. in September 1939 from Cambridge, was almost the opposite character to Knox, but together they complimented each other. Turing was described as disorganised, absent minded and almost inarticulate, being unable to convey his ideas to others. But he was also an original thinker and a very practical inventor, teaching himself many skills in order to turn his ideas into practical solutions.

So when Knox discussed with Turing his idea for a general Enigma decyphering machine, he understood immediately and started work on the design of one. One month later in October, he had completed the design of the Bombe, and GC&CS had contracted the British Tabulating Machine Company to manufacture them. A month later in November, Turing had laid out the entire theoretical framework for tackling the Enigma in all its variations.

The Bombe, although its name was an acknowledgement to the Polish Bomba, was a very different machine. There were superficial similarities in that each used interconnected Enigma type machines, but the concept was completely different with a different architecture and on a much larger scale.

The basic idea behind the Bombe was that of matching an unknown cypher text to a ‘crib’. A crib was the known or guessed at plaintext that would appear when the cypher text was decyphered. For example, the message from a weather station would probably contain the word German word "WETTER", meaning weather, or "VORHERSAGE", meaning forecast; the longer the crib, the better the results.

But what was clever about the Bombe was that it would work irrespective of the plugboard settings. When the Bombe stopped with a positive result, it had deduced the rotor order and the message key, and enabled the cryptographers to deduce the plug board connections. From these, the complete daily setting was known and the cryptographers job was done. Other messages from the same day could be swiftly deciphered by clerks on Typex machines wired as Enigmas. That was the theory anyway.

The first Bombe was delivered on 14 March 1940 and it was huge; measuring over six feet in height, seven feet wide and weighing over a ton, containing the equivalent of thirty interconnected Enigmas. Although the first results were encouraging, its performance was disappointing, with numerous false ‘stops’, each one needing to be checked out; a time consuming process.

Gordon Welchman, another Cambridge mathematician, devised an improvement to the Bombe called the diagonal board, which was a wired matrix of all 26 alphabetic characters and became known as the ‘spider’. This enabled the Bombe to detect most of its own ‘false’ stops and continue running, thus vastly reducing the time wasted checking them. A further modification by Turing to the diagonal board permitted the simultaneous scanning of all 26 plugboard connections during each test. These modifications were incorporated into all new Bombes, the first of which was delivered on 8 August 1940.

The Bombe continued to be developed throughout WW2, with several specialised models such as the Abwehr Bombe for dealing with Abwehr signals (These were encyphered on a unique four rotor Enigma with unique multiple turnover rotors). In total Britain built 210 three rotor Bombes.
 

The naval problem remains

Naval Enigma was still proving intractable in 1940 with too many unknowns, such as the short signal book, bigram tables etc. The basic problem was that the Bombe required a ‘crib’ with which to compare the message cypher text. The probable plaintext contents, or gist, of the message was needed for this, which could not be known or guessed at unless there were enough decyphered messages available. These would only become available if the messages could be decyphered. There simply were not enough of these to provide cribs. This vicious circle needed to be broken, and the knowledge that even quite small ships such as the Polares used the naval Enigma gave birth to an idea.
 
 
 

emason
24-06-2011, 17:28
The German Naval Enigma (Part 2)
 

The cryptographer’s problem

On 1 July 1940, the Kriegsmarine introduced new Naval Enigma bigram tables. This meant that no new messages could be broken until the new bigram tables are either captured or reconstructed. But the bigram tables can reconstructed only if the Naval Enigma settings can be worked out with the Bombe. For this, a crib (the probable plaintext) is required, which is unlikely because many messages need to be decyphered beforehand. Somehow or other, it was imperative that the new bigram tables be obtained.
 

A desperate solution

After it was discovered that the Polares was using the naval Enigma, it was proposed that another German vessel be captured and the tables and other cypher material grabbed before they could be destroyed. These would become known as ‘pinches’. Several ideas were considered, including a rather fanciful one by Ian Fleming, called Operation Ruthless, which was dropped after initially being accepted. However, before any opportunity presented itself, there occurred an incident which delivered valuable material into British hands.
 

Operation Claymore

On 4 March 1941 at 04.30, Force ‘Rebel’ consisting of HMS Somali and four other destroyers of the 6th Destroyer Flotilla and two troop carriers took part in a commando raid on the Lofoten Islands, off the NW coast of Norway, codename Operation Claymore. The object of the raid was to destroy fish oil supplies, factories and generally to make a nuisance of themselves.

At about 06.20, a trawler was seen by lookouts from the Somali about two miles away. Somali’s opening fire was returned by the trawler, but Somali’s next shots hit the trawler in the wheelhouse killing the Captain and two others. Her boiler was also hit and she started to steam around in circles on fire eventually running aground. However, Somali, on hearing gunfire ashore, left the trawler whose name was now known to be the Krebs, to see if the commandos needed her assistance. It was not until 09.00 that Somali returned to find the Krebs afloat with a white flag waving. A boarding party led by Lieutenant Sir Marshall Warmington was sent over and rescued five injured seamen. The Captain’s cabin was searched and cypher documents together with two Enigma rotors were discovered.

What he had found were the Enigma settings for the month of April, which contained the rotor settings and the plugbord connections, but not the bigram tables.

Possibly, an opportunity was lost when HMS Tartar sank the 10,000 ton Hamburg without first searching her.
 
101360


A cunning plan

At Bletchley Park, Harry Hinsley was engaged in signals traffic analysis when he realised that German weather ships operating in the Arctic were transmitting their reports in the Dolphin cypher (Heimisch), the main Naval Enigma cypher. Although the signals could not be read, it was obvious that they must be carrying a naval Enigma machine and several months cypher documents. So a plan was devised to capture one of these ships.

A force of three cruisers (Edinburgh, Birmingham and Manchester) and four destroyers (Nestor, Bedouin, Eskimo and Somali), under the command of Vice-Admiral Holland, were assigned for the task. The size of the force was necessary to search a large area of water for the target weather ship, which had been identified as the Muenchen. The date for the capture, 7 May 1941, had been chosen so that the Muenchen’s Enigma documents would be current for the next two months.

The Muenchen was first spotted by the Edinburgh and the destroyers raced at full speed towards her. The most obvious problem would be that, as soon as the ship came under attack, the crew would destroy all its cypher documents and jettison the Enigma machine. However, it was hoped that if the initial attack was swift and intense enough, the crew, most of whom were civilian fishermen, would be more intent on saving themselves than the Enigma material. And it nearly worked, with most of the crew abandoning the ship in two boats, but the machine and documents had already been put into a weighted canvass bag in preparation for jettisoning.

The Somali raced to the Muenchen, and changing from full ahead to full astern, her boarding party at the ready, came to a full stop against the weather ship. Her boarding party, again let by Lieutenant Warmington, jumped aboard the Muenchen directly from the deck of Somali and secured the ship intact. A thorough search revealed that the weighted canvass bag had been thrown overboard, but the Short Weather Signal Book and the entire Enigma settings for the Dolphin cypher for June were found locked in the Captain’s cabin.

Two days after the Muenchen’s capture, there occurred another serendipitous capture of importance.
 

The Bulldog bites

On 9 May 1941, U-110 under the command of Fritz-Julius Lemp (who had previously sunk the liner Athenia), together with U-210, attacked convoy OB318 300 miles SW of Greenland. After several ships had been hit by torpedoes, U-110 was detected by HMS Aubretia who attacked her with depth charges, causing much damage, amongst which was:-
1. The batteries were damaged and giving off chlorine gas.
2. The rudders were inoperable.
3. None of the depth gauges were working.
4. The pressurised air containers for blowing the ballast tanks were leaking.

The ballast tanks were blown while there was still enough air left in them, but with no idea of their depth or if they were sinking or rising, everyone just hoped it would work and waited. Then the boat started a rocking motion which told them they had surfaced, and the abandon ship order was given and everyone piled out.

They were met with the sight of HMS Aubretia, HMS Bulldog and HMS Broadway bearing down on them. Commander Baker-Cresswell aboard the Bulldog, seeing the U-boat crew running along the deck, apparently towards their gun, ordered his crew to open fire, when only 100 yards away. When the U-boat crew started jumping overboard he ordered cease fire.

HMS Broadway, charging in on a collision course with the U-boat, apparently with the intention of ramming, just avoided it at the last minute, but damaged herself in a glancing collision with it. A boarding party, led by Sub-Lieutenant David Balme, from the Bulldog found everything intact inside and took an hour and a half to strip her of every document and paper they could lay their hands on, including a complete Enigma machine with all its rotors.

Meanwhile on the Bulldog, the German prisoners were hurried out of sight so that they could not witness the boarding of their boat, having assumed it had sunk. U-110 was taken under tow by Bulldog to Iceland. When the Admiralty was informed of what had happened, they signalled back "Your operation is to be referred to as Operation Primrose in all future signals. Reference to it is to be prefaced by Top Secret and signals to be made only in cypher". In another signal they sent "Operation Primrose is to be treated with the greatest secrecy and few people allowed to know as possible." Later they signalled "Am sending submarine expert immediately by Sunderland flying boat".

The next day, while under tow, U-110 sank in a strong wind. On being informed, the Admiralty signalled "Primrose having sunk makes it no, repeat no less important that fact of having had her in our hands should remain secret. This fact is to be rigorously impressed on all who have any knowledge of the facts."

The sinking of the U-110 was in retrospect fortunate because it preserved the secret of her capture, otherwise if she had been brought into a port it would have been there for all to see, and would not have stayed a secret for long. In view of the Admiralty’s signals, it would probably have been necessary to sink the U-110 out of sight anyway.
 

Another setback

Instead of using Latitude and Longitude for specifying locations in the world’s oceans, the Kriegsmarine used position charts, which divided the seas up into squares, each of which was given a code. This was a more secure system for the U-boats, and was used to define rendezvous points etc. On 1 June 1941, the codes for locations in the North Atlantic were changed.

With the documents from the Muenchen, Dolphin signals were being read currently until 15 June when new bigram tables were introduced. Although they could be reconstructed over time, it was important to get hold of the new tables, or reconstruct them quickly, so another ‘pinch’ was organised. Again a weather ship was the obvious target, and this time the Lauenburg was chosen.
 

The weather ship Lauenburg

The trawler Lauenburg sailed from Tronheim on 25 May to take station east of Jan Mayen Island in the Arctic. It was thought that she would be carrying Enigma material for the months of June and July. This time B.P. appointed one of its own staff, Lieutenant Allon Bacon RNVR, to accompany the boarding party to ensure that no document of importance was overlooked, and to search for any other that may be of value.

On 25 June, the cruiser Nigeria, and three destroyers Tartar, Bedouin and Jupiter sailed from Scapa Flow heading for the Faeroe Islands where they refuelled, then continued towards Jan Mayen Island. Passing to the southeast of it on 28 June, they then changed course to ENE. Their search pattern was hampered by both fog and icebergs. However, at 18.50 a mast was spotted behind an iceberg and the warships changed course towards it.

On being given instructions to fire at, but not hit the Lauenburg, the Tartar’s Chief Gunners Mate replied "Christ, that should be easy!" On board the Lauenburg, after coming under fire, all but two of the crew took to the boats. The two remaining threw the Enigma overboard and started to burn documents in the ship’s furnace, by which time the Tartar had drawn alongside and her boarding party, led by Lieutenant Hugh Wilson, were already climbing aboard.

They found papers and documents strewn all over the wheelhouse and charthouse floors. Everything with writing on it was gathered together and after an hour they had filled thirteen sacks with it. It was all transferred to the Tartar, where Lt. Allon Bacon spent the next couple of days sorting it, amongst which he found the complete Enigma settings for the whole of July.

101362
 

The captured documents

Krebs
The documents recovered from the Krebs were the daily Enigma settings for the month of April. This enabled the cryptographers to read all the Dolphin traffic for that month. Although the messages were of little importance, they did help the cryptographers in reconstructing the bigram tables. But the rotors were not new, as Bletchley already had them.

Muenchen
The Muenchen provided the Short Weather Signal Book and the settings for the month of June. The signal book provided a source of cribs for the cryptographers with which to set a menu for the Bombes. The June settings obviously enabled them to read the June Dolphin traffic currently. Together they enabled the cryptographer to lay out the groundwork for decrypting all future Dolphin traffic.

U-110
Although the Enigma machine recovered from U-110 was the first naval machine to be recovered intact, it was of little use since its internal wiring and that of all its rotors was already known. Also recovered were the bigram tables, but as these had already been reconstructed they were of little use, except to confirm the accuracy of the reconstructed tables.

The most important items recovered from the U-110 were the Short Signal Book, and the Offizier (or Officer only) settings for June. The Short Signal Book would provide cribs for the Bombes, and did so for a long time until it was replaced. The Officer settings enabled the cryptographers to work out how to break future Officer signals.

Officer signals were used by the BdU and U-boat commanders to communicate high grade signals. When they had such a message to send, they first encyphered it with the Enigma set to the Officer settings, then gave the cypher text to the usual operator to encypher again using the normal Dolphin settings for the day. These signals were often used by the BdU to tell the U-boat commanders of a change in the Naval Enigma procedures, and so were of great interest to the cryptographers, and enabled them to defeat measures introduced by the Kriegsmarine to make their Enigma signals more secure.

Lauenburg
The July Enigma settings captured from the Lauenburg enabled the cryptographers to reconstruct the new bigram tables, and the location charts introduced on 1 June, and to read the July Dolphin signals currently.

From now on, until the end of the war, Bletchley Park was able to read Dolphin currently with delays from a few hours to a few days, whatever changes were made to it. One such change occurred on 29 November when new bigram tables were introduced. Although Dolphin could still be broken with the few Bombes available, it was a lengthy process and messages lose their operational value with the delay.
 

Operations Anklet and Archery

Royal Navy ship’s commanders had been instructed to capture cypher and signals documents from enemy ships whenever possible. Such an opportunity arose during Operation Anklet in the Lofoten Islands, and Operation Archery on the coast of southwest Norway.

Operation Anklet (26 December 1941)
The main purpose of Operation Anklet was to transport commandos to four landing positions in the Lofoten Islands, where they would establish themselves for a couple of months, in order to cut communications between those islands and German troops in northern Norway.

As the naval force of thirteen destroyers, corvettes and minesweepers, led by HMS Arethusa, steamed into Vestfjord, an armed trawler was spotted in the distance, which was approached by HMS Ashanti who opened fire on it. The trawler was the Geier (VP5102), which was hailed by a Norwegian officer aboard Ashanti ordering them to line up on deck with their hands up, which most of them did.

As the Ashanti neared the trawler, her guns which should have already been trained on the Geier, started to traverse towards her. The trawler’s crew, thinking they were about to fired upon, jumped overboard. Captain Onslow of the Ashanti on seeing this, but not having seen the late movement of his guns, thinking that scuttling charges had been set, gave the order to open fire, but cancelling it almost immediately when he realised what had happened.

Fortunately one of the shells fired had passed straight though the radio room without exploding, much to the alarm of the operator who abandoned ship before he could transmit a distress signal, or destroy any documents. An Enigma machine, both sets of bigram tables and various signal and code books were recovered.

Operation Archery (27 December 1941)
The purpose of this operation was to mount a commando raid (later called the Vargsoy Raid) on German bases on two islands off the southwest coast of Norway. As the sea between these islands and the mainland was frequently used by escorted German coastal convoys, it had the potential for a pinch, so Lieutenant Allon Bacon from Bletchley Park, who had boarded the Lauenburg, sailed again but this time in HMS Onslow. The attack was delayed for 24 hours due to bad weather.

In short, two small convoys were attacked. The first, escorted by the armed trawler Foehn, was boarded by HMS Onslow and everything recovered. The second, escorted by another armed trawler Donner (VP5904), was boarded by HMS Offa but everything was lost in the sea while transferring it back to the Offa.
 
101361


Results

With all the Enigma material captured during 1941, the main naval cypher Dolphin (Heimisch) was read currently for the rest of the war, with delays only when procedures or tables were changed. The main use to which this treasure trove of naval information was put was mainly in the Battle of the Atlantic, routing convoys away from concentrations of U-boats. But there were other successes.

On 22 November, the German supply ship Atlantis was sunk in the South Atlantic by HMS Devonshire after a rendezvous signal was read. On 31 November, another supply ship the Python, which had picked up Atlantis’s survivors, suffered a similar fate at the hands of HMS Dorsetshire, again in the South Atlantic. (More on these and other sinkings later.)

There were failures also when, on 28 September, the submarine Clyde attempted to ambush three U-Boats at a rendezvous in the Cape Verde Islands. When all three U-Boats escape, Doenitz noted in his War Diary: ‘The most likely explanation is that our cypher has been compromised.’

 
A warning

Bletchley Park had several indications starting in mid 1941 that a new four rotor Enigma machine would be introduced, when decrypted signals made reference to it. Also when U-570 was captured off Iceland in August 1941, a wooden Enigma case was found aboard with four slots in the lid where there were normally only three. When a signal, accidentally transmitted in four rotor cypher was retransmitted in the usual three rotor Dolphin cypher, B.P. was able to work out the wiring of the fourth rotor.

This new machine started to be used operationally on 1 February 1942 when a new network called Triton, which B.P. called Shark, for North Atlantic U-Boats was introduced. As if this wasn’t bad enough, in January 1942 the Short Weather Signals Book was replaced, depriving Bletchley Park of cribs for the Bombes.
 

John Odom
24-06-2011, 20:22
Another most interesting piece, Bill!

emason
24-06-2011, 20:29
Thank you John, I am glad you enjoyed it.

emason
08-07-2011, 18:24
The German Naval Enigma (Part 3)
 

The situation in February 1942

After months of uncertainty and anticipation, and following the introduction of a new Short Weather Signal Book in January which deprived B.P. of cribs, the dreaded day arrived on 1 February 1942 when BdU (U-Boat Command) separated the North Atlantic U-Boats from other U-Boats and vessels by introducing a new network exclusive to them called ‘Triton’.

This new network, which B.P. called ‘Shark’, now used the new four rotor Enigma machine. With no Short Weather Signal Book to provide cribs, and no four rotor Bombes with which to break into Shark, there began a ten month long ‘blackout’ of North Atlantic U-Boat Enigma signals. Consequently, losses of ships from trans-Atlantic convoys soon started to rise alarmingly.
 

The four rotor Enigma

The four rotor machine (designated model M4) was a modification of the three rotor machine, made by squeezing the fourth rotor and reflector into the space previously occupied by the reflector alone. Thus both the fourth rotor and the new reflector were very thin, which meant that the new rotor was not interchangeable with other rotors. This was fortunate for B.P. otherwise the number of rotor orders would have increased from 336 to 3024. The new rotor could be manually rotated but not advanced by movement of the third rotor as there was no mechanism in the original three rotor machine to do this. The new reflector could be removed and replaced if necessary. This was of course unknown to B.P.

What was also unknown to them was that the new machine could emulate a three rotor machine and encrypt signals for the Dolphin network, which was now called ‘Hydra’ by the Germans, although B.P. still referred to it as Dolphin. It could do this only if the ring setting on the fourth rotor was ‘Z’ and the rotor position set to ‘A’, because, at these settings, the new rotor and new reflector were designed to behave in the same way as the old three rotor reflector. The ability to do this was an operational necessity, otherwise U-Boats would have had to carry both types of machine, if their cruise straddled the changeover date of 1 February.

This ability, although an ingenious solution, was eventually to prove Shark’s Achille’s heel (explained later). The first indication of this, although its significance was not understood at the time by B.P., was in December 1941 when a U-Boat signalman sent an undecypherable signal on the Dolphin network, followed by another signal which when decrypted said "My previous decyphers with setting ‘B’ "; meaning that he had mistakenly set the fourth rotor to ‘B’ instead of ‘A’. He then resent it encyphered with the correct setting. It was mistakes like this which enabled B.P. to determine the wiring of the fourth rotor before it came into use.
 

Operational Intelligence during the blackout

The Kriegsmarine continued to use the Hydra (Dolphin) key for the Baltic, Mediterranean, Black Sea and Home waters. Dolphin was being read currently by B.P. and Werftschlussel, the German dockyard cypher, a non machine hand cypher, had been broken and was also being read currently. About 14,000 Dolphin and dockyard signals each month were being read by Hut 8.

The submarine tracking room
The Submarine Tracking Room (STR) was a part of the Admiralty’s Operational Intelligence Centre (OIC). Its objective, as its name suggests, was to track each and every U-Boat in the North Atlantic, from the time it left its base to the time it arrived back. It derived its information from D/F, decrypted Dolphin signals, and decrypted dockyard signals and other sources. The main purpose of the STR at this time was to route convoys away from concentrations of U-boats.

The dockyard signals told them about the construction of U-Boats, how many were being built, when they would be ready, their sea trials, crew training and when they came into service. It also told them about U-Boats returning from operations, what repairs were needed and when they would be ready for further service. It was also a rich source of cribs, since dockyard signals were often repeated on the Dolphin network.

From the Dolphin signals they could read when a U-Boat either left of arrived at its base, because it was always escorted by vessels which reported these events on the Dolphin network. From this the identity and number of U-boats at sea was known. It also indicated which U-boats were lost by their non return.

A ‘Shark’ signal from a U-boat, although unintelligible, gave away its position through land based D/F intercepts. From this, the STR could estimate its route and speed, and thus its probable destination. Because U-boat signals were formalized, the length of a ‘Shark’ signal gave a clue as to its content. Thus a weather report could be distinguished from a sighting report which was different from a position report. When a convoy was sighted, the necessary sending of many signals alerted the OIC to the fact.

When ‘Shark’ deprived the OIC of operational intelligence about individual U-Boats, the STR used all these sources and the knowledge already gained about the different U-Boat types, their average speed and endurance, to estimate the position and numbers of U-Boats and whether they posed a threat to a convoy. It was from these estimates that a ‘working fiction’ was used, which was used as though it were true, until proved otherwise.

Direction Finding (D/F)
For locating U-Boats, the STR relied heavily on land based D/F. Because many signals from U-Boats were of short duration, a coordinated effort was required so that a signal could be identified with a known position of origin.

When a signal was detected at one of the radio listening stations, the D/F stations would be alerted immediately by direct land line to take a bearing on the signal at that frequency. The bearings were sent to the STR who plotted them on their huge wall map and obtained the U-Boat’s position. D/F stations were established in Britain, Iceland, Newfoundland, Bermuda, Freetown, Ascension, Cape Town and later along the east coast of the US.

[Aside: After the Royal Navy had lost contact with the Bismarck in May 1941, it was D/F which located her after she had sent several long signals on the large ship frequency. Although the signal could not be read and the position obtained was inconclusive, it did indicate that she was heading towards France. This was confirmed when operational control for the Bismarck was passed from Gruppe Nord in Kiel to Gruppe West in France which used a different frequency.]
 

BdU’s suspicions about Enigma

In the early days of Dolphin being read, particularly in 1941, the Admiralty made careless use of Ultra information, with action being taken upon it without providing a credible alternative source to mask the origins of their intelligence. Apart from Atlantis and Python, there were too many instances of Royal Navy ships and submarines turning up at U-Boat rendezvous points in remote areas of the Atlantic.

Throughout WW2, Doenitz had suspicions that German naval signals were being read by the enemy, and instigated several inquiries into their security, and each time he was assured that it was virtually impossible for anyone to break into the Enigma cypher unless they had all the settings and all the key documents. Even with these, reading would only be possible for a short period until the settings expired.

For example, an inquiry into how the supply ships Atlantis and Python came to be sunk in November 1941 concluded that Naval Enigma was probably not compromised. The investigators reasoning was, that if British cryptographers had been reading German Enigma messages, the Royal Navy would have caught up with the Tirpitz in January 1942 as she moved from Germany to Norway, and with Scharnhorst and Gneisenau as they moved from France to Germany in February 1942.

This was the problem with the inquiries, they looked at events narrowly instead of being part of a big picture in which a general trend would have appeared. Had they looked more generally, they would have seen the number of convoy interceptions and ships sunk in the North Atlantic decline in the latter half of 1941 and January 1942, then increase again in February 1942 when Shark was introduced. Before February 1942 only one in ten convoys was being intercepted, after that the number rose to one in three.

Perhaps the reason this was not obvious to them was the decrease in North Atlantic sinkings was masked by the increase in sinkings off the United States eastern seaboard; and after February 1942 the reverse was true. If they had looked at the North Atlantic alone, and their change of fortunes after Shark was introduced, the alarm bells should have rung that Dolphin was being read, and with no changes made to it, would continue to be read.

BdU attributed the decrease in Atlantic convoy sinkings to greater numbers of escort ships and reduced numbers of Atlantic U-Boats, some of which had been transferred to the United States eastern seaboard. The progressive rise in Atlantic sinkings from February 1942 onwards was attributed to increased penetration of the Allied Naval Cypher No.3 (the convoy cypher) by their Bdienst at about this time. (Bdienst continued to read much of this cypher with little delay until June 1943.)

It seems as though they were in denial that their Enigma signals could be read. They always found reasons to explain away any evidence which pointed the finger at the security of their signals. The Admiralty was fortunate that the Germans themselves provided the credible alternatives to the source of their Enigma intelligence; dockyard spies, careless talk, enemy direction finding etc. were all considered likely causes.

The Stitchwort
Still not fully convinced by the reassurances given to him by the investigators, Doenitz demanded an emergency method of encyphering signals should he have further reason to suspect a breach of security. For this the Stichwort, or Keyword, was devised where each of the first five letters of the word signified a change in the settings. This was given to every U-Boat commander on departure to be used only when instructed.

It is best described by example. If the keyword was BERLIN:-
1. The ‘B’ referred to the rotors to be used. ‘B’ being the second letter of the alphabet specified that two was to be added to the rotor identifying number. Thus rotor II became IV, rotor V became VII, and rotor VIII became II.
2. The ‘ERL’ referred to the ring settings. In this case, five (‘E’) was added to the ring setting for the rotor in position one, eighteen (‘R’) to the second, and twelve (‘L’) to the third.
3. The ‘I’ referred to the plugboard. ‘I’ being the ninth letter, told the operator to add nine to each of the plugboard connections.

However impressive this additional complication seemed, it was not Bombe proof if a correct crib was used. To the cryptographers in Hut 8 (the naval section of B.P.) it was more of a nuisance than anything else.
 

Progress into Shark

Between February and December 1942, B.P. decyphered only three Shark signals. These were on the paired days of April 23 and 24 when it was surmised that a decrypted Dockyard message had been re-encyphered into Shark; and 14 March when the crib was guessed at when it was announced that Doenitz had been made an Admiral. But each time it took six, three rotor Bombes, seventeen days to discover the key.

From the loss of 48 ships at 277,000 tons in January 1942, the numbers had risen to 109 ships at 720,000 tons in November 1942, the highest monthly tonnage total in the whole of WW2. With more U-Boats coming into service and little progress being made into Shark, crisis level had been reached.

On 22 November 1942 the OIC sent B.P. a memorandum, asking if perhaps "a little more attention" might be paid to the Shark problem. The Battle of the Atlantic, it said "is the one campaign which Bletchley Park are not at present influencing to any marked extent - and it is the only one in which the war can be lost unless B.P. do help. I do not think that this is an exaggeration." - an understated but desperate plea. It was not as though B.P. were ignoring the problem of breaking into Shark, it was just that their problems seemed insurmountable until the four rotor Bombes became available.

But just two days later, Hut 8 received what must have been the answer to their prayers – the new Short Weather Signal Book and the Short Naval Signal Book, retrieved three weeks earlier when U-559 was captured in the Mediterranean. This did not enable them immediately to decypher Shark, but it enabled them to devise probable cribs with which to test on the Bombes.

The breakthrough came on 13 December when, after a short weather signal was decrypted, an incredible blunder by the Germans was discovered: U-Boats were still sending weather reports with their four rotor Enigma set to three rotor mode! That is with the fourth rotor set to ‘A’ with a ring setting of ‘Z’.

Once the short weather signal was decrypted the daily settings would emerge, then all that was needed to break a day’s four rotor Shark signals was to test the fourth rotor in each of its 26 positions. Thus cribs derived from the short weather reports could be used to break into Shark messages on existing three rotor Bombes – provided a short weather crib could be found!

After ten and a half months, the long wait was finally over. A very relieved First Sea Lord wrote a cautionary note to the US Chief of Naval Operations on the evening of 13 December:

" As the results of months of the most strenuous endeavour a few days’ U-boat traffic will be readable in the immediate future and this may lead to better results in the near future.
You will, I am sure, appreciate the care necessary in making use of this information to prevent suspicion being aroused as to its source. We have found this especially difficult when action by routeing authorities outside the Admiralty is required.
It would be a tragedy if we had to start all over again on what would undoubtedly be a still more difficult problem."

emason
26-07-2011, 16:59
The German Naval Enigma (Part 4)
 
 
Reading Shark

The North Atlantic U-Boat cypher Shark (Triton) was finally broken on 13 December 1942, when B.P. discovered that short weather signals were being sent with the four rotor Enigma set to three rotor mode, and thus could be broken with the existing three rotor Bombes - provided a short weather crib could be found.

Some Shark signals were broken fast enough for operational use, but delays from 24 to 72 hours were common. One result from the decrypts was that it became clear that the Allied Naval Cypher No 3 was compromised. In fact, the B-Dienst had been reading this cypher (among others) with considerable success for many months, frequently reading about convoy sailing dates, their routes and composition some 10 to 20 hours in advance of sailing. Naval Cyphers 3 and 4 were subsequently replaced by Naval Cypher 5 on 10 June 1943. Also of interest to them was the Admiralty’s estimates of U-Boat numbers and positions, which they attributed mainly to D/F.

On 10 March 1943 the BdU brought into force a new short-signal weather signal book, and B.P feared that it would make Shark unreadable again, as they would be denied Short Weather report cribs for the Bombes. It was expected that it would only be read with a long delay on a few days each month, and this would improve only after several months work. However, this situation lasted only nine days until 19 March when they resumed their regular reading of the Shark traffic with the usual interruptions and delays.

Several factors contributed to this remarkable achievement. One was that B.P. in March 1943 now had 60 Bombes available to them. Another was that having read Shark traffic since the previous December, B.P. now had the experience to work out alternative methods of producing cribs. It had been helped by the Short Naval Signal Book (SSB) and other material retrieved from U-559 being still in force.

The Short Signal Book was used by the U-boats for making sighting and battle reports, and at this period in time the convoy battles were at their height, producing a wealth of such short signals to use as cribs. In fact, from 10 March to the end of June 1943, Shark was broken on 90 days out of the 112, most of which were read with the aid of these short-signal reports, which like the short weather signals, were encyphered in three rotor mode.

In May 1943, the U-boats were withdrawn from the North Atlantic, so ending the convoy battles. This also meant that the number of short signals available diminished greatly, reducing the cribs regularly used as means of entry into Shark. However, when these short signals diminished, B.P. was able to profit from the growing German habit of re-encvphering in the Shark settings, signals which had previously been transmitted in three rotor cyphers such as Dolphin.

These re-encypherments provided cribs into Shark that were longer and more reliable than the few short signal still available, but they also took much longer to test on the three rotor Bombes. But when the high speed, four rotor Bombes were introduced in June 1943, this was speeded up until, from August 1943, the breaking of Shark became almost a daily routine.

In an official history of B.P. it stated: "It indicated the end of the great Shark battle. From now onwards, there were always enough cribs and enough Bombes to keep the situation under control."

 
The four rotor Bombes

B.P.’s four rotor Bombes entered service in June 1943, followed shortly by the US Navy’s Bombes in August 1943. From September onwards, Shark was generally broken within 24 hours. Shark decryption was shared between the US and B.P. for these months. But in December 1943, in order to prevent duplication of effort and maximise Bombe usage, the routine decryption of Shark was transferred entirely to the US Navy's Op-20-G in Nebraska Avenue, Washington, because they had many more Bombes (made by the National Cash Register Company) in operation and were faster and more reliable than the British models.

The US Navy also placed its spare Bombes at the disposal of B.P. for work on other keys in a joint programme based on intimate relations and direct communications between the cryptanalysts. These communications became possible when the Combined Cypher Machine (CCM) was installed in November between the Allies, Britain, Canada and USA, making it no more difficult for B.P. to use Washington as it was to use a Bombe outstation such as Stanmore or Eastcote in England.

The radio signals interception was still carried out by the Y Service and the messages sent to Washington for decryption. B.P. continued with research into naval Enigma, while the US dealt with the routine decryption.

 
Using Shark

The main use to which Shark intelligence was put was probably in the re-routeing of convoys, but it had many other uses. It helped in the sinking of the Scharnhorst in December 1943, and armed raiders such as Atlantis were tracked down and sunk with its assistance. In 1943 the US Navy employed Ultra offensively in its many sinkings of U-tankers (Milk Cows) which caused much alarm in B.P. who feared that the Germans would suspect that their Enigma cypher was compromised (which they did) and introduce major changes that would render Shark unreadable (which they didn’t).

Additionally, Doenitz received some startling news which heightened his suspicions. On 10 August 1943, a report from Switzerland warned that the Americans are reading Naval Enigma. "Over the last few months, Germany’s naval cyphers, which are used to give operational orders to the U-boats, have been successfully broken. All orders are being read currently. The source is a Swiss American in an important secretarial position in the U.S. Navy Department."

However, Doenitz was told that the "continuous current reading of our W/T traffic is out of the question." Doenitz was unconvinced, reasoning that between 12 June - 1 August 1943 thirteen out of twenty one U-boat meetings at sea had been undisturbed, but between 3-11 August, all ten meetings had been interrupted. This led him to believe that Naval Enigma messages might have been read at the end of July 1943 and the beginning of August 1943, but he was sure that they could not have been read thereafter because of new security measures introduced, the new Stitchwort – a keyword given to U-boat commanders prior to departure, which changed the daily basic Enigma settings - a vain hope.

On 18 August, a further alarming report from Switzerland stated that the informant was "related to our Military Attache, and often travelled to London with the US navy delegation, so he should be well informed", adding "The British Naval Intelligence Office are giving a lot of help [to the Royal Navy] in the fight against the U-boats. A special office has specialised in dealing with codebreaking since the war broke out. For several months it has been very successful. They can now read German Admiralty orders to the U-boat commanders. This is a great help when it comes to hunting U-boats.".

The accuracy and currency of this information indicates that the reports were genuine. However, the identity of the informant has never been established despite the clues contained in the reports.

 
Naval Enigma developments

From 1 July 1943, the BdU introduced a new fourth rotor (Gamma) and new reflector (Caesar) into the four rotor Enigma machine. These were alternatives to the original fourth rotor (Beta) and reflector (Bruno). Despite these developments B.P. continued to break the settings. One reason for this was that changes of fourth rotor and reflector were made by the Germans only once per month, so once B.P. had determined the wiring of the new rotor and reflector, which it did within a few days, it was only when making the first break of each month that delays were encountered.

In May 1945, new bigram tables were brought in at the same time as the initial rotor setting (Grundstellung) was changed several times per day. Consequently, the bigram tables could not be reconstructed using existing procedures. A new system was devised to deal with the problem, but it would have taken several months for it to have become operational. The author of The History of Hut Eight wrote: "It is, I think, possible that we should not have survived this bigram table change. It would probably have been necessary to capture the bigram tables." But fortunately, this was far too late to have made any difference.


Conclusion

In their highly successful penetration of the German naval cyphers, in many ways B.P. was lucky. They were lucky that the Germans made many fundamental security errors in their operation of the Enigma machine that the cryptanalysts of B.P. were able to exploit.

1. The original Heimish (Dolphin) network was too large and used too frequently, giving B.P. an abundance of intercepted messages that could be analysed. For example, all weather ships used it even though the information contained in the signals sent and received didn’t warrant the security that Enigma offered.

2. Messages were repeated in one cypher that previously been sent in another, such as weather reports and dockyard messages, so providing many cribs.

3. Many messages were formalized so that the contents could be partially guessed at. There was one weather station that repeatedly began every report with "Weather forecast Biscay", providing the perfect crib.

4. Many operator errors were made. For example, when a Dolphin message encyphered on the four rotor machine was sent with the fourth rotor incorrectly set, it was re-sent with the correct setting; enabling B.P. to determine the wiring of the fourth rotor and reflector even before Shark came into use.

5. Changes to their cyphers were made piecemeal which gave B.P. less trouble to overcome than had changes been made simultaneously. For example, had the bigram tables been changed at the same time as Shark was introduced, the already difficult task of decryption might have proved impossible, even with the material captured from U-559.

6. Although Doenitz constantly questioned the security of the naval cyphers, the Germans were generally too confident that Enigma was unbreakable and too ready to attribute obvious signal security breaches to other causes.

The German confidence in Enigma was not entirely misplaced. In fact, several British cryptanalysts have said they might never have penetrated the naval Enigma had it been properly operated. It was lucky for them that it wasn’t.

It is generally recognised that the breaking of the Enigma in all its variations was the biggest secret of WW2. But perhaps an even bigger secret was that it was German operating procedures and errors which made this possible.
 
 

Dave Hutson
26-07-2011, 18:24
Hi Bill,

Great read and although not WWII vintage takes me back to our Cryptography training in the fifties. Brings back a lot of memories of the daily ritual of setting up the rotors and the agony of reset if you got one wrong.

Dave H

brian james
26-07-2011, 21:35
Absolutely fascinating read..The 'Enigmas' would have to be considered the first Naval 'computer' wouldn't they?....Great stuff.....

emason
02-08-2011, 17:31
The Abwehr Enigma

 
Background

In March 1940 the Radio Security Service (RSS), charged with intercepting illicit wireless in the UK, had intercepted and broken some simple hand cyphers used by a Hamburg station, which was found to controlled by the Abwehr – the German military intelligence and counter intelligence service.

Bletchley set up a section to deal with this, and by December had broken the main hand cyphers used by them. This enabled B.P. to give advance warning to MI5 of the arrival in Britain of German agents. From this and the wireless traffic it became apparent that the Abwehr were operating a considerable espionage network across Europe and elsewhere. In fact, their networks extended from South America to the far East; from Scandanavia to Africa; the whole of the occupied countries from Salonika to Warsaw, with an extensive network in the main cities in the Balkans.

The agents transmitted their messages using the code given to them by the Abwehr to their controllers who in turn, decoded and analysed the messages before transmitting them on to Germany after encyphering them on their Enigma machines.

It was thought that if this Enigma traffic could be read then it might be possible to predict Germany’s strategic thinking, what future operations she could be planning, and therefore to where her armed forces might be moved. It also might be possible to feed misinformation into this network to affect military and political decisions in Britain’s favour.
 

The Abwehr machine

The Abwehr has several espionage networks operating in Turkey, Spain, Portugal and the Balkans; each with their own settings and key. The preamble and indicator to each message was disguised so as to resemble that of an entirely different network. The Abwehr Enigma machine had no plugboard but was unique in that it had four rotors with each one being unique to the Abwehr and each having multiple turnovers.

The section in B.P. that broke the Abwehr hand cyphers had made no progress into the Enigma traffic, so the task was given to Dillwyn Knox’s section ISK (Illicit Services Knox) around June 1941. Its first task was to sort out the various networks and which messages could be attributed to each one.

Once this was completed it became apparent that, almost unbelievably, they were still using the double encyphered message key indicator, which had been abandoned by the other services years ago as being too insecure, albeit that it was now a four digit key (for 4 rotors).

When this was deduced, it was seen (after much hard thinking and flashes of inspiration) that the rotors always turned over within the indicator. For this to happen the rotors must have multiple turnover points. In fact the rotors had 11, 15 and 17 turnovers, meaning that there was at least one turnover every four characters entered. From this the rotor wiring was worked out, but it still remained to discover how the rotor ring related to the turnovers. To do this, messages would have to be broken first.

That first message was broken on 8 December 1941, and from this the Abwehr machine held no more secrets, it was broken. After this, it became almost routine to read agent’s reports and what the controllers were telling Berlin and vice versa.
 

The Spanish network

The Spanish network was of particular interest because of Spain’s Fascist sympathies and its proximity to Gibraltar, and it was known that German agents were already operating there. This network became so well known to B.P. that a complete picture of its operations was known, with the names of every member of the Abwehr staff at the Madrid headquarters with their cover names and real functions, even the names of their dogs, together with similar knowledge of many of the staff at outlaying stations.

So when in February 1942, a new network sprang up from Spain, it drew attention to itself. This network, based in Algeciras (five miles west of Gibraltar, had outstations reporting to it from Ceuta in Morocco (20 miles across the Straits, south of Gibraltar) and Tetuan, also in Morocco. Its purpose was the surveillance of the Straits of Gibraltar. They sent daily reports on shipping movements and the arrival and departure of aircraft at Gibraltar to the main Abwehr office in Madrid, who then encyphered it on their Enigma machine and transmitted it to Berlin. (One agent reported the name of a departing battleship as "Beware of the Propellers").

In March 1942 the agents started to talk about a new surveillance apparatus that would come into use in April. The Germans had given their surveillance operation the codename of "Bodden" which was the name of the stretch of water between islands in the Baltic and the mainland. One of these islands was Peenemunde, where presumably the apparatus was developed and tested.

Combining this intelligence and aerial reconnaissance photographs of the sites, Dr. R.V. Jones, the head of scientific intelligence, concluded that it was an infra-red detection system to detect ships transiting the Straits. It consisted of three parallel infra-red searchlights on the southern side if the Straits directed northwards, to be detected from Algeciras. In addition, another large infra-red detector at Ageciras would detect ships by the heat radiated from their funnels.

Once this was known, the next decision was what to do about it, as Operation TORCH was in the planning and this could be a threat to its success. Suggestions of blowing the sites up were overruled in favour of a diplomatic solution. So the British Ambassador, Sir Samuel Hoare, had a meeting with Franco, telling him that the British Government was aware that he had allowed Germans in Spanish uniforms to set up surveillance sites, thereby infringing Spain’s neutrality.

On 29 May, an Abwehr message from Madrid to Berlin was decyphered, stating that the British knew about Operation Bodden and had complained to Franco about it. Eventually Project Bodden was abandoned in time to prevent it endangering Operation Torch.
 

The Twenty Committee

As preparation for Operation Sealion (the invasion of Britain), German agents were landed in England and all were promptly captured. In 1941 it was suggested that the captured agents should be ‘turned’ and allowed to transmit misinformation back to their controllers. Those agents who did not agree to this were executed - quite an incentive! The agents, and the intelligence they transmitted, were controlled and coordinated by the "Twenty Committee", so called because they operated a double cross system (XX).

The agents transmitted their messages using the hand code given to them by the Abwehr to their controllers. They in turn decoded and analysed the messages before transmitting them on to Germany after encyphering them on their Enigma machines. By reading this traffic it was possible to see if the agent’s misinformation was being accepted as true.

One agent codename SNOW, of whom the Germans thought highly, transmitted directly to Berlin from Wandsworth prison. His radio only transmitted a weak signal, so a German trawler was positioned in the North Sea to act as a relay station. This boat was equipped with an Abwehr machine and relayed SNOW’s reports encyphered. This boat was considered vital to B.P. because it provided high quality cribs, and therefore was allowed to continue its work unmolested, much to the Royal Navy’s chagrin.
 

Operation Mincemeat

Later in the war, when Operation Husky (the invasion of Sicily) was being planned, it became necessary to convince the Germans that the landings were to take place elsewhere. For this, Operation Mincemeat was put into action. The story is too well known to recount here in detail, but briefly, it consisted of planting a body in the sea off the coast of Spain with a briefcase chained to its waist containing documents about landings that were planned for Greece and Sardinia, with a diversionary attack on Sicily. The main landing was to be in Greece, and the codename for Sardinia was Husky. It was assumed that the Spanish would recover the body and allow the Germans to see the briefcase contents.

The first indication that the Germans had seen the contents and documents was when it was mentioned in an Abwehr message that Kuhlenthal, the Head of the Madrid Office, had gone to Berlin urgently to "discuss his Spanish report".

The knowledge that the deception plan had been accepted as genuine was read on 14 May when General Alfred Jodl informed his senior Mediterranean commanders and Naval Staff Operations that "Enemy landing on a large scale is projected in the near future in both the east and west Mediterranean".

Further confirmation came when an Abwehr intercept from Rhodes stated "that the Allied attack would be directed against Cape Araxo and Kalamata" in Greece. This same warning was passed on to the Salonika, Belgrade and Sofia Abwehr offices.
 

Operation Fortitude

This was the D-Day deception plan to make it appear that the forthcoming invasion, which could not be disguised, was to take place in the Pas de Calais area instead of Normandy. To this end, a notional First US Army Group – FUSAG, under General Patton, was created as assembling in the Kent and Dover areas.

The agent most concerned with this was ‘Garbo’, who ran a notional network of agents who reported on the FUSAG ‘build up’. It was these ‘reports’ which Garbo transmitted to his controller in Madrid who sent them directly to Berlin, where they went directly to Hitler. By monitoring this Abwehr traffic it was possible to track the progress of the deception and fine tune it where necessary.

The purpose of Garbo’s reports was to feed the Germans information leading them to believe that the main attack would be in the Pas de Calais area and that a diversionary attack was to be made first in the Cherbourg area. Even when it was obvious after D-Day that Normandy was the main attack, the agents explained that, because Normandy was unexpectedly successful, it had been decided to reinforce it. Long after D-Day, several German Panzer divisions were kept in the Pas de Calais area for an attack which didn’t materialise.

GARBO’s standing was such that his reports continued to be believed long after D-Day. So it was with considerable pleasure that Bletchley intercepted an Abwehr message from Berlin to Garbo’s controller in Madrid which said that his reports "have been confirmed almost without exception and are to be described as especially valuable". He was so highly regarded that he was awarded an Iron Cross 2nd class in December 1944.
 

The V-weapons

One of the agent’s jobs was to report on the accuracy of the V1 and V2 weapons. They falsely reported that many of the impact points were to the north and west of London. The Germans reduced the range so the majority now tended to fall short of their intended targets.
 

Conclusion

By the end of the war 140,000 Abwehr messages were solved from networks around the world. It could be argued that breaking the Abwehr Enigma was the most important contribution to the war. It often provided information of strategic importance. For example, on 28 February 1942, Churchill reluctantly backed Tito instead of the Chetniks when he learned, from decrypted Abwehr signals from Yugoslavia, that the Chetniks were being forced out of Bosnia by the Partisans.
 
 

John Odom
02-08-2011, 20:20
Another great report! Thanks again!

emason
08-08-2011, 20:45
Thank you John, I am glad you found it interesting. There are so many fascinating stories associated with the Enigma in all its configurations and uses that it's hard to select only a couple to relate. I will take up the story again and hope to keep posting them from time to time.

John Odom
08-08-2011, 21:14
Remember that many more read and enjoy, than post thanks. I know that the thanks helps bring on the next installment! And most of this has not been readily available to the Joe in the street.

emason
20-10-2011, 17:44
Enigma and the V-Weapons

By the end of 1941 the technical section of Air Intelligence (AI) thought that they had all the latest developments in German aircraft covered from their inspection of crashed aircraft. But at the end of 1941 the FW190 made its appearance to the great surprise of everyone, especially AI. This forced them to give more attention to obtaining advance intelligence about the German Air Force (GAF) and its new technical developments.

In the first half of 1942, Enigma started to provide intelligence on German technical developments. The Japanese diplomatic decrypts of traffic to Tokyo also yielded further information. Other sources include POW interrogations, photo reconnaissance, agents and captured documents.

Rocket projectiles and unmanned aircraft had been connected with the test site at Peenemunde in the Oslo Report of 1939. In January 1943 photo reconnaissance of Peenemunde began, which showed large scale construction going on but no indication of their purpose.

The first contributions to the V-weapons intelligence picture came from decrypts of diplomatic messages from the Foreign Ministry to Buenos Aries on the 10 and 13 August 1943 which mentioned reprisal weapons in general including large guns. (Large guns probably refer to the V-3)

For a long time, intelligence was both sparse, inaccurate and fragmentary, just giving tantalising glimpses of what might be coming. Consequently the development of V-weapons was confused between rockets, pilotless aircraft, glider bombs, jet and rocket propelled aircraft, resulting in no clear picture as to what was actually being developed, and therefore what form any threat would take.

-----"-----
 
The V-1 Flying Bomb (FZG-76)

General

The V-1 flying bomb was given the code name of FlakZielGerat 76 (FZG-76), which means Anti-aircraft Target Apparatus, to conceal its true purpose. Because its air frame was made by the Fieseler Aircraft Company, it was sometimes called the Fi-103. It became known as the V-1 when it came into operation.
 

The Intelligence

Air Intelligence (AI) had been keeping a watch on the 14th and 15th Company of Signals Experimental Regiment ever since their involvement in 1941 with the development of the navigation beams became known (Knickerbein etc.). In one of their intercepted signals in April 1943 the first mention of Peenemunde was made, indicating that the GAF had an experimental radar station there. From then on, the 14th and 15th Companies received special attention.

In the autumn of 1943, Enigma revealed that the 14th Company of Signals had positioned detachments along the Baltic coast. The detachments started to transmit signals in the same type of low grade code used previously in northern France in 1941 when reporting on new radar installations. As this code had already been broken, the signals soon revealed the radar plots of objects flying in an ENE direction from Peenemunde.

The plots gave a series of ranges and bearings from which the course of the objects could be plotted, and their capabilities could be determined. The information gained from these intercepted signals included the time of launch, the height at which they flew, their speed, the intended target point and the actual impact area. A.I. virtually had a ringside seat at the trials of the object, from which they were able to monitor the development of the object and assess the threat continuously.

Enough was now known about Peenemunde to make it an important bombing target. On 17/18 August a heavy raid of 600 bombers attacked Peenemunde causing heavy damage and loss of life. Unfortunately, many hundreds of conscripted foreign workers were killed when misdirected bombs fell on their quarters.

On 7 September 1943, two Enigma signals mentioned a "Flakzielgerat 76" (FZG-76), which it was thought could refer to a pilotless aircraft mentioned in the Oslo Report (http://www.worldnavalships.com/forums/showthread.php?t=7197&highlight=oslo) which had associated FZ development with glider bombs and pilotless aircraft.

In October, a series of signals with further course plots showed that these were associated with an aircraft, not a rocket, and that it was a weapon and unmanned. It was also being test fired from a place called Zempin, 12 miles SE of Peenemunde. From these data it was concluded that it was a winged missile. With this information the Central Interpretation Unit (CIU) were asked to check back on photographs previously taken of Peenemunde for a small aircraft.

On 13 November they identified one in a photograph taken on 23 of June. Now they knew what they were looking for, further examples were found on 22 June and 30 September photographs. From measurements it appeared to have a wingspan of 20ft and be propelled by some kind of jet or rocket engine.

Other intercepted Enigma signals revealed that Zempin was associated with the FZG-76. Knowing the regular time of launching, PR were asked to take photographs of specific points at specific times. On 28 November one such flight succeeded in photographing Peenemunde where it revealed a small aircraft at the bottom of a ramp ready for launching. The same series of photographs showed the same arrangement of buildings at Zempin as found at the ‘ski’ sites in northern France (see below). Although it had been thought probable, it was now confirmed that the French ski sites were to be used for launching the flying bomb.

In January 1944 the plots started to show an improvement in the objects performance. By the middle of March, they showed that 40% would hit their target, and by end of March 60% would hit.

In April and May, the activity and test firings at Peenemunde and Zempin greatly increased. This was interpreted as either renewed urgency being given to the trials, or that development had reached such a stage that a sustained rate of fire was being tested. Either way it indicated that their operational deployment was not far away. It was also noticed that the height at which they flew had been steadily reduced. (This was due to problems with the barometric fuel pressure regulator.)

At the end of May, the British were allowed to examine the wreckage of two flying bombs that had crashed in Swedish territory. They found that the bomb was propelled by a pulse jet using low grade aviation fuel, and that it was controlled by three gyroscopes and a magnetic compass.
 

Developments in northern France

By the end of October 1943, agents from within France had reported several sites in northern France where the building of concrete strips and posts were aligned on London. On 3 November, the Photo Reconnaissance Unit (PRU) began to photograph these sites, and by the end of the month 75 such sites had been identified. They were located in a corridor 10 miles from the coast, 30 miles wide, stretching 200 miles from the Pas de Calais, and a further seven were found in the Cherbourg area. All were aligned within 7 degrees of London. They became known as ski sites from their characteristic shape.

Enigma had revealed that Flak Regiment 155 (W), which had been involved with Peenemunde and Zempin, was also involved in France and seemed to be connected with a number of new large structures which had been reported by French agents. By February 1944, eight of these had been located. They were intended to be supply depots for the "ski" launching sites. But none were completed as it was thought, correctly, that they would be too obvious a target for Allied bombers, so they were left unfinished to act as decoys.

The real depots were hidden deep in caves.

[Note: Flak Regiment 155 (W) was commanded by Oberst (Colonel) Wachtel and became known as the Wachtel organisation. Wachtel was responsible for the whole V1 operation, from testing at Penemunde to deployment in France, including supply and launching.]

 
Assessment at end of 1943

Technical assessment
The technical data on the flying bomb was assessed as follows:
1. It was 17-22 feet long with a wingspan of 17-20ft; propelled probably by rocket, and capable of carrying a warhead of 1,000 to 3,000lbs.
2. It flew at 6,500ft generally, at a speed of between 200 and 400mph, and having a range of 120-140 miles.
3. It was controlled magnetically.

It was seen from the plots that the flying bomb was neither reliable nor accurate enough to be an effective weapon, even against such a large target as London, as only about one in six launched would reach its intended target.

Operational assessment
From the evidence available it was calculated that a single launching site could sustain a firing rate of one every 30 minutes until its supply of 20 flying bombs ran out, but it would take two days to fully re-supply the site. Thus one site could fire twenty bombs every three days. As there were now 87 sites discovered, it meant that theoretically, about 600 flying bombs per day could be launched against London, but it was believed that this number could not be sustained as it well exceeded Germany’s capacity to manufacture them. (It had been planned for a monthly total of 5,000 bombs by May 1944, but this was subjected to several downward revisions.)

The threat was considered so serious that it was advised that alternatives to the Overlord embarkation ports be found that were outside the range of the bombs.

 
Countermeasures

As part of the ongoing bombing campaign against German aircraft production, the Fieseler Works at Kassel had been bombed in October 1943. Unbeknown to the Allies, this was where the prototype flying bombs were being made for testing at Peenemunde, and had the effect of severely disrupting the supply and putting the timetable for its introduction back by months. Because of this, the main production facilities were widely dispersed to minimise disruption in the event of successful Allied bombing raids.

By December 1943, enough was known about the flying bomb to start thinking about countermeasures, about seven months before it became operational.

Offensive measures
The most obvious measure was to bomb the launching sites. Large scale bombing of the launching sites started on 21 December by the US 9th Army Air Force and 2nd Tactical Air Force. Much damage was done, and repeat attacks were made to ensure that repairs were delayed. These attacks continued until April 1944 and overall, 20,000 tons of bombs were been dropped.

The Germans decided to abandon most of these sites, leaving them as decoys, and started to build simpler, modified sites which were very difficult to locate. Essentially, they just built the foundations then left, returning a few days before they were needed and assembled pre-fabricated buildings and ramps that took only two days to erect and become operational. Unlike the ski sites, they were not self sufficient and required frequent replenishment from large supply depots.

When the presence of these new launching sites became known, a huge PR program was initiated with the object of photographing the whole of northern France, an area covering over 5,000 square miles. French agents had reported about a hundred of the sites, of which 31 were confirmed.

The bombing of places of manufacture and assembly was considered impractical as they were so well dispersed with locations mainly unknown.

Defensive measures
Once the bombs were launched they could be shot down. This could be done by aircraft, but faster fighters to intercept them were required. They could also be shot down by AA guns, but these needed new predictors and more sensitive proximity fuses, both of which were being manufactured in America. Barrage balloons were also deployed.

-----"-----

The V-1 attacks begin

The FZG-76 became known as the V-1 (Retaliation Weapon One) as soon as it became operational, but was popularly known as the doodlebug or buzzbomb from the sound it made. The initial attack was delayed by the bombing of communications for the D-Day landings in Normandy which disrupted the delivery of components required to complete many of the launching sites. The delivery of V-1s to already completed sites was also disrupted.

It was not until 13 June, seven days after the Normandy landings began, that the first V-1s were launched against London. It proved to be a bit of a damp squib. Of the ten bombs launched, five crashed shortly after take off; one went missing, probably crashing in the Channel; four crashed in SE England, with only one reaching London, crashing in Bethnal Green in the East End.

But this hiccough was soon corrected, and by 16 June, 244 V-1s had been launched from 55 sites, 144 reached England of which 73 reached the Greater London area. The bombs were launched in salvoes to swamp the defences and they flew at heights varying between 2,000ft and 3,000ft. By 22 June about 1,260 V-1s had been launched, of which 300 had been shot down and 450 had landed in London.

The reaction
The first reaction was to bomb the suspected V-1 supply depots and remaining ski sites, but because these had been abandoned as decoys, there was no reduction in the intensity of attacks. It was not until the end of June that it was realised that they were being supplied from elsewhere. The new modified launching sites were difficult to locate from the air, and as soon as they were destroyed, new ones sprang up. Radar tracks of the bombs gave an indication of from where they were launched. By the end of June, 63 had been located.

Enigma decrypts indicated the location of the new supply sites. One mentioned a cavern in Nucourt, others were St Leu d’Esserent and Rilly la Montagne. These three sites were more than just supply depots, they were actually used for assembling the V-1s from components arriving by rail from the factories. The modified launching sites were supplied from these depots with completed V-1s by road at night.

From Enigma, the number of V-1s assembled and distributed by St Leu d’Esserent was learned, and by comparing with the numbers launched, it became clear that this site was the main supply base. Consequently, it was heavily bombed on 4 and 7 July by 617 Squadron with Tallboy bombs. Enigma then showed that the tunnels had collapsed and approach roads to the depot were blocked and supplies were being diverted to Nucourt. Nucourt was bombed on 10 July and twice more on 15/16 July, resulting in its roof collapsing burying large numbers of flying bombs. On 17 July Rilly la Montagne was attacked. As a result of these raids there was significant but temporary reduction in the V-1 firing rate.

Deception
The double agents in London, who were controlled by MI5, were asked to report on the times and locations of where the V-1s struck. This caused a dilemma because if false information were given it could be discovered by PR and the agent’s credibility blown. On the other hand, they could not be given the truth as this would be aiding the enemy. This was resolved by correctly reporting the impact points but coupled with times of bombs falling short. It was hoped that this would give the impression the bombs were overshooting, in the hope that their range would be reduced and as a result, many more would fall short of London.

Unknown to Air Intelligence, there were two reasons why this should not work. One was that several bombs had been fitted with radio transmitters to enable the Germans to determine the points of impact accurately, and these showed that they tended to fall short. The other was that a Me262 jet had performed a PR flight over London to record the damage done by the bombs.

There were also three reasons why the deception did work. The first was that the credibility of the agents was such that they were regarded as being absolutely reliable, leading the Germans to suspect that their D/F of the V-1 transmitters was faulty. The second was that only North London was photographed in the PR flight as the southern half was obscured by cloud. As German PR had not photographed London for over three years, it appeared that the heavy damage done to north London had been done by the V-1s, thus supporting the agent’s reports. But in truth, most of the damage had been done in the heavy raids of 1941. The third reason was that it suited the Germans to believe they were succeeding in their campaign.


July-September 1944

In the first five weeks of attacks, about 3,000 bombs came within range of the defences which were able to shoot down about 43%. This improved to 50% when the defences were redistributed on 13 July, with most of the 800 AA guns being moved to coastal areas, leaving the area behind them as far as London free for the fighters, which also operated over the Channel.

[Note: The fastest and most successful fighters used were the Spitfire XIV and Tempest V, although others were used. They were modified to increase their speed by stripping out the armour, removing paint and polishing the surfaces, and modifying the engines to accept 150 octane fuel. This gained about 30 mph giving them a small advantage in speed over the V-1. But there were not enough of them, so a wing of Mustang IIIs were borrowed from Second Tactical Air Force. From the end of July the Meteor jet fighter entered service, but only in small numbers at this stage.]

About this time, the new proximity shells and radar predictors from America started to arrive in numbers. Thereafter, the number shot down improved rapidly until mid August when on average, about 75% were shot down. On 28 August, of the 97 bombs approaching England, 13 were shot down by fighters over the Channel; 63 were shot down by AA guns, with a further 10 by fighters as they approached London; two collided with balloons; three fell outside London and only four reached London.

The Allied advance from Normandy gradually overran the launching sites, when on 1 September the last V-1 from France was launched. But surprisingly the V-1s continued to arrive. Since 8 July, a number of bombs were tracked as coming from the east, approaching London along the River Thames, thus avoiding most of the defences which were oriented towards bombs approaching from France.

Towards the end of July, 25% of bombs were arriving this way, but no launching sites had been discovered either in Belgium or Holland. These bombs were all launched at night time and flew at the low altitude of 1,000ft. Because only the exhaust flame could be seen in the dark, rangefinding was very difficult, but a special rangefinder was developed for the intercepting fighters. It was concluded, mostly from Enigma decrypts, that these bombs were air launched from modified He111s of Gruppe 111/KG3 based in Holland. By monitoring signals from this Gruppe it became possible to give advance warning to the RAF of flying bomb attacks. It was also discovered that the Gruppe were to be reinforced by two more Gruppen by October.

On Christmas Eve 1944, fifty He111s launched V-1s towards Manchester from a position off the Yorkshire/Lincolnshire coast, thus bypassing the defences. Thirty crossed the coast, eleven came down within fifteen miles of the city centre, ten within ten miles, and only one landed in the city itself.

In total, about 1,200 were air launched against London, 90 against Southampton, and 50 against Manchester. The final attacks came on 13 January 1945, with an isolated attack on 29 March 1945.

The Germans continued their V-1 attacks against, mainly Antwerp, Liege and Brussels, until March 1945.

-----"-----
 
Statistics

The first V-1 attack against London was launched on 13 June 1944 from northern France, and the last ground launched attack on 1 September 1944. In this time period, the average daily rate was about 100 launchings. 1,176 V1s were launched from specially adapted He111s, to compensate for loss of ground sites.

Overall. 8,617 were launched of which 7,488 were detected, others presumed to have crashed prematurely. 3,957 were brought down by the defences, and 2,420 reached their targets. The British casualties, almost entirely civilians, were 5,500 killed, with 16,000 seriously injured.

Another 2,448 V-1s were launched against Antwerp, Liege and Brussels.

 
Postscript

V-1 technical data
1. It was 25feet 4inches long with a wingspan of 17ft 4 inches; propelled by an Argus pulse jet, and carried a warhead of 2,000lbs.
2. It flew at 1,000-4,000ft, at a speed of up to 400mph, and had a range of 120-140 miles.
3. It was controlled by autopilot with gyros and magnetic compass. The autopilot only allowed for small corrections in course after launching.

The pulse jet had insufficient thrust for take off, so the V-1 was launched from a ramp by a steam catapult powered by a mixture of hydrogen peroxide (T-Stoff) and sodium permanganate (Z-Stoff). Launching speed was about 250 mph until the V-1 reached its operating altitude where it slowly accelerated until it was travelling at 350mph when it reached the English coast and 400mph over London.

As designed, it was too fast to be intercepted except by the fastest of fighters. If it had been only 50mph faster, only a jet fighter could have caught it. It was most susceptible to AA fire because it flew straight and level and took no evasive action. This was particularly so when the new proximity shells came into service, and doubly so with the new American 90mm AA guns with their SCR 584 radar predictors.

In some respects the V-1 was the ideal bombing weapon. It was cheap and simple to mass produce. It was, barring accidents, only of danger to the enemy. It could be remotely launched out of the range of enemy weapons except aircraft, whose bombs were mostly ineffective against the launching sites. On the down side, it was inaccurate but good enough to act as a ‘retaliation weapon’ in reprisal for Allied bombing of German cities.
-----"-----
 
 
 
 

Mitch Hinde
20-10-2011, 19:35
Hi Bill

Have read all your posts in conjunction with Hugh Sebag Montefiore's "Enigma" and found them both extremely informative.
Incidentally, todays obituaries in the Daily Telegraph includes Gil Hayward who contributed to the construction of the machines at Bletchley Park.

Mitch Hinde

emason
21-10-2011, 17:24
Another 2,448 V-1s were launched against Antwerp, Liege and Brussels.


This is not correct, as these figures are for actual hits on the cities. The number of launchings was very much higher, over 8,000. Reliable figures are hard to come by, and it depends on how large an area is chosen to record a hit.

The same is true of London's figures. It depends on whether you count hits in central London, Greater London, the London postal district, or the London Civil Defence area.

emason
22-10-2011, 17:01
Hi Bill
Have read all your posts in conjunction with Hugh Sebag Montefiore's "Enigma" and found them both extremely informative.
Incidentally, todays obituaries in the Daily Telegraph includes Gil Hayward who contributed to the construction of the machines at Bletchley Park.
Mitch Hinde

Hello Mitch, the Hugh Sebag Montefiore book is an excellent one and very readable with many explanations of decryption, if only I could understand all of it!

Thanks for reading the posts and hope you didn't find too many errors in them.

Mitch Hinde
23-10-2011, 17:21
Hi Bill

BBC2 Tuesday evening 2100.
Code Breakers:
Bletchley Park's Lost Heroes.
Should be interesting.

Mitch Hinde

emason
26-11-2011, 18:39
From post #38 - V1 flying bombs

Deception
The double agents in London, who were controlled by MI5, were asked to report on the times and locations of where the V-1s struck. This caused a dilemma because if false information were given it could be discovered by PR and the agent’s credibility blown. On the other hand, they could not be given the truth as this would be aiding the enemy. This was resolved by correctly reporting the impact points but coupled with times of bombs falling short. It was hoped that this would give the impression the bombs were overshooting, in the hope that their range would be reduced and as a result, many more would fall short of London.
 
There was only one agent involved in this, and he wasn't German. He was an Englishman from Durham named Eddie Chapman.

Eddie Chapman was, in the 1930's a professional save blower who fled to Jersey to avoid arrest. But in Jersey he was recognised and temporarily avoided arrrest by jumping through a closed window of a restaurant. He had the misfortune to be in jail when the Germans invaded the Channel Islands.

As a ruse to be released from jail, he wrote to the German Commandant offering his services to the Germans. Nothing happened until he was transferred to a prison in France and had been there for a couple of months. His offer was accepted and he began training as an agent and saboteur. After training, he was given the code name Fritz and, in 1942, parachuted back to England where he immediately reported himself to the police, eventually coming under the control of MI5.

MI5 gave him the code name of ZigZag. He soon received orders by radio from his German controllers to blow up the De Havilland Mosquito factory. With the help of MI5 and scenery painters who made dummy damaged transformers and painted holes in walls, he planted explosives in a safe position and detonated them. This act of 'sabotage' was reported in the Press and, after confirmation by a photo reconnaisance aircraft, his German controllers were suitably impressed.

He was recalled to Germany via Portugal travelling on a merchant ship as a signed on seaman. The ship's captain was told by MI5 not to be surprised if his new recruit jumped ship in Lisbon. Once there Chapman reported to the German embassy where he was given a bomb disguised as a lump of coal and told to plant it in his ship's coal bunker. Instead he told the Captain of the ship and gave him the 'lump of coal'. The Germans never found out and assumed the ship was sunk. In doing this he proved his allegiance to both sides.

He was then taken to Germany and awarded the Iron Cross for his services, and then was taken for a long holiday in Norway where he was given all he asked for. After this was over he was sent to England again for his second assignment - to report on the damage the flying bombs were doing, for which he was to be paid £100,000. Again he arrived by parachute and reported to the police.

His price to MI5 for sending false information back to Germany was his freedom and his criminal record wiped clean. He ended the war a comparatively rich man, but boredom and his love of danger lured him back into criminal ways. After various escapades, he retired to Spain, where all British criminals seem to go, and died there in 1997.

He was the subject of a recent BBC program, and a book by Ben Macintyre called "Agent ZigZag".

harry.gibbon
23-05-2013, 16:09
The following excerpt is taken from an article entitled; Battle of the Atlantic: Mersey was only place for this final tribute.
by Peter Elson, The Liverpool Post May 23 2013

--------------------------------------------------------------


On May 25 and May 26, from 9.30pm until midnight, two high-intensity lasers will project from the Liver Building, transmitting the phrase “Thank You” in Morse Code

This is a nod to British mathematical genius Alan Turing, who helped crack the Enigma Code which saved many lives during the Battle of the Atlantic.

Source Liverpool Daily Post (http://www.liverpooldailypost.co.uk/liverpool-news/regional-news/2013/05/23/battle-of-the-atlantic-mersey-was-only-place-for-this-final-tribute-99623-33369778/), where the full article outlining the commerative programme can be read.


Little h

Rupert
23-05-2013, 16:31
I'm all for him being recognised, but it seems an odd way of doing it. He had no involvement in morse itself, or lasers as far as I know

harry.gibbon
23-05-2013, 17:37
I'm all for him being recognised, but it seems an odd way of doing it. He had no involvement in morse itself, or lasers as far as I know

Perhaps it stems from an earlier tribute last year, see following excerpts:-


Remembering Alan Turing
October 2012

Computer Art Image of the Month

Alan Turing Year 2012 continues apace with a variety of events inspired by the great contribution made by the mathematician and code breaker to the history of computer science and modern biology.

This month we continue to explore Turing’s influence on the strand of art that is interested in the intersection of arts and technology.

Due to the secret nature of the work Turing (and others) did during the Second World War, and the prominence of American mega-systems and companies subsequently, Britain’s contribution to computing innovation can only too easily be overlooked.

I think it right therefore that 2012, the centenary of his birth, has been designated ‘Alan Turing Year.’ Of course such achievements involve many people but I hope that the focus on Alan Turing’s fascinating work helps to educate those unfamiliar with the tremendous pioneering role Britain played and continues to play in this field.

After seeing Sir Tim Berners-Lee make an appearance in the London 2012 Olympic Opening ceremony, I believe it is indeed an exciting time for the convergence of computing history and culture in this country.

The future of Bletchley Park and its preservation as an important world heritage and educational site is now assured, thanks to major support from Google as well as other partners. (Earlier this year Turing was even honoured with a Google Doodle - a playable Turing Machine.

Last month we looked at Patrick Tresset who was inspired by the concept of machine intelligence put forward by Turing in 1950 and this month we’re considering the work of artists / curators Craig Morrison and Joel Cockrill who have been commissioned by the Arts Council of Wales to produce a laser and light installation honouring Turing’s life and legacy. Appropriately entitled Thank You, Craig and Joel’s piece will be shown at the blinc digital arts festival in Conway, North Wales, which this year is dedicated to the memory of Alan Turing.

Craig and Joel are the team who last year produced amazing digital projections onto the 500 year old walls of Conwy Castle and Plas Mawr, a Elizabethan era mansion, seen by over 8,000 people.
-----------------
Described by Joel as ‘an emotional response’ to Turing’s life and work, Thank You will use programmed ‘hyperboloids’, or rolling spheres, to sweep across sea and sky above the castle towers flickering at a frequency calculated using Morse code projecting the text ‘Thank You’ towards the heavens.

Source; BCS The Chartered Institute for IT (http://www.bcs.org/content/conWebDoc/48180) where the full article can be read and which includes an image/artists impression of the lasers.


Little h


Little h

emason
23-05-2013, 18:12
While Turing's contribution to decrypting the naval enigma was immense, his contribution to the attack on Tunny was minimal.

Perhaps the preceeding few posts might be more at home in the Enigma thread, as the subject of this thread has no connection to the Battle of the Atlantic - Tunny being an Army cipher.



Moderator Note: Last four posts moved to this thread as suggested.

harry.gibbon
23-05-2013, 20:08
While Turing's contribution to decrypting the naval enigma was immense, his contribution to the attack on Tunny was minimal.

Perhaps the preceeding few posts might be more at home in the Enigma thread, as the subject of this thread has no connection to the Battle of the Atlantic - Tunny being an Army cipher.



Moderator Note: Last four posts moved to this thread as suggested.


Sorry Bill and Terry,

Entirely my fault, did the search using 'Turing' and thought I had opened this thread with 'Enigma' in the middle of the thread title.:o I refuse to be backclassed to Chicksands though - Bletchley for 6 months was long enough :p

Little h