This is Bletchley Park Mansion, known to the codebreakers and all who worked on the site simply as "The House".
Dedication stone for the memorial to those who worked on the site.
Memorial to the workers on the site. This is sited outside the building which now houses the Bombe machine. Can you see the message in the inscription?
Turing Bombe
Front view of the reconstructed Bombe. The rotors are arranged in 12 vertical banks of 3. Each vertical bank represents a 3 rotor Enigma cypher machine. There are three horizontal banks of 12 so 36 parallel operations can be run on the machine simultaneously. The centre row has an extra bank of three rotors at the right hand side, these are used to indicate the result when the machine enters a stop state indicating that a possible match has been found.
The rotors operate in a reciprocal manner, Each have 26 positions corresponding to the letters of the alphabet. Each drum was internally wired to operate parallel checking operations simultaneously. Wiring is terminated on 104 brushes arranged concentrically on the reverse. The brushes provide an electrical contact and run against a fixed commutator mounted around the centre pivot for each drum. When the machine is operating, the top drum rotates continuously, the middle drum advances by one letter position for each rotation of the top drum, the bottom drum advances by one letter stop for each full rotation of the middle drum. This emulates and checks all the possible top wheel and ring combinations on a 3 wheel Enigma machine.
Enigma Enciphering Machine.
The number of possible ring settings on such a machine is 26 x 26 x 26 (26^3) = 17,576, however enciphering was further complicated in that the wheels could abe located in different positions when the machine was set up for the day, the "wheel order". This adds a further permutation of 6 possible combinations according to how they were arranged, this brings the number to 105,456 possible start positions for the basic 3 wheel enciphering machine. Enigma was developed in the 1920's and constantly enhanced throughout the 30's. Another permutation was added by way of using 3 rings from a set of 5 which increase the number of possible "wheel orders" to 60.
The last part of the encryption process uses a front mounted alphabetic plug board on the machine. This allowed the operator to further scramble enciphered letters. If A was connected to S, S would be produced if a letter was encoded as A and A produced if the output was an S. Typically the operators would use 5 or 6 plug links. If 6 are used the possible combinations of plug positions is 100,391,791,500. Multiply this out with the combinations available from the wheels and you start to see why the Germans thought that the system was fully secure.
This Wiki has further information and documents the development of the many variants that were produced:
Bombe
This is the Bombe machines answer to the Enigmas plug board. This is set up along with the start positions of the rotors in accordance with the crib and resultant menus produced by the manual decryption teams who worked on the intercepted messages.
This is the rear view of the Bombe with the chassis hinged open, the panel on the right has the sensing relays that are driven by the rotors on the front of the machine.
This is a closer view of the back chassis of the Bombe showing the bearings for the rotors and the automatic lubrication system employed, this was fed from a tray of lubricant mounted under the machine.
Checking machine this was used to quickly check the settings produced by the Bombe machine when it reached a "stop" state. The machine was set up according to the results from the Bombe. The encoded message was then checked to see if the resultant output was plain German text. If this was the case, the run was declared at end. The text could be passed back to the analysts to further refine the decrypt. The Bombe could be stripped and made ready for the next menu to be received and run.
The Bombes were operated by WRNS from various huts around the site. On average it took about 30 minutes to set up the rotors and plugboard on a Bombe from the menu's supplied by the decrypt teams prior to starting the run. I would imagine that the noise in the various Bombe rooms with multiple machines running at the same time would be similar to a weaving shed. The women worked 8 hour shifts on their feet to monitor the operation of the machines and deal with false stops and breakdowns. Interestingly, the Bombe rooms were the first to be fitted with fluorescent lights outside of government buildings which were the only other establishments using them at the time. Veterans have commented that they would emerge from a shift covered in oil and smelling of electrical arcing from the rotors and commutators.
A couple of views of the British TYPEX machine
Here is a close up, not a very clear picture as the machine is under a concave perspex cover and the picture has picked up the reflection from the fluorescent room lighting.
The British TYPEX machine was a 5 wheel encryption and decryption device it was fitted with a printed tape output so the machine could produce hard copy of resultant decoded data from encrypted data entered via the keyboard. The machine could emulate 3 and 4 wheel German Enigma machines. We used these machines for our most secure encrypted communications. The 5 wheel system we used was never compromised by the Germans.
The Bombe display room also contains a section devoted to Alan Turing who was instrumental in developing the techniques used for the original manual decryption process and converting this into an electro mechanical operation.
The sculpture is constructed from multiple layers of slate which represent the complexity of the the levels that he had to solve to achieve the breakthrough. He is depicted seated, studying the operation of an Enigma cipher machine. Quite a striking image, looks "digital", which I suppose is part of the intention as well. A brilliant mind, the man was persecuted by the establishment that he served based solely on his sexuality and it is forever to the shame of the Nation that it took over 60 years before a pardon was granted.
Onward to Colossus
German High Command used a different enciphering system, this was called Lorenz and unlike Enigma it used the 32 symbol teleprinter Baudot code. Baudot output is produced as a five channel data stream. The machines used 12 wheels to encode the 5 channel data from tape. The number of starting permutations possible from this arrangement are staggering and were beyond the scope of the existing Bombe machines. At the speed that the mechanical Bombes ran at it would take months to process a single data stream produced by this method of encoding. Another solution was required.
12 ring Lorenz Cipher Machine
Signals were first heard in 1940 which caused Bletchley to prick it's ears up and investigate. A small team was set up to look at possible ways of cracking the code. The output from Lorenz is effectively a digital data stream with various logic algorithms applied during the encryption process. It took two years of intensive research to manually work out the logic that was used in the encryption process. The Post Office Engineering Research Department were asked if they could produce a machine to emulate the logic process. The machine was called Tunny and could be programmed with the logic data produced by the manual decrypt process. The enciphered text could then be fed into the machine simultaneously and the logic applied by the machine, resultant output is decoded text.
The Tunny is made up from a number of interlinked units.
Teleprinter Stations
Plug board for configuring logic settings on the Tunny prior to a run.
Dual channel teleprinter tape drive and synchronisation and logic circuitry. This was known as the "Heath Robinson" by the WRNS and other operators of the machine. One tape processed the logic data stream from the manually worked out wheel positions provided by the decrypt team and the other ran the enciphered radio message to be decrypted.
This solution worked and could successfully decode the encrypted teleprinter data back into plain German text. The method had a number of drawbacks though. Both channels of the tape input on the Heath Robinson had to be perfectly synchronised or the run would fail. The speeds that the tape had to run at to process the data (1000cps) made the synchronisation critical and frequent dropouts were experienced. It took up to 6 weeks to produce the logic menus manually from the intercepted radio messages so resultant decrypts although accurate were useless as the information was out of date.
Colossus
The machine worked though and the next step was to refine the process. Tommy Flowers a brilliant engineer from the Post Office Research Centre was asked to look at the problems and see if there was anything that could be improved on. What he came up with was what we now know as Colossus. The Heath Robinson solution for running the ring operations from a continuous tape loop was replaced by emulating the ring operations using electronic valves. This replaced the need for a dual synchronised tape system. With the ring emulations now running electronically in the valve circuits, a single channel higher speed tape input could be used to input the encrypted message stream. This ran at 5,000cps with no dropouts.
Front view of Colossus
Beadstead high speed tape input used for inputting the encoded teleprinter data.
Rear view of Colossus
Side view of Colossus showing the dual rack construction to hold the various banks of valves and other circuitry.
Close up of one of the valve banks used to emulate the ring operations of the Lorenz.
The first Colossus was delivered to Bletchley just before Christmas in 1943. By January it was operational and deciphering Lorenz encoded data. The machine reduced the processing time from weeks to hours and was operational in time to confirm that the German High Command had swallowed the deception plan put in place before D Day that convinced them that the main attack would be in the Pas de Calais and not Normandy. The Germans used dedicated land line based teleprinter lines for the bulk of their communications of Lorenz. After D Day the Allies targeted this infrastructure which forced the Germans to use radio the amount of traffic available for intercept and decryption went through the roof. To cater for this, ten Colossi were built and operated from various locations on the Bletchley site. By the end of the war 63 Million characters of German text had been intercepted and decoded by a team of 550 operatives at Bletchley and the various Lorenz intercept stations.
It is interesting to note that the team involved with breaking the Lorenz code did not actually see a Lorenz cipher machine until just before the end of the war when one was captured and returned to the park for analysis. The team had been breaking the code for two and a half years! A remarkable achievement when you think that the whole system was reverse engineered from an enciphered teleprinter tape.
This is a link to the late Tony Sales Codes and Ciphers site which documents the Lorenz cipher system and development of the Tunny decryption system and ultimately the Colossus rebuild which he undertook when he was director of the Bletchley trust.
The Lorenz Cipher System and how Bletchley Park broke it
After VJ Day eight of the Colossus machines were dismantled at Bletchley. Two were transported to Eastcote where the Government Code and Cypher Section were based after the war. Some of the key personnel from the Bletchley site were tranferred there also. This organisation eventually morphed into GCHQ and moved into pupose built accomodation at Cheltenham where they are still based. The remaining two Colossus machines were dismantled in the 1960's and all the technical documentation destroyed.
The work that went on at Bletchley and the technical developments made by the various departments both in the techniques of cipher breaking and the hardware developed to aid in that process were top secret. Churchill regarded the work as greater than top secret and designated and referred to the project as ULTRA. Nothing was revealed about the codebreakers or the systems they developed until 30 years after the end of the WWII. Some of the veteran operatives that are still with us will not talk of their work at Bletchley.
One of the direct consequences of this allowed the Americans to lay claim to designing the worlds first electronic computer ENIAC which was first run in 1946. This myth was continued until information started to emerge in the late 1970's and was finally put to bed when the replica Colossus built to original specification was run on the 50th anniversary of the switch on of ENIAC in 1996 finally proving that the technology was in operational at Bletchley two years before the American imposter.
