Recorded below is the definition of what we have called the target machine.
The simplified definition is, that we are rebuilding a Basic machine as defined in a BTM parts list dated 23.10.43 plus later amendments. Basic in this case means the later series of machines that can be built up into various models as compared with the earlier ones which were detailed specifically to the type in question. 68 machines were built to this standard.
Quoting the specification as defined by Bletchley Park. - On 31st January 1944, the first "New Standard" machine was delivered, and was installed at Out Station Eastcote. It was numbered 250, [named Nice] this series being quite distinct from previous types. A separate set of numbers was commenced at the works. The New Standard was basically the same as the Old Standard, i.e. speed approximately the same; number of enigmas; and type of drums, but it used Siemens B.T.M. relays and was less flexible.
Our machine is modelled on an almost identical machine number 297 and named Atlanta. This was delivered to the American bay at Out Station Eastcote. in July 1944.
The basic machine is a 3 wheel 36 Enigma version using Siemens type sense relays but allows for certain variations within the parts schedule. The first is the type of commutator used behind the 36 fast drums. When run for 24 hours a day, 7 days per week the piano wire drum brushes wore grooves in the face of the commutators. When these had to be replaced it took a lot of 'out of service' time to undo and redo the 3848 or so commutator connections. BTM tried to improve on this by inventing replaceable commutators for the fast positions which were in effect thin additional commutators bolted to the face of the original ones. The theory being that the wearing face could be replaced by just undoing four bolts. Two RAF engineers have independently told me that whilst the idea was good in theory, reliability suffered badly because the connections where the original and additional commutators came together in 104 places relied solely on the pressure created by just the 4 screws. Apparently the Bakelite warped and the brass to brass connections became unreliable. Apparently many machines reverted to the original design. Bearing in mind that we don't expect anything like 24 hour working we have decided to take the ex RAF maintenance engineers advice and keep to the original simple arrangement. This also has the benefit of considerably less manufacturing effort.
The parts schedule also allows for two different 'point' numbers, either 30 or 39. This needs a little explaining. The Bombe mechanism is not a true cyclometer and cannot carry between adjacent commutator segments of the upper drum and at the same time continue sensing without loss of time. In reality the carry takes from 13 character times or points to as little as 4 depending on the model. A point being 1/26th of the rotation of the top drums or 1 sense position. What actually happens is that 26 sense action will take place on the top drum and then it will 'idle' over a given number of points or sense positions whilst the carry mechanism ratchets the middle row of drums on one position. When this carry action is complete sensing will recommence for 26 points. The 'start sensing' position is not important to the code breaking process so long as all 26 positions are tested before the carry takes place. To complete the story the slow or bottom drums also ratchet on one position at the same time as the middle drums in one case out of 26.
Getting back to the 30 versus 39 point options. The former is capable of performing a carry in 4 points whereas the latter needs 13. All other gearing being equal a 30 point machine is therefore 30% faster than a 39 point but at a cost. In order to perform a faster carry, sharper cams have to be fitted together with extra gearing and various other items. Forty eight extra spur gears plus all their supporting shafts and bearings etc. can be left out of a 39 point machine. It has therefore been decided to make our Target machine a 39 point. What is the impact I hear people ask? Well a 39 point will cover all 26 x 26 x 26 sense positions in 20.1 minutes. But a stop should be found on average in half this time of 10 minutes and 3 seconds. A 30 point machine would complete a typical job 30% faster in just under 8 minutes or just over 3 minutes faster. In the demonstration environment, which we aim to achieve, this small saving in time can hardly be justified when set against the considerable amount of extra manufacturing effort.
The target machine is therefore a 39 point machine. Incidentally some ex Bletchley Park people called this a 13 point carry machine which is another way of saying the same thing.
The brief specification is as follows
If we leave aside the early prototypes and experimental machines, there were two types of Bombe Frame produced by BTM during WW II. The first type was used as part of the first '3 wheel' Bombe delivered between March 1940 and March 1943.
By March 1943 the urgent need to break German 4 wheel Enigma codes could begin in anger when the first BTM 4 Wheel Bombes were delivered. It was these machines that brought about the second type of frame which added a major heavy section at the right hand end. This section was required to mount the additional 36 'Very Fast' Fourth Wheels. It is this second type of frame that has been re-built.
By January 1944 it had been realised that Bletchley Park together with the Americans had sufficient numbers of 4 Wheel Bombes but needed more 3 Wheel versions. BTM chose not to revert to the first frame but to standardise on the second type which could be used in either 3 Wheel and BTM 4 Wheel assemblies. A version of BTM Bombe was also used with the Post Office/Mawdsley 4th wheel attachment known as W.W. (Wyne-Williams) or Cobra. This used the earlier frame.
Our re-built Bombe will be built as a later type of 3 Wheel Bombe; virtually the same as those which would have been delivered to Out-station Eastcote commencing January 1944.
Return to Bombe Index