|
|
|
|
|
|
|
Rotor Enigma K
Railway variant of Enigma K (A27)
- wanted item
During World War II (WWII), Nazi Germany used a
special variant of the commercial Enigma K cipher machine
for use by the Reichsbahn (German Railway).
It was a standard Enigma K with rewired rotors
and a rewired UKW.
Enigma traffic from the Reichsbahn was first encountered
by the codebreakers at Bletchley Park
on 25 July 1940 and all messages
were decrypted until the traffic ceased a month later, on 27 August 1940.
The wiring was recovered cryptanalytically [3].
|
According to internal BP reports, Colonel Tiltman was responsible
for breaking Railway Enigma during this period.
Early in the next year, on 23 January 1941, railway traffic was intercepted
again, originating from Eastern Europe, Russia and the Balkans.
Two weeks later, on 7 February 1941, the traffic was broken for the first time.
BP named the traffic Rocket, but later renamed it Rocket I.
Although there were initially no known images of this Railway Enigma variant,
it is now certain it was a standard Enigma K
with rewired cipher rotors and a rewired reflector (UKW). 1
|
|
|
|
In his 'Report on E Operations of the GC&CS'
[2], US codebreaker
William Friedman
claimed that Railway Enigma had a moving UKW, 2
but this is not the case.
As the machine only has one notch on each rotor, the
UKW would hardly ever advance, even if it was driven. The only
machines known to have a movable UKW are the
Zählwerk Enigma (A28) and
Enigma G (G31),
both of which have multiple turnover notches that cause frequent stepping
of the UKW. It is likely that Friedman meant that the UKW was settable,
which is the case with commercial Enigma D
and Enigma K.
|
-
In September 2023, an original Rocket I Railway Enigma was put up for auction
at Bonhams in London [6]. It allowed researchers to verify the wiring for the
first time [4][5] and compare it to the recovered wiring [3].
-
Moving means that the UKW is driven by the cipher rotor to its right.
|
After the first successful breaks in 1941, the railway traffic key was
named Rocket. It was mainly used on railway traffic networks in Eastern
Europe, Russia and the Balkans. It was later renamed Rocket I in
order to discriminate it from other railway networks.
Breaking Rocket I was relatively easy for BP, and the network provided
good intelligence about production and movement of supplies.
Nevertheless, problems with Rocket I were reported on 19 September 1944,
and it wasn't before 28 October 1944 that BP gained entry into the traffic again.
The blackout was apparently caused by eccentricities in the cribs
during this period [1].
The problems with Rocket I illustrate that even a standard Enigma K,
without the Army's plugboard (Steckerbrett) could be hard to break if
the contents of the messages were less predictable. It also shows that the
Bombes were virtually useless without good cribs.
|
In September 1942, a similar key appeared for Western Europe. The new
key was called Rocket II and was only broken once when it was using
the Rocket I machines. Apparently, the traffic consisted of practicing
messages only. Judging from the characteristics of these messages, the
(practicing) traffic continued until May 1944, after which a new key,
and possible another machine, was introduced. They new key was called
Rocket III and remained unbroken for quite some time.
For a long time it was unclear whether Enigma was used at all.
After capturing some key sheets in August 1944, some of the older
messages were decrypted. It became clear that Enigma was used, but
that the contents of these messages were sufficiently 'obscure'
so that cribs could not be used easily.
It is most likely that Rocket II and III used the
(military) Service Enigma (Enigma I).
Rocket II was later renamed Blunderbuss.
|
|
Below is the wiring of the rotors of the standard Railway Enigma (Rocket I), as
recovered in 2023 by Patrick Hayes, from Enigma K with serial number K438
that turned up for auction at Bonhams in London (UK) [4]. 1
This wiring has meanwhile been confirmed independently by Detlev Gross
who measured the UKW with serial number K456 [5].
This is the correct wiring of the rotors:
|
|
Rotor
|
ABCDEFGHIJKLMNOPQRSTUVWXYZ
|
Notch
|
Turnover
|
#
|
|
|
|
ETW
|
QWERTZUIOASDFGHJKPYXCVBNML
|
|
|
|
|
I
|
EVLPKUDJHTGSZFRABWYICOXNMQ
|
G
|
Y
|
1
|
|
II
|
HXMQKGJTSCZFLBERNAWYIDOVPU
|
M
|
E
|
1
|
|
III
|
JHDBSKYPZNMVXURECLIGQOAWTF
|
V
|
N
|
1
|
|
UKW
|
DNSAJQIPGEXRWBVHFLCZYOMKUT
|
|
|
|
|
Note that the physical wiring is different from the wiring
that was recovered cryptanalytically by Bletchley Park
during the war, which was based on incorrect assumptions.
They are equivalent however, provided that appropriate adjustments are made
to the ring settings for a given daily key.
In the same vein, the turnover positions of rotors I and III are swapped as
result of a misidentification by BP, rather than a physical swapping of
the alphabet rings by the Germans.
BP was aware of both differences, at least by 1944,
and knew which corrections were needed.
Below is the wiring as it was cryptanalytically recovered during the war by
the codebreakers at Bletchley Park (UK).
Full details were given by Philip Marks in Cryptologia of 2015/1
in the article
Enigma Wiring Data: Interpreting Allied Conventions from WWII
[3].
|
|
Rotor
|
ABCDEFGHIJKLMNOPQRSTUVWXYZ
|
Notch
|
Turnover
|
#
|
|
|
|
ETW
|
QWERTZUIOASDFGHJKPYXCVBNML
|
|
|
|
|
I
|
JGDQOXUSCAMIFRVTPNEWKBLZYH
|
V
|
N
|
1
|
|
II
|
NTZPSFBOKMWRCJDIVLAEYUXHGQ
|
M
|
E
|
1
|
|
III
|
JVIUBHTCDYAKEQZPOSGXNRMWFL
|
G
|
Y
|
1
|
|
UKW
|
QYHOGNECVPUZTFDJAXWMKISRBL
|
|
|
|
|
-
Initially the wiring measured by Patrick Hayes [4] appeared to be different
from the wiring measured by Detlev Gross [5],
but it later turned out the the UKW had a offset of 13 positions when it
was measured.
|
- David Hamer, Geoff Sullivan and Frode Weierud,
Enigma Variations: An Extended Family of Machines
Cryptologia, July 1998, Volume XXII, Number 3.
- William Friedman, Report on E Operations of the GC&CS at Bletchley Park
Signal Security Agency, Washington. 12 August 1943. pp. 39, 67.
NARA, RG 457, NSA Historical Collection, Box 1126, Nr. 3620. 1
- Philip Marks, Enigma Wiring Data: Interpreting Allied Conventions from WWII
Cryptologia Volume 39, 2015, Issue 1.
- Patrick Hayes, Correct wiring of the Railway Enigma K (A27) with serial number K438
Personal correspondence, 17 April 2023.
- Detlev Gross, Owner of UKW of Railway Enigma K456
Personal correspondence.
- Bonhams, 'The Railway Enigma': a very rare H&R K-model Enigma cipher machine
Auction: Instruments of Science and Technology Online.
12 September 2023.
Enigma K with serial number K438.
|
-
Declassified by NARA on 6 June 2003.
|
|
|
|
Any links shown in red are currently unavailable.
If you like the information on this website, why not make a donation?
© Crypto Museum. Created: Saturday 18 September 2010. Last changed: Monday, 30 October 2023 - 08:14 CET.
|
 |
|
|
|
| |
