150 | 14 TUNNy: HITlER’S BIGGEST fISH
the wheel rotated. The operator could adjust the cams, sliding any that he selected sideways, so
that the selected cams became inoperative and no longer pushed the switch when they passed
it (Fig. 14.4). As the wheel turns, it now no longer produces a uniform stream of pulses, but a
pattern of pulses and non-pulses (crosses and dots). Once the operator had adjusted the cams
in this way, their arrangement around the wheel, with some operative and some inoperative,
was called the ‘wheel pattern’.
It was the pattern of the cams around the wheels that produced the chi-stream and the psi-
stream. Whenever a key was pressed at the keyboard (or a letter read in from the tape in ‘auto’
mode), it caused the five chi-wheels to turn in unison, just far enough for one cam on each
wheel to pass its switch. Depending on whether or not that cam was operative, a cross or a dot
was produced.
Operative Inoperativefigure 14.4 A wheel cam in the operative (left) and inoperative position.
Redrawn by Jack Copeland and Dustin Parry from ‘General Report on Tunny’. All rights reserved.
Suppose, for example, that the cam at the first chi-wheel’s switch produced a dot (no pulse)
and the cam on the second likewise produced no pulse (a dot) at its switch, but the cams on the
third and fourth both produced a cross (pulse), and the cam on the fifth produced a dot: then
the letter that the chi-wheels produced at this point in their rotation was ••xx• (N). The five
psi-wheels also contributed a letter (or other keyboard character) and this was in effect added
to N to produce a character of the key-stream. Under the rules of Tunny addition, it makes no
difference whether the letter produced by the chi-wheels and the letter produced by the psi-
wheels are regarded as being added to each other, with the resulting letter then being added to
the plaintext letter, or whether the letters produced by the wheels are regarded as being added
successively to the plaintext letter as shown in Fig. 14.3.
There was an important complication in the motion of some of the wheels. Although the
chi-wheels always moved round by one cam every time a key was pressed at the keyboard (or a
letter arrived from the tape, or from the radio receiver), the psi-wheels moved irregularly. The
psi-wheels might all move round one cam in unison with the chi-wheels, or they might all stand
still, missing an opportunity to move. This irregular motion of the psi-wheels was described as
‘staggering’ at Bletchley Park. The motions of the two motor wheels determined whether the
psi-wheels moved with the chi-wheels, or stood still. Tunny’s designers presumably believed
that this irregular motion of the psi-wheels would enhance the security of the machine—but in
fact it turned out to be Tunny’s crucial weakness.
The first break into Tunny
When messages from the unknown new machine were first intercepted, nothing of their content
could be read. But in August 1941 Colonel John Tiltman (Fig. 14.5) scored a tremendous suc-
cess, managing to break a message about 4000 characters long. One of codebreaking’s legends,
Tiltman had made a series of major breakthroughs against Japanese military ciphers before