Only the message key – the positions the Enigma machine’s codewheels were turned to before starting encipherment or decipherment – changed with each message. The outer settings, i.e., the plugboard connections and the “Grund” (the positions at which the wheels were to be set in order to encipher the procedure indicator group to obtain the message key) were changed daily, usually at noon. The plugboard connections were then left in place until the next day’s key change, except when an “Offizier” message had to be enciphered or deciphered. The inner settings, i.e., the wheel order and ring settings, were changed every other day (except that in months with 31 days, there would be one 3-day period with the same inner settings, usually at the end of the month), and left in place until the next key change.

As to the time needed to encipher a message: in Appendix III to “Action This Day,” Ralph Erskine, after describing the “Kenngruppenbuch” indicator system used by most of the naval Enigma key-nets, tells of a monitored conversation in which a captured U-boat radio operator told another prisoner that at top speed, working with another radioman as a team (one man to key in the plaintext letters and the other to write down the ciphertext letters as they lit up on the lampboard), it used to take him at least five minutes to encipher a message of 40 groups (160 letters).

To this it should be added that some time would have been needed to have prepared the message plaintext – writing out numbers as words, replacing punctuation with certain letters, etc., before enciphering it with the Enigma. Also, before an outgoing message could be sent, the radio personnel had to set the wheels of the Enigma back to the message key and then decipher the encrypted text, in order to verify that it had been enciphered correctly.

For an outgoing message, the tasks of selecting the indicator groups, enciphering them according to the Kenngruppenbuch procedure, and using the unenciphered procedure indicator group to determine the message key all would have been done beforehand shortly after the daily change of keys, in order to save time. The practice was to prepare at the beginning of each day a number of cipher forms for use that day, each with a different pair of indicator groups, and fill in on each form both the unenciphered and enciphered indicator groups and also the resulting message key and the date for which it was valid. For incoming messages, however, the task of deciphering the indicators and using them to obtain the message keys had to be done on a message-by-message basis.

As to the time it would take to transmit a message, Arthur O. Bauer, in this article
[www.cdvandt.org]
(based on a chapter from his book “Funkpeilung als alliierte Waffe gegen deutsche U-Boote 1939-1945”), says that Kriegsmarine radio operators transmitted at a speed of from 60 to 70 letters per minute. He also says that based on a speed of 65 letters per minute, three sample messages averaging about 225 letters each would take an average of four minutes and 18 seconds each to transmit. Presumably this figure includes the time not just to transmit the message text itself, but also things like call signs, procedural signs, and the message preamble consisting of the time/date group, serial number and word count. Otherwise, the arithmetic doesn’t seem to come out right.

An example from U-628’s KTB might give a rough idea of the time it took for an exchange of messages:
[www.uboatarchive.net]
U-628’s commander (Hasenschar) composes a “very very urgent” message to B.d.U. time-stamped 1935 asking for authorization to hunt a lone northeast-headed ship. After being enciphered, the message is transmitted 17 minutes later at 1952. After it has been received and deciphered at U-boat headquarters, B.d.U. makes his reply, time-stamped 2039, stating that while there are no restricted areas in the North Atlantic at that time, the steamer in question is almost certainly a neutral Irish ship and is not to be attacked. 36 minutes later, at 2115, the enciphered reply is received by U-628.


Dan Girard