It was a form of Polyalphabetic cipher. The motion was governed by 3 (or possibly 4) with positive and negative lug settings as with the Hagelin machines. The motion was the sum of the positive settings, subject to an overlap principle, similar in principle to the M-209 device. Otto Buggisch did not know the cycle of the motion of the wheels, or the details of the construction by which they acted to vary the motion when the crank was turned.^[9]

Additional security was provided by using only 36 strips at one time, leaving about 4 slots blank. When these slots reached the enciphering position, a random letter was chosen and inserted in the cipher text, and the plain text letter was enciphered by the next strip that came to the enciphering position.^[9]

No ideas were ever formulated on the total number of strips to be used, or the frequency of settings changes. Preliminary tests by Dr Doering and Otto Buggisch, gave the machine a high security rating. However it was just as bulky as the Enigma cipher device, and could not print letters, which was then the chief improvement desired. For these reasons it was rejected, and only a lab model was ever built.^[10][9]

This cipher machine was based on Hagelin encipherment, but included a mechanism for variable stepping the Hagelin wheels. The device had six pin wheels which were mutually prime. The first five of these wheels had kicks of 1,2,4,8 and 10 respectively. The sixth wheel made these kicks positive and negative. The enciphering cycle of one letter, consisted of three elements:

1. This took place if, and only if, the sixth wheel had an active pin the motion index position. If this was the case, then all the following occurred: Wheel 1 moved one step. Each of the remaining four wheels moved one step unless the wheel to its left had an active pin in the motion index position in which case it would move two steps.
2. A key kick was generated which was the sum of all the kicks of wheels which had active pins in the kick index position. However, if the sixth wheel had an active pin in kick index position, the key kick would be 25 minus the sum of all the other kicks. Under such a circumstance, the key would complement itself.
3. Identical to step 1, except when it occurred whether or not Wheel 6 had an active pin in the motion index position. In this step, Wheel 6 also stepped one or two positions, depending on the state of Wheel 5.

The original specifications called for a lightweight, durable machine to be used by units forward of the division. Menzer designed it to provide a cipher tape, being keyboard operated in order to improve encryption speed. As a result of the keyboard operation, he was able to redesign the arrangement of letters on the print wheels to flatten the cipher frequency count.

During the war there was shortages of aluminum and magnesium resulting in the machine weighing between 12 and 15 kilograms, which was too heavy for field use. Removal of the keyboard would have made the machine lighter, but the design of the print wheels prevented their being directly used for encipherment. Production stopped because no one knew what to do. About 1000 machines were built and these were distributed to the Abwehr, which began using them in 1944.^[2] The Luftwaffe supposedly used these for 10 figure traffic, which was possibly for weather reports.^[11]

Two alphabets were written on the slide rule, 13 characters of each on the fixed base, and 13 characters of each on the top and the bottom of the sliding strip. The latter were so written that only one alphabet at a time was in phase. Alphabets could be changed as often as required.

In use, the slide was pulled to the right until it stopped, winding the spring that drove the mechanism. Pressing a button released the slide to move left. When at either or both of reading positions, ${\displaystyle A_{1}}$ ${\displaystyle A_{2}}$, the pinks were all inactive, the slide stopped and the encipherment took place. If the step came from ${\displaystyle A_{2}}$ alone, or ${\displaystyle A_{1}}$ and ${\displaystyle A_{2}}$ together, the slide took an additional step. When the slide stopped, either the top or the bottom alphabet would be in phase and the cipher value could be read off. Pressing the button again would allow the strip to slide left to its next stop.

Many Enigma devices were planned to be replaced with the Schlüssselkasten. It had a fairly high level of security. Given the alphabets on the side rule, it was possible to recover the pin patterns with a crib of about 30 letters. Without the crib, computer assistance would have been necessary and large quantities of cipher would have been required to recover the alphabets. A modification was considered in which two 26-character alphabets were slid against one another, rather than the 13 character segments. This would have increased the device security considerably, since more text would have been required to recover the alphabets. It would, however, have simplified recovery of the pin pattern after alphabet recovery. The solution of a single message was most unlikely^[2][12][13]

Various methods of key generation were used. On Chilean links, an 11-letter key word was numbered as for a transposition key, with the first digit of a two-digit numbers dropped. This key was extended by appending a two-digit group count and a four-digit time group:

A N T O F O G A S T A

1 6 0 7 4 8 5 2 9 1 3 1 2 1 4 4 0

On other links, a Fibonacci sequence of 100-125 digits would be generated through various manipulations of date, time and a secret number. If a message were longer that the key, it would be reversed as many as times as necessary. Key generation tables were also used.

In use, the key constituted the input to an autoclave. After aligning the alphabets according to a prearranged system or according to an indicator in the message, a stylus was inserted into the hole corresponding to the first key digit, and the lower disk was rotated clockwise until the stylus was stopped by the end of the notch. The plain text was then found on the upper disk and its cypher value read off the lower disk. The stylus was then placed in the whole corresponding to the second digit of the key, and the same procedure was repeated for the second letter of the text. Thus, the true key at any point in the same cypher was equal to the sum of all previous key inputs (mod 26).^[2]