|
|
|
|
|
|
|
T&N Menzer HSE →
Since the SG-39 had to be backward compatible with the
Naval Enigma M4 — the Kriegsmarine specifically
requested that —
it had three driven cipher rotors with 26 contact points each, plus a fourth
settable (but not driven) rotor (i.e. stator). Furthermore, the rotors had
stepping notches that were compatible with Naval Enigma M4.
In contrast with the Enigma however, SG-39 had two plugboards:
One between the keyboard and the first rotor, and one at the reflector
(UKW). Furthermore the electrically wired rotors were driven by
a set of Hagelin-style pin-wheels of length 21, 23 and 25.
Each pin-wheel position had a pin to make it active or inactive,
resulting in an irregular stepping pattern of the rotors.
|
|
|
Because of these features, the machine can be regarded as a hybrid between
an Enigma rotor machine and a
Hagelin pin-and-lug (pin-wheel) cipher machine.
As this combination made the machine cryptographically much stronger than
the individual Enigma or Hagelin, it was thought
to be a suitable candidate to replace the Enigma in due course.
However, due to the inability to decide on the final features of the machine,
and the selection of the SG-41 (Hitlermühle)
as an Enigma replacement candidate,
the SG-39 was not taken into production.
Shortly after the end of the war, a TICOM inspection team
visited the T&N factory in Frankfurt-am-Main,
and identified three prototypes, of which only one was complete (and missing).
The most incomplete prototype was taken to the US
and is shown in the images above and below.
Although the SG-39 was never taken into production, some of its ideas
were reused after the war (1953-1957) for the development of the so-called
High-Speed Enigma, in cooperation with the newly established
German cipher authority ZfCh. 2
In particular the combined use of electrical cipher rotors and a
stepping mechanism driven by Hagelin-type pin-wheels, became one of
the design features. The HSE was used by the
German Foreign Office from 1957 to the mid-1960s.
➤ More about the High Speed Enigma
|
 |
-
Heereswaffenamt was the central office for the technical development
and production of weapons, ammunition and equipment for the German Army
(arms research), located in Berlin. It had several departments, of which
Prüf (also written as 'Pruef') was responsible for development and testing
[9].
-
At the time still known as Studiengesellschaft zur Förderung
Wissenschaftlicher Arbeit mbh. (SFWA) — Society for Scientific Work.
In 1959 it officially became the
Zentralstelle für das Chiffrierwesen (ZfCh).
|
Although there were different development versions of the SG-39,
the final machine was backward compatible with
Military Enigma I and Naval Enigma M4.
The drawing below is an educated guess of the SG-39's appearance, based on
the post-war TICOM
interrogation of Wachtmeister Otto Buggisch,
who had worked at OKH/Chi and OKW/Chi during the war, and was able to give
a detailed account of the machine's capabilities [5].
It is not entirely clear whether the reflector (UKW) was replaceable
(like UKW-D),
or whether it was wired to another plugboard at the right.
In any case, the design is pretty similar to the drawing below, which was
extracted from German Patent DE 1,045,692.
It is believed that it describes the final version of the SG-39.
The patent was filed by T&N well after the war,
in August 1952, just before they started the development of the
High Speed Enigma (HSE). It is likely that this was done
to secure the patent rights to their wartime developments, which formed
the starting point for the development of the HSE in 1953.
|
| | ME1 | End plate 1 (ETW) | TW | Keyboard |
| | ME2 | End plate 2 (to UKW) | B | Mode selector |
| | MA | Stator | TA | Plugboard (Steckerbrett) |
| | MB | Cipher rotor 3 | TU | Reflector (UKW) |
| | MC | Cipher rotor 2 | U | Electronic 26→5 encoder |
| | MD | Cipher rotor 1 | DA | Plaintext printer |
| | RA | Pin-wheel 1 | DB | Ciphertext printer |
| | RB | Pin-wheel 2 | M | Motor |
| | RC | Pin-wheel 3 | 44 | Turn-over notches |
| | 97 | Mechanical 26→5 encoder | 151 | Start-stop clutch |
|
In the patent drawing, the two printers are shown side-by-side at the left
side of the machine. In practice, the space to the left of the keyboard is
limited, so it is also possible that the two printers were placed
behind the keyboard and on top of each other, or that they were combined
in one integrated assembly.
The keyboard has 28 keys plus a spacebar, whereas the rotors have only 26
contact points for the 26 letters of the latin alphabet (A-Z). It is likely
that the letter 'J' was used as a SPACE-character, as described in
Patent DE 975,036 [III].
In that case – when in cipher mode – pressing the spacebar actually
selects the letter 'J'. In this mode, the key 'J' is
mechanically blocked. In case a 'J' was required in the text, the user
had to substitute it for an 'I', which does not affect the
legibility of the text.
As a result, only the letters A-Z appear in the ciphertext.
When deciphering, the spacebar is mechanically blocked, whilst the
deciphered letter 'J' is automatically replaced by a space.
The function of the keys with the white dots at the bottom left and bottom
right of the keyboard is currently unknown. It is possible that these are
leftovers from versions that were teleprinter-compatible. In that case they
might, for example, have produced the ITA-2 codes for
CR and LF.
It is likely that the plugboards were placed at the right side of the machine.
As the space at the right side of the keyboard is limited, we assume that the
plugboards were placed vertically, and that they were accessible from the
right side of the device, possibly behind a lid in the enclosure.
The diagram above shows an educated guess of what the plugboards might have
looked like. At the bottom is the regular plugboard that is inserted between
the keyboard and the entry disc (ETW). Although in the patent [III]
the keyboard is drawn as a single-ended one, the description does not
specifically claim that. Therefore, we assume that the SG-39 had a double-ended
(reciprocal) plugboard, just like the Enigma, which is
cryptographically weak by design. In that case, not all patch cables had
to be installed. Any unsteckered letters are then self-steckered.
In case the plugboard was single-ended, each letter could be patched to
each other letter (rather than being swapped in pairs), which would have been
cryptographically much stronger and way more difficult to break, without
loosing compatibility with Enigma. This would have been very
similar to the second design of the Enigma Steckerbrett,
which was never released.
Note however, that in that case, all 26 patch cables must be installed
on the plugbord at all times.
In the drawing above, the plugboard of the rewirable reflector (UKW) is shown
at the top. It has 26 single banana-type sockets — one for each letter —
all of which must be patched. That means that all 13 strap wires must
be installed on the UKW plugboard at all times.
➤ History of the Enigma plugboard
|
|
|
Differences with Enigma M4
|
|
|
- Motor-driven
- Faster operation
- Irregular rotor stepping
- Stepping controlled by pin-wheels (in addition to rotor notches)
- Printed output (on two separate printers)
- Configurable reflector (UKW)
- Single-ended plugboard 1
|
-
This is hypothetical. It is shown in the patent [III] as a
single-ended plugboard, but it is not claimed as such in the description.
Furthermore, none of the interrogated eyewitnesses have claimed that it
was single ended [5][6][7]. A single-ended plugboard is
cryptographically much stronger than a double-ended (reciprocal)
one. For backward compatibility with Enigma, it had to be used
in double-ended configuration.
|
|
During the development of the SG-39, the properties and specifications of
the machine were significantly changed a number of times. Since there are
no known designators for the various incarnations of the machine, we will
denote them here as a, b, c and d, in analogy to [1] and [5].
|
- Combined rotor and pin-wheel machine
This version has three cipher rotors without stepping notches and without
a ring-setting (Ringstellung).
The rotors have 32 contact points each (rather than 26).
Rotor stepping is entirely controlled by three pin-wheels (N1, N2 and N3)
of length 21, 23 and 25 respectively.
The reflector is pluggable
(like Enigma's UKW-D) and can be swapped easily.
A prototype of this version was demonstrated in 1942 [5]. It was equipped
with an automatic morse transmitter.
Note that this version is not compatible with the Enigma.
- Backward compatible with Military Enigma I
The wiring of the three rotors (I, II and III) is now identical to the
wiring of the Enigma I rotors.
Furthermore, the rotors now have a ring-setting
(Ringstellung) and a turn-over notch. 1
In addition, a plugboard is inserted before rotor I.
Rotor stepping is controlled by the stepping notch on the circumference
of the rotors (complete with the double-stepping of the middle rotor [4]),
and also by the pin-wheels. 2 If the two coincide, a single step is made.
This version is backward compatible with the 3-rotor Enigma I.
- Backward compatible with Naval Enigma M4
With this version an extra rotor is inserted between rotor III and the
reflector (UKW), just like the extra rotor (ZW) on the
Enigma M4.
As this extra rotor is not driven, it is also known as
a stator. It makes this version backward compatible with
military Enigma I
and Naval Enigma M4.
Also new in this version is that the notch of rotor III causes rotor I
to make a step, but this does not affect backward compatibility with
Enigma.
A prototype of this version was demonstrated in 1944 [5].
- Modified rotor stepping
Exact details of this version are unknown, but according to a post-war
TICOM interview with Otto Buggisch of OKW/Chi,
he said that the was told by Dr. Stein (OKW/Chi) in the summer of 1944,
that the reciprocal influencing of the rotors was to be altered.
The machine remained backward compatible with
Enigma however [5].
|
|
-
The turn-over notch is constructed differently, but is functionally
compatible with the turn-over notch of an Enigma rotor, complete with
double stepping of the middle rotor as described by David Hamer in [4].
-
For backward compatibility with Enigma, all pins of the pin-wheel that
drives the rightmost rotor (1) should be set to active, whilst the pins of
the other two pin-wheels must be set to inactive.
|
During World War II (WWII) several cipher machines were
used by the German Armed Forces at various levels of command, including the
Lorenz SZ-40/42,
the Siemens T-52 (Geheimshreiber)
and the Enigma. Of these machines, Enigma
is arguably the most well-known one, since more than 35,000 units were
used during the war for tactical traffic.
At several moments in time however,
the security of the Enigma was questioned and replacement
candidates were evaluated.
One of the potential candidates was
Schlüsselgerät 39 (SG-39) of which the development had been initiated in 1939.
Although the invention of the SG-39 is often attributed to
Fritz Menzer, it appears to have been
a joint development of the Heereswaffenamt
(Wa Prüf 7) and the
manufacturer Telefonbau und Normalzeit
(T&N), in which
Menzer undoubtedly played an important part.
Over time, various incarnations of the machine were built and evaluated,
as listed above.
At some point it was decided to make the SG-39 backward compatible with
Naval Enigma M4. By doing so, the SG-39 would be backward
compatible with all plugboard Enigma cipher machines, as the M4
is backward compatible with the military Enigma I when the 4th
rotor (ZW) is set to 'A'.
|
| |
Simplified Enigma M4 circuit diagram.
Move the mouse over the diagram to see
what happens when the letter 'Q' is pressed.
|
Note that only the first three rotors (1, 2 and 3) are driven.
Rotor 1 makes a single step on each key-press.
Rotor 2 only steps when the notch on rotor 1 is engaged.
Likewise, rotor 3 steps when the notch on rotor 2 is
engaged. This means that rotor 3 hardly moves at all.
This odometer-like stepping behaviour is known a regular stepping
or Enigma stepping.
The extra rotor (ZW) never steps, although it can be set manually
to any of 26 positions. For this reason, the 4th rotor (ZW) is also known as
a stator (rather than a rotor).
The reflector (UKW) is fixed and does not move.
It cannot be rewired in the field,
but there were two differently wired ones (known as b and c).
|
The plugboard (Steckerbrett) is reciprocal, which means that letters are swapped
in pairs. Un-steckered letters are patched to themselves. Whenever a key is
pressed, the current passes the plugboard, the ETW, rotors 1, 2, 3 and ZW.
It then hits the reflector (UKW), which returns the current via a different
path through rotors ZW, 3, 2, 1, ETW and plugboard, until a
lamp lights up.
Move the mouse over the diagram above to see when happens when a
key is pressed. An unfortunate side-effect of the reflector (UKW)
is that a letter can never be enciphered to itself.
|
|
|
This was one of the weaknesses of Enigma that were exploited
by Allied codebreakers during WWII. Other weaknesses were the regular
odometer-style stepping of the rotors, the reciprocal nature of the
plugboard, and the ring setting (Ringstellung) which hardly does anything at
all.
➤ More about Enigma M4
|
One of the improvements of the SG-39 over the Enigma, was
the addition of two strip printers. One printer was connected directly to the
keyboard and printed the plaintext (PT). The other one replaced the Enigma's
lamp panel, and printed the ciphertext (CT) directly to paper.
The rest of the machine was functionally equivalent to
Enigma M4, since it had to be
backward compatible. A further improvement over the Enigma was the use of a
configurable reflector, similar to UKW-D.
|
| |
Simplified SG-39 circuit diagram.
Move the mouse over the diagram to see
what happens when the letter 'Q' is pressed.
|
By far the biggest improvement over Enigma however, is the addition of three
Hagelin-type pin-wheels (N1-N3) that introduce irregular
stepping and make the machine far less predictable.
It is currently unclear how exactly the pin-wheels
controlled the stepping of the cipher rotors, but it is likely that
the stepping pattern was altered several times during the machine's
development. The diagram above shows the stepping behaviour (in red)
as it was described by Otto Buggisch in [5].
All three pin-wheels make a single step on each key-press.
Each pin-wheel has a different number of positions (21, 23 and 25),
which are prime to 26. This is done to obtain the longest possible
cipher period.
Furthermore, each pin-wheel position has a pin that can be set to active
or inactive. Whenever an active pin hits a sensing mechanism, it
causes the corresponding cipher rotor(s) to make a single step.
As a result, all cipher rotors may step frequently and irregularly.
To maintain backward compatibily with Enigma, the three
cipher rotors have settable rings plus the usual stepping notches at the
specified positions (identical to Enigma rotors).
This stepping works in addition to the
stepping caused by the pin-wheels, but if they coincide, only a single
step is made. When compatibility with Enigma was required,
all pins of the rightmost pin-wheel (N1) had to be set to active, whilst
all pins of the other two pin-wheels had to be set to inactive. This
causes the rightmost rotor (1) to step on each keypress, whilst the
stepping of rotors 2 and 3 is no longer controlled by the pin-wheels,
but only by the notches on rotors 1 and 2.
|
|
It was the intension to supply the SG-39 with two sets of rotors:
one set that was compatible with Enigma and one that was uniquely wired
for the SG-39. The former was only used when compatibility with
Enigma was required, in which case the pin-wheels had to be set as described
in the previous paragraph. The other rotor set was for using the machine
in SG-39 configuration.
|
|
The SG-39 was equipped with two strip printers: one for the plaintext and one
for the ciphertext.
These printers were driven by a 5-bit digital signal known as the
ITA-2 or Baudot standard.
Conversion of the 26 keys from the keyboard into the
5-bit ITA-2 code for the plaintext printer,
was done mechanically by the keyboard mechanism.
For the ciphertext printer,
conversion of the 26 output signals to ITA-2 was
done by means of a diode matrix.
Additional circuitry was present to insert a SPACE after every 5th character.
This SPACE is not part of the ciphertext and is therefore not encrypted.
It is likely that standard 5-bit strip printers from teleprinter
production were used.
|
|
Although the SG-39 was initially intended as a replacement candidate for
the Enigma, the developers as well as the customer
appeared to be indecisive on the final specifications of the machine.
Initially the machine was designed for telegraphy. Then it had to be
compatible with Enigma, and then they couldn't decide as to
whether it should be made compatible with teleprinter equipment.
|
It had become heavy and bulky and required an external power supply.
Furthermore, another machine on which
Fritz Menzer had worked,
Schlüsselgerät 41 (SG-41) – nicknamed Hitlermühle (Hitler Mill) –
had meanwhile been chosen as a potential Enigma replacement.
But like the SG-39, the SG-41 was found much too heavy.
The development of the SG-39 was abandoned some time in 1943, but was
resumed in 1944, although the machine was never completed.
After the war, on 24 June 1945,
a TICOM investigation team led by Major Howard C. Barlow visited the T&N
factory in Frankfurt/Main in the hope to seize German cipher technology.
He identified three partially complete prototypes.
|
|
|
The two most complete ones appeared not to be present, as they had been
shipped to the Army Depot in Tauberbischofsheim in March of that year [3].
Barlow did not take any action to retrieve the missing machines. He thought
it wouldn't be worth the 150-mile trip.
Eventually, he retured home with the most
incomplete prototype of which the rotors and the printers were missing.
The images at the top of this page are from that prototype.
After the war, around 1953, the German Government was in need of cipher
machines for its diplomatic and military traffic. Several companies were
asked to develop equipment for a variety of purposes and classification levels.
At T&N, the SG-39 project was resurected. Using new technologies that had
meanwhile emerged, they decided to increase the number of rotors to 8 and the
number of pin-wheels to 7. In addition, no less then 10 static discs (stators)
were placed in between the rotors. The rotors as well as the stators contained
a total of 26 removable and interchangeable wiring cores. The machine was called High Speed Enigma (HSE),
and was used by the German Foreign Office from 1957 until the mid-1960s.
It is shown in the image above.
➤ More about the High Speed Enigma
|
- German patent DE 853,007}
Schaltungsanordung zur Verzögerung von Schaltvorgängen
Alexander Wirth on behalf of T&N, filed 19 March 1943.
Although this patent is not directly related to cipher machines, it
shows the electronic circuit that is used in the SG-39 and also in
the HSE, to stretch the control pulse to the solenoid of the clutch.
This is done to bridge the time between the start signal and the
closing of the impulse contact on the machine's main axle.
- German Patent DE 974,447 — Verschlüsselungsgerät
Karl Grundlfinger on behalf of T&N, filed 15 May 1943.
As far as we know this is the only SG-39 related patent that was filed
during WWII. It describes a machine with 4 rotors, pin-wheels, a built-in
keyboard and two printers. It also describes the use of 32→5 encoders
and 5→32 decoders. Note that in this version of the machine, the rotors
have 32 contact points rather than 26. It is likely that this version was
intended for use with teleprinter equipment (telex).
- German Patent DE 975,036 — Chiffriergerät
Karl Grundlfinger on behalf of T&N, filed 17 August 1952.
This patent is related to the SG-39, but was filed after the war
in 1952, just before T&N started the development of the
High Speed Enigma (1953).
It describes the use of three cipher rotors (MA, MB, MC), a plugboard (TA)
and a pluggable reflector (TU).
The patent also describes how the letter 'J' could be used
as a SPACE character.
- German Patent DE 1,045,692 — Chiffriergerät
Karl Grundlfinger & Franz Burkhard on behalf of T&N, filed 16 August 1952.
This patent is related to the SG-39, but was filed after the war
in 1952, just before T&N started the development of the
High Speed Enigma (1953).
It shows the combined use of cipher rotors and pin-wheels,
and a mechanical 5-bit encoder for the keyboard.
- German Patent DE 1,101,027
Scheibenförmige Kontaktvorrichtung für elektrische Chiffriergeräte
Franz Burkhard on behalf of T&N, filed 16 August 1952.
This patent is related to the SG-39, but was filed after the war
in 1952, just before T&N started the development
of the High Speed Enigma (1953).
It describes a removable cipher rotor with 26 contacts that are marked
with the letters of the Latin alphabet (A-Z).
It is likely that rotors of this type were used in the SG-39.
|
- Schlüsselgerät 39
- SG-39
- Gerät 39
- Cipher Device 39
- Coding Equipment No. 39
- Enigma-machine 39
|
- Wikipedia, Schlüsselgerät 39
Accessed 21 June 2026
- Wikipedia (Germany), Schlüsselgerät 39
Accessed 21 June 2026
- David P. Mowry, German Cipher Machines of World War II
NSA, Center for Cryptologic History. Revised 2014. pp. 18-22.
- David Hamer: Actions involved in the 'double stepping' of the middle rotor 1
Cryptologia, January 1997, Volume XX, Number 1.
- Final Report Written by Wachtmeister Otto Buggisch of OKH/Chi and OKW/Chi
TICOM report I-137. 8 October 1945.
- Encipering Devices worked on by Dr. LIEBKNECHT at Wa Pruef 7
TICOM report I-57. 2 August 1945. pp. 8-9.
- Fritz Menzer, Cipher Device 39
Extracted from TICOM report DF-174A.
TICOM, December 1949. pp. 7-13.
- Fritz Menzer, Calculation of the period with Device 39 Period
Extracted from TICOM report DF-174A.
TICOM, December 1949. pp. 15-21.
- Wikipedia (Germany), Heereswaffenamt
Accessed 15 August 2026.
|
-
Reproduced here with kind permission from the author.
|
|
|
|
Any links shown in red are currently unavailable.
If you like the information on this website, why not make a donation?
© Crypto Museum. Created: Sunday 21 June 2026. Last changed: Saturday, 15 August 2026 - 16:15 CET.
|
|
|
|
|
|
![Incomplete SG-39 seen from the front. Photograph obtained from NSA [3].](img/sg39_front_large.jpg)
![Incomplete SG-39 seen from the front. Photograph obtained from NSA [3].](img/sg39_front_thumb.jpg "image # sg39_front_large.jpg")
![Incomplete SG-39 seen from the right side. Photograph obtained from NSA [3].](img/sg39_side_thumb.jpg "image # sg39_side_large.jpg")
![Incomplete SG-39 seen from the right side. Photograph obtained from NSA [3].](img/sg39_side_large.jpg)
![High Speed Enigma Mark II (1957). Photograph kindly provided by Klaus Kopacz [1]. Click for further details.](../hse/img/TN_HSE_mk2_small.jpg)
