|
|
|
|
|
|
|
Rotor Enigma Tree D
The Enigma D (A26) and its history are described in detail
on our Enigma D page. On this page we will take a closer
look at the machine with serial number A343, its history and any
differences with other known surviving Enigma D units.
The image on the right shows the machine in its present state before
cleaning and restoration. When it was found,
a sheet with typed instructions in Swedish
was fitted inside the case lid by means of four thumbtacks.
The sheet was removed to reveal the original
Enigma shield and the address
tag of Chiffriermaschinen AG (ChiMaAG).
Considering its age, the machine is in excellent condition. Although there
are signs of intensive usage, the surface is
clean and the celluloid windows over the rotors
are still in one piece and show no sign of discolouration.
This means that the lid was kept closed when the machine was not in use.
It is also clear that the machine was serviced
regularly during its operational life.
|
|
|
|
The machine has several features that are not present on comparable machines.
Some of these appear to be aftermarket upgrades that were installed many
years ago, probably shortly after the machine was purchased in 1926.
Looking inside the machine, in particular at the
entry disc (ETW), it becomes clear that it saw
serious usage during its operational life. When the machine was rediscovered
in 2019 (and sold in 2025) it was fitted with three rotors (I, II and III)
from a different machine (A327) that had also been owned by the original
customer S.T.A.B. (Ivar Kreuger).
|
-
Although the name 'Enigma D' is often used in literature —
it makes sense, as 'D' follows after 'C' — this name was never used
by the manufacturer (ChiMaAG) in official correspondence. Instead,
they always used the model number A26.
|
The image below gives an overview of the features of the A343,
in particular the ones that are different from other known
Enigma D specimen. Note that the rotors originally belonged
to the machine with serial number A327, and that the lamp
panel was originally part of the A345.
|
Svenska Tändsticks AB (S.T.A.B.) was a Swedish multinational company that
controlled ~ 70% of the worldwide production of matches at the time.
It was owned by Ivar Kreuger, who became known as The Match King.
Through his holding company Kreuger & Toll [14] – and several of its subsidiaries –
Kreuger controlled more than 400 companies and banks worldwide [2].
As telephone conversations could easily be tapped, and overseas telegrams
could easily be intercepted in those days — they were generally sent over radio —
Kreuger secured his confidential business communications by means of
Enigma cipher machines.
From surviving documents it is known that STAB owned the following
Enigma cipher machines [9]:
|
In this report we will only look at the five Enigma D (A26) machines,
which are incompatible with the other models.
A343 and A344 were supplied to Direktor Schernikau, the respresentative
of STAB in Berlin. They were delivered at Behrenstrasse 43/44; the
address of the Disconto-Gesellschaft (later: Deutsche Bank), which was
also owned and controlled by Kreuger & Toll [2][6][13].
The snippet above was extracted from a
memorandum of 24 October 1935
of Heimsoeth und Rinke – the successor of ChiMaAG 1 –
which lists the sales of Enigma machines within Germany.
Two days earlier, on 21 October 1935,
another memo
was issued, in which the sales to all countries are listed [12].
From this we learn that A345 was
supplied to an unspecified customer in Russia.
As the lamp panel of the A345 was found inside the A343,
it is likely that the A345 also belonged to STAB, and that it was delivered
to Kreuger's subsidiary in St. Petersburg (Russia).
A343, A344 and A345 were not the only Enigma D machines used
by STAB. The same memo
shows that another four Enigma D machines (A324, A325, A327 and A328) 2
had been supplied to Sweden earlier in 1926. As part of the Histocrypt
proceedings of 2018, Anders Wik published an overview of Enigma machines supplied
to Sweden [5], in which he lists A327 and A328 as STAB machines.
This means that a total of five Enigma D machines were used by STAB.
It is likely that all five machines had the same wiring of the rotors.
As Enigma D was the standard commercial model in 1926, the rotors
of these machines were normally delivered with the
standard commercial wiring.
When the A343 surfaced in November 2025, it was found with the rotors
of the A327 — one of the Enigma D machines that had been purchased by STAB
in early 1926.
Although the wiring of these rotors is similar to the
commercial wiring, it is not identical.
Furthermore, the name S.T.A.B. and the serial number
are engraved in red, which was probably done to show that they
contain alternative wiring.
As Kreuger was a significant player in international business, he had
offices is several countries, including Sweden, France and Russia.
It is known that A344 was investigated in 1927 in Paris by Arvid Gerhard Damm,
an employee of competitor AB Cryptograph. 3
He filed his report in August 1927 [7].
We assume that STAB's Enigma D machines were used at the following
locations:
|
A327 ? A328 ? A343 Sweden Stockholm A344 France Paris A345 Russia St. Petersburg
|
|
Kreuger's financial conduct was questionable, to put it mildly,
and is described by some as a Ponzi scheme [4].
During the Great Depression (1929-1939), his empire
collapsed and in 1932 he was found dead in the bedroom of his flat in Paris.
According to the police, he had committed suicide, although this was disputed
30 years later by his family and by others,
after classified documents had been released [2].
Nevertheless, most of Kreuger's companies survived and many of them still
exist to this day, including STAB which is now known as
Swedish Match AB [3].
|
|
|
Saved from the scrap heap
|
|
|
|
The present owner of the A343 obtained it from the estate of a Swedish
family that has no known connection with the Kreuger family or with STAB.
The previous owner had reportedly 'rescued' the machine from a scrap heap.
In 2019, after he had passed away, his family had to
clear the house and rediscovered the A343 in the attic.
In 2025 it was decided to sell it to the current owner [1].
|
|
In the following sections, each part of the machine is further described.
This is done to give an accurate account of the state in which the machine
changed hands in 2025.
|
|
The wooden case is in excellent condition and does not need any rework.
Below are detailed photographs of all sides.
The following anomalies are noted:
|
Each of the above points will be further discussed below.
It seems likely that the machine was initially supplied with a leather
carrying strap mounted at the right side of the wooden case. The
scuff marks of the metal brackets
are still visible. It was probably removed soon after the machine was
purchased, as there no discolouration is visible at this position.
The lid retaining bracket is an important feature that holds the
lid of the wooden case in a sloped position when it is open.
It is missing from the A343, which means that the
lid has no end-stop, which causes stress
to the hinges and screws. It can be solved by placing an
object of appropriate height under the opened lid, or by making
a reproduction of the original retaining bracket.
|
When the A343 was found in 2025, it had a paper sheet with typed operating
instructions in Swedish fitted inside the case lid. It was held in place
by four brass thumbtacks in the corners. As the original decals could be
felt behind the paper, it was decided to carefully remove the thumbtacks
and preserve the typed instructions. The sheet had already been removed
at least once, as two tiny thumbtack holes can be spotted in each corner.
Below is a high resolution scan of the sheet.
Click the image to view it in closeup. The photographs below show how the
sheet was originally fitted inside the case lid.
Translated into English (with help from ChatGPT) it reads:
|
Memorandum
for the enciphering of telegrams.
1) The person who has received information about the key for STAB's cipher
is responsible for keeping it strictly secret. He shall therefore
personally set the machine according to the current key and thereafter
either personally carry out the enciphering or deciphering, or himself
supervise the person to whom he delegates this work. Immediately after the
work has been completed, he shall personally reset the machine so that the
key used is completely erased. If necessary, the ciphering wheels shall be
removed from the machine and stored in a safe.
2) When using plain language, a space between each word is indicated with
'X', in the manner described on pages 11-12 of the manual.
3) Numbers in plain language are indicated by inserting 'Y' before and after
the number according to the description on page 12, where the following
letters are read as the digits written beneath them:
A B C D E F G H I J
1 2 3 4 5 6 7 8 9 0
4) For code telegrams, only 5-letter code groups are used, preferably
Bentley's. A code word may not be divided, and each 5-letter code word
must correspond exactly to a 5-letter group in the cipher telegram.
Numbers in code telegrams are always given in code.
5) A longer message containing secret information should as a rule be
divided so that only the secret part of the telegram is enciphered,
while the rest is telegraphed in code or plain language without enciphe-
ring. The enciphered portion is written in the form of 5-letter groups;
the non-enciphered portion, if in code, in the form of 10-letter groups.
6) Note: The first 10 letters, i.e., the first two 5-letter groups
of a cipher telegram or of the enciphered part of a telegram (see the
preceding point), serve a special purpose and shall therefore not be
typed into the machine, whether for enciphering or deciphering.
7) If, when pressing a key, for example a lamp lights up, the faulty
lamp is located in the following way: keep one key — for example the
X key — pressed down; then press the other keys in succession until,
when pressing one of them, no lamp lights up. If this happens, for
example when pressing the K key, then the X lamp is the faulty one.
(Note: provided that during the search the B key is kept correctly
pressed, the wheels do not move during the search.)
Point 4 above, mentions the use of Bentley's for creating telegrams.
This refers to Bentley's Complete Phrase Code
— a codebook, compiled in 1909 by E.L. Bentley, for shortening telegrams
with a compression ratio of at least 50% [10].
Each five-letter group in the book replaces a word, an expression or a complete
sentence of the original plaintext.
Codebooks were common in those days for lowering the cost of telegrams.
Bently's codebook is very similar (but not identical) to other codebooks of the era,
such as the Peterson Code (1929) and
The New Boe Code (1937).
➤ Read Bentley's Complete Phrase Code
|
|
The body of the machine is in very good condition. It shows the usual
signs of usage and wear, but there is no need for any rework. The
celluloid windows, through which the rotor positions are visible, are
clear and have no cracks. The alphabet film on the lamp panel
is also in good shape (see below).
The keyboard is in very good state and shows no signs of fading or corrosion.
|
|
Scratches on the top surface
around the position of the knob, reveal that
it was often left in the wrong position when opening the lid.
To put it in the north-south position, it must be placed
in between the settings 'dkl' and 'aus'.
This problem is fixed on later versions of the Enigma machine, such as
Enigma K (A27) and military Enigma I, where the knob is
fitted to the lid.
|
-
It is thought that most users placed the power selector in a fixed
position and removed the knob for convenience. In any case the knob
was missing from the surviving A319, A320 and A327.
|
|
The lamp film (or alphabet film)
inside the A343 is in good condition, especially when considering
its age. It has shrunk somewhat, but this is not unusual for celluloid films
after near 100 years.
|
In fact, the lamp film of contemporary machine A319
had shrunk so much, that it had to be replaced [8].
This is clearly not the case here.
Although there are some cracks
– resulting from the use of the wrong type of light bulbs –
the film is good enough to remain in place.
The inside of the lamp panel carries serial number A345,
which indicaties that it was probably swapped at some point.
Enigma D with serial number A345 was also owned by STAB.
➤ All machines owned by STAB
|
|
|
When the machine was originally supplied, it had the
usual lock at the front
of the wooden case. This lock is used only to keep the lid closed, and does
not need a key. Like with all early commercial Enigma machines, the lock
is made of highly polished nickel-plated brass. The main part of the lock
is mounted in the bottom half of the case, with a mating pin mounted in
the case lid.
With the A343, this mating pin is broken-off
(and missing), which is not unusual.
It was also missing from the A319 and was
probably caused by repeatedly dropping the lid onto the base, whilst the
lock was in 'closed' position. With the A343, the original lock was not needed
anyway, as an additional lock is mounted to its left.
It mates with a pin that is also mounted in the lid.
Contrary to the original lock, this lock needs a key.
Unfortunately the key for this lock is not present.
The extra lock is clearly an aftermarket upgrade, as the pencil
markings are still visible at the front.
It is unknown when it was added,
but it was probably done shortly after it was puchased in 1926.
A similar lock was added to the
left side of the metal lid of the machine
itself. It prevents access to the internal parts (i.e. the cipher rotors)
by an unauthorised person. The lock itself is
mounted to the inside of the metal lid,
whilst the mating counterpart is fitted to a
wooden block inside the base of the wooden case.
Judging from the scratch marks,
this is also an aftermarket upgrade.
|
|
Like with other Enigma cipher machines, the serial number of the A343 is
stamped or engraved at several positions. The list below shows the
positions in which serial numbers were found. In most cases they match
that of the machine, but there are some anomalies as explained below.
|
The most obvious place for a serial number is at the
front of the wooden case,
directly below the lock. On the A343 the number is present, but is barely
visible, probably caused by frequent and intensive handling of the machine
when it was in active use. The number is also
stencilled inside the case lid,
at the centre of the lower plinth. On the A319 is was not present
at this location.
The serial number is usually stamped in the bottom panel
of the machine. When the bare machine (removed from the wooden case) is turned over,
the number can be found close to the rear edge.
Inside the machine, the serial number is present at the
left side of the leftmost stepping stopper.
The number is also present at the
inside of the lamp panel,
either at the left or the right side. In this case the
number A345 is present at the right side,
which doesn't match the serial number of the machine (A343).
There are several plausable explanations for this: (1) a mistake was made
by the manufacturer, (2) the case shell of the A345 was mistakenly fitted to
the A343 after servicing the machines, or (3) one machine was used as a parts
donor when refurbishing the others.
In this case, (2) is perhaps the most likely, as the machine with
number A345 also belonged to STAB.
The serial number is also engraved into the right side of each rotor, but as
these are removable, they are easily swapped with those from another machine.
In this case, the original UKW is present (A343), but the rotors
are from the A327 — another machine that belonged to STAB.
|
The A343 was found with three rotors (I, II and III), but their serial
numbers do not match that of the machine. All three rotors are engraved
with S.T.A.B. A327 — another Enigma D that belonged to STAB.
Also note that the text is engraved in red rather than white, which
was probably done to indicate that it contains non-standard wiring.
The image below shows the right side of rotor II (2).
The ring setting (German: Ringstellung) is similar to that on later machines,
but the short stub – that allows the spring to be lifted with a tumb nail –
is missing. Furthermore, it should be noted that the stepping notch 1
is attached to the rotor body rather than to the alphabet ring (as with
later machines). This means that when altering the Ringstellung, the
notch stays in the same position. As a result, the cryptographic effect of the
Ringstellung is null.
This was clearly a misfeature of the Enigma D (A26), which was fixed in the
Enigma K (A27) that appeared in 1927.
On all three cipher rotors, the stepping notch is located at the position
of the letter 'G' (when the Ringstellung is set at 'A'). The notch is
'sensed' by a stepping lever at the rear of the rotor stack. A transition
of the adjacent rotor occurs when the letter 'Z' is in the window prior
to a key press. Note that this is different from later machines like
Enigma K (A27), where the stepping of rotor I (where the notch is also
at the position of the letter 'G') occurs at the letter 'Y'. This difference of
one position is caused by the fact that the window is at a different position.
On Enigma D, it shows the letter at the dead top, while on later machines
it shows the letter in front of it.
The reflector, or Umkehrwalze (UKW),
is the one that was originally
supplied with the machine (A343) and is engraved in white, indicating that
it holds the standard commercial wiring.
|
 |
-
The stepping notch of a rotor controls the stepping of the rotor to its left.
|
|
As the A343 was found with the rotors of the A327 (both owned by STAB), it
is likely that all Enigma D machines owned by STAB had the same wiring of
the rotors. This wiring is different from the standard commercial wiring,
although there are some similarities. Both the standard and the special
STAB-wiring are shown below.
|
|
|
Standard commercial wiring
|
|
|
Below is the wiring of Enigma D and its rotors, as described in 1927
by Hugh Foss in his report
The Reciprocal Enigma [11].
The wiring is identical to the wiring of the later Enigma K (A27)
and can therefore be classed as 'commercial wiring'.
Note that on Enigma D rotors, the turnover notch is attached to
the rotor body, rather than the alphabet ring (index)
as on Enigma K rotors. This means that when the
ring setting is changed, the turnover of the adjacent rotor will happen at
a different letter. Furthermore, the position of the notch was identical
on all three rotors (I, II, III).
Also note that the window is at the centre top of the rotor, rather than
one position in front of it, as on later machines. This means that the
turnover occurs one letter later than on, say, Enigma K.
Turnover = RingSetting - 1
The table below shows the position of the notch when the ring is set to 'A'.
In that case, the notch is located at the position of the letter 'G', and
a turnover of the adjacent rotor will occur when the letter 'Y' in visible
in the window.
When the ring is set to 'B', the turnover will occur at 'A', etc.
|
| Rotor | ABCDEFGHIJKLMNOPQRSTUVWXYZ | Notch 1 | Turnover 1 | # |
|
|
| ETW | QWERTZUIOASDFGHJKPYXCVBNML | |
|
|
| I | LPGSZMHAEOQKVXRFYBUTNICJDW | G | Z | 1 |
| II | SLVGBTFXJQOHEWIRZYAMKPCNDU | G | Z | 1 |
| III | CJGDPSHKTURAWZXFMYNQOBVLIE | G | Z | 1 |
| UKW 2 | IMETCGFRAYSQBZXWLHKDVUPOJN | |
|
|
|
|
Below is the wiring of the rotors that were found in the A343.
These rotors are engraved with the text 'S.T.A.B. A327.
As the A327 was also owned by STAB, it is likely that
all Enigma D machines used by STAB had this wiring.
The differences with the standard wiring are shown below in red.
|
| Rotor | ABCDEFGHIJKLMNOPQRSTUVWXYZ | Notch | Turnover | # |
|
|
| ETW | QWERTZUIOASDFGHJKPYXCVBNML | |
|
|
| I | LZGSPMHAEOQKUXRFYBVTNICJDW | G | Z | 1 |
| II | SLVJBTFXGQOHEWIRCYAMKPZNDU | G | Z | 1 |
| III | CJGDPSAKTURHWZXFMQNYOBVLIE | G | Z | 1 |
| UKW | IMETCGFRAYSQBZXWLHKDVUPOJN | |
|
|
|
|
The wiring of the entry disc (ETW) was not changed, and the wiring of
the reflector (UKW) was also left unaltered.
It is likely that the rotors were supplied with this
wiring by ChiMaAG, as it differs only marginally from the
standard wiring. In each rotor, only two
letter pairs were swapped:
|
-
Notch and Turnover position shown when Ringstellung is at 'A'.
Notch is the physical position of the notch on the rotor. Turnover
is the position visible in the window before the turnover occurs.
Note that on Enigma D, the window is at the centre top of the rotor
rather than one position in front of it, as on later machines.
-
This wiring is measured using the contact at the white dot as the
reference point. When the UKW is installed in the machine with its
ring setting at 'A' and the letter 'A' visible through the window in the lid,
this is the contact at the centre top of the UKW.
|
|
As the four M4 bolts were missing from the bottom of the wooden case,
the machine could be removed quite easily. The photographs below represent
the condition of the machine when it was received (i.e. before cleaning
and oiling). In order to access the interior, place the
power selector in the north-south position
(in between 'dkl' and 'aus') and loosen the
two thumb-screws at the front edge. The hinged lid can now be opened.
Note that the knob of the power selector and the two power terminals
to its right, protrude the lid and remain in
place when the lid is opened.
|
As the A343 is not part of our collection and was only briefly at
Crypto Museum for this investigation, we have made no attempt to
restore it in any way, other than non-invasive cleaning of the
interior and the rotors and applying a few drops of oil.
One of the recessed screws of the lamp panel was missing, along with
the mating nut, but these were later
found inside the machine and
were refitted. Another bolt and nut were found and stowed for now.
When loosening the two thumb screws at the front edge of the hinged
top lid of the machine, it was noted that the rightmost one could
fall out of its pocket.
This confirms that the machine saw serious usuage
and that the lid had been opened many times to set the rotors.
Due to excessive wear of the rightmost pocket, the screw is no longer
held in place. This was solved by adding a
hard-paper fixating ring
to the shaft of the thumb screw.
The lamp panel was somewhat loose and could be moved, which was solved
by tightening the faxating screws at the bottom panel. It is possible
that a previous owner had (temporarily) removed the lamp panel in a
repair attempt, and had failed to fully tighten them again.
To carry out a first rough test, the machine was fitted with new light
bulbs and an improvised battery was installed. Although there are still
a substantial number of contact problems (e.g. the switches under the
keyboard) some of the lamps lit up when a key was pressed. During this
test it was noted that one of the pins of the UKW was binding. To
prevent further damage, the pin was freed and all pins were oiled.
|
- Leather grip and brackets missing
- Case lid retaining bracket missing
Recessed screw and nut missing from lamp panel Mounting screws missing from case bottom - Bakelite screw tops missing from power terminals
- Case lock pin broken off
Lamp panel loose - Lamps missing from lamp panel
- Lamps missing from spare light bulb bracket
- Keys missing from both key locks
Lid hold down thumb-screw not held in place Contacts pin(s) of UKW binding - Keyboard contacts need cleaning
|
- Instruction sheet removed from case lid and preserved
- Case interior and machine exterior superficially cleaned
- Rotors cleaned and oiled
- UKW cleaned and oiled
- Rotor spindle cleaned and oiled
- Machine interior cleaned and oiled
- Lamp panel fixated by tightening the screws at the bottom
- Missing screw from lamp panel found and refitted
- Machine fitted inside case with modern screws (4 × M4)
- Rightmost lid hold down thumb-screw fixated with paper ring
- Contact pins of UKW oiled
|
Device Rotor cipher Machine Brand Enigma Type D Model A26 Designator Ch.8 Serial number A343 Manufacturer ChiMaAG Country Germany Year 1926 Customer S.T.A.B. (Ivar Kreuger), Sweden Rotors 3 (removable) Contacts 26 Turnovers 1 per rotor Reflector Settable (26 positions) Wiring see above Stepping Regular (Enigma stepping) Plugboard no Dimensions 290 × 280 × 155 mm (incl. wooden case) Weight 8.6 kg (incl. wooden case)
➤ More about Enigma D
|
- Glen Miranker — owner of Enigma D (A26) with serial number A343
Crypto Museum, November 2025.
- Wikipedia, Ivar Kreuger
Visited 26 November 2025.
- Wikipedia, Swedish Match
Visited 26 November 2025.
- Wikipedia, Ponzi scheme
Visited 26 November 2025.
- Anders Wik, The First Classical Enigmas,
Swedish Views on Enigma Development 1924-1930
Histocrypt Proceedings, 15 May 2018.
- Frode Weierud, Enigma D versus Zähwerk Enigma
CryptoCellar Research. 15 September 2024. 6 pages.
Personal correspondence September 2022, February 2024.
- Arvid Gerhard Damm, Preliminary statement regarding glow lamp cipher machine Enigma
Preliminärt utlåtande angående "Glühlampen-Chiffriermaschine Enigma" (Swedish).
Paris (France), 1 August 1927. [Serial number A344 is mentioned on page 4].
- Paul Reuvers and Marc Simons, Enigma D with serial number A319
Crypto Museum, February 2024.
- Frode Weierud, Personal correspondence
Crypto Cellar Research, November 2025.
- E.L. Bentley, Bently's Complete Phrase Code
13 December 1909.
- Hugh Foss, The Reciprocal Enigma 1
TNA, HW25/14. Undated, but probably 1927/28.
- Memorandum, List of commercial glow-lamp Enigma machines 1926-1935
Heimsoeth und Rinke, 21 October 1935.
Document kindly provided by Frode Weierud [9].
- Memorandum, List of commercial glow-lamp Enigma machines in Germany 1926-1935
Heimsoeth und Rinke, 24 October 1935.
Document kindly provided by Frode Weierud [9].
- Wikipedia, Kreuger & Toll
Visited 29 November 2025.
|
|
|
|
Any links shown in red are currently unavailable.
If you like the information on this website, why not make a donation?
© Crypto Museum. Created: Wednesday 26 November 2025. Last changed: Friday, 05 December 2025 - 11:52 CET.
|
|
|
|
|
| |
![Extract from memorandum of 24 October 1935 [13]. Click to read the full memorandum. Document kindly provided by Frode Weierud [9].](img/19351024_a343_a344_h.png)
![Extract from memorandum of 21 October 1935 [12]. Click to read the full memorandum. Document kindly provided by Frode Weierud [9].](img/19351021_a327_a328_a345.png)
