Teletypewriter Communication Codes

Gil Smith
gil@vauxelectronics.com
2001


(Document Notes)



Abstract

Preliminary -- 5/01 gil smith Corrections or comments to gil@vauxelectronics.com

This information is pulled from a variety of sources, such as various emails of the greenkeys group. For more discussion of teletypewriter code development, see:
http://www.nadcomm.com/fiveunit/fiveunits.htm
http://fido.wps.com/texts/codes/index.html
http://www.science.uva.nl/faculteit/museum/DWcodes.html




FIVE-UNIT CODES: USTTY and ITA2 (aka BAUDOT)

There were a few variations in character codes for five-level teletypewriter machines. The two most-common character codes were ITA2 and USTTY (a variation of ITA2).

The USTTY and ITA2 5-level teletypewriter codes are commonly referred to as "Baudot" codes. While this is technically incorrect, these popular 5-level codes evolved from the work of Jean Maurice Emile Baudot of France -- it seems fitting to accept the defacto reference to "Baudot" as implying USTTY or ITA2 codes, since they were the 5-level codes that saw practical use in teletypewriter systems. However, the true Baudot code dates to around 1874, when Baudot designed the "Baudot Multiplex System," a printing telegraph. The system used a 5-level code generated by a device with five keys, operated with two left-hand fingers, and three right-hand fingers -- this required great skill on the part of the operator who entered the code directly. However, it was still a major improvement in communications -- prior to Baudot's design, communication was carried out using Morse code with a telegraph key. The 5-level "Baudot" code was actually designed by Johann Gauss and Wilhelm Weber. Used primarily in France, the Baudot Multiplex System, also made inroads in Britain. The original Baudot code defined the familiar structure of a 5-level code set, using LTRS and FIGS case shifting, and became known as the International Telegraph Alphabet 1 (ITA1). Another recognition of Baudot's contribution to data communications is the term "baud," which refers to bits-per-second speed of serial data.

Around 1901, Donald Murray of New Zealand developed an automatic telegraphy system, using a typewriter-like keyboard mechanism, and a variation on the original Baudot 5-level code. While Baudot's code was designed with finger-actuation in mind, Murray's code was designed for mechanization, to minimize machine wear for frequently-occuring charaters. The Murray system employed a keyboard perforator, which allowed an operator to manually punch a paper tape, and a tape transmitter for sending the message from the punched tape. At the receiving end of the line, a printing mechanism would print on a paper tape, and/or a reperforator could be used to make a perforated copy of the message. Early British Creed machines were used with the Murray system.

Around 1930, the CCITT (International Telegraph and Telephone Consultative Committee) introduced the International Telegraph Alphabet 2 (ITA2). The United States standardized on a variation of ITA2 called the American Teletypewriter code or USTTY. ITA2 and the USTTY variant became the basis for 5-level teletypewriter codes until 7-level ASCII code debuted (in an upper-case-only form) in 1963, and finally matured in 1967 to the form still used today.

Common to all of these 5-level codes is the shifting of keys using a FIGS or LTRS code. The same 5-level (5-bit) code is used to represent a lowercase symbol (LTRS case) or an uppercase symbol (FIGS case). When the LTRS or FIGS character is transmitted, it defines whether subsequent characters are to be interpreted as lowercase or uppercase. The receiving machine must remember the case until it is next changed.

The ITA2 code was an international standard, used in applications such as Telex service. The shifted-D character could be interpreted as WRU (Who-Are-You). The shifted-F, G, and H characters were technically undefined, but had commonly-used symbols.

USTTY was a variation of ITA2, and was the prevalent code used in the United States. The letters D, F, G, H, J, S, V, and Z had different FIGS symbols than ITA2, and, for some reason, BELL and ' were opposite of ITA2. There were also variations for specific applications -- that is, machines that had primarily a USTTY or ITA2 code set, with just a few changed keys. J and H were said to have the most commonly-changed FIGS characters. USTTY was also the base code used for conversion of five-level to seven-level codes for later dial TWX service.

The shifted-H symbol (#/STOP/blank) was sometimes used in USTTY machines as a "motor off" control code; the next character would turn the motor back on. Since some characters are missed until the motor is up to speed, a BLANK (NULL) character would be sent to start the motor, with perhaps a dozen additional BLANKs sent as filler to allow the motor speed to stabilize before reception of the message.

Some teletypewriter machines "unshift-on-space," which is to say they revert to LTRS state after a space character is received. Some machines unshift at the the end of a line; this machine-specific behavior is not presumed by the code. Transmitted messages typically used a CR LF LTRS LTRS sequence at the end of a line (eg: at 72 chars) -- the CR LF positioned printing at the start of the next line, and the LTRS LTRS not only provided an explicit unshift, but it allowed the carriage time to fully return to the home position. 


            USTTY        ITA2     Fractions  
    LTRS    FIGS         FIGS       FIGS           other FIGS variations  
    ----    ----      ----------    ----           ------------------------------------  
     A       -         -             -  
     B       ?         ?            5/8  
     C       :         :            1/8            WRU     (degree?)  
     D       $         # (WRU)       $  
     E       3         3             3  
     F       !         @ (undef)    1/4            #   %   (blank)  
     G       &         * (undef)     &             @  
     H       #         $ (undef)     #             STOP    (blank)   (British-pound)  
     I       8         8             8  
     J       '         BELL          '             ,  
     K       (         (            1/2  
     L       )         )            3/4  
     M       .         .             ?  
     N       ,         ,            7/8            \  
     O       9         9             9  
     P       0         0             0  
     Q       1         1             1  
     R       4         4             4  
     S      BELL       '            BELL                   (blank)  
     T       5         5             5  
     U       7         7             7  
     V       ;         =            3/8  
     W       2         2             2  
     X       /         /             /  
     Y       6         6             6  
     Z       "         +             "

Translation of USTTY code to ASCII is shown below. A different ASCII character is needed depending on the current case (LTRS or FIGS): 


           USTTY                         ASCII translation  
 ---------------------------           ---------------------  
 hex   binary    LTRS   FIGS              LTRS        FIGS  
 ----  -----    -----  -----     |     ---------   ---------  
 0x00  00000    BLANK  BLANK     |     0x00  NUL   0x00  NUL  
 0x01  00001      E      3       |     0x45  E     0x33  3  
 0x02  00010      LF     LF      |     0x0A  LF    0x0A  LF  
 0x03  00011      A      -       |     0x41  A     0x2D  -  
 0x04  00100      SP     SP      |     0x20  SP    0x20  SP  
 0x05  00101      S     BELL     |     0x53  S     0x07  BEL  
 0x06  00110      I      8       |     0x49  I     0x38  8  
 0x07  00111      U      7       |     0x55  U     0x37  7  
 0x08  01000      CR     CR      |     0x0D  CR    0x0D  CR  
 0x09  01001      D      $       |     0x44  D     0x24  $  
 0x0A  01010      R      4       |     0x52  R     0x34  4  
 0x0B  01011      J      '       |     0x4A  J     0x27  '  
 0x0C  01100      N      ,       |     0x4E  N     0x2C  ,  
 0x0D  01101      F      !       |     0x46  F     0x21  !  
 0x0E  01110      C      :       |     0x43  C     0x3A  :  
 0x0F  01111      K      (       |     0x4B  K     0x28  (  
 0x10  10000      T      5       |     0x54  T     0x35  5  
 0x11  10001      Z      "       |     0x5A  Z     0x22  "  
 0x12  10010      L      )       |     0x4C  L     0x29  )  
 0x13  10011      W      2       |     0x57  W     0x32  2  
 0x14  10100      H      #       |     0x48  H     0x23  #  
 0x15  10101      Y      6       |     0x59  Y     0x36  6  
 0x16  10110      P      0       |     0x50  P     0x30  0  
 0x17  10111      Q      1       |     0x51  Q     0x31  1  
 0x18  11000      O      9       |     0x4F  O     0x39  9  
 0x19  11001      B      ?       |     0x42  B     0x3F  ?  
 0x1A  11010      G      &       |     0x47  G     0x26  &  
 0x1B  11011     FIGS   FIGS     |     0x0E  SO    0x0E  SO  
 0x1C  11100      M      .       |     0x4D  M     0x2E  .  
 0x1D  11101      X      /       |     0x58  X     0x2F  /  
 0x1E  11110      V      ;       |     0x56  V     0x3B  ;  
 0x1F  11111     LTRS   LTRS     |     0x0F  SI    0x0F  SI


FIVE-UNIT WEATHER CODES

Starting in the 1940s, weather reports were sent hourly on landline circuits at 60 WPM. The weather, or "WX" symbols were used on model 15 and 19 machines with a special WX type basket. The WX code had special FIGS symbols representing cloud cover and wind direction.

The WX service did not set their machines to unshift-on-space, which would have required a FIGS before each group of numerals (some weather codes were in groups of five numerals). The end-of-line sequence was typically LTRS CR LF LTRS, or, if the next line began with numerals, LTRS CR LF FIGS.

From a model 15 manual, the weather symbol set is defined as: 


 Lowercase       Uppercase Weather symbols       Definition  
  (ltrs)                  (figs)  
 ---------    ----------------------------------------------------------------------------  
     A        (up arrow)                         North wind  
     B        (circle w/ vertical & horiz line)  Overcast -- More than 9/10 covered  
     C        (circle)                           Sky clear --  less than 1/10 cloud cover  
     D        (upper-right arrow)                Northeast wind  
     F        (right arrow)                      East wind  
     G        (lower-right arrow)                Southeast wind  
     H        (down arrow)                       South wind  
     J        (lower-left arrow)                 Southwest wind  
     K        (left arrow)                       West wind  
     L        (upper-left arrow)                 Northwest wind  
     N        (circle w/ two vertical lines)     Broken clouds -- More than 1/2 covered  
     V        (circle w/ one vertical line)      Scattered clouds -- less than 1/2 covered  
     Z        +


The remainder of the upper case symbols are the same as USTTY.



SIX-UNIT TELETYPESETTING (TTS) CODE

The six-level TTS (Teletypesetting) code was introduced by Walter Morey, to address the need for full upper and lowercase letters for newspaper and other publishing applications. The Teletype Model 20 is a six-level teletypewriter based on a Model 15 machine.

(need table)





SEVEN/EIGHT-UNIT CODES: ASCII-63 to ASCII-67 (ITA-5)

ASCII-63, introduced in 1963, was a 7-level (7-bit) code which did away with the LTRS/FIGS case shifting of the original 5-level codes. ASCII-63 only had uppercase letters -- upper and lowercase letters were both defined in the ASCII-67 (1967) version, which is still in use today. I understand that ASCII-67 was accepted as an international version of ASCII, and called ITA-5 (International Telegraph Alphabet No. 5). The ASCII-67 code, and a translation to 5-level USTTY, are shown below.

ASCII is technically a 7-bit code, but it is often referred to as an 8-bit code, since it is usually transmitted as 8-bits, with the 8th bit either space, mark, or even-parity. "8-level" ASCII teletype machines usually sent even parity ascii from the keyboard, but some (especially early units) sent bit-8 always mark. Model 33 and 35 machines were uppercase-only, and, except for DEL, characters from col-4 printed as col-3. 


                         ASCII-67   
     -----------------------------------------------   
         col-1        col-2       col-3       col-4   
     -------------   --------    -------     -------   
     0x00  ^@ NUL    0x20  SP    0x40  @     0x60  `  
     0x01  ^A SOH    0x21  !     0x41  A     0x61  a  
     0x02  ^B STX    0x22  "     0x42  B     0x62  b  
     0x03  ^C ETX    0x23  #     0x43  C     0x63  c  
     0x04  ^D EOT    0x24  $     0x44  D     0x64  d  
     0x05  ^E ENQ    0x25  %     0x45  E     0x65  e  
     0x06  ^F ACK    0x26  &     0x46  F     0x66  f  
     0x07  ^G BEL    0x27  '     0x47  G     0x67  g  
     0x08  ^H BS     0x28  (     0x48  H     0x68  h  
     0x09  ^I HT     0x29  )     0x49  I     0x69  i  
     0x0A  ^J LF     0x2A  *     0x4A  J     0x6A  j  
     0x0B  ^K VT     0x2B  +     0x4B  K     0x6B  k  
     0x0C  ^L FF     0x2C  ,     0x4C  L     0x6C  l  
     0x0D  ^M CR     0x2D  -     0x4D  M     0x6D  m  
     0x0E  ^N SO     0x2E  .     0x4E  N     0x6E  n   
     0x0F  ^O SI     0x2F  /     0x4F  O     0x6F  o  
     0x10  ^P DLE    0x30  0     0x50  P     0x70  p  
     0x11  ^Q DC1    0x31  1     0x51  Q     0x71  q  
     0x12  ^R DC2    0x32  2     0x52  R     0x72  r  
     0x13  ^S DC3    0x33  3     0x53  S     0x73  s  
     0x14  ^T DC4    0x34  4     0x54  T     0x74  t  
     0x15  ^U NAK    0x35  5     0x55  U     0x75  u  
     0x16  ^V SYN    0x36  6     0x56  V     0x76  v  
     0x17  ^W ETB    0x37  7     0x57  W     0x77  w  
     0x18  ^X CAN    0x38  8     0x58  X     0x78  x  
     0x19  ^Y EM     0x39  9     0x59  Y     0x79  y  
     0x1A  ^Z SUB    0x3A  :     0x5A  Z     0x7A  z  
     0x1B  ^[ ESC    0x3B  ;     0x5B  [     0x7B  {  
     0x1C  ^\ FS     0x3C  <     0x5C  \     0x7C  |  
     0x1D  ^] GS     0x3D  =     0x5D  ]     0x7D  }  
     0x1E  ^^ RS     0x3E  >     0x5E  ^     0x7E  ~  
     0x1F  ^_ US     0x3F  ?     0x5F  _     0x7F  DEL  

       ^x is control key and character key together  


            ASCII-67                                        USTTY translation  
   -----------------------------         -----------------------------------------------------  
   col-1   col-2   col-3   col-4            col-1         col-2         col-3         col-4   
   ------  -----   -----   -----         -----------   -----------   -----------   -----------  
   ^@ NUL    SP      @       `      |    N 0x00 BLANK  N 0x04  SP       --            --    
   ^A SOH    !       A       a      |       --         F 0x0D  !     L 0x03  A     L 0x03  A  
   ^B STX    "       B       b      |       --         F 0x11  "     L 0x19  B     L 0x19  B  
   ^C ETX    #       C       c      |       --         F 0x14  #     L 0x0E  C     L 0x0E  C  
   ^D EOT    $       D       d      |       --         F 0x09  $     L 0x09  D     L 0x09  D  
   ^E ENQ    %       E       e      |       --            --         L 0x01  E     L 0x01  E  
   ^F ACK    &       F       f      |       --         F 0x1A  &     L 0x0D  F     L 0x0D  F  
   ^G BEL    '       G       g      |    F 0x05 BELL   F 0x0B  '     L 0x1A  G     L 0x1A  G  
   ^H BS     (       H       h      |       --         F 0x0F  (     L 0x14  H     L 0x14  H  
   ^I HT     )       I       i      |       --         F 0x12  )     L 0x06  I     L 0x06  I  
   ^J LF     *       J       j      |    N 0x02  LF       --         L 0x0B  J     L 0x0B  J  
   ^K VT     +       K       k      |       --            --         L 0x0F  K     L 0x0F  K  
   ^L FF     ,       L       l      |       --         F 0x0C  ,     L 0x12  L     L 0x12  L  
   ^M CR     -       M       m      |    N 0x08  CR    F 0x03  -     L 0x1C  M     L 0x1C  M  
   ^N SO     .       N       n      |    N 0x1B FIGS   F 0x1C  .     L 0x0C  N     L 0x0C  N  
   ^O SI     /       O       o      |    N 0x1F LTRS   F 0x1D  /     L 0x18  O     L 0x18  O  
   ^P DLE    0       P       p      |       --         F 0x16  0     L 0x16  P     L 0x16  P  
   ^Q DC1    1       Q       q      |       --         F 0x17  1     L 0x17  Q     L 0x17  Q  
   ^R DC2    2       R       r      |       --         F 0x13  2     L 0x0A  R     L 0x0A  R  
   ^S DC3    3       S       s      |       --         F 0x01  3     L 0x05  S     L 0x05  S  
   ^T DC4    4       T       t      |       --         F 0x0A  4     L 0x10  T     L 0x10  T  
   ^U NAK    5       U       u      |       --         F 0x10  5     L 0x07  U     L 0x07  U  
   ^V SYN    6       V       v      |       --         F 0x15  6     L 0x1E  V     L 0x1E  V  
   ^W ETB    7       W       w      |       --         F 0x07  7     L 0x13  W     L 0x13  W  
   ^X CAN    8       X       x      |       --         F 0x06  8     L 0x1D  X     L 0x1D  X  
   ^Y EM     9       Y       y      |       --         F 0x18  9     L 0x15  Y     L 0x15  Y  
   ^Z SUB    :       Z       z      |       --         F 0x0E  :     L 0x11  Z     L 0x11  Z  
   ^[ ESC    ;       [       {      |       --         F 0x1E  ;        --            --   
   ^\ FS     <       \       |      |       --            --            --            --   
   ^] GS     =       ]       }      |       --            --            --            --   
   ^^ RS     >       ^       ~      |       --            --            --            --   
   ^_ US     ?       _      DEL     |       --         F 0x19  ?        --            --


One possible USTTY translation of ASCII-67 (shown above):
- change case as needed by sending LTRS or FIGS: L=LTRS, F=FIGS, N=No-Change
- ascii lowercase chars converted to uppercase
- map control chars NUL, BEL, LF and CR
- append LTRS LTRS after each CR for mechanical TTY machines to return fully
- use SO to send FIGS, SI to send LTRS (not needed unless testing shifts)
- no translation for other control chars, or ( ) < > [ ] { } ? % * + = @ ` _ \ ^ | ~
(send BLANK or nothing)


ASCII-63 was mostly identical to the current ASCII-67 version. The definitions of the control characters (col-1 above) varied between the two versions, as defined below. Also, in ASCII-63, the upper 32 positions (col-4) were undefined, except for three: RUB (0x7F), ACK (0x7C), and ESC (0x7E). There are inconsistent references to an ALT-MODE char (0x7D) in ASCII-63. In the 1967 version, RUB became DEL and stayed in the same position, but ACK and ESC moved into the control character area (col-1). In ASCII-67, ^ replaced the up-arrow symbol, and _ replaced the left-arrow.

ASCII-63 and ASCII-67 are the common variants of ASCII, but there appear to have been some transitional versions as well: in the Teletype Model 33 manual, there are references to a 1965 version of ASCII, that had SS in place of SUB (0x1A), \ for @ (0x40), ~ for \ (0x5C), an odd character in place of | (0x7C), and | for ~ (0x7E). A Teletype code card for M33 and M35 machines indicatesa 1966 version of ASCII, though the printable characters shown on the card were identical in all versions.

The DC1 and DC3 control characters have been used as aux-device start/stop commands (eg: auto-tape-reader option in Teletype M33asr). DC2/DC4 have also been used as auto-tape-punch on/off commands in the M33asr.

DC1 and DC3 are also referred to as XON and XOFF respectively. XON and XOFF are widely used today as software-handshaking flow-control characters for throttling data transfer into a serial buffer. 


      ASCII-63 Control Characters    ASCII-67 Control Characters  
      ---------------------------    ---------------------------  
      NULL = Null char               NUL = Null char  
      SOM = Start of Message         SOH = Start of Heading  
      EOA = End of Address           STX = Start of Text  
      EOM = End of Message           ETX = End of Text  
      EOT = End of Transmission      EOT = End of Transmission  
      WRU = Who Are You              ENQ = Enquiry  
      RU  = Are You                  ACK = Acknowledge  
      BELL = Signal Bell             BEL = Signal Bell  
      FE0 = Format Effector 0        BS = Back Space  
      HT/SK = Horizontal Tab         HT = Horizontal Tab  
      LF = Line Feed                 LF = Line Feed (aka NL = New Line)  
      VTAB = Vertical Tab            VT = Vertical Tab  
      FF = Form Feed                 FF = Form Feed  
      CR = Carriage Return           CR = Carriage Return  
      SO = Shift Out                 SO = Shift Out (special)  
      SI = Shift In                  SI = Shift In (normal)  
      DC0 = Device Control 0         DLE = Data Link Escape  
      DC1 = Device Control 1         DC1 = Device Control 1 (aka XON) (M33 tape-reader-on)  
      DC2 = Device Control 2         DC2 = Device Control 2 (M33 tape-punch-on)  
      DC3 = Device Control 3         DC3 = Device Control 3 (aka XOFF) (M33 tape-reader-off)  
      DC4 = Device Control 4         DC4 = Device Control 4 (M33 tape-punch-off)  
      ERR = Error                    NAK = Negative Acknowledge  
      SYNC = Synchronous             SYN = Synchronous Idle  
      LEM = Logical End of Medium    ETB = End of Transmission Block  
      S0 = Special Application 0     CAN = Cancel  
      S1 = Special Application 1     EM = End of Medium (Media?)  
      S2 = Special Application 2     SUB = Substitute  
      S3 = Special Application 3     ESC = Escape  
      S4 = Special Application 4     FS = Field Separator  
      S5 = Special Application 5     GS = Group Separator  
      S6 = Special Application 6     RS = Record Separator  
      S7 = Special Application 7     US = Unit Separator  
      RUB = Rubout                   DEL = Delete  
      ACK = Acknowledge              (position 0x7C became |)  
      ALT-MODE = Alternate Mode      (position 0x7D became })  
      ESC = Escape                   (position 0x7E became ~)




Document Notes

Created by Gil Smith, July 2001. The original file is smith--teletype-codes.txt.