• Please review our updated Terms and Rules here

Period serial port pinout for serial devices [1980s]

ardent-blue

Experienced Member
Joined
Jan 2, 2015
Messages
479
Folks, I have been unsuccessful in grokking the odd IBM PS/2 Model 90 DB25 [or DE25] pinout. I have found period user manuals and articles for various third party computers and hobbyist pubs. Seems like the serial standards were less inclusive in the mid 80s, and some manufacturers made some decisions to simplify DTS to RTS wiring by using some of the higher pins. I assume the second channel of the older RS232 standard were utilized. Then some issues with DTE to DTE, or DCE, or god knows what...

Anyone here with some information on serial device pinouts for late 80s computers / devices?
 
What you have to remember is that the RS232 standard is a shopping list and implementers were free to choose which pins they implemented. It was also designed to connect a modem (Data Communications Equipment) to a computer terminal (Data Terminal Equipment). However the DB25 RS232 you find on computers pretty much mirrors the original IBM PC serial card. So if you download the manual for that you are good to go.
 
Folks, I re-read the Serial Port portion of the HITR, it makes no mention of DTE or DCE for the serial port on the IBM Model 90.

According to this page, Difference between DTE and DCE -BOTH- the computer -AND printer can be DTE. In addition... IBM had the serial port connected Rolm phone, a CBX connected voice / data device that could transfer data at 64K [but not for PSTN, unfortunately]. I have never found an RP240 or RP244, this has a DB25 port...

So... you ask what I'm trying to do... Deduce -WHY- IBM chose to populate Serial 1 [DB25] with 15 pins 2-13, 20, 22, and 24.

AST Serial Port DTE to NEC 7700 Series Serial DTE Printer Page 100
" In this case, the printer uses SRTS (Secondary Request To Send, pin 19) instead of pin 4 (RTS); apart from that exception, Figure 8-3 follows the ideal DTE input and sequence rules.
The advantage of the interface shown in Figure 8-3 is that every pin is driven by its functional counterpart on the other side of the interface; the functionally related signal pairs include DTR/DSR, RTS/CTS, and TxD/RxD. This should also hold true for the interface you design."

My SWAG, IBM chose parts of the earlier primary/secondary channels to support whatever supported devices.
 
The secondary (low-speed) channel was sometimes used with synchronous setups, for modem control. Usually run not much faster than 1200 bps.
I'm sure there's more to the modem control, but I'd buy that for a dollar... Let me find the two channel DB-25 pinout, identify the populated pin functions, and leave things up to interested onlookers if they have an external serial device.

As for myself, I can't remember the last time I used a DB-25 serial, it must have been in the late 1990s.

Ah-ha! RS-232C: The Interface By Larry Kraemer Micro Cornucopia, Number 25, August-September 1985 pages 36-39 physical. Blasted Gooble... If you want PDF, here it is...

"Synchronous data transmissions are normally more intelligent than asynchronous (smarter operators). Packet networks such as X.25 often transfer data synchronously." IBM did love them some X.25 in the day...

"Speeds up to 345.6K bits per second are supported using IBM Enhanced EIA-232-D which requires a special shielded cable up to 20 feet long." And that's about it. Methinks IBM is confused over EIA 232 C and D.

Ow. X.25 was supported by adapters like the Portmaster... Maybe an external modem, for you-all, this would be a slamn-dunk, but IBM did dark and mysterious things because... they could. Another bout of surfing for long forgotten arcania... I quit binge drinking too young...
 
Last edited:
Note also, that sync comms requires external clocking. Hence pins 17 and 24. cf. Bell 208A modem.

Most of my serial stuff is 25-pin D-sub. Some cables using the old Bell System Bakelite hoods. Patch boxes and signal monitors tend to be 25 pin.
 
Trying to find 786x modem documentation, there was an IBM management program ? CNM or whatnot. The period IBM modems did async or sync. More finger ninjitsu...
 
Note also, that sync comms requires external clocking. Hence pins 17 and 24. cf. Bell 208A modem.
Transmitter Clock (pin 24) provides the modem with transmit signal element timing information (optional). If
the DTE sources the transmit signal element timing on this pin, the modem can be configured to use this clock
signal rather than its own internal clock. This signal pertains to synchronous communications only.

Ring Indicator (pin 22) gives an indication of a ring being received on the telephone line. This circuit carriers
signal information to indicate the modem is receiving call signals (ring/tones). The modem turns on this signal
whenever an incoming ring is detected. When the modem is configured for AutoSync, this pin carries a signal to
indicate the end of a synchronous frame, and does not monitor incoming rings.

The truth is out there... How IBM implemented sync comms is unclear.
 
Folks, I re-read the Serial Port portion of the HITR, it makes no mention of DTE or DCE for the serial port on the IBM Model 90.

So... you ask what I'm trying to do... Deduce -WHY- IBM chose to populate Serial 1 [DB25] with 15 pins 2-13, 20, 22, and 24.
Ah well its not a modem, so its DTE, Data Terminal Equipment. Some multi-user computers had ports wired as DCE so you could plug terminals in with straight through lead.

Why 2--13 , 20, 22 & 24, well looking at the manual you linked to, it didn't. It implemented 2-8, 20 22. Why? well these are the minimum to needed to provide modem communications to a remote computer, so:-

Data Transfer
Flow Control
Incoming Call notification and answer.

these were the common interface signals. I remeber the rules for null modem leads :-

2->3
3->2
7->7
6->8+20
8+20->6

the other pins were mainly used for synchronous comms, where you needed clocks, and might have secondary data flows, so for example on ISDN there is a second links to allow "out of band" call set up
 
Back
Top