Loading ...
Sorry, an error occurred while loading the content.

Altair M6850 configuration, cables, and Handshaking.

Expand Messages
  • W Tom
    I ve studied the 88-2SIO Rev 0 manual, compared two 88-2SIO board revisions, and two types of internal cables. My 88-2SIO notes are are on a web page
    Message 1 of 11 , Jul 28, 2010
    • 0 Attachment
      I've studied the 88-2SIO Rev 0 manual, compared two 88-2SIO board revisions, and two types of internal cables.

      My 88-2SIO notes are are on a web page

      I hope to find a Rev 1 manual. Next, I'll check the turnkey and 88-UIO M6850 signals and cables.

      The 2SIO provides a non-standard RS-232 interface at the 25-pin connector on the Altair back panel. Each revision is a little different. This is what I found.

      The board is basically DTE with most handshaking signal disabled. The Rev 0 internal cable has some issues. The REV 1 board has one jumper I might cut. The Rev 0 manual has a NOTICE! that I have questions about accuracy.

      The Rev 0 manual does not specify the exact setting and cable shipped. Factory boards assume CTS and DCD are not provided by the external device. The boards have undocumented jumpers to provide these DTE input signals to the M6850.

      The manual has a notice similar to:

      NOTICE!! 1. If the Data Carrier Detect and Clear To Send inputs are not connected, they must be jumpered to Ground.

      I think this is wrong or outdated. The factory boards have these lines connected to +5V. Ground is RS-232 undefined, and +5V is Low.

      A possible problem is that the factory internal cables can route +5V from the 2SIO to OUTPUTS of the device connected. The line drivers can probably handle the voltage, but I don't see how that could be good.

      I have one example where someone added a resistor, probably to limit current. I think not connecting CTS and DCD in the internal cable is a good idea.

      The Rev 1 cable would be hard to change so I might cut one board jumper. The Rev 1 boards provide ground and 5V on two reserved lines.

      Both revisions swap transmit and receive so a device with a 3-wire cable thinks the board is DCE. The swap is done to a DTE console and use 3 wires to connect to a DTE board.

      I'm going to add handshaking and use more than three wires in the cable, so default configuration and internal cable configuration must be understood. Any comments on 2SIO configuration?


      W Tom S
      http://tom12.tech.officelive.com/882SIO.aspx 











       









    • steve
      Tom, If you had received your 2SIO manual from me, you would have seen my hand written addendum to that notice ( NOTICE!! 1. If the Data Carrier Detect and
      Message 2 of 11 , Jul 28, 2010
      • 0 Attachment
        Tom,

        If you had received your 2SIO manual from me, you would have seen my hand written addendum to that notice ("NOTICE!! 1. If the Data Carrier Detect and Clear To Send inputs are not connected, they must be jumpered to Ground.)"

        My addendum said THEY SHOULD BE JUMPERED TO GROUND FOR TTY AND TTL OPERATION, BUT JUMPERED TO +5 VOLTS FOR RS232 OPERATION.

        It was written by me when I worked at MITS, and that's the very same manual that I scanned for sale on my web site.

        I believe the statement to be true, but I haven't thought about it for 30 years. DCD and CTS must certainly be jumpered either high or low if they're not used, but without checking the schematic, I can't say for sure right now which way is really correct. If I recall correctly, there's an extra inverter involved in one configuration that's not present in the other configuration, and that's why the jumpers aren't always the same. I added that note because it was easily overlooked when troubleshooting non-functioning boards.

        steve
        ============================

        --- In altaircomputerclub@yahoogroups.com, "W Tom" <yahoo@...> wrote:
        ...
        > The manual has a notice similar to:
        > NOTICE!! 1. If the Data Carrier Detect and Clear To Send inputs are not connected, they must be jumpered to Ground. I think this is wrong or outdated. The factory boards have these lines connected to +5V. Ground is RS-232 undefined, and +5V is Low.
        >
        > ...
        > W Tom S
        > http://tom12.tech.officelive.com/882SIO.aspx
        > <http://tom12.tech.officelive.com/882SIO.aspx>
        >
      • W Tom
        I wanted some one to double check my observations, since hardware is not my specialty. My 2SIO attached notes are original hand written notes. The notes match
        Message 3 of 11 , Jul 28, 2010
        • 0 Attachment
          I wanted some one to double check my observations, since hardware is not my specialty. My 2SIO attached notes are original hand written notes. The notes match what I am finding on factory boards.

          Am I interpreting 14-pin chips correctly? I start at the dot and go anti-clockwise. The signals are connected to pin 14. Is that Vcc?

          The jumpers are shown on the page http://tom12.tech.officelive.com/882SIO.aspx 

          The Rev 0 jumpers are at the Molex connector .
          The Rev 1 jumpers are at the line driver IC.

          Steve, Do you have a REV 1 manual?

          W. Tom S


          --- In altaircomputerclub@yahoogroups.com, "steve" <alltare@...> wrote:
          >
          >
          >
          > Tom,
          >
          > If you had received your 2SIO manual from me, you would have seen my hand written addendum to that notice ("NOTICE!! 1. If the Data Carrier Detect and Clear To Send inputs are not connected, they must be jumpered to Ground.)"
          >
          > My addendum said THEY SHOULD BE JUMPERED TO GROUND FOR TTY AND TTL OPERATION, BUT JUMPERED TO +5 VOLTS FOR RS232 OPERATION.
          >
          > It was written by me when I worked at MITS, and that's the very same manual that I scanned for sale on my web site.
          >
          > I believe the statement to be true, but I haven't thought about it for 30 years. DCD and CTS must certainly be jumpered either high or low if they're not used, but without checking the schematic, I can't say for sure right now which way is really correct. If I recall correctly, there's an extra inverter involved in one configuration that's not present in the other configuration, and that's why the jumpers aren't always the same. I added that note because it was easily overlooked when troubleshooting non-functioning boards.
          >
          > steve
          > ============================
          >
          > --- In altaircomputerclub@yahoogroups.com, "W Tom" yahoo@ wrote:
          > ...
          > > The manual has a notice similar to:
          > > NOTICE!! 1. If the Data Carrier Detect and Clear To Send inputs are not connected, they must be jumpered to Ground. I think this is wrong or outdated. The factory boards have these lines connected to +5V. Ground is RS-232 undefined, and +5V is Low.
          > >
          > > ...
          > > W Tom S
          > > http://tom12.tech.officelive.com/882SIO.aspx
          > > <http://tom12.tech.officelive.com/882SIO.aspx>
          > >
          >
        • geistmarty
          ... Hi All; TOM, I think what part of the confusion is --- is where the VCC or Ground is to be connected..... IF you use Ground then Do NOT connect D1 or D2 or
          Message 4 of 11 , Jul 28, 2010
          • 0 Attachment
            --- In altaircomputerclub@yahoogroups.com, "W Tom" <yahoo@...> wrote:
            >
            > I wanted some one to double check my observations, since hardware is not
            > my specialty. My 2SIO attached notes are original hand written notes.
            > The notes match what I am finding on factory boards.
            >
            > Am I interpreting 14-pin chips correctly? I start at the dot and go
            > anti-clockwise. The signals are connected to pin 14. Is that Vcc?
            >
            > The jumpers are shown on the page
            > http://tom12.tech.officelive.com/882SIO.aspx
            > <http://tom12.tech.officelive.com/882SIO.aspx>
            >
            > The Rev 0 jumpers are at the Molex connector .
            > The Rev 1 jumpers are at the line driver IC.
            >
            > Steve, Do you have a REV 1 manual?
            >
            > W. Tom S
            >
            >
            > --- In altaircomputerclub@yahoogroups.com, "steve" <alltare@> wrote:
            > >
            > >
            > >
            > > Tom,
            > >
            > > If you had received your 2SIO manual from me, you would have seen my
            > hand written addendum to that notice ("NOTICE!! 1. If the Data Carrier
            > Detect and Clear To Send inputs are not connected, they must be jumpered
            > to Ground.)"
            > >
            > > My addendum said THEY SHOULD BE JUMPERED TO GROUND FOR TTY AND TTL
            > OPERATION, BUT JUMPERED TO +5 VOLTS FOR RS232 OPERATION.
            > >
            > > It was written by me when I worked at MITS, and that's the very same
            > manual that I scanned for sale on my web site.
            > >
            > > I believe the statement to be true, but I haven't thought about it for
            > 30 years. DCD and CTS must certainly be jumpered either high or low if
            > they're not used, but without checking the schematic, I can't say for
            > sure right now which way is really correct. If I recall correctly,
            > there's an extra inverter involved in one configuration that's not
            > present in the other configuration, and that's why the jumpers aren't
            > always the same. I added that note because it was easily overlooked
            > when troubleshooting non-functioning boards.
            > >
            > > steve
            > > ============================
            > >
            > > --- In altaircomputerclub@yahoogroups.com, "W Tom" yahoo@ wrote:
            > > ...
            > > > The manual has a notice similar to:
            > > > NOTICE!! 1. If the Data Carrier Detect and Clear To Send inputs are
            > not connected, they must be jumpered to Ground. I think this is wrong
            > or outdated. The factory boards have these lines connected to +5V.
            > Ground is RS-232 undefined, and +5V is Low.
            > > >
            > > > ...
            > > > W Tom S
            > > > http://tom12.tech.officelive.com/882SIO.aspx
            > > > <http://tom12.tech.officelive.com/882SIO.aspx>
            > > >
            > >
            >
            Hi All;
            TOM, I think what part of the confusion is --- is where the VCC or Ground is to be connected..... IF you use Ground then Do NOT connect D1 or D2 or if using the Second Port The E1 and E2 to Anything ELSE But Tie them to Ground.... But IF you Go Thru the 1489's then the INPUT to the 1489's is tied to VCC (+5)... So it depends on where the tie in point is....
            THANK YOU Marty
          • steve
            TOM- SEE REPLIES EMBEDDED BELOW ... DIDN T I JUST CONFIRM THAT IN MY PREVIOUS MESSAGE? ... WHEN LOOKING AT ALL CHIPS FROM ABOVE (NOT FROM THE BOTTOM SIDE), PIN
            Message 5 of 11 , Jul 28, 2010
            • 0 Attachment
              TOM-
              SEE REPLIES EMBEDDED BELOW

              --- In altaircomputerclub@yahoogroups.com, "W Tom" <yahoo@...> wrote:
              >
              > I wanted some one to double check my observations, since hardware
              > is not my specialty. ...

              DIDN'T I JUST CONFIRM THAT IN MY PREVIOUS MESSAGE?

              > Am I interpreting 14-pin chips correctly? I start at the dot and go
              > anti-clockwise. The signals are connected to pin 14. Is that Vcc?

              WHEN LOOKING AT ALL CHIPS FROM ABOVE (NOT FROM THE BOTTOM SIDE), PIN 1 IS INDICATED BY A DOT OR ARROW OR NOTCH OR OTHER UNIQUE MARKING. IF THE NOTCH OR MARKING IS CENTERED AT THE END OF A DIP, THEN IT INDICATES THE END FROM WHICH YOU BEGIN COUNTING COUNTERCLOCKWISE. THE TOP LEFT PIN IS ALWAYS PIN 1. PIN 14 (THE TOP RIGHT PIN OF A 14 PIN PACKAGE) IS USUALLY Vcc, BUT NOT NECESSARILY. WHAT DO YOU MEAN BY "THE SIGNALS ARE CONNECTED TO PIN 14"? WHAT SIGNALS? (Vcc ISN'T A SIGNAL, IT'S +5VOLTS). IF REAL SIGNAL LINES ARE CONNECTED TO PIN 14, BUT Vcc IS NOT, THEN THE Vcc PIN IS SOMEWHWERE ELSE. YOU DIDN'T GIVE A CHIP NUMBER, SO WE CAN'T HELP WITH ITS PINOUTS. YOU CAN GOOGLE ANY CHIP FOR ITS DATA SHEET. JUST TYPE THE NUMBER IN YOUR SEARCH BAR, FOLLOWED BY 'DATA SHEET'.
              ...

              > Steve, Do you have a REV 1 manual?

              MY 2SIO MANUAL IS DATED MARCH 1977, WHICH WOULD MAKE IT REV 0. IT DOESN'T MENTION RIBBON CABLES OR IDC CONNECTORS- JUST THE ORIGINAL MOLEX CONNECTORS AND CABLES. DESPITE PHYSICAL DIFFERENCES, I THINK THE CIRCUITRY IS PRETTY MUCH IDENTICAL.
              ...

              STEVE
              ==========================


              > --- In altaircomputerclub@yahoogroups.com, "steve" <alltare@> wrote:
              > >
              > >
              > >
              > > Tom,
              > >
              > > If you had received your 2SIO manual from me, you would have seen my
              > hand written addendum to that notice ("NOTICE!! 1. If the Data Carrier
              > Detect and Clear To Send inputs are not connected, they must be jumpered
              > to Ground.)"
              > >
              > > My addendum said THEY SHOULD BE JUMPERED TO GROUND FOR TTY AND TTL
              > OPERATION, BUT JUMPERED TO +5 VOLTS FOR RS232 OPERATION.
              > >
              > > It was written by me when I worked at MITS, and that's the very same
              > manual that I scanned for sale on my web site.
              > >
              > > I believe the statement to be true, but I haven't thought about it for
              > 30 years. DCD and CTS must certainly be jumpered either high or low if
              > they're not used, but without checking the schematic, I can't say for
              > sure right now which way is really correct. If I recall correctly,
              > there's an extra inverter involved in one configuration that's not
              > present in the other configuration, and that's why the jumpers aren't
              > always the same. I added that note because it was easily overlooked
              > when troubleshooting non-functioning boards.
              > >
              > > steve
              > > ============================
              > >
              > > --- In altaircomputerclub@yahoogroups.com, "W Tom" yahoo@ wrote:
              > > ...
              > > > The manual has a notice similar to:
              > > > NOTICE!! 1. If the Data Carrier Detect and Clear To Send inputs are
              > not connected, they must be jumpered to Ground. I think this is wrong
              > or outdated. The factory boards have these lines connected to +5V.
              > Ground is RS-232 undefined, and +5V is Low.
              > > >
              > > > ...
              > > > W Tom S
              > > > http://tom12.tech.officelive.com/882SIO.aspx
              > > > <http://tom12.tech.officelive.com/882SIO.aspx>
              > > >
              > >
              >
            • W Tom
              STEVE, NO NEED TO SHOUT, lower case is more readable. ... No! We confirmed that I have better notes and compare them to multiple board samples. I did Google
              Message 6 of 11 , Jul 28, 2010
              • 0 Attachment
                STEVE,

                NO NEED TO SHOUT, lower case is more readable.


                > >
                > > I wanted some one to double check my observations, since hardware
                > > is not my specialty. ...
                >
                > DIDN'T I JUST CONFIRM THAT IN MY PREVIOUS MESSAGE?

                No! We confirmed that I have better notes and compare them to multiple board samples. I did Google the chips and standards, check lines with a meter, and make reference images before I asked the experts. I noticed that GND is an undefined voltage level and +5V is Space or a low bit.

                I missed the inverter. What do you get when you invert undefined? My notes did say to make the change at the Molex and that is what I saw looking at the board.

                Marty, you got right.

                > TOM, I think what part of the confusion is --- is where the VCC or Ground is to
                > be connected..... IF you use Ground then Do NOT connect D1 or D2 or if using the
                > Second Port The E1 and E2 to Anything ELSE But Tie them to Ground.... But
                > IF you Go Thru the 1489's then the INPUT to the 1489's is tied to VCC (+5)...
                > So it depends on where the tie in point is....

                I pulled out another 2SIO manual and found more notes.

                "MITS manual was unclear in it's note to jumper DCD & CTS to GND. These signals must be jumpered to +5V at the interface connector because it gets inverted before it gets to the 6850."

                > MY 2SIO MANUAL IS DATED MARCH 1977, WHICH WOULD MAKE IT REV 0. IT DOESN'T MENTION RIBBON CABLES OR IDC CONNECTORS- JUST THE ORIGINAL MOLEX CONNECTORS AND CABLES. DESPITE PHYSICAL DIFFERENCES, I THINK THE CIRCUITRY IS PRETTY MUCH IDENTICAL.

                The Rev 0 & Rev 1 circuitry is pretty much the same. The internal cables and mods are not well described, that is why I am studying boards and cables, then documenting the results.

                The differences are at the back connector:

                Rev 1 has ground on 25-10 (RS-232 Reserved) and +5V on 25-9 (RS-232 Reserved). These pins are not connected on a Rev 0 cable.

                Rev 0 has +5V on 25-5 (CTS) and 25-8 (DCD)
                Rev 1 has  +5V on 25-5 (CTS) and no connection to 25-8 (DCD)

                Rev 0 connects +5V at the interface connector.
                Rev 1 connects +5V at the back of a 1489 line driver to I1 & I6

                The circuitry is pretty much the same, but jumper location and internal cabling produce different non-standard connections.

                Both revisions swap transmit & receive, but not handshaking signals. Does the manual or notes mention that small detail?

                THANK YOU Marty!

                W Tom S
                 


                --- In altaircomputerclub@yahoogroups.com, "steve" <alltare@...> wrote:
                >
                >
                >
                > TOM-
                > SEE REPLIES EMBEDDED BELOW
                >
                > --- In altaircomputerclub@yahoogroups.com, "W Tom" yahoo@ wrote:
                > >
                > > I wanted some one to double check my observations, since hardware
                > > is not my specialty. ...
                >
                > DIDN'T I JUST CONFIRM THAT IN MY PREVIOUS MESSAGE?
                >
                > > Am I interpreting 14-pin chips correctly? I start at the dot and go
                > > anti-clockwise. The signals are connected to pin 14. Is that Vcc?
                >
                > WHEN LOOKING AT ALL CHIPS FROM ABOVE (NOT FROM THE BOTTOM SIDE), PIN 1 IS INDICATED BY A DOT OR ARROW OR NOTCH OR OTHER UNIQUE MARKING. IF THE NOTCH OR MARKING IS CENTERED AT THE END OF A DIP, THEN IT INDICATES THE END FROM WHICH YOU BEGIN COUNTING COUNTERCLOCKWISE. THE TOP LEFT PIN IS ALWAYS PIN 1. PIN 14 (THE TOP RIGHT PIN OF A 14 PIN PACKAGE) IS USUALLY Vcc, BUT NOT NECESSARILY. WHAT DO YOU MEAN BY "THE SIGNALS ARE CONNECTED TO PIN 14"? WHAT SIGNALS? (Vcc ISN'T A SIGNAL, IT'S +5VOLTS). IF REAL SIGNAL LINES ARE CONNECTED TO PIN 14, BUT Vcc IS NOT, THEN THE Vcc PIN IS SOMEWHWERE ELSE. YOU DIDN'T GIVE A CHIP NUMBER, SO WE CAN'T HELP WITH ITS PINOUTS. YOU CAN GOOGLE ANY CHIP FOR ITS DATA SHEET. JUST TYPE THE NUMBER IN YOUR SEARCH BAR, FOLLOWED BY 'DATA SHEET'.
                > ...
                >
                > > Steve, Do you have a REV 1 manual?
                >
                > MY 2SIO MANUAL IS DATED MARCH 1977, WHICH WOULD MAKE IT REV 0. IT DOESN'T MENTION RIBBON CABLES OR IDC CONNECTORS- JUST THE ORIGINAL MOLEX CONNECTORS AND CABLES. DESPITE PHYSICAL DIFFERENCES, I THINK THE CIRCUITRY IS PRETTY MUCH IDENTICAL.
                > ...
                >
                > STEVE
                > ==========================
                >
                >
                > > --- In altaircomputerclub@yahoogroups.com, "steve" <alltare@> wrote:
                > > >
                > > >
                > > >
                > > > Tom,
                > > >
                > > > If you had received your 2SIO manual from me, you would have seen my
                > > hand written addendum to that notice ("NOTICE!! 1. If the Data Carrier
                > > Detect and Clear To Send inputs are not connected, they must be jumpered
                > > to Ground.)"
                > > >
                > > > My addendum said THEY SHOULD BE JUMPERED TO GROUND FOR TTY AND TTL
                > > OPERATION, BUT JUMPERED TO +5 VOLTS FOR RS232 OPERATION.
                > > >
                > > > It was written by me when I worked at MITS, and that's the very same
                > > manual that I scanned for sale on my web site.
                > > >
                > > > I believe the statement to be true, but I haven't thought about it for
                > > 30 years. DCD and CTS must certainly be jumpered either high or low if
                > > they're not used, but without checking the schematic, I can't say for
                > > sure right now which way is really correct. If I recall correctly,
                > > there's an extra inverter involved in one configuration that's not
                > > present in the other configuration, and that's why the jumpers aren't
                > > always the same. I added that note because it was easily overlooked
                > > when troubleshooting non-functioning boards.
                > > >
                > > > steve
                > > > ============================
                > > >
                > > > --- In altaircomputerclub@yahoogroups.com, "W Tom" yahoo@ wrote:
                > > > ...
                > > > > The manual has a notice similar to:
                > > > > NOTICE!! 1. If the Data Carrier Detect and Clear To Send inputs are
                > > not connected, they must be jumpered to Ground. I think this is wrong
                > > or outdated. The factory boards have these lines connected to +5V.
                > > Ground is RS-232 undefined, and +5V is Low.
                > > > >
                > > > > ...
                > > > > W Tom S
                > > > > http://tom12.tech.officelive.com/882SIO.aspx
                > > > > <http://tom12.tech.officelive.com/882SIO.aspx>
                > > > >
                > > >
                > >
                >
              • steve
                ... The intent was to differentiate MY TEXT from yours- not to shout. Sorry if you didn t realize that, but you re right- it does look like shouting and all
                Message 7 of 11 , Jul 28, 2010
                • 0 Attachment
                  --- In altaircomputerclub@yahoogroups.com, "W Tom" <yahoo@...> wrote:
                  >
                  > STEVE,
                  >
                  > NO NEED TO SHOUT, lower case is more readable.
                  >

                  The intent was to differentiate MY TEXT from yours- not to shout.  Sorry if you didn't realize that, but you're right- it does look like shouting and all caps text is hard to read, too.  In the future I'll us color, if you'll agree to view my embedded comments in rich text rather than plain text mode.

                  Steve

                • Steve Piette
                  ... The board isn t really a DTE or DCE until you consider the cable from the board connector to the back panel. It s the wiring of that cable that configures
                  Message 8 of 11 , Jul 28, 2010
                  • 0 Attachment
                    On 7/28/2010 11:28 AM, W Tom wrote:
                    I've studied the 88-2SIO Rev 0 manual, compared two 88-2SIO board revisions, and two types of internal cables.

                    My 88-2SIO notes are are on a web page

                    I hope to find a Rev 1 manual. Next, I'll check the turnkey and 88-UIO M6850 signals and cables.

                    The 2SIO provides a non-standard RS-232 interface at the 25-pin connector on the Altair back panel. Each revision is a little different. This is what I found.

                    The board is basically DTE with most handshaking signal disabled. The Rev 0 internal cable has some issues. The REV 1 board has one jumper I might cut. The Rev 0 manual has a NOTICE! that I have questions about accuracy.

                    The board isn't really a DTE or DCE until you consider the cable from the board connector to the back panel. It's the wiring of that cable that configures the interface as either a DCE or DTE. Based on the manual the Rev 0 boards don't change the signals on the Molex connector so I have to assume the difference was in the wiring of the Molex connector to DB-25 connector. On the Rev 1 boards where they use a IDC type connector, that looks to have 26 pins, they could have used standard IDC connectors and flat ribbon cable between a DB-25 and the edge connector and if so they would have to change the on-board wiring depending what's connected to the port to be a DTE or DCE.


                    The Rev 0 manual does not specify the exact setting and cable shipped. Factory boards assume CTS and DCD are not provided by the external device. The boards have undocumented jumpers to provide these DTE input signals to the M6850.

                    The available Rev 0 manual defines the signals on the Molex connector as:

                     1 - CTS
                     2 - DCD
                     3 - RTS
                     4 - GND
                     5 - (TTL TXD)
                     6 -
                     7 - RXD
                     8 - TXD
                     9 - (TTL RXD)
                    10 - Tied to GND

                    It leaves the connections between the board and your device as an exercise for the reader.

                    The Rev 1 board (based on what you have on your web page) looks like they were thinking about using ribbon cables based on which pins they assigned the signals to.

                     1 -
                     2 - RXD
                     3 - TXD
                     4 - RTS
                     5 - CTS
                     6 -
                     7 - GND
                     8 -
                     9 - DCD
                    10 - Tied to GND

                    Note the interesting mapping. :)

                    It's interesting to note that this is a different arrangement than how the pins are used on the Turnkey board. So a cable for the 2SIO would be wrong when used with the Turnkey board.

                    The Turnkey board connector signals are:

                     1 - (TTL RTS)
                     2 - [TTY CL]
                     3 - -18v
                     4 - RXD
                     5 - DCD
                     6 - CTS
                     7 - External Clock
                     8 - GND
                     9 - RTS
                    10 - TXD

                    I'm sure there was a good reason for the difference that has been lost but I bet it drove the purchasing, manufacturing and service people nuts.


                    The manual has a notice simi! lar to:

                    NOTICE!! 1. If the Data Carrier Detect and Clear To Send inputs are not connected, they must be jumpered to Ground.

                    I think this is wrong or outdated. The factory boards have these lines connected to +5V. Ground is RS-232 undefined, and +5V is Low.

                    Somebody got confused. If you were going to do it at the 6850, tieing DCD and CTS to ground would enable them. If you going to do it at the S1 connector you have to pull them high. If you look at the Turnkey board you see jumpers to do just that.  They are connected via pull-up resistors to +9 and the MC1489 RS-232 receiver inverts them. The resistors are weak (4.7k) pull-ups so that an external device can actually pull them down when connected without burning anything up.


                    A possible problem is that the factory internal cables can route +5V from the 2SIO to OUTPUTS of the device connected. The line drivers can probably handle the voltage, but I don't see how that could be good.

                    I have one example where someone added a resistor, probably to limit current. I think not connecting CTS and DCD in the internal cable is a good idea.

                    The Rev 1 cable would be hard to change so I might cut one board jumper. The Rev 1 boards provide ground and 5V on two reserved lines.

                    Both revisions swap transmit and receive so a device with a 3-wire cable thinks the board is DCE. The swap is done to a DTE console and use 3 wires to connect to a! DTE board.

                    I'm going to add handshaking and use more than thr ee wires in the cable, so default configuration and internal cable configuration must be understood. Any comments on 2SIO configuration?


                    W Tom S
                    http://tom12.tech.officelive.com/882SIO.aspx

                  • W Tom
                    Steve P, Thank you very much for clarifying the issues. I would like to provide correct information on my page and needed feedback on some issues and people to
                    Message 9 of 11 , Jul 29, 2010
                    • 0 Attachment
                      Steve P,

                      Thank you very much for clarifying the issues. I would like to provide
                      correct information on my page and needed feedback on some issues and
                      people to point out what is not clear.

                      For anyone who did not understand, I'm only considering RS-232
                      configurations that were shipped from MITS. I'm ignoring TTY, Kits, and
                      user mods.

                      > The board isn't really a DTE or DCE until you consider the cable from
                      > the board connector to the back panel. It's the wiring of that cable
                      > that configures the interface as either a DCE or DTE

                      That is the point I was trying to make and need to explain better. If
                      you map the signals from the 10-pin straight through to the 25-pin
                      cable. the implementation is DTE.

                      If transmit and receive are swapped in the internal cable, the board
                      looks like DCE from the viewpoint of a 3-wire cable. If you use more
                      than three wires, then the boards looks part DTE, part DCE, part
                      unimplemented, and +5V on some lines.

                      > Based on the
                      > manual the Rev 0 boards don't change the signals on the Molex
                      connector
                      > so I have to assume the difference was in the wiring of the Molex
                      > connector to DB-25 connector.

                      Yes! The transmit/receive swap is barely noticeable in a table and is
                      easy to miss.

                      > It leaves the connections between the board and your device as an
                      > exercise for the reader.

                      Not a fun exercise for me. I'm using factory cables and not liking
                      everything done at the factory. I'm going to need to add handshaking
                      lines on one port and it is nice to know what is connected in the
                      default configuration shipped by MITS.

                      The exercise is two part - internal cable and external cable. To
                      configure the external cable, you must know which board revision and
                      what internal cable is used.


                      > Note the interesting mapping. :)

                      > The Turnkey board connector signals are:
                      >
                      > 1 - (TTL RTS)
                      > 2 - [TTY CL]
                      > 3 - -18v
                      > 4 - RXD
                      > 5 - DCD
                      > 6 - CTS
                      > 7 - External Clock
                      > 8 - GND
                      > 9 - RTS
                      > 10 - TXD
                      >
                      > I'm sure there was a good reason for the difference that has been lost
                      > but I bet it drove the purchasing, manufacturing and service people
                      nuts..

                      It was driving me nuts too, but now at least I know the issues exist.

                      Thanks for the Turnkey board connector signals (10-pin Molex) Now I will
                      see where signals go on the back panel 25-pin connector. I'll also check
                      the 88-UIO and cable.

                      W. Tom S

                      --- In altaircomputerclub@yahoogroups.com, Steve Piette <steve@...>
                      wrote:
                      >
                      > On 7/28/2010 11:28 AM, W Tom wrote:
                      > >
                      > >
                      > > I've studied the 88-2SIO Rev 0 manual, compared two 88-2SIO board
                      > > revisions, and two types of internal cables.
                      > >
                      > > My 88-2SIO notes are are on a web page
                      > > <http://tom12.tech.officelive.com/882SIO.aspx>
                      > >
                      > > I hope to find a Rev 1 manual. Next, I'll check the turnkey and
                      88-UIO
                      > > M6850 signals and cables.
                      > >
                      > > The 2SIO provides a non-standard RS-232 interface at the 25-pin
                      > > connector on the Altair back panel. Each revision is a little
                      > > different. This is what I found.
                      > >
                      > > The board is basically DTE with most handshaking signal disabled.
                      The
                      > > Rev 0 internal cable has some issues. The REV 1 board has one jumper
                      I
                      > > might cut. The Rev 0 manual has a NOTICE! that I have questions
                      about
                      > > accuracy.
                      >
                      > The board isn't really a DTE or DCE until you consider the cable from
                      > the board connector to the back panel. It's the wiring of that cable
                      > that configures the interface as either a DCE or DTE. Based on the
                      > manual the Rev 0 boards don't change the signals on the Molex
                      connector
                      > so I have to assume the difference was in the wiring of the Molex
                      > connector to DB-25 connector. On the Rev 1 boards where they use a IDC
                      > type connector, that looks to have 26 pins, they could have used
                      > standard IDC connectors and flat ribbon cable between a DB-25 and the
                      > edge connector and if so they would have to change the on-board wiring
                      > depending what's connected to the port to be a DTE or DCE.
                      >
                      > >
                      > > The Rev 0 manual does not specify the exact setting and cable
                      shipped.
                      > > Factory boards assume CTS and DCD are not provided by the external
                      > > device. The boards have undocumented jumpers to provide these DTE
                      > > input signals to the M6850.
                      >
                      > The available Rev 0 manual defines the signals on the Molex connector
                      as:
                      >
                      > 1 - CTS
                      > 2 - DCD
                      > 3 - RTS
                      > 4 - GND
                      > 5 - (TTL TXD)
                      > 6 -
                      > 7 - RXD
                      > 8 - TXD
                      > 9 - (TTL RXD)
                      > 10 - Tied to GND
                      >
                      > It leaves the connections between the board and your device as an
                      > exercise for the reader.
                      >
                      > The Rev 1 board (based on what you have on your web page) looks like
                      > they were thinking about using ribbon cables based on which pins they
                      > assigned the signals to.
                      >
                      > 1 -
                      > 2 - RXD
                      > 3 - TXD
                      > 4 - RTS
                      > 5 - CTS
                      > 6 -
                      > 7 - GND
                      > 8 -
                      > 9 - DCD
                      > 10 - Tied to GND
                      >
                      > Note the interesting mapping. :)
                      >
                      > It's interesting to note that this is a different arrangement than how
                      > the pins are used on the Turnkey board. So a cable for the 2SIO would
                      be
                      > wrong when used with the Turnkey board.
                      >
                      > The Turnkey board connector signals are:
                      >
                      > 1 - (TTL RTS)
                      > 2 - [TTY CL]
                      > 3 - -18v
                      > 4 - RXD
                      > 5 - DCD
                      > 6 - CTS
                      > 7 - External Clock
                      > 8 - GND
                      > 9 - RTS
                      > 10 - TXD
                      >
                      > I'm sure there was a good reason for the difference that has been lost
                      > but I bet it drove the purchasing, manufacturing and service people
                      nuts.
                      >
                      > >
                      > > The manual has a notice simi! lar to:
                      > >
                      > > *NOTICE!! 1. If the Data Carrier Detect and Clear To Send inputs are
                      > > not connected, they must be jumpered to Ground. *
                      > >
                      > > I think this is wrong or outdated. The factory boards have these
                      lines
                      > > connected to +5V. Ground is RS-232 undefined, and +5V is Low.
                      >
                      > Somebody got confused. If you were going to do it at the 6850, tieing
                      > DCD and CTS to ground would enable them. If you going to do it at the
                      S1
                      > connector you have to pull them high. If you look at the Turnkey board
                      > you see jumpers to do just that. They are connected via pull-up
                      > resistors to +9 and the MC1489 RS-232 receiver inverts them. The
                      > resistors are weak (4.7k) pull-ups so that an external device can
                      > actually pull them down when connected without burning anything up.
                      >
                      > >
                      > > A possible problem is that the factory internal cables can route +5V
                      > > from the 2SIO to OUTPUTS of the device connected. The line drivers
                      can
                      > > probably handle the voltage, but I don't see how that could be good.
                      > >
                      > > I have one example where someone added a resistor, probably to limit
                      > > current. I think not connecting CTS and DCD in the internal cable is
                      a
                      > > good idea.
                      > >
                      > > The Rev 1 cable would be hard to change so I might cut one board
                      > > jumper. The Rev 1 boards provide ground and 5V on two reserved
                      lines.
                      > >
                      > > Both revisions swap transmit and receive so a device with a 3-wire
                      > > cable thinks the board is DCE. The swap is done to a DTE console and
                      > > use 3 wires to connect to a! DTE board.
                      > >
                      > > I'm going to add handshaking and use more than thr ee wires in the
                      > > cable, so default configuration and internal cable configuration
                      must
                      > > be understood. Any comments on 2SIO configuration?
                      > >
                      > >
                      > > W Tom S
                      > > http://tom12.tech.officelive.com/882SIO.aspx
                      >
                    • Steve Piette
                      ... I d suggest showing the board strapping in one table and the cable mapping in a different one. Maybe somebody will remember if there were multiple
                      Message 10 of 11 , Jul 29, 2010
                      • 0 Attachment
                        On 7/29/2010 3:54 AM, W Tom wrote:
                        > Steve P,
                        >
                        > Thank you very much for clarifying the issues. I would like to provide
                        > correct information on my page and needed feedback on some issues and
                        > people to point out what is not clear.
                        >
                        > For anyone who did not understand, I'm only considering RS-232
                        > configurations that were shipped from MITS. I'm ignoring TTY, Kits, and
                        > user mods.
                        >
                        >

                        I'd suggest showing the board strapping in one table and the cable
                        mapping in a different one.
                        Maybe somebody will remember if there were multiple different factory
                        cables depending if it was a terminal, modem, or printer connected to
                        the port.

                        >> The board isn't really a DTE or DCE until you consider the cable from
                        >> the board connector to the back panel. It's the wiring of that cable
                        >> that configures the interface as either a DCE or DTE
                        >>
                        > That is the point I was trying to make and need to explain better. If
                        > you map the signals from the 10-pin straight through to the 25-pin
                        > cable. the implementation is DTE.
                        >

                        On the Rev 0 boards you can't map the signals straight through from the
                        10 pin Molex connector to the DB-25.
                        You can map out the factory cable but there may be other factory cables
                        that implement a DCE configuration.

                        The Rev 1 boards do have a logical mapping if you assume the factory
                        cable was a 26 pin ribbon cable with
                        dual row 26 pin Berg connector on one end and a IDC DB-25 on the other end.

                        > If transmit and receive are swapped in the internal cable, the board
                        > looks like DCE from the viewpoint of a 3-wire cable. If you use more
                        > than three wires, then the boards looks part DTE, part DCE, part
                        > unimplemented, and +5V on some lines.
                        >
                        >
                        >> Based on the manual the Rev 0 boards don't change the signals on the
                        >> Molex connector so I have to assume the difference was in the wiring
                        >> of the Molex connector to DB-25 connector.
                        >>
                        > Yes! The transmit/receive swap is barely noticeable in a table and is
                        > easy to miss.
                        >
                        >

                        Where in the manual is that table. I clearly didn't see it.

                        >> It leaves the connections between the board and your device as an
                        >> exercise for the reader.
                        >>
                        > Not a fun exercise for me. I'm using factory cables and not liking
                        > everything done at the factory. I'm going to need to add handshaking
                        > lines on one port and it is nice to know what is connected in the
                        > default configuration shipped by MITS.
                        >

                        Handshaking for hardware flow control (RTS/CTS) or to implement DTR/DCD?

                        > The exercise is two part - internal cable and external cable. To
                        > configure the external cable, you must know which board revision and
                        > what internal cable is used.
                        >
                        >
                        >> The Rev 1 board (based on what you have on your web page) looks like
                        >> they were thinking about using ribbon cables based on which pins they
                        >> assigned the signals to.
                        >>
                        >> 1 -
                        >> 2 - RXD
                        >> 3 - TXD
                        >> 4 - RTS
                        >> 5 - CTS
                        >> 6 -
                        >> 7 - GND
                        >> 8 -
                        >> 9 - DCD
                        >> 10 - Tied to GND
                        >>
                        >> Note the interesting mapping. :)
                        >>
                        >>
                        >> The Turnkey board connector signals are:
                        >>
                        >> 1 - (TTL RTS)
                        >> 2 - [TTY CL]
                        >> 3 - -18v
                        >> 4 - RXD
                        >> 5 - DCD
                        >> 6 - CTS
                        >> 7 - External Clock
                        >> 8 - GND
                        >> 9 - RTS
                        >> 10 - TXD
                        >>
                        >> I'm sure there was a good reason for the difference that has been lost
                        >> but I bet it drove the purchasing, manufacturing and service people
                        >> nuts..
                        >>
                        >>
                        >
                        > It was driving me nuts too, but now at least I know the issues exist.
                        > Thanks for the Turnkey board connector signals (10-pin Molex) Now I will
                        > see where signals go on the back panel 25-pin connector. I'll also check
                        > the 88-UIO and cable.
                        >
                        > W. Tom S

                        Since I don't have a 2SIO board I can't verify something shown in the
                        schematics that doesn't make sense.
                        The schematic on page 1 shows DCD going from the 6850 to point D1/E1 and
                        on page 2 it shows D1/E1 as being grounded.
                        For CTS it's D2/E2. I think that's an error in the schematics since they
                        have to connect either the 1489 RS-232 receiver
                        or an inverter to them. So I don't think that D1/E1 and D2/E2 are really
                        grounded on the board via PCB traces as shown.
                        If they are then they would need to be cut to use DCD and CTS.

                        I didn't explain very well that the manual saying you need to ground DCD
                        and CTS makes perfect sense if the engineer was thinking about grounding
                        those signals at the D1/E1 and D2/E2 connection points using jumpers
                        rather than at the S1/S2 connector points.
                      • W Tom
                        Thanks again Steve P, The picture is getting clearer to me. I will improve my notes. I ve been asked what my problem is, because understanding RS-232 is easy.
                        Message 11 of 11 , Jul 29, 2010
                        • 0 Attachment
                          Thanks again Steve P,

                          The picture is getting clearer to me. I will improve my notes. I've been asked what my problem is, because understanding RS-232 is easy. From what I read, I'm not the first person with RS-232 interface questions. My mistake for using the term SPACE or MARK. I found these terms are known to be confusing.


                          >
                          > I'd suggest showing the board strapping in one table and the cable
                          > mapping in a different one.
                          > Maybe somebody will remember if there were multiple different factory
                          > cables depending if it was a terminal, modem, or printer connected to
                          > the port.

                          That is my plan, after I understand what is going on. What I have now, is the first draft. With a little help from my friends in this group, I'll improve the presentation.


                          > Since I don't have a 2SIO board I can't verify something shown in the
                          > schematics that doesn't make sense.
                          > The schematic on page 1 shows DCD going from the 6850 to point D1/E1 and
                          > on page 2 it shows D1/E1 as being grounded.
                          > For CTS it's D2/E2. I think that's an error in the schematics since they
                          > have to connect either the 1489 RS-232 receiver
                          > or an inverter to them. So I don't think that D1/E1 and D2/E2 are really
                          > grounded on the board via PCB traces as shown.
                          > If they are then they would need to be cut to use DCD and CTS.
                          >

                          Are my images good enough to see the board front? Would images of the board back be useful?


                          > On the Rev 0 boards you can't map the signals straight through from the
                          > 10 pin Molex connector to the DB-25.
                          > You can map out the factory cable but there may be other factory cables
                          > that implement a DCE configuration.

                          Bad description on my part. By straight through I meant (Molex)transmit to 25-2 transmit, (Molex)receive to 25-3


                          > > Yes! The transmit/receive swap is barely noticeable in a table and is
                          > > easy to miss.
                          > >
                          > >
                          >
                          > Where in the manual is that table. I clearly didn't see it.


                          (Capitalization from MITS Manual)

                          THE FOLLOWING CHART INDICATES THE NORMAL INTER-CONNECTIONS TO BE MADE IF YOU WANT TO WIRE YOUR BOARD FOR RS-232 COMPATIBILITY.

                          SIGNAL DESCRIPTION PIN NUMBER

                          Receive 2
                          Request to Send 4
                          Clear to Send 5
                          Transmit 3
                          Carrier Detect 8
                          Data Terminal Ready 20
                          Signal Ground 7

                          The signal descriptions are for the 10-pin Molex.
                          The pin numbers are for the DB-25.

                          Note that 25-2 is Transmitted Data in the standard.
                          25-3 is Receive.

                          The pins 25-4 (RTS), 25-5 (CTS), 25-7 (SG), 25-8 (CD), and 25-20(DTR)are labeled correctly for the DB-25. I believe pins are named from the perspective of DTE. Looking at the 25-pin connector at the back panel, I see DTE pinouts except for transmit and receive.

                          Looking at the table, I see PIN 2 and PIN 3. It is an exercise for the reader to notice the SIGNAL DESCRIPTION is swapped only for pins 2 & 3.


                          > Handshaking for hardware flow control (RTS/CTS) or to implement DTR/DCD?
                          >

                          I may be using the term Handshaking incorrectly. All I want is one handshaking signal (DCD) to send to the RCPM BYE program. In my configuration there is no modem to handshake with. I need one signal to approximate a phone connecting and hanging up.


                          > I didn't explain very well that the manual saying you need to ground DCD
                          > and CTS makes perfect sense if the engineer was thinking about grounding
                          > those signals at the D1/E1 and D2/E2 connection points using jumpers
                          > rather than at the S1/S2 connector points.

                          I knew the NOTICE! had to have a better explanation. It assumes the reader is a hardware engineer with the big picture. Somewhere, I read that the mod was make at the Molex. The NOTICE! does not provide the context or match what MITS did on my board samples.

                          Both GND and +5V are the right answer, you just have to know the right question.


                          W Tom S


                          --- In altaircomputerclub@yahoogroups.com, Steve Piette <steve@...> wrote:
                          >
                          > On 7/29/2010 3:54 AM, W Tom wrote:
                          > > Steve P,
                          > >
                          > > Thank you very much for clarifying the issues. I would like to provide
                          > > correct information on my page and needed feedback on some issues and
                          > > people to point out what is not clear.
                          > >
                          > > For anyone who did not understand, I'm only considering RS-232
                          > > configurations that were shipped from MITS. I'm ignoring TTY, Kits, and
                          > > user mods.
                          > >
                          > >
                          >
                          > I'd suggest showing the board strapping in one table and the cable
                          > mapping in a different one.
                          > Maybe somebody will remember if there were multiple different factory
                          > cables depending if it was a terminal, modem, or printer connected to
                          > the port.
                          >
                          > >> The board isn't really a DTE or DCE until you consider the cable from
                          > >> the board connector to the back panel. It's the wiring of that cable
                          > >> that configures the interface as either a DCE or DTE
                          > >>
                          > > That is the point I was trying to make and need to explain better. If
                          > > you map the signals from the 10-pin straight through to the 25-pin
                          > > cable. the implementation is DTE.
                          > >
                          >
                          > On the Rev 0 boards you can't map the signals straight through from the
                          > 10 pin Molex connector to the DB-25.
                          > You can map out the factory cable but there may be other factory cables
                          > that implement a DCE configuration.
                          >
                          > The Rev 1 boards do have a logical mapping if you assume the factory
                          > cable was a 26 pin ribbon cable with
                          > dual row 26 pin Berg connector on one end and a IDC DB-25 on the other end.
                          >
                          > > If transmit and receive are swapped in the internal cable, the board
                          > > looks like DCE from the viewpoint of a 3-wire cable. If you use more
                          > > than three wires, then the boards looks part DTE, part DCE, part
                          > > unimplemented, and +5V on some lines.
                          > >
                          > >
                          > >> Based on the manual the Rev 0 boards don't change the signals on the
                          > >> Molex connector so I have to assume the difference was in the wiring
                          > >> of the Molex connector to DB-25 connector.
                          > >>
                          > > Yes! The transmit/receive swap is barely noticeable in a table and is
                          > > easy to miss.
                          > >
                          > >
                          >
                          > Where in the manual is that table. I clearly didn't see it.
                          >
                          > >> It leaves the connections between the board and your device as an
                          > >> exercise for the reader.
                          > >>
                          > > Not a fun exercise for me. I'm using factory cables and not liking
                          > > everything done at the factory. I'm going to need to add handshaking
                          > > lines on one port and it is nice to know what is connected in the
                          > > default configuration shipped by MITS.
                          > >
                          >
                          > Handshaking for hardware flow control (RTS/CTS) or to implement DTR/DCD?
                          >
                          > > The exercise is two part - internal cable and external cable. To
                          > > configure the external cable, you must know which board revision and
                          > > what internal cable is used.
                          > >
                          > >
                          > >> The Rev 1 board (based on what you have on your web page) looks like
                          > >> they were thinking about using ribbon cables based on which pins they
                          > >> assigned the signals to.
                          > >>
                          > >> 1 -
                          > >> 2 - RXD
                          > >> 3 - TXD
                          > >> 4 - RTS
                          > >> 5 - CTS
                          > >> 6 -
                          > >> 7 - GND
                          > >> 8 -
                          > >> 9 - DCD
                          > >> 10 - Tied to GND
                          > >>
                          > >> Note the interesting mapping. :)
                          > >>
                          > >>
                          > >> The Turnkey board connector signals are:
                          > >>
                          > >> 1 - (TTL RTS)
                          > >> 2 - [TTY CL]
                          > >> 3 - -18v
                          > >> 4 - RXD
                          > >> 5 - DCD
                          > >> 6 - CTS
                          > >> 7 - External Clock
                          > >> 8 - GND
                          > >> 9 - RTS
                          > >> 10 - TXD
                          > >>
                          > >> I'm sure there was a good reason for the difference that has been lost
                          > >> but I bet it drove the purchasing, manufacturing and service people
                          > >> nuts..
                          > >>
                          > >>
                          > >
                          > > It was driving me nuts too, but now at least I know the issues exist.
                          > > Thanks for the Turnkey board connector signals (10-pin Molex) Now I will
                          > > see where signals go on the back panel 25-pin connector. I'll also check
                          > > the 88-UIO and cable.
                          > >
                          > > W. Tom S
                          >
                          > Since I don't have a 2SIO board I can't verify something shown in the
                          > schematics that doesn't make sense.
                          > The schematic on page 1 shows DCD going from the 6850 to point D1/E1 and
                          > on page 2 it shows D1/E1 as being grounded.
                          > For CTS it's D2/E2. I think that's an error in the schematics since they
                          > have to connect either the 1489 RS-232 receiver
                          > or an inverter to them. So I don't think that D1/E1 and D2/E2 are really
                          > grounded on the board via PCB traces as shown.
                          > If they are then they would need to be cut to use DCD and CTS.
                          >
                          > I didn't explain very well that the manual saying you need to ground DCD
                          > and CTS makes perfect sense if the engineer was thinking about grounding
                          > those signals at the D1/E1 and D2/E2 connection points using jumpers
                          > rather than at the S1/S2 connector points.
                          >
                        Your message has been successfully submitted and would be delivered to recipients shortly.