Image Map Image Map
Page 3 of 3 FirstFirst 123
Results 21 to 29 of 29

Thread: No serial output on PDP-11

  1. #21
    Join Date
    Dec 2012
    Location
    Colorado
    Posts
    186
    Blog Entries
    1

    Default

    It is best if you stop thinking of this system as an 11/34. There is almost nothing left that is recognizable as an 11/34 except for the enclosure and the front panel. A J11 CPU will only look at the halt and run signals, and is hard coded to expect a DEC style serial interface at the standard PDP-11 console port address.

  2. #22
    Join Date
    Jun 2012
    Location
    UK - Worcester
    Posts
    3,543

    Default

    So, something is running on the bus that the SIO is connected to...

    Have you got the means to check if pins 35 and 36 go LOW simultaneously (e.g. a logic analyser)?

    This will definitively answer the question as to whether this SIO is actually being addressed or not.

    If you have access to a logic analyser, we can then check to see what is being written to the SIO.

    Did you check the transmitter and receiver clocks, linking RTS to CTS and checking the transmit data pin of the SIO?

    Dave

  3. #23
    Join Date
    Sep 2019
    Location
    Zurich, CH
    Posts
    107

    Default

    Quote Originally Posted by Al Kossow View Post
    J11 ODT expects to talk to a DL11 compatible UART.
    If you don't have one, you're stuck.
    Z80 peripherals aren't compatible.
    If there are no other serial interfaces, they may have glued something onto the Z80 to look like a DL11, but you
    won't know until you dig into the hardware further.
    The other thing I would do is dump any firmware in the system so you can look for clues in the code.
    Right. Good point. But I would assume it is, or a lot of things would become "interesting".
    The Z80 peripherals are not connected directly to the J11, I believe. If they were, there wouldn't be much need for the Z80 CPU. I'd say it sounds similar to what you have on the 11/9x CPU, where you have a Z80 CPU with peripherals, and this whole thing is sitting on the J11 bus, and looks like a DL11.

  4. #24
    Join Date
    Sep 2019
    Location
    Zurich, CH
    Posts
    107

    Default

    Quote Originally Posted by Nick7928 View Post
    Hello,

    Here I have included a photo of all of the boards that came the machine. https://imgur.com/a/MWuPGBA

    The TX and RX lines from the serial port go to the bottom Z80 SIO chips, and I don't know where the TX and RX lines go on the second SIO
    The wires are from your probes? Did you ever have this thing working, or is this the initial attempt?

    I would suspect modem signals is not the "thing", but of course I could be wrong. But, assuming you are probing the serial pins on the SIO, and not seeing any output there, then we are not talking about any problem at the conversion to RS-232 levels, where you otherwise often can have the problem.
    That sortof leaves this to either a broken SIO, or else you are not getting the Tx clock. So again - check the clocks fed into the SIO. If you have the clocks there, and the pins for data and chip select and so on all are "alive", then it's possible the SIO itself has broken. (I hope you have been careful about those wires you put on there...)

  5. #25

    Default

    Thank you so much for the help everyone! I really appreciate it!

    At the moment, I currently have a logic analyzer (The small 8 channel USB ones). Is there any way to test clocks with a logic analyzer?
    I've been really careful about where I put my wires and have made many precautions to make sure that I am not shorting anything.

  6. #26
    Join Date
    Sep 2019
    Location
    Zurich, CH
    Posts
    107

    Default

    Quote Originally Posted by Nick7928 View Post
    Thank you so much for the help everyone! I really appreciate it!

    At the moment, I currently have a logic analyzer (The small 8 channel USB ones). Is there any way to test clocks with a logic analyzer?
    I've been really careful about where I put my wires and have made many precautions to make sure that I am not shorting anything.
    Sure. A logic analyzer is basically just a kind of oscilloscope. Definitely good enough to be usable to check for clocks.
    Exactly how you do it is more a question on how you configure and use your analyzer. Hard for us to tell.

  7. #27
    Join Date
    Jun 2012
    Location
    UK - Worcester
    Posts
    3,543

    Default

    Yes,

    If you have a 24 MHz set of probes connected to your logic analyser (and it is capable of resolving to this) then you should be able to measure up to 2.4 MHz clocks relatively easily (and perhaps a bit higher if you push it).

    You should be able to put your logic analyser into 'single shot manual mode'. Hit the button to collect a trace.

    If you can tell us exactly what you have (manufacturer, make, model etc.) we can have a look at the user manual for it and advise better.

    Dave

  8. #28

    Default

    Hello,

    I have a small USB logic analyzer like the one in this website https://www.sparkfun.com/products/15033

    I am running Sigrok Pulseview on it to collect data

  9. #29

    Default

    My guess would be that this is some sort of remote diagnostic support designed by Atex, where the serial console from the CPU is fed to the Z80 and processed in some manner, possibly including the video controller. In your collection of parts, is there any switch labeled something like "Local / Remote"? If so, you may need to have all of the cables and distribution panels in place to convince the Z80 to output on the "console" port.

    OTOH, it may be that the Z80 inhibits all console output until it detects that the operating system is loaded, so they don't "scare" the users with weird console output. But that would make troubleshooting difficult, so there would probably be some way (switch, jumper, special control character on the console port, etc.) to get the full output.

    This oddness may not be new with their J11-based systems. It is possible that the 11/34 systems had the same sort of thing going on (but in that case it would be on an extra card, since the 11/34 was a complicated enough board that it would be silly to replicate it just to put remote console stuff on the same board).

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •