Image Map Image Map
Page 1 of 7 12345 ... LastLast
Results 1 to 10 of 62

Thread: pdp-8/e and RX01 restoration.

  1. #1

    Default pdp-8/e and RX01 restoration.

    Hi
    I am restoring a pdp-8/e plus a RX01.
    The 8/e is basically running.
    It will execute the normal toggle in tests.
    Serial I/O works and does the echo test ok.

    Adding the RX01 and its RX8 Controller has thrown up some issues.
    The dip switches on the RX8 are set to decode 75 the standard device address.
    The device address does not decode as it should because BUS I/O PAUSE L is not asserted.
    I/O instructions for example 6751 should have the 6 decoded by the CPU on the Timing Board.
    This should give rise to BUS I/O PAUSE L

    So moving to the timing board
    I/O PAUSE L is produced in E7.
    All signals are present with the exception of I/O TIME.
    I/O time comes from the I/O time flip flop.
    This has FETCH on the D input and a signal from the inverted side of the bi-stable that generates STROBE, connected to the clock input.
    FETCH is present but nothing seems to be happening on the clock input.
    No clock, so whats on D won't get transferred to the output.
    Pin 5 on the flip flop should be high to enable I/O PAUSE L.
    What I don't understand is if there is no STROBE how can memory be working?

    Comments Please

  2. #2

    Default

    If you look at Figure 3-19 in the 8/e Maintenance Manual Volume 1, it shows separate Memory Timing and Processor Timing outputs. It is quite possible that most of the timing signals are working, but the IOT signal timing is not. When an IOT instruction is decoded the timing circuit behavior changes. You can see the timing signal signal differences in the two Timing charts in the 8/e Engineering Drawings. Section 3.21 describes the I/O Timing Control. You can see the STROBE signal on figure 3-30 that clocks the I/O TIME (1) FF E06.

    I am not an expert in interpreting DEC schematics, but I will give it a try.
    For the I/O TIME (1) FF E06 to clock, its C pin 3 is driven by E05 pin 6 and needs to transition from L to H.
    That means that pins 1, 2, 4, & 5 on E05 all need to be H, and then one needs to go L.
    E34 pin 14 is STROBE CLEAR and drives E05 pin 1. My guess is that it goes L to H.
    E05 pin 2 looks like a power on clear?
    E05 pins 4 & 5 are driven by E05 pin 8, STROBE H.
    E05 pins 9, 10, & 12 are driven by E05 pin 6.
    E05 pin 13 is driven by E39 pin 13, TP1 L.

    I would look at the inputs to both halves of E05 and STROBE H to see what they are doing.
    Member of the Rhode Island Computer Museum
    http://www.ricomputermuseum.org

  3. #3

    Default

    Panelman cleaned the gold fingers on the processor boards and the backplane connectors, and now the RX8E controller board will decode IOT instructions.
    Executing the RX8E INIT instruction or pressing the CLEAR switch will cause both disks to recalibrate.
    Member of the Rhode Island Computer Museum
    http://www.ricomputermuseum.org

  4. #4

    Default

    Quote Originally Posted by m_thompson View Post
    Panelman cleaned the gold fingers on the processor boards and the backplane connectors, and now the RX8E controller board will decode IOT instructions.
    Executing the RX8E INIT instruction or pressing the CLEAR switch will cause both disks to recalibrate.
    Only just returned to tackling this one.
    Wife came home on Wednesday from hospital with failed old hip repair replaced with a new hip.
    She's one determined lady and has been doing the exercises and is already somewhat independent.

    As ever you try to eliminate things.
    I have an RX02 in good order and have set the switches to be an RX01 working with an RX8.
    I'm just making sure the diagnostics have remained intact in core.
    Ill see how far I get with that.

    I'm expecting to get anther M8330 mid next week.

  5. #5

    Default

    If you had another M8650 you could run serialdisk and boot OS/8 from an emulated disk drive, actually four emulated drives. The performance is quite good, and probably as fast as the RX01. There are batch files on the RK05 diag pack to build bootable RX01s.
    Member of the Rhode Island Computer Museum
    http://www.ricomputermuseum.org

  6. #6

    Default

    Quote Originally Posted by m_thompson View Post
    If you had another M8650 you could run serialdisk and boot OS/8 from an emulated disk drive, actually four emulated drives. The performance is quite good, and probably as fast as the RX01. There are batch files on the RK05 diag pack to build bootable RX01s.
    OK.. A few things have changed. I got as far as the RX01 attempting to boot and probably only lacking a bootable flloppty

    I then had a look at SerialDisk It needs Linux or Cygwin under windows plus talk of compiling. Not my area of expertise.

    Next onto dumprest yep once again C sources etc.

    I have neither compilers or skills in C so that eliminates those two.

    On top of that I have been able to establish that the RIM loader works just fine but having tried several RIM format BIN loaders and found they appear to load but will never load anything themselves.

    So at t he moment its RIM format programs only.

    I did manage to load and run a RIM format FOCAL language and had it running for a while.
    Keyboard I/O and program execution worked.

    So I'm adding back memory 4K at a time at the moment.

    R

  7. #7

    Default

    Does anybody know how the EMA links should set on 4 x 4K G104 boards
    For example
    EMA
    0 1 2

    Field 0 IN IN IN

    Field 1 ? ? ?

    Field 2 ? ? ?

    Field 3 ? ? ?

  8. #8

    Default

    I didn't find a chart for the jumpers in any of the documentation.

    The field selection circuit is show in figure 3-51 in the PDP-8e Maintenance Manual Volume 1

    If I understand how the circuit works, to select the core for Field 0, EMA0, EMA1, and EMA2 will all be high.
    Pins 13, 14, and 2 of E1 will all be high.
    The input pins 9, 10, 12, and 13 of E7 need to be high to have pin 8 low to have FIELD H signal from pin 6 of E7 high.
    For the output pins of E1 and E6 to be high we need one input high and one input low.
    We have one input pin high from the signals on E1, so the other input needs to be low.
    My guess is that for Field 0 all three EMA jumpers are in.

    To select the core for Field 1, EMA0 and EMA1 will be high, and EMA2 will all be low.
    In this case pins 13 and 14 of E1 are high, and pin 2 is low.
    For the output pins of E1 and E6 to be high we need one input high and one input low.
    My guess is that for Field 1 the EMA2 jumper is out.

    It looks like the EMA* jumper is removed when that EMA signal is active.

    Field EMA0 EMA1 EMA2
    0 IN IN IN
    1 IN IN OUT
    2 IN OUT IN
    3 IN OUT OUT
    Member of the Rhode Island Computer Museum
    http://www.ricomputermuseum.org

  9. #9

    Default

    Thanks
    I'll set the jumpers as per your table and see if I can get a memory test to run.

    Toggling in the RIM loader and using it to load RIM programs seems to work but once loaded the programs are not always stable.
    The binary loader doesn't load anything. I am not sure that using Hyperterm in text upload mode is the right way to try to do it.

    I really need something that simulates the ASR 33 tape reader closely.

    I may be reduced to wading about in a C of existing code.

    Rod

  10. #10
    Join Date
    Aug 2010
    Location
    Silicon Valley USA
    Posts
    632
    Blog Entries
    4

    Default

    Quote Originally Posted by Panelman View Post
    Toggling in the RIM loader and using it to load RIM programs seems to work but once loaded the programs are not always stable.
    The binary loader doesn't load anything. I am not sure that using Hyperterm in text upload mode is the right way to try to do it.
    Hyperterm in text upload mode probably is not going to work very well. I use TeraTerm, switching between 7b mode (for text) and 8b mode (for binary download) when necessary. Binary file download, with an added 1ms per character delay at 9600baud, works reliably for me on my 8m. Both in sending files thru the RIM loader (which gets loaded into memory thru my M847 diode bootloader card) and downloading the binary loader and then sending BIN format binaries. No need for any custom programming, unless you really want to.

    FYI I have loadable RIM and BIN binaries and listings of most all key 8e/f/m DEC diagnostics here: https://www.ak6dn.com/PDP-8/MAINDEC/

    Don
    Last edited by AK6DN; August 25th, 2017 at 02:38 PM.

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
  •