1. This forum section is a read-only archive which contains old newsgroup posts. If you wish to post a query, please do so in one of our main forum sections (here). This way you will get a faster, better response from the members on Motherboard Point.

2nd Sun 3/80 waking up another beauty

Discussion in 'Sun Hardware' started by David Reichelt, Nov 21, 2008.

  1. hi fellows,

    got an other sun 3/80 in a very fine condition. no scratches, no broken parts.
    and the best, it´s equipped with a cg6 framebuffer ^^

    after cleaning the psu with compressed air, i connected the serial port a with
    my PehCeh running teraterm. even after 20 minutes of running i dont get ANY
    letters on the terminal.

    jumpers for the serial ports are in rs-232 possition, so that can´t by the
    problem. i also tried a type 4 keyboard with a 13w3 monitor. same result here
    :-( even stop + n does not help.

    is there anybody who has an idea how to get this machine working again?

    with best regards,

    david

    p.s. die obp chip is labeled 525-1041-05, where can i find more information
    about this?
     
    David Reichelt, Nov 21, 2008
    #1
    1. Advertising

  2. David Reichelt <> writes:

    >got an other sun 3/80 in a very fine condition. no scratches, no broken parts.
    >and the best, it´s equipped with a cg6 framebuffer ^^


    >after cleaning the psu with compressed air, i connected the serial port a with
    >my PehCeh running teraterm. even after 20 minutes of running i dont get ANY
    >letters on the terminal.


    Do you have a Sun keyboard connected to it?

    It's likely that the NVRAM chip has lost its juice; but I'm not sure what
    the consequences of that are; what's the system doign when you turn it on?

    (I had one 3/80 as my desktop many years ago)

    Casper
    --
    Expressed in this posting are my opinions. They are in no way related
    to opinions held by my employer, Sun Microsystems.
    Statements on Sun products included here are not gospel and may
    be fiction rather than truth.
     
    Casper H.S. Dik, Nov 21, 2008
    #2
    1. Advertising

  3. Casper H.S. Dik schrieb:
    > David Reichelt <> writes:
    >
    >> got an other sun 3/80 in a very fine condition. no scratches, no broken parts.
    >> and the best, it´s equipped with a cg6 framebuffer ^^

    >
    >> after cleaning the psu with compressed air, i connected the serial port a with
    >> my PehCeh running teraterm. even after 20 minutes of running i dont get ANY
    >> letters on the terminal.

    >
    > Do you have a Sun keyboard connected to it?
    >
    > It's likely that the NVRAM chip has lost its juice; but I'm not sure what
    > the consequences of that are; what's the system doign when you turn it on?
    >
    > (I had one 3/80 as my desktop many years ago)
    >
    > Casper



    Hi Casper,

    I tried with and without keyboard(type 4 and 5). afaik, a Sun workstation
    powered up without keyboard goes to diag mod on ttya, right?

    You are right, the nvram is dead, but even with a dead nvram it ~should~ boot to
    the obp, at least to show me that the ethernet adress (mac) is set to FF:FF:FF:FF.

    If I turn the system on, the psu fan starts and a red light in the psu lights
    up. If a keyboard is connected, I hear a beep.

    Is this beep just a keyboard check or the first step of a diag mod?

    Kind regards,

    David
     
    David Reichelt, Nov 21, 2008
    #3
  4. David Reichelt <> writes:

    >I tried with and without keyboard(type 4 and 5). afaik, a Sun workstation
    >powered up without keyboard goes to diag mod on ttya, right?


    If you don't connect a keyboard, it should talk to ttya.

    You could try start the system and use L1-N (Stop-N); use the Stop/L1
    key as you would a shift/control key.

    >You are right, the nvram is dead, but even with a dead nvram it ~should~ boot to
    >the obp, at least to show me that the ethernet adress (mac) is set to FF:FF:FF:FF.


    >If I turn the system on, the psu fan starts and a red light in the psu lights
    >up. If a keyboard is connected, I hear a beep.


    >Is this beep just a keyboard check or the first step of a diag mod?



    That's a long time ao; I don't really know; I seem to remember that if
    something was wrong, it would display a code in the leds on the keyboard.

    Casper
    --
    Expressed in this posting are my opinions. They are in no way related
    to opinions held by my employer, Sun Microsystems.
    Statements on Sun products included here are not gospel and may
    be fiction rather than truth.
     
    Casper H.S. Dik, Nov 21, 2008
    #4
  5. David Reichelt <> wrote:

    > after cleaning the psu with compressed air, i connected the serial port a with
    > my PehCeh running teraterm. even after 20 minutes of running i dont get ANY
    > letters on the terminal.


    Du you have a proper serial cable, ie. one which supports hardware
    handshake?
    >
    > jumpers for the serial ports are in rs-232 possition, so that can´t by the
    > problem.


    Which jumpers, and in what position exactly?
    Have you pulled the graphics card?

    > i also tried a type 4 keyboard with a 13w3 monitor. same result here
    > :-( even stop + n does not help.


    IIRC, Stop-N doesn't do anything useful on a Sun3/xx.
    You may want to check out
    <http://sunsolve.sun.com/handbook_pub/validateUser.do?target=Systems/Sun3/CPU_01_Sun-PROM-Monitor>
    or, even better, Document 800-5027 if you can still find it somewhere.

    > is there anybody who has an idea how to get this machine working again?


    Firstly: Use a proper serial cable. If you don't have one, make at least
    sure that it does at least loopback or partial handshaking.
    Secondly: Pull the framebuffer.

    If it still doesn't work: Come back.

    > p.s. die obp chip is labeled 525-1041-05, where can i find more information
    > about this?


    That's Boot PROM revision 3.0. You should consider upgrading to 3.0.3.


    Martin
    --
    "For the Snark's a peculiar creature, that won't
    Be caught in a commonplace way.
    Do all that you know, and try all that you don't;
    Not a chance must be wasted to-day!"
     
    Martin Etteldorf, Nov 21, 2008
    #5
  6. Martin Etteldorf schrieb:
    > David Reichelt <> wrote:
    >
    >> after cleaning the psu with compressed air, i connected the serial port a with
    >> my PehCeh running teraterm. even after 20 minutes of running i dont get ANY
    >> letters on the terminal.

    >
    > Du you have a proper serial cable, ie. one which supports hardware
    > handshake?


    The wiring looks like

    D-SUB9 D-SUB25
    Ground Ground
    1 8
    2 2
    3 3
    4 6
    5 7
    6 20
    7 5
    8 4
    9 22

    I tested the cable just minutes ago with my other sun 3/80. It runs fine. Is my
    wiring also right in your eyes?

    >> jumpers for the serial ports are in rs-232 possition, so that can´t by the
    >> problem.

    >
    > Which jumpers, and in what position exactly?
    > Have you pulled the graphics card?


    j043 1-2
    j044 1-2
    j045 2-3
    j020 20mhz
    j1000 open

    Framebuffer is not longer in the box, i had to remove it to check all jumper.

    >
    >> i also tried a type 4 keyboard with a 13w3 monitor. same result here
    >> :-( even stop + n does not help.

    >
    > IIRC, Stop-N doesn't do anything useful on a Sun3/xx.
    > You may want to check out
    >

    <http://sunsolve.sun.com/handbook_pub/validateUser.do?target=Systems/Sun3/CPU_01_Sun-PROM-Monitor>
    > or, even better, Document 800-5027 if you can still find it somewhere.


    I have a collection of documentation. I will search for it
    >
    >> is there anybody who has an idea how to get this machine working again?

    >
    > Firstly: Use a proper serial cable. If you don't have one, make at least
    > sure that it does at least loopback or partial handshaking.
    > Secondly: Pull the framebuffer.


    done

    > If it still doesn't work: Come back.
    >
    >> p.s. die obp chip is labeled 525-1041-05, where can i find more information
    >> about this?

    >
    > That's Boot PROM revision 3.0. You should consider upgrading to 3.0.3.


    I would love to, but the mainboard is 08rev50, so if we can believe Heiko Krupp
    (http://www.sun3arc.org/hardpatches/3.80-tuning/80tuning.phtml) I need to
    upgrade to 12rev50 first. Can you confirm that information?
     
    David Reichelt, Nov 21, 2008
    #6
  7. David Reichelt

    DoN. Nichols Guest

    On 2008-11-21, Casper H.S Dik <> wrote:
    > David Reichelt <> writes:
    >
    >>I tried with and without keyboard(type 4 and 5). afaik, a Sun workstation
    >>powered up without keyboard goes to diag mod on ttya, right?

    >
    > If you don't connect a keyboard, it should talk to ttya.
    >
    > You could try start the system and use L1-N (Stop-N); use the Stop/L1
    > key as you would a shift/control key.
    >
    >>You are right, the nvram is dead, but even with a dead nvram it ~should~ boot to
    >>the obp, at least to show me that the ethernet adress (mac) is set to FF:FF:FF:FF.

    >
    >>If I turn the system on, the psu fan starts and a red light in the psu lights
    >>up. If a keyboard is connected, I hear a beep.

    >
    >>Is this beep just a keyboard check or the first step of a diag mod?

    >
    >
    > That's a long time ao; I don't really know; I seem to remember that if
    > something was wrong, it would display a code in the leds on the keyboard.


    And the Sun-3 systems tended to not have the OBP, but rather a
    simpler bit of firmware, which had as the entire prompt a '> ' instead
    of "ok ".

    The hostid in the NVRAM should start with 4200 (with four more
    characters which are the serial number).

    If the start of the hostid is wrong, the system won't know where
    to look for various I/O devices and such. So -- you need to start with
    42, and probably set the "00" to something like "03" so it won't match
    any issued hostid. Yes, the number should be 525-1031-01 according to
    my very old FEH.

    I think that the 3/80 did not understand the L1-A (STOP-A) key
    sequence during boot.

    To examine or change a parameter in the NVRAM, you had to use a
    'q' command followed by the hex address, and type in the new value
    followed by a <CR> to change the value. To exit this mode you type a
    space followed by a <CR>, or a '.' and <CR>

    Beware possible typos in the following.

    The addresses for the parameters were:

    0x14 The amount of installed memory (in hex)

    0x15 The amount of memory to test at boot (in hex)

    0x16 Monitor screen size
    00 1152x900
    12 1024x1024
    13 1600x1280
    14 1440x1440
    15 1024x768

    0x17 Watchdog reset action
    00 Watchdog reset returns to PROM monitor
    12 Watchdog reset causes a Power On Reset (default)

    0x18 Operating system boot device
    00 Polls devices (default)
    12 uses EEPROM/NVRAM specified device

    0x19 - 0x1a SunOs boot device (in ASCII)
    xy 78 79 id 69 64
    xd 78 64 gn 67 6e
    sd 73 64 le 6c 65
    ie 69 65

    0x1b, 0x1c, 0x1d (controller, unit, partition)
    00 00 00 (default)

    0x1f Primary terminal
    00 Monochrome frame buffer
    10 Serial port A
    11 Serial port B
    12 VMEbus and Sun 3/60 P4 colro frame buffers
    (configure locations 0x60c - 0x613 when VX and
    MVX graphics options are installed)
    20 P4 color frame buffer

    0x20 Poser-up banner
    00 Sun Banner
    12 Custom Banner

    0x21 Keyboard click
    00 click is OFF
    12 Click is ON (Default)

    0x22 - 0x23 Diagnostic boot device in ASCII
    (same as 0x19 to save my typing)

    0x24, 0x25, 0x26 (controller unit partition)
    00 00 00 (default)

    0x28 - 0x4f (diagnostic boot path)
    40 bytes in ASCII for path

    0x50 (Hi Res # columns
    50 80 columns
    78 120 columns

    0x51 Hi Res # Rows
    22 34 rows
    30 48 rows

    0x58 Serial port A Default Baud Rate
    00 uses 9600 baud
    12 uses EEPROM/NVRAM defined baud rate

    0x59 - 0x5a Serial Port A baud rate
    1200 baud 04 b0
    4800 baud 12 c0
    9600 haud 25 80

    0x5b Serial port A DTR/RTS
    00 Asserts DTR and RTS signals
    12 Does not assert above

    0x60 Serial port B default baud rate
    00 uses 9600 baud
    12 uses EEPROM/NVRAM defined baud rate

    In the diag position, port B is set to outptu at 1200
    baud. The setting of location 0x60 - 0x62 is ignored

    0x61 - 0x62 Serial port B baud rates (as above in 0x59 - 0x5a

    0x63 Serial prot B DTR/RTS

    as in 0x5b above

    [ skip over 386i only values ]

    0x18f Logo Type
    00 Sun
    06 3D for CG6
    12 Custom


    0x492 Password Mode Select
    5E fully secure mode
    01 Command Secure mode
    All else -- non-secure mode

    0x493 - 0x49a Password bytes

    0x60c VX & MVX options boot code

    0x610 - 0x613 VX & MVX options bus tyep

    0x70b Sun 3/80 power-on mode
    06 normal boot
    12 diagnostic mode
    All else -- full diagnostic boot

    The Open PROM started with the 4/25 apparently.

    Since the locations for the HostID and the MAC address are not
    shown here, you'll have to dig through one of the sites dedicated to
    recovering such data.

    Good Luck,
    DoN.

    --
    Email: <> | Voice (all times): (703) 938-4564
    (too) near Washington D.C. | http://www.d-and-d.com/dnichols/DoN.html
    --- Black Holes are where God is dividing by zero ---
     
    DoN. Nichols, Nov 22, 2008
    #7
  8. DoN. Nichols schrieb:
    > On 2008-11-21, Casper H.S Dik <> wrote:
    >> David Reichelt <> writes:
    >>
    >>> I tried with and without keyboard(type 4 and 5). afaik, a Sun workstation
    >>> powered up without keyboard goes to diag mod on ttya, right?

    >> If you don't connect a keyboard, it should talk to ttya.
    >>
    >> You could try start the system and use L1-N (Stop-N); use the Stop/L1
    >> key as you would a shift/control key.
    >>
    >>> You are right, the nvram is dead, but even with a dead nvram it ~should~ boot to
    >>> the obp, at least to show me that the ethernet adress (mac) is set to FF:FF:FF:FF.
    >>> If I turn the system on, the psu fan starts and a red light in the psu lights
    >>> up. If a keyboard is connected, I hear a beep.
    >>> Is this beep just a keyboard check or the first step of a diag mod?

    >>
    >> That's a long time ao; I don't really know; I seem to remember that if
    >> something was wrong, it would display a code in the leds on the keyboard.

    >
    > And the Sun-3 systems tended to not have the OBP, but rather a
    > simpler bit of firmware, which had as the entire prompt a '> ' instead
    > of "ok ".


    Of course, you are right. I think OBP started with the Sun4 series.

    > The hostid in the NVRAM should start with 4200 (with four more
    > characters which are the serial number).
    >
    > If the start of the hostid is wrong, the system won't know where
    > to look for various I/O devices and such. So -- you need to start with
    > 42, and probably set the "00" to something like "03" so it won't match
    > any issued hostid. Yes, the number should be 525-1031-01 according to
    > my very old FEH.


    If the hostid is wrong(e.g. because the nvram is dead), will the system start up
    anyway? Or do i have the program the nvram in an other 3/80?

    > I think that the 3/80 did not understand the L1-A (STOP-A) key
    > sequence during boot.
    >


    Now I know, that you right again with this point.

    > To examine or change a parameter in the NVRAM, you had to use a
    > 'q' command followed by the hex address, and type in the new value
    > followed by a <CR> to change the value. To exit this mode you type a
    > space followed by a <CR>, or a '.' and <CR>
    >
    > Beware possible typos in the following.
    >
    > The addresses for the parameters were:
    >
    > 0x14 The amount of installed memory (in hex)
    >
    > 0x15 The amount of memory to test at boot (in hex)
    >
    > 0x16 Monitor screen size
    > 00 1152x900
    > 12 1024x1024
    > 13 1600x1280
    > 14 1440x1440
    > 15 1024x768
    >
    > 0x17 Watchdog reset action
    > 00 Watchdog reset returns to PROM monitor
    > 12 Watchdog reset causes a Power On Reset (default)
    >
    > 0x18 Operating system boot device
    > 00 Polls devices (default)
    > 12 uses EEPROM/NVRAM specified device
    >
    > 0x19 - 0x1a SunOs boot device (in ASCII)
    > xy 78 79 id 69 64
    > xd 78 64 gn 67 6e
    > sd 73 64 le 6c 65
    > ie 69 65
    >
    > 0x1b, 0x1c, 0x1d (controller, unit, partition)
    > 00 00 00 (default)
    >
    > 0x1f Primary terminal
    > 00 Monochrome frame buffer
    > 10 Serial port A
    > 11 Serial port B
    > 12 VMEbus and Sun 3/60 P4 colro frame buffers
    > (configure locations 0x60c - 0x613 when VX and
    > MVX graphics options are installed)
    > 20 P4 color frame buffer
    >
    > 0x20 Poser-up banner
    > 00 Sun Banner
    > 12 Custom Banner
    >
    > 0x21 Keyboard click
    > 00 click is OFF
    > 12 Click is ON (Default)
    >
    > 0x22 - 0x23 Diagnostic boot device in ASCII
    > (same as 0x19 to save my typing)
    >
    > 0x24, 0x25, 0x26 (controller unit partition)
    > 00 00 00 (default)
    >
    > 0x28 - 0x4f (diagnostic boot path)
    > 40 bytes in ASCII for path
    >
    > 0x50 (Hi Res # columns
    > 50 80 columns
    > 78 120 columns
    >
    > 0x51 Hi Res # Rows
    > 22 34 rows
    > 30 48 rows
    >
    > 0x58 Serial port A Default Baud Rate
    > 00 uses 9600 baud
    > 12 uses EEPROM/NVRAM defined baud rate
    >
    > 0x59 - 0x5a Serial Port A baud rate
    > 1200 baud 04 b0
    > 4800 baud 12 c0
    > 9600 haud 25 80
    >
    > 0x5b Serial port A DTR/RTS
    > 00 Asserts DTR and RTS signals
    > 12 Does not assert above
    >
    > 0x60 Serial port B default baud rate
    > 00 uses 9600 baud
    > 12 uses EEPROM/NVRAM defined baud rate
    >
    > In the diag position, port B is set to outptu at 1200
    > baud. The setting of location 0x60 - 0x62 is ignored
    >
    > 0x61 - 0x62 Serial port B baud rates (as above in 0x59 - 0x5a
    >
    > 0x63 Serial prot B DTR/RTS
    >
    > as in 0x5b above
    >
    > [ skip over 386i only values ]
    >
    > 0x18f Logo Type
    > 00 Sun
    > 06 3D for CG6
    > 12 Custom
    >
    >
    > 0x492 Password Mode Select
    > 5E fully secure mode
    > 01 Command Secure mode
    > All else -- non-secure mode
    >
    > 0x493 - 0x49a Password bytes
    >
    > 0x60c VX & MVX options boot code
    >
    > 0x610 - 0x613 VX & MVX options bus tyep
    >
    > 0x70b Sun 3/80 power-on mode
    > 06 normal boot
    > 12 diagnostic mode
    > All else -- full diagnostic boot
    >
    > The Open PROM started with the 4/25 apparently.
    >
    > Since the locations for the HostID and the MAC address are not
    > shown here, you'll have to dig through one of the sites dedicated to
    > recovering such data.


    At the monent i get any response from the system, the information above will to
    prove itself more valuable then gold(or oil ;-), but i still have no response
    from the box at all.

    >
    > Good Luck,
    > DoN.
    >


    Thx DoN, i will stay on mission :)
     
    David Reichelt, Nov 22, 2008
    #8
  9. David Reichelt

    Winston Guest

    David Reichelt <> writes:
    > after cleaning the psu with compressed air, i connected the serial port a
    > with my PehCeh running teraterm. even after 20 minutes of running i dont
    > get ANY letters on the terminal.


    Your PeeCee/teraterm is configured for 9600, even parity, and for hardware
    flow control, I hope? If you still get nothing, try switching your serial
    cable from straight-through to crossover, or vice versa.
    -WBE
     
    Winston, Nov 22, 2008
    #9
  10. I never had a 3/80, but IIRC the 3/60 had a few LEDs next to the diag
    switch on the back which would tell you a little about startup failures.
     
    Reginald Beardsley, Nov 22, 2008
    #10
  11. Winston schrieb:
    > David Reichelt <> writes:
    >> after cleaning the psu with compressed air, i connected the serial port a
    >> with my PehCeh running teraterm. even after 20 minutes of running i dont
    >> get ANY letters on the terminal.

    >
    > Your PeeCee/teraterm is configured for 9600, even parity, and for hardware
    > flow control, I hope? If you still get nothing, try switching your serial
    > cable from straight-through to crossover, or vice versa.
    > -WBE

    Hi Winston,

    Teraterm is configured like you said, serial cable is crossover. An other 3/80
    talks to me viva this cable right now, but i also tried straight-through,
    nothing happens :-(

    Best regards,
    David
     
    David Reichelt, Nov 22, 2008
    #11
  12. David Reichelt schrieb:
    > Martin Etteldorf schrieb:
    > > David Reichelt <> wrote:
    > >
    > >> after cleaning the psu with compressed air, i connected the serial

    > port a with
    > >> my PehCeh running teraterm. even after 20 minutes of running i dont

    > get ANY
    > >> letters on the terminal.

    > >
    > > Du you have a proper serial cable, ie. one which supports hardware
    > > handshake?

    >
    > The wiring looks like
    >
    > D-SUB9 D-SUB25
    > Ground Ground
    > 1 8
    > 2 2
    > 3 3
    > 4 6
    > 5 7
    > 6 20
    > 7 5
    > 8 4
    > 9 22
    >
    > I tested the cable just minutes ago with my other sun 3/80. It runs
    > fine. Is my wiring also right in your eyes?
    >
    > >> jumpers for the serial ports are in rs-232 possition, so that can´t

    > by the
    > >> problem.

    > >
    > > Which jumpers, and in what position exactly?
    > > Have you pulled the graphics card?

    >
    > j043 1-2
    > j044 1-2
    > j045 2-3
    > j020 20mhz
    > j1000 open
    >
    > Framebuffer is not longer in the box, i had to remove it to check all
    > jumper.
    >
    > >
    > >> i also tried a type 4 keyboard with a 13w3 monitor. same result here
    > >> :-( even stop + n does not help.

    > >
    > > IIRC, Stop-N doesn't do anything useful on a Sun3/xx.
    > > You may want to check out
    > >

    >

    <http://sunsolve.sun.com/handbook_pub/validateUser.do?target=Systems/Sun3/CPU_01_Sun-PROM-Monitor>

    >
    > > or, even better, Document 800-5027 if you can still find it somewhere.

    >
    > I have a collection of documentation. I will search for it
    > >
    > >> is there anybody who has an idea how to get this machine working again?

    > >
    > > Firstly: Use a proper serial cable. If you don't have one, make at least
    > > sure that it does at least loopback or partial handshaking.
    > > Secondly: Pull the framebuffer.

    >
    > done
    >
    > > If it still doesn't work: Come back.
    > >
    > >> p.s. die obp chip is labeled 525-1041-05, where can i find more

    > information
    > >> about this?

    > >
    > > That's Boot PROM revision 3.0. You should consider upgrading to 3.0.3.

    >
    > I would love to, but the mainboard is 08rev50, so if we can believe
    > Heiko Krupp
    > (http://www.sun3arc.org/hardpatches/3.80-tuning/80tuning.phtml) I need
    > to upgrade to 12rev50 first. Can you confirm that information?



    Today a had some more time, so i replaced everything in the box with working
    parts. Here are the results:

    PSU can´t be the problem, even with new psu the system does not come up. The PSU
    is the box is able to start a SS2.

    The CG6 Framebuffer works in me other 3/80 :))) really nice to see SUNVIEW with
    colors. BTW it´s not that slow ^^ with 16MB RAM, u can work fast!

    I also changed the PROM and the NVRAM chip. Again nothing happens. Both chips
    run in my other sun 3/80. NVRAM is dead(we knew it).

    So the problem has to be on the systemboard. I did not change the ram and the
    CPU etc. But I checked if everything fits right in it´s place.

    Would it be worth a try to switch the cpu? i have a third board here around,
    wish went up in flames two years ago. I am a litte bit afraid to do that,
    because i don´t really know how bad the third board is damaged.
     
    David Reichelt, Nov 22, 2008
    #12
  13. Reginald Beardsley schrieb:
    > I never had a 3/80, but IIRC the 3/60 had a few LEDs next to the diag
    > switch on the back which would tell you a little about startup failures.


    sadly, the sun 3/80 does not have any diag LEDs. but maybe there is an other way
    to get some diag information.
     
    David Reichelt, Nov 22, 2008
    #13
  14. David Reichelt

    DoN. Nichols Guest

    On 2008-11-22, David Reichelt <> wrote:
    > DoN. Nichols schrieb:
    >> On 2008-11-21, Casper H.S Dik <> wrote:


    [ ... ]

    >>> That's a long time ao; I don't really know; I seem to remember that if
    >>> something was wrong, it would display a code in the leds on the keyboard.

    >>
    >> And the Sun-3 systems tended to not have the OBP, but rather a
    >> simpler bit of firmware, which had as the entire prompt a '> ' instead
    >> of "ok ".

    >
    > Of course, you are right. I think OBP started with the Sun4 series.
    >
    >> The hostid in the NVRAM should start with 4200 (with four more
    >> characters which are the serial number).
    >>
    >> If the start of the hostid is wrong, the system won't know where
    >> to look for various I/O devices and such. So -- you need to start with
    >> 42, and probably set the "00" to something like "03" so it won't match
    >> any issued hostid. Yes, the number should be 525-1031-01 according to
    >> my very old FEH.

    >
    > If the hostid is wrong(e.g. because the nvram is dead), will the system start up
    > anyway?


    I'm not sure. I've never had my hands on a 3/80 long enough to
    do anything with it. I just almost bought one at a hamfest once. :)

    > Or do i have the program the nvram in an other 3/80?


    If it won't start with a bad NVRAM, how will you program it in
    the other system, since you will have to remove the working NVRAM from
    it to put the questionable one in.

    Now -- what you *could* do as a test is to remove the NVRAM from
    the working system and put it in the "new" one, and see whether it makes
    any difference. If so -- then working towards fixing the NVRAM is worth
    while.

    Note that some of the parameters and addresses which I copied
    suggest that TTYB is used as the console in certain conditions -- and
    one of three likely baud rates. (IIRC, 0x00 as a speed parameter
    produced no valid baud rate -- or at least not a documented one.)

    If you can fix it with a good NVRAM, what *I* would do is take
    the time to wire-wrap up a device to allow reading and programming the
    NVRAM out of the 3/80. Then, you could start by noting down the hostid
    and MAC address on the good one, then dumping the entire contents to a
    file, and identifying which addresses contained those two parameters
    (converting to hex of comparison, of course). Then take the file, copy
    it into the other NVRAM, (after surgery to give it a new external
    battery), and change the hostid and MAC address just enough to avoid
    conflicts. For the MAC address, the general advice used to be to get a
    cheap old PC ethernet card, steel the MAC address from it, and crush the
    bipolar PROM which holds the address in the card, so you can never have
    a duplicate address. Not sure how easy it is to do that these days,
    unless you have a cache of old PC ethernet cards, or no someone else who
    has them. :)

    Or -- you *could* contact Sun with the barcode from the existing
    NVRAM, and pay them to supply you with a new one programmed with the
    original parameters (determined from the barcode). Again, I would not
    do this until I had verified that the system worked with the other NVRAM
    installed.

    [ ... ]

    >> To examine or change a parameter in the NVRAM, you had to use a
    >> 'q' command followed by the hex address, and type in the new value
    >> followed by a <CR> to change the value. To exit this mode you type a
    >> space followed by a <CR>, or a '.' and <CR>


    [ ... ]

    >> 0x58 Serial port A Default Baud Rate
    >> 00 uses 9600 baud
    >> 12 uses EEPROM/NVRAM defined baud rate
    >>
    >> 0x59 - 0x5a Serial Port A baud rate
    >> 1200 baud 04 b0
    >> 4800 baud 12 c0
    >> 9600 haud 25 80
    >>
    >> 0x5b Serial port A DTR/RTS
    >> 00 Asserts DTR and RTS signals
    >> 12 Does not assert above
    >>
    >> 0x60 Serial port B default baud rate
    >> 00 uses 9600 baud
    >> 12 uses EEPROM/NVRAM defined baud rate
    >>
    >> In the diag position, port B is set to output at 1200
    >> baud. The setting of location 0x60 - 0x62 is ignored


    This suggests that if you can set the diag position, you will be
    using TTYB as the console at 1200 baud.

    >> 0x61 - 0x62 Serial port B baud rates (as above in 0x59 - 0x5a
    >>
    >> 0x63 Serial prot B DTR/RTS
    >>
    >> as in 0x5b above


    [ ... ]

    >> Since the locations for the HostID and the MAC address are not
    >> shown here, you'll have to dig through one of the sites dedicated to
    >> recovering such data.

    >
    > At the monent i get any response from the system, the information above will to
    > prove itself more valuable then gold(or oil ;-), but i still have no response
    > from the box at all.
    >
    >>
    >> Good Luck,
    >> DoN.
    >>

    >
    > Thx DoN, i will stay on mission :)


    Again -- good luck.
    DoN.

    --
    Email: <> | Voice (all times): (703) 938-4564
    (too) near Washington D.C. | http://www.d-and-d.com/dnichols/DoN.html
    --- Black Holes are where God is dividing by zero ---
     
    DoN. Nichols, Nov 23, 2008
    #14
  15. David Reichelt <> wrote:

    > The CG6 Framebuffer works in me other 3/80 :)))


    You may want to test the X11R6.8 I built some two years ago:
    http://www.sun3arc.org/precompiled/X11R6.8.tar.gz

    I never had a chance to test it on a CG6-equipped 3/80.



    Martin
    --
    "For the Snark's a peculiar creature, that won't
    Be caught in a commonplace way.
    Do all that you know, and try all that you don't;
    Not a chance must be wasted to-day!"
     
    Martin Etteldorf, Nov 24, 2008
    #15
  16. Martin Etteldorf schrieb:
    > David Reichelt <> wrote:
    >
    >> The CG6 Framebuffer works in me other 3/80 :)))

    >
    > You may want to test the X11R6.8 I built some two years ago:
    > http://www.sun3arc.org/precompiled/X11R6.8.tar.gz
    >
    > I never had a chance to test it on a CG6-equipped 3/80.
    >
    >
    >
    > Martin

    Martin Etteldorf schrieb:
    > David Reichelt <> wrote:
    >
    >> The CG6 Framebuffer works in me other 3/80 :)))

    >
    > You may want to test the X11R6.8 I built some two years ago:
    > http://www.sun3arc.org/precompiled/X11R6.8.tar.gz
    >
    > I never had a chance to test it on a CG6-equipped 3/80.
    >


    So the CG6 is quite rare, right? the man who gave it in my handy told me, that
    the model code is FGX-8V4 and that he bought it 1989 for around 40.000 DM.

    I will text your build :)

    Do you have any ideas how to make some further diags on the other systemboard?
    I did some additional testing this weekend(see above).
     
    David Reichelt, Nov 24, 2008
    #16
  17. David Reichelt <> wrote:
    > Martin Etteldorf schrieb:
    > > David Reichelt <> wrote:
    > >
    > >> The CG6 Framebuffer works in me other 3/80 :)))

    > >
    > > You may want to test the X11R6.8 I built some two years ago:
    > > http://www.sun3arc.org/precompiled/X11R6.8.tar.gz
    > >
    > > I never had a chance to test it on a CG6-equipped 3/80.
    > >
    > >
    > >
    > > Martin

    > Martin Etteldorf schrieb:
    > > David Reichelt <> wrote:
    > >
    > >> The CG6 Framebuffer works in me other 3/80 :)))

    > >
    > > You may want to test the X11R6.8 I built some two years ago:
    > > http://www.sun3arc.org/precompiled/X11R6.8.tar.gz
    > >
    > > I never had a chance to test it on a CG6-equipped 3/80.
    > >


    Methinks, something ain't quite right with your quoting.

    > So the CG6 is quite rare, right?


    Not more so than the 3/80 itself. The CG6 was quite popular,
    especially in the last years of the machine's existence.

    > the man who gave it in my handy told me, that
    > the model code is FGX-8V4 and that he bought it 1989 for around 40.000 DM.


    That's a bit exaggerated. The list prices when these systems came out
    were:
    3/80-FM 15.470.-DM (Mono Framebuffer, MG4, Option 501-1402)
    3/80-FC 24.500.-DM (Color Framebuffer, CG4, Option 501-1443)
    3/80-FGX 33.530.-DM (Color Framebuffer, CG6, Option 501-1505)

    You of course didn't by them for the list price. Discounts of at least
    25% were common, and sometimes these machines were even thrown in as a
    christmas present when you bought a certain amount of some more serious
    hardware.

    > I will text your build :)
    >
    > Do you have any ideas how to make some further diags on the other systemboard?


    You mean, like, without an oscilloscope and some measuring equipment?
    Difficult.
    I'd start pulling and re-seating every socketed chip, including the
    memory modules. Clean the contacts where it deems necessary. Check
    for cold soldering spots and disconnected patch wires.



    Martin
    --
    "For the Snark's a peculiar creature, that won't
    Be caught in a commonplace way.
    Do all that you know, and try all that you don't;
    Not a chance must be wasted to-day!"
     
    Martin Etteldorf, Nov 25, 2008
    #17
  18. Martin Etteldorf schrieb:
    > David Reichelt <> wrote:
    >> Martin Etteldorf schrieb:
    >>> David Reichelt <> wrote:
    >>>
    >>>> The CG6 Framebuffer works in me other 3/80 :)))
    >>> You may want to test the X11R6.8 I built some two years ago:
    >>> http://www.sun3arc.org/precompiled/X11R6.8.tar.gz
    >>>
    >>> I never had a chance to test it on a CG6-equipped 3/80.
    >>>
    >>>
    >>>
    >>> Martin

    >> Martin Etteldorf schrieb:
    >>> David Reichelt <> wrote:
    >>>
    >>>> The CG6 Framebuffer works in me other 3/80 :)))
    >>> You may want to test the X11R6.8 I built some two years ago:
    >>> http://www.sun3arc.org/precompiled/X11R6.8.tar.gz
    >>>
    >>> I never had a chance to test it on a CG6-equipped 3/80.
    >>>

    >
    > Methinks, something ain't quite right with your quoting.



    I promise improvement

    >> So the CG6 is quite rare, right?

    >
    > Not more so than the 3/80 itself. The CG6 was quite popular,
    > especially in the last years of the machine's existence.
    >
    >> the man who gave it in my handy told me, that
    >> the model code is FGX-8V4 and that he bought it 1989 for around 40.000 DM.

    >
    > That's a bit exaggerated. The list prices when these systems came out
    > were:
    > 3/80-FM 15.470.-DM (Mono Framebuffer, MG4, Option 501-1402)
    > 3/80-FC 24.500.-DM (Color Framebuffer, CG4, Option 501-1443)
    > 3/80-FGX 33.530.-DM (Color Framebuffer, CG6, Option 501-1505)


    Always nice to get the right facts :) You dont have a printed pricelist to
    collect, do you? Even a electronic copy would be nice :)

    > You of course didn't by them for the list price. Discounts of at least
    > 25% were common, and sometimes these machines were even thrown in as a
    > christmas present when you bought a certain amount of some more serious
    > hardware.


    I heard about that. That must have been apply especial universities.

    >> I will text your build :)
    >>
    >> Do you have any ideas how to make some further diags on the other systemboard?

    >
    > You mean, like, without an oscilloscope and some measuring equipment?


    I dont have the knowledge for this part of the job. but i know a real geek:

    http://john.ccac.rwth-aachen.de:8000/alf/
    http://john.ccac.rwth-aachen.de:8000/misc/sun3_60.html

    I will ask him about the amount of time and how much the parts costed

    > Difficult.
    > I'd start pulling and re-seating every socketed chip, including the
    > memory modules. Clean the contacts where it deems necessary. Check
    > for cold soldering spots and disconnected patch wires.


    That´s a good advice! The shipping could have done some damage. I will check
    everything carefully, in the hope not do make some new defect!

    After this work is done, i will report!

    With best regards,
    David
     
    David Reichelt, Nov 25, 2008
    #18
  19. David Reichelt

    DoN. Nichols Guest

    On 2008-11-25, David Reichelt <> wrote:
    > Martin Etteldorf schrieb:


    [ ... ]

    >> Difficult.
    >> I'd start pulling and re-seating every socketed chip, including the
    >> memory modules. Clean the contacts where it deems necessary. Check
    >> for cold soldering spots and disconnected patch wires.

    >
    > That´s a good advice! The shipping could have done some damage. I will check
    > everything carefully, in the hope not do make some new defect!


    O.K. I case you don't already know this, you should ideally
    wear a grounding wrist strap to prevent static discharge while you're
    handling these components. Have something static free to set them down
    on if you remove them fully from the machine. Typically -- black or
    pink foam plastics are good for the purpose, or the semi-metalized bags
    that disk drives and the like come in.

    In the absence of an official grounding strap, you can put some
    wire around your wrist, connect it through about a 1 megohm resistor to
    the metal chassis of the computer, and then you won't develop a static
    charge relative to the computer.

    Or -- if you are really careful, you can do this by always
    touching one hand to the chassis (and holding it there) before and while
    touching the chip or card with the other. This takes care, but it can
    work well if you are careful to do it every time you check something.
    Remember that we are getting to the time of the year when static is more
    of a problem (unless you are in Australia or somewhere else in the
    Southern hemisphere -- unlikely, but possible with a remote connection
    for your e-mail.

    You can test the system with only some of the memory present. I
    forget whether the 3/80 wanted memory added in blocks of four or would
    accept them in blocks of only two. But this could help you to eliminate
    problems caused by a bad SIMM locking up the system bus.

    Good Luck,
    DoN.

    --
    Email: <> | Voice (all times): (703) 938-4564
    (too) near Washington D.C. | http://www.d-and-d.com/dnichols/DoN.html
    --- Black Holes are where God is dividing by zero ---
     
    DoN. Nichols, Nov 26, 2008
    #19
  20. DoN. Nichols schrieb:
    > On 2008-11-25, David Reichelt <> wrote:
    >> Martin Etteldorf schrieb:

    >
    > [ ... ]
    >
    >>> Difficult.
    >>> I'd start pulling and re-seating every socketed chip, including the
    >>> memory modules. Clean the contacts where it deems necessary. Check
    >>> for cold soldering spots and disconnected patch wires.

    >> That´s a good advice! The shipping could have done some damage. I will check
    >> everything carefully, in the hope not do make some new defect!

    >
    > O.K. I case you don't already know this, you should ideally
    > wear a grounding wrist strap to prevent static discharge while you're
    > handling these components. Have something static free to set them down
    > on if you remove them fully from the machine. Typically -- black or
    > pink foam plastics are good for the purpose, or the semi-metalized bags
    > that disk drives and the like come in.
    >
    > In the absence of an official grounding strap, you can put some
    > wire around your wrist, connect it through about a 1 megohm resistor to
    > the metal chassis of the computer, and then you won't develop a static
    > charge relative to the computer.
    >
    > Or -- if you are really careful, you can do this by always
    > touching one hand to the chassis (and holding it there) before and while
    > touching the chip or card with the other. This takes care, but it can
    > work well if you are careful to do it every time you check something.
    > Remember that we are getting to the time of the year when static is more
    > of a problem (unless you are in Australia or somewhere else in the
    > Southern hemisphere -- unlikely, but possible with a remote connection
    > for your e-mail.
    >
    > You can test the system with only some of the memory present. I
    > forget whether the 3/80 wanted memory added in blocks of four or would
    > accept them in blocks of only two. But this could help you to eliminate
    > problems caused by a bad SIMM locking up the system bus.
    >
    > Good Luck,
    > DoN.
    >


    Thx for the hints DoN.
    I had some time to do the job. i checked everything twice, but again nothing
    happens :-(

    Today a had some time to think again while sitting in a trafficjam.

    The idea is, that i have a 3rd sun 3/80 board here around. It went up in flames,
    when i tried to power it up, but this was years ago. I think i posted the
    "event" here...

    Today I read the following lines:
    http://www.nabble.com/Sun-3-80-magic-smoke-td17169022.html

    Man, what a luck! Last week I went to the dump, I even had the board in my car,
    but then saved it. ^^

    The only part that is visual destroy is a cap(C17119). So if I am real lucker, I
    am able to fix the 3rd board.

    Cross your fingers.
     
    David Reichelt, Dec 2, 2008
    #20
    1. Advertising

Want to reply to this thread or ask your own question?

It takes just 2 minutes to sign up (and it's free!). Just click the sign up button to choose a username and then you can ask your own questions on the forum.
Similar Threads
  1. *Vanguard*
    Replies:
    11
    Views:
    622
    *Vanguard*
    Jun 18, 2004
  2. Replies:
    2
    Views:
    397
  3. J Cod

    seductive beauty of screens ?

    J Cod, Jul 3, 2005, in forum: Laptops
    Replies:
    0
    Views:
    356
    J Cod
    Jul 3, 2005
  4. Nikolaj Johansen
    Replies:
    0
    Views:
    819
    Nikolaj Johansen
    Dec 12, 2003
  5. Suze
    Replies:
    4
    Views:
    1,107
    dorayme
    Aug 28, 2011
Loading...

Share This Page