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

ELF II in a coma....

Expand Messages
  • 64536
    I recently dug out my ELF II for some tinerking. I have the 4K card, the Giant Board, the ROM BASIC, and an EPROM Programmer, and the ASCII keyboard. But it
    Message 1 of 12 , Aug 20, 2007
    • 0 Attachment
      I recently dug out my ELF II for some tinerking. I have the 4K card,
      the Giant Board, the ROM BASIC, and an EPROM Programmer, and the ASCII
      keyboard. But it seems to not accept hex input - I'm just trying it
      out with the 4K card and the giant board. I've tried different jumper
      combinations on the giant card (including J12)- I remember how kludgey
      that was from my high school days - I have all the jumpers connected
      to a little dipswitch control panel to make it easier. I've traced
      the keyboard signals and input key signals all the way back to the
      1802. So I replaced it and it still doesn't work- just shows "00" in
      the display. It may be some strange jumper setting/combination - but
      I've followed the old manuals to no avail.

      Notes in my manual show I took it out and tested it (and it worked)
      back in 2002. Something must have happened while it was sitting on a
      shelf...

      Any ideas on how I can wake the sleeping beauty?
    • Al W,
      John, It might be better to start the troubleshooting by removing the more complicated items such as the 4k card and the giant board leaving a basic 256 byte
      Message 2 of 12 , Aug 20, 2007
      • 0 Attachment
        John,

        It might be better to start the troubleshooting by removing the more complicated items
        such as the 4k card and the giant board leaving a basic 256 byte Elf II. This would
        eliminate them as a possible cause of the problem and make diagnostics easier.

        Common shelf aging problems are oxidation of switches, oxidation of IC sockets and
        memory loss of the operator. ;o)

        An easy thing to try is to gently raise the socketed IC's one at a time and re-insert them.
        No need to take them all the way out and be careful not to generate static electricity while
        doing this. One should always be careful with static electricity around an Elf anyway. Re-
        inserting the IC's will usually break up the oxidation of the pins.

        Switch contact oxidation can often be broken loose by flexing the pushbutton switches a
        few times or operating the toggle switches.

        Take a new look at your jumper "enhancements". Maybe one of them is not made
        correctly. It's rare but the CMOS keyboard decoder could have been zapped.

        The Elf II is pretty simple to operate but sometimes the operator just needs a refresher
        reading of the manual. ;o)

        Al Winfrey
        ....

        --- In cosmacelf@yahoogroups.com, "64536" <john_crane_97@...> wrote:
        >
        > I recently dug out my ELF II for some tinerking. I have the 4K card,
        > the Giant Board, the ROM BASIC, and an EPROM Programmer, and the ASCII
        > keyboard. But it seems to not accept hex input - I'm just trying it
        > out with the 4K card and the giant board. I've tried different jumper
        > combinations on the giant card (including J12)- I remember how kludgey
        > that was from my high school days - I have all the jumpers connected
        > to a little dipswitch control panel to make it easier. I've traced
        > the keyboard signals and input key signals all the way back to the
        > 1802. So I replaced it and it still doesn't work- just shows "00" in
        > the display. It may be some strange jumper setting/combination - but
        > I've followed the old manuals to no avail.
        >
        > Notes in my manual show I took it out and tested it (and it worked)
        > back in 2002. Something must have happened while it was sitting on a
        > shelf...
        >
        > Any ideas on how I can wake the sleeping beauty?
        >
      • 64536
        Al, Thanks for the tips. I ve found that the motherboard no longer has the 2101 ram chips, so I guess I m stuck with running on the 4K card. Can t seem to
        Message 3 of 12 , Aug 21, 2007
        • 0 Attachment
          Al,

          Thanks for the tips. I've found that the motherboard no longer has
          the 2101 ram chips, so I guess I'm stuck with running on the 4K
          card. Can't seem to find 2101 chips anywhere these days! I'm
          presuming (hoping) that the machine can run on just the 4K card
          addressed to start @0000 and doesn't need the 2101s. I remember
          that the 2101s were supposed to get addressed just after the 4K card.

          I have a switch mounted on J12 on the Giant Board (to switch between
          using the hex keypad and using the rs232 as they share a line to the
          CPU), it could be faulty/oxidized. I'll have to check the
          dipswitches for oxidation too for the other jumpers. I think
          whatever the problem is it's something that's preventing the CPU
          from accepting keyboard input and not the keyboard entry circuity
          itself (I'm thinking a faulty J12 could do this). I used a logic
          probe and traced all the key presses & toggle switch flips sending
          signals all the way back to the 1802, so I feel confident that part
          is OK. Of course, I only know that signals are present - not that
          they are the correct ones. I suppose the keyboard circuitry could
          be zapped and is just sending the wrong signals along. I could see
          blips on the data bus too when I hit "input", but after a couple, I
          always get a pulse train on the data bus. This was when I was just
          loading data from the hex key pad with the "load" key up. So I have
          no idea where all that data is coming from since the CPU isn't in
          run mode.

          There are very few sockets anywhere in the system. I originally
          built the machine when I was in high school and on the cheap so
          everything was soldered directly to the board. Over the years I've
          replaced a couple of chips and I used sockets for those. I've
          already done the socket/chip cleaning routine on them. You're
          definitely right about that. I've seen IC leads from chips in other
          machines that were BLACK!


          -John

          --- In cosmacelf@yahoogroups.com, "Al W," <wa9hsl@...> wrote:
          >
          > John,
          >
          > It might be better to start the troubleshooting by removing the
          more complicated items
          > such as the 4k card and the giant board leaving a basic 256 byte
          Elf II. This would
          > eliminate them as a possible cause of the problem and make
          diagnostics easier.
          >
          > Common shelf aging problems are oxidation of switches, oxidation
          of IC sockets and
          > memory loss of the operator. ;o)
          >
          > An easy thing to try is to gently raise the socketed IC's one at a
          time and re-insert them.
          > No need to take them all the way out and be careful not to
          generate static electricity while
          > doing this. One should always be careful with static electricity
          around an Elf anyway. Re-
          > inserting the IC's will usually break up the oxidation of the
          pins.
          >
          > Switch contact oxidation can often be broken loose by flexing the
          pushbutton switches a
          > few times or operating the toggle switches.
          >
          > Take a new look at your jumper "enhancements". Maybe one of them
          is not made
          > correctly. It's rare but the CMOS keyboard decoder could have been
          zapped.
          >
          > The Elf II is pretty simple to operate but sometimes the operator
          just needs a refresher
          > reading of the manual. ;o)
          >
          > Al Winfrey
          > ....
          >
          > --- In cosmacelf@yahoogroups.com, "64536" <john_crane_97@> wrote:
          > >
          > > I recently dug out my ELF II for some tinerking. I have the 4K
          card,
          > > the Giant Board, the ROM BASIC, and an EPROM Programmer, and the
          ASCII
          > > keyboard. But it seems to not accept hex input - I'm just
          trying it
          > > out with the 4K card and the giant board. I've tried different
          jumper
          > > combinations on the giant card (including J12)- I remember how
          kludgey
          > > that was from my high school days - I have all the jumpers
          connected
          > > to a little dipswitch control panel to make it easier. I've
          traced
          > > the keyboard signals and input key signals all the way back to
          the
          > > 1802. So I replaced it and it still doesn't work- just
          shows "00" in
          > > the display. It may be some strange jumper setting/combination -
          but
          > > I've followed the old manuals to no avail.
          > >
          > > Notes in my manual show I took it out and tested it (and it
          worked)
          > > back in 2002. Something must have happened while it was sitting
          on a
          > > shelf...
          > >
          > > Any ideas on how I can wake the sleeping beauty?
          > >
          >
        • Steve Valin
          John, Can you be more specific as to what you are seeing? Are you putting the machine into input mode, or trying to run out of a ROM monitor? The RS-232
          Message 4 of 12 , Aug 21, 2007
          • 0 Attachment
            John,

            Can you be more specific as to what you are seeing?
            Are you putting the machine into "input" mode, or
            trying to run out of a ROM monitor? The RS-232 input
            would be ROM monitor only, so I'm a little confused
            about that.

            If you are in "input" mode (single cycle DMA into
            SRAM), what do you see on the display when you enter a
            byte? Have you tried entering a walking one's pattern
            (01, 02, 04, 08, 10, 20, 40, 80)? You should be able
            to probe the outputs of the keyboard decoder and latch
            to see if what you enteres is being decoded. The way
            things operate in "input" mode is that your decoded
            keyboard entry sits at the input of the 4016 switches,
            then you hit the input switch, which will enable the
            4016 switches, write this data to SRAM (pointed to by
            R0 address), then read from this SRAM location and
            latch it to the displays. The key point here is that
            the displays will reflect the contents of SRAM, and
            not what's on the keyboard decoder bus. This means
            that the whole path has to be working in order to get
            the correct byte displayed. On my two ELF II's, I had
            to replace the keyboard decoders on both, as well as a
            2101 part on each (stuck sram cells inside the part).

            Since you don't have the 2101's and are using a plug
            in board, there is a lot more things that have to
            work! Maybe if you state where you are located, you
            may be near someone here that can help. I am in
            northern CA (bay area and sierra foothills).

            -steve


            --- 64536 <john_crane_97@...> wrote:

            > Al,
            >
            > Thanks for the tips. I've found that the
            > motherboard no longer has
            > the 2101 ram chips, so I guess I'm stuck with
            > running on the 4K
            > card. Can't seem to find 2101 chips anywhere these
            > days! I'm
            > presuming (hoping) that the machine can run on just
            > the 4K card
            > addressed to start @0000 and doesn't need the 2101s.
            > I remember
            > that the 2101s were supposed to get addressed just
            > after the 4K card.
            >
            > I have a switch mounted on J12 on the Giant Board
            > (to switch between
            > using the hex keypad and using the rs232 as they
            > share a line to the
            > CPU), it could be faulty/oxidized. I'll have to
            > check the
            > dipswitches for oxidation too for the other jumpers.
            > I think
            > whatever the problem is it's something that's
            > preventing the CPU
            > from accepting keyboard input and not the keyboard
            > entry circuity
            > itself (I'm thinking a faulty J12 could do this). I
            > used a logic
            > probe and traced all the key presses & toggle switch
            > flips sending
            > signals all the way back to the 1802, so I feel
            > confident that part
            > is OK. Of course, I only know that signals are
            > present - not that
            > they are the correct ones. I suppose the keyboard
            > circuitry could
            > be zapped and is just sending the wrong signals
            > along. I could see
            > blips on the data bus too when I hit "input", but
            > after a couple, I
            > always get a pulse train on the data bus. This was
            > when I was just
            > loading data from the hex key pad with the "load"
            > key up. So I have
            > no idea where all that data is coming from since the
            > CPU isn't in
            > run mode.
            >
            > There are very few sockets anywhere in the system.
            > I originally
            > built the machine when I was in high school and on
            > the cheap so
            > everything was soldered directly to the board. Over
            > the years I've
            > replaced a couple of chips and I used sockets for
            > those. I've
            > already done the socket/chip cleaning routine on
            > them. You're
            > definitely right about that. I've seen IC leads
            > from chips in other
            > machines that were BLACK!
            >
            >
            > -John
            >
            > --- In cosmacelf@yahoogroups.com, "Al W,"
            > <wa9hsl@...> wrote:
            > >
            > > John,
            > >
            > > It might be better to start the troubleshooting by
            > removing the
            > more complicated items
            > > such as the 4k card and the giant board leaving a
            > basic 256 byte
            > Elf II. This would
            > > eliminate them as a possible cause of the problem
            > and make
            > diagnostics easier.
            > >
            > > Common shelf aging problems are oxidation of
            > switches, oxidation
            > of IC sockets and
            > > memory loss of the operator. ;o)
            > >
            > > An easy thing to try is to gently raise the
            > socketed IC's one at a
            > time and re-insert them.
            > > No need to take them all the way out and be
            > careful not to
            > generate static electricity while
            > > doing this. One should always be careful with
            > static electricity
            > around an Elf anyway. Re-
            > > inserting the IC's will usually break up the
            > oxidation of the
            > pins.
            > >
            > > Switch contact oxidation can often be broken loose
            > by flexing the
            > pushbutton switches a
            > > few times or operating the toggle switches.
            > >
            > > Take a new look at your jumper "enhancements".
            > Maybe one of them
            > is not made
            > > correctly. It's rare but the CMOS keyboard decoder
            > could have been
            > zapped.
            > >
            > > The Elf II is pretty simple to operate but
            > sometimes the operator
            > just needs a refresher
            > > reading of the manual. ;o)
            > >
            > > Al Winfrey
            > > ....
            > >
            > > --- In cosmacelf@yahoogroups.com, "64536"
            > <john_crane_97@> wrote:
            > > >
            > > > I recently dug out my ELF II for some tinerking.
            > I have the 4K
            > card,
            > > > the Giant Board, the ROM BASIC, and an EPROM
            > Programmer, and the
            > ASCII
            > > > keyboard. But it seems to not accept hex input
            > - I'm just
            > trying it
            > > > out with the 4K card and the giant board. I've
            > tried different
            > jumper
            > > > combinations on the giant card (including J12)-
            > I remember how
            > kludgey
            > > > that was from my high school days - I have all
            > the jumpers
            > connected
            > > > to a little dipswitch control panel to make it
            > easier. I've
            > traced
            > > > the keyboard signals and input key signals all
            > the way back to
            > the
            > > > 1802. So I replaced it and it still doesn't
            > work- just
            > shows "00" in
            > > > the display. It may be some strange jumper
            > setting/combination -
            > but
            > > > I've followed the old manuals to no avail.
            > > >
            > > > Notes in my manual show I took it out and tested
            > it (and it
            > worked)
            > > > back in 2002. Something must have happened
            > while it was sitting
            > on a
            > > > shelf...
            > > >
            > > > Any ideas on how I can wake the sleeping beauty?
            > > >
            > >
            >
            >
            >




            ____________________________________________________________________________________
            Boardwalk for $500? In 2007? Ha! Play Monopoly Here and Now (it's updated for today's economy) at Yahoo! Games.
            http://get.games.yahoo.com/proddesc?gamekey=monopolyherenow
          • erd_6502
            ... They have a $20 minimum order, but Unicorn Electronics has them for $1.49 each... http://198.170.117.30/IC/STATIC.html ... Just after? That must be a
            Message 5 of 12 , Aug 21, 2007
            • 0 Attachment
              --- In cosmacelf@yahoogroups.com, "64536" <john_crane_97@...> wrote:
              >
              > Al,
              >
              > Thanks for the tips. I've found that the motherboard no longer has
              > the 2101 ram chips, so I guess I'm stuck with running on the 4K
              > card. Can't seem to find 2101 chips anywhere these days!

              They have a $20 minimum order, but Unicorn Electronics has them for
              $1.49 each...

              http://198.170.117.30/IC/STATIC.html

              > I'm
              > presuming (hoping) that the machine can run on just the 4K card
              > addressed to start @0000 and doesn't need the 2101s. I remember
              > that the 2101s were supposed to get addressed just after the 4K card.

              Just after? That must be a jumper thing, then.

              I would think that if you could strap the onboard 256 bytes to be
              _after_ the Big Board's 4K (which would make sense from the standpoint
              of simplifying the decode logic), then you wouldn't _need_ the onboard
              256 bytes in the first place. I can see that being a "feature" back
              in the day - those 2101s/5101s/1822s were a bit pricey compared to 1K
              and 4K SRAMs after a while.

              > I have a switch mounted on J12 on the Giant Board (to switch between
              > using the hex keypad and using the rs232 as they share a line to the
              > CPU), it could be faulty/oxidized.

              Toggle switches don't tend to oxidize much (unlike sockets and IC
              pins), but that should be easy to check with a voltmeter.

              > I think
              > whatever the problem is it's something that's preventing the CPU
              > from accepting keyboard input and not the keyboard entry circuity
              > itself...

              From what I remember about the ElfII's keypad circuit, the first
              keypress latches 4 bits, then the second keypress gates the present
              key from the 72C922 and the previous latched 4 bits to the 1802.
              Other than that, the load-mode circuitry is similar to the original
              Elf. If you think the 72C922 circuit is working correctly, then the
              problem might be in the /DMA_IN circuit or any buffers/analog switch
              chips on the bus.

              I'd make sure first that the load and reset and run buttons
              (switches?) are presenting the right signals to the 1802, or else you
              won't see normal behavior. Then I'd debug the keypad circuit, then
              memory protect.

              If you want to debug the board without the Big Board and you don't
              have 2101s (or CMOS 5101s or RCA 1822s), you could consider
              fabricating a plug-in SRAM replacement from perfboard and some
              machined-pin socket pins and a 6116 2K SRAM (ignoring 1.75K of the
              chip). If you don't have a 6116 in your parts box, they were common
              in 1980s microcomputers, or are inexpensive to buy from places like BG
              Micro. Alternately, you could track down a 6264 or 62256 and "waste"
              even more of the chip, but they might be easier to find.

              Something else you could do is to consider building what the older
              brother of my Elf mentor did 30 years ago - he debugged his Quest Elf
              with a debounced pushbutton switch on the clock to the 1802 and used a
              volt meter as a logic probe. All the timing diagrams are still
              available - eight clock ticks equals one machine cycle - just step
              through everything. You could probably speed up the debugging with an
              LED logic probe or two - there are designs for a CMOS 4049-based one
              in Don Lancaster's "CMOS Cookbook", IIRC. A few Gompers clips or flea
              clips might make that task more efficient, but it's not essential to
              get started.

              Fortunately, the 1802 is easier to debug than most machines, in my
              experience. If you do remove a chip and replace it with a socket, my
              recommendation is to use high-quality machined-pin sockets. They are
              usually between $0.15-$1.00 each, depending on size, but very
              reliable. I buy them by the tube when I see them at hamfests, and I
              buy larger ones in needed quantities from places like BG Micro,
              especially when they have periodic sales.

              Good luck with everything,

              -ethan
            • Lee Hart
              ... I have lots of them. They are actually 5101; the CMOS version of the 2101, but are pin-for-pin compatible and take lots less power. I ll gladly send you
              Message 6 of 12 , Aug 22, 2007
              • 0 Attachment
                <john_crane_97@...> wrote:
                > I've found that the motherboard no longer has the 2101 RAM chips,
                > so I guess I'm stuck with running on the 4K card. Can't seem to
                > find 2101 chips anywhere these days!

                I have lots of them. They are actually 5101; the CMOS version of the
                2101, but are pin-for-pin compatible and take lots less power. I'll
                gladly send you two for the cost of postage if you send me a stamped
                self-addressed envelope to return them in.

                A useful trick for debugging an 1802 system is to pull the DMA-IN or
                DMA-OUT pin low. The 1802 will then sequentially output addresses on the
                MA0-8 lines, and either read (DMA-OUT) or write (DMA-IN) to memory in
                every cycle. An oscilloscope or even a simple logic probe can check the
                various lines to see if they are toggling, and if not, find where the
                problem lies.
                --
                "Never doubt that the work of a small group of thoughtful, committed
                citizens can change the world. Indeed, it's the only thing that ever
                has!" -- Margaret Mead
                --
                Lee A. Hart, 814 8th Ave N, Sartell MN 56377, leeahart_at_earthlink.net
              • 64536
                Hi Steve, Thanks for the help. My Giant Board has a built in monitor, but I can t access it because I can t get any response from the keyboard. The Elf II has
                Message 7 of 12 , Aug 23, 2007
                • 0 Attachment
                  Hi Steve,

                  Thanks for the help.

                  My Giant Board has a built in monitor, but I can't access it because
                  I can't get any response from the keyboard. The Elf II has three
                  toggle switches (LOAD RUN and MEMORY PROTECT), and a hexadecimal
                  keypad with an INPUT key and dual character LED display. What
                  normally happens is that you hit the "LOAD" toggle key, type in a
                  hex byte on the keypad and then hit the INPUT key. The byte you
                  typed on the keypad should appear on the LED display.

                  This doesn't happen on my ELF. I just get "00" in the LED display.
                  What I did to diagnose the problem was to trace the signals from the
                  keys and toggles all the way back to the 1802. The hope was to find
                  a bad chip (maybe a buffer) that wasn't sending the signal along.
                  Everything seemed fine. Now I know (thanks to you!) that the backend
                  needs to work too as the LEDS simply show the SRAM contents.

                  I haven't had the time in the past few days to dig deeper, but my
                  next steps will be to inspect/replace the switch I installed on J12,
                  find some 2101s and remove the 4K card. And then reconfigure the
                  main board so I can remove the Giant Board, so I can test the "bare
                  bones" system. The Giant Board requires some hardware mods ( a few
                  jumpers changed and a diode removed I think) So I'll have to spend a
                  few minutes with a soldering iron to reconfigure things.

                  Sorry, I a bit far from California. I'm in Conroe, TX. (just north
                  of Houston).

                  Also, if anyone is interested, I came across issues #1-30 of IPSO
                  FACTO, issues #1-9 of CLUB 1802, and some assorted QUEST newsletters
                  buried in the box with my ELF. When I have some more time, I'll
                  make some scans for posterity.

                  -J

                  --- In cosmacelf@yahoogroups.com, Steve Valin <sjvalin@...> wrote:
                  >
                  > John,
                  >
                  > Can you be more specific as to what you are seeing?
                  > Are you putting the machine into "input" mode, or
                  > trying to run out of a ROM monitor? The RS-232 input
                  > would be ROM monitor only, so I'm a little confused
                  > about that.
                  >
                  > If you are in "input" mode (single cycle DMA into
                  > SRAM), what do you see on the display when you enter a
                  > byte? Have you tried entering a walking one's pattern
                  > (01, 02, 04, 08, 10, 20, 40, 80)? You should be able
                  > to probe the outputs of the keyboard decoder and latch
                  > to see if what you enteres is being decoded. The way
                  > things operate in "input" mode is that your decoded
                  > keyboard entry sits at the input of the 4016 switches,
                  > then you hit the input switch, which will enable the
                  > 4016 switches, write this data to SRAM (pointed to by
                  > R0 address), then read from this SRAM location and
                  > latch it to the displays. The key point here is that
                  > the displays will reflect the contents of SRAM, and
                  > not what's on the keyboard decoder bus. This means
                  > that the whole path has to be working in order to get
                  > the correct byte displayed. On my two ELF II's, I had
                  > to replace the keyboard decoders on both, as well as a
                  > 2101 part on each (stuck sram cells inside the part).
                  >
                  > Since you don't have the 2101's and are using a plug
                  > in board, there is a lot more things that have to
                  > work! Maybe if you state where you are located, you
                  > may be near someone here that can help. I am in
                  > northern CA (bay area and sierra foothills).
                  >
                  > -steve
                  >
                  >
                  > --- 64536 <john_crane_97@...> wrote:
                  >
                  > > Al,
                  > >
                  > > Thanks for the tips. I've found that the
                  > > motherboard no longer has
                  > > the 2101 ram chips, so I guess I'm stuck with
                  > > running on the 4K
                  > > card. Can't seem to find 2101 chips anywhere these
                  > > days! I'm
                  > > presuming (hoping) that the machine can run on just
                  > > the 4K card
                  > > addressed to start @0000 and doesn't need the 2101s.
                  > > I remember
                  > > that the 2101s were supposed to get addressed just
                  > > after the 4K card.
                  > >
                  > > I have a switch mounted on J12 on the Giant Board
                  > > (to switch between
                  > > using the hex keypad and using the rs232 as they
                  > > share a line to the
                  > > CPU), it could be faulty/oxidized. I'll have to
                  > > check the
                  > > dipswitches for oxidation too for the other jumpers.
                  > > I think
                  > > whatever the problem is it's something that's
                  > > preventing the CPU
                  > > from accepting keyboard input and not the keyboard
                  > > entry circuity
                  > > itself (I'm thinking a faulty J12 could do this). I
                  > > used a logic
                  > > probe and traced all the key presses & toggle switch
                  > > flips sending
                  > > signals all the way back to the 1802, so I feel
                  > > confident that part
                  > > is OK. Of course, I only know that signals are
                  > > present - not that
                  > > they are the correct ones. I suppose the keyboard
                  > > circuitry could
                  > > be zapped and is just sending the wrong signals
                  > > along. I could see
                  > > blips on the data bus too when I hit "input", but
                  > > after a couple, I
                  > > always get a pulse train on the data bus. This was
                  > > when I was just
                  > > loading data from the hex key pad with the "load"
                  > > key up. So I have
                  > > no idea where all that data is coming from since the
                  > > CPU isn't in
                  > > run mode.
                  > >
                  > > There are very few sockets anywhere in the system.
                  > > I originally
                  > > built the machine when I was in high school and on
                  > > the cheap so
                  > > everything was soldered directly to the board. Over
                  > > the years I've
                  > > replaced a couple of chips and I used sockets for
                  > > those. I've
                  > > already done the socket/chip cleaning routine on
                  > > them. You're
                  > > definitely right about that. I've seen IC leads
                  > > from chips in other
                  > > machines that were BLACK!
                  > >
                  > >
                  > > -John
                  > >
                  > > --- In cosmacelf@yahoogroups.com, "Al W,"
                  > > <wa9hsl@> wrote:
                  > > >
                  > > > John,
                  > > >
                  > > > It might be better to start the troubleshooting by
                  > > removing the
                  > > more complicated items
                  > > > such as the 4k card and the giant board leaving a
                  > > basic 256 byte
                  > > Elf II. This would
                  > > > eliminate them as a possible cause of the problem
                  > > and make
                  > > diagnostics easier.
                  > > >
                  > > > Common shelf aging problems are oxidation of
                  > > switches, oxidation
                  > > of IC sockets and
                  > > > memory loss of the operator. ;o)
                  > > >
                  > > > An easy thing to try is to gently raise the
                  > > socketed IC's one at a
                  > > time and re-insert them.
                  > > > No need to take them all the way out and be
                  > > careful not to
                  > > generate static electricity while
                  > > > doing this. One should always be careful with
                  > > static electricity
                  > > around an Elf anyway. Re-
                  > > > inserting the IC's will usually break up the
                  > > oxidation of the
                  > > pins.
                  > > >
                  > > > Switch contact oxidation can often be broken loose
                  > > by flexing the
                  > > pushbutton switches a
                  > > > few times or operating the toggle switches.
                  > > >
                  > > > Take a new look at your jumper "enhancements".
                  > > Maybe one of them
                  > > is not made
                  > > > correctly. It's rare but the CMOS keyboard decoder
                  > > could have been
                  > > zapped.
                  > > >
                  > > > The Elf II is pretty simple to operate but
                  > > sometimes the operator
                  > > just needs a refresher
                  > > > reading of the manual. ;o)
                  > > >
                  > > > Al Winfrey
                  > > > ....
                  > > >
                  > > > --- In cosmacelf@yahoogroups.com, "64536"
                  > > <john_crane_97@> wrote:
                  > > > >
                  > > > > I recently dug out my ELF II for some tinerking.
                  > > I have the 4K
                  > > card,
                  > > > > the Giant Board, the ROM BASIC, and an EPROM
                  > > Programmer, and the
                  > > ASCII
                  > > > > keyboard. But it seems to not accept hex input
                  > > - I'm just
                  > > trying it
                  > > > > out with the 4K card and the giant board. I've
                  > > tried different
                  > > jumper
                  > > > > combinations on the giant card (including J12)-
                  > > I remember how
                  > > kludgey
                  > > > > that was from my high school days - I have all
                  > > the jumpers
                  > > connected
                  > > > > to a little dipswitch control panel to make it
                  > > easier. I've
                  > > traced
                  > > > > the keyboard signals and input key signals all
                  > > the way back to
                  > > the
                  > > > > 1802. So I replaced it and it still doesn't
                  > > work- just
                  > > shows "00" in
                  > > > > the display. It may be some strange jumper
                  > > setting/combination -
                  > > but
                  > > > > I've followed the old manuals to no avail.
                  > > > >
                  > > > > Notes in my manual show I took it out and tested
                  > > it (and it
                  > > worked)
                  > > > > back in 2002. Something must have happened
                  > > while it was sitting
                  > > on a
                  > > > > shelf...
                  > > > >
                  > > > > Any ideas on how I can wake the sleeping beauty?
                  > > > >
                  > > >
                  > >
                  > >
                  > >
                  >
                  >
                  >
                  >
                  >
                  _____________________________________________________________________
                  _______________
                  > Boardwalk for $500? In 2007? Ha! Play Monopoly Here and Now (it's
                  updated for today's economy) at Yahoo! Games.
                  > http://get.games.yahoo.com/proddesc?gamekey=monopolyherenow
                  >
                • Andrew Wasson
                  Hey J, Probably a stupid question but have you ensured that your memory protect toggle isn t active? That will prevent you from changing the data for whatever
                  Message 8 of 12 , Aug 23, 2007
                  • 0 Attachment
                    Hey J,
                    Probably a stupid question but have you ensured that your memory
                    protect toggle isn't active? That will prevent you from changing the
                    data for whatever address you are on.

                    Andrew

                    --- In cosmacelf@yahoogroups.com, "64536" <john_crane_97@...> wrote:
                    >
                    > Hi Steve,
                    >
                    > Thanks for the help.
                    >
                    > My Giant Board has a built in monitor, but I can't access it because
                    > I can't get any response from the keyboard. The Elf II has three
                    > toggle switches (LOAD RUN and MEMORY PROTECT), and a hexadecimal
                    > keypad with an INPUT key and dual character LED display. What
                    > normally happens is that you hit the "LOAD" toggle key, type in a
                    > hex byte on the keypad and then hit the INPUT key. The byte you
                    > typed on the keypad should appear on the LED display.
                    >
                    > This doesn't happen on my ELF. I just get "00" in the LED display.
                    > What I did to diagnose the problem was to trace the signals from the
                    > keys and toggles all the way back to the 1802. The hope was to find
                    > a bad chip (maybe a buffer) that wasn't sending the signal along.
                    > Everything seemed fine. Now I know (thanks to you!) that the backend
                    > needs to work too as the LEDS simply show the SRAM contents.
                    >
                    > I haven't had the time in the past few days to dig deeper, but my
                    > next steps will be to inspect/replace the switch I installed on J12,
                    > find some 2101s and remove the 4K card. And then reconfigure the
                    > main board so I can remove the Giant Board, so I can test the "bare
                    > bones" system. The Giant Board requires some hardware mods ( a few
                    > jumpers changed and a diode removed I think) So I'll have to spend a
                    > few minutes with a soldering iron to reconfigure things.
                    >
                    > Sorry, I a bit far from California. I'm in Conroe, TX. (just north
                    > of Houston).
                    >
                    > Also, if anyone is interested, I came across issues #1-30 of IPSO
                    > FACTO, issues #1-9 of CLUB 1802, and some assorted QUEST newsletters
                    > buried in the box with my ELF. When I have some more time, I'll
                    > make some scans for posterity.
                    >
                    > -J
                    >
                    > --- In cosmacelf@yahoogroups.com, Steve Valin <sjvalin@> wrote:
                    > >
                    > > John,
                    > >
                    > > Can you be more specific as to what you are seeing?
                    > > Are you putting the machine into "input" mode, or
                    > > trying to run out of a ROM monitor? The RS-232 input
                    > > would be ROM monitor only, so I'm a little confused
                    > > about that.
                    > >
                    > > If you are in "input" mode (single cycle DMA into
                    > > SRAM), what do you see on the display when you enter a
                    > > byte? Have you tried entering a walking one's pattern
                    > > (01, 02, 04, 08, 10, 20, 40, 80)? You should be able
                    > > to probe the outputs of the keyboard decoder and latch
                    > > to see if what you enteres is being decoded. The way
                    > > things operate in "input" mode is that your decoded
                    > > keyboard entry sits at the input of the 4016 switches,
                    > > then you hit the input switch, which will enable the
                    > > 4016 switches, write this data to SRAM (pointed to by
                    > > R0 address), then read from this SRAM location and
                    > > latch it to the displays. The key point here is that
                    > > the displays will reflect the contents of SRAM, and
                    > > not what's on the keyboard decoder bus. This means
                    > > that the whole path has to be working in order to get
                    > > the correct byte displayed. On my two ELF II's, I had
                    > > to replace the keyboard decoders on both, as well as a
                    > > 2101 part on each (stuck sram cells inside the part).
                    > >
                    > > Since you don't have the 2101's and are using a plug
                    > > in board, there is a lot more things that have to
                    > > work! Maybe if you state where you are located, you
                    > > may be near someone here that can help. I am in
                    > > northern CA (bay area and sierra foothills).
                    > >
                    > > -steve
                    > >
                    > >
                    > > --- 64536 <john_crane_97@> wrote:
                    > >
                    > > > Al,
                    > > >
                    > > > Thanks for the tips. I've found that the
                    > > > motherboard no longer has
                    > > > the 2101 ram chips, so I guess I'm stuck with
                    > > > running on the 4K
                    > > > card. Can't seem to find 2101 chips anywhere these
                    > > > days! I'm
                    > > > presuming (hoping) that the machine can run on just
                    > > > the 4K card
                    > > > addressed to start @0000 and doesn't need the 2101s.
                    > > > I remember
                    > > > that the 2101s were supposed to get addressed just
                    > > > after the 4K card.
                    > > >
                    > > > I have a switch mounted on J12 on the Giant Board
                    > > > (to switch between
                    > > > using the hex keypad and using the rs232 as they
                    > > > share a line to the
                    > > > CPU), it could be faulty/oxidized. I'll have to
                    > > > check the
                    > > > dipswitches for oxidation too for the other jumpers.
                    > > > I think
                    > > > whatever the problem is it's something that's
                    > > > preventing the CPU
                    > > > from accepting keyboard input and not the keyboard
                    > > > entry circuity
                    > > > itself (I'm thinking a faulty J12 could do this). I
                    > > > used a logic
                    > > > probe and traced all the key presses & toggle switch
                    > > > flips sending
                    > > > signals all the way back to the 1802, so I feel
                    > > > confident that part
                    > > > is OK. Of course, I only know that signals are
                    > > > present - not that
                    > > > they are the correct ones. I suppose the keyboard
                    > > > circuitry could
                    > > > be zapped and is just sending the wrong signals
                    > > > along. I could see
                    > > > blips on the data bus too when I hit "input", but
                    > > > after a couple, I
                    > > > always get a pulse train on the data bus. This was
                    > > > when I was just
                    > > > loading data from the hex key pad with the "load"
                    > > > key up. So I have
                    > > > no idea where all that data is coming from since the
                    > > > CPU isn't in
                    > > > run mode.
                    > > >
                    > > > There are very few sockets anywhere in the system.
                    > > > I originally
                    > > > built the machine when I was in high school and on
                    > > > the cheap so
                    > > > everything was soldered directly to the board. Over
                    > > > the years I've
                    > > > replaced a couple of chips and I used sockets for
                    > > > those. I've
                    > > > already done the socket/chip cleaning routine on
                    > > > them. You're
                    > > > definitely right about that. I've seen IC leads
                    > > > from chips in other
                    > > > machines that were BLACK!
                    > > >
                    > > >
                    > > > -John
                    > > >
                    > > > --- In cosmacelf@yahoogroups.com, "Al W,"
                    > > > <wa9hsl@> wrote:
                    > > > >
                    > > > > John,
                    > > > >
                    > > > > It might be better to start the troubleshooting by
                    > > > removing the
                    > > > more complicated items
                    > > > > such as the 4k card and the giant board leaving a
                    > > > basic 256 byte
                    > > > Elf II. This would
                    > > > > eliminate them as a possible cause of the problem
                    > > > and make
                    > > > diagnostics easier.
                    > > > >
                    > > > > Common shelf aging problems are oxidation of
                    > > > switches, oxidation
                    > > > of IC sockets and
                    > > > > memory loss of the operator. ;o)
                    > > > >
                    > > > > An easy thing to try is to gently raise the
                    > > > socketed IC's one at a
                    > > > time and re-insert them.
                    > > > > No need to take them all the way out and be
                    > > > careful not to
                    > > > generate static electricity while
                    > > > > doing this. One should always be careful with
                    > > > static electricity
                    > > > around an Elf anyway. Re-
                    > > > > inserting the IC's will usually break up the
                    > > > oxidation of the
                    > > > pins.
                    > > > >
                    > > > > Switch contact oxidation can often be broken loose
                    > > > by flexing the
                    > > > pushbutton switches a
                    > > > > few times or operating the toggle switches.
                    > > > >
                    > > > > Take a new look at your jumper "enhancements".
                    > > > Maybe one of them
                    > > > is not made
                    > > > > correctly. It's rare but the CMOS keyboard decoder
                    > > > could have been
                    > > > zapped.
                    > > > >
                    > > > > The Elf II is pretty simple to operate but
                    > > > sometimes the operator
                    > > > just needs a refresher
                    > > > > reading of the manual. ;o)
                    > > > >
                    > > > > Al Winfrey
                    > > > > ....
                    > > > >
                    > > > > --- In cosmacelf@yahoogroups.com, "64536"
                    > > > <john_crane_97@> wrote:
                    > > > > >
                    > > > > > I recently dug out my ELF II for some tinerking.
                    > > > I have the 4K
                    > > > card,
                    > > > > > the Giant Board, the ROM BASIC, and an EPROM
                    > > > Programmer, and the
                    > > > ASCII
                    > > > > > keyboard. But it seems to not accept hex input
                    > > > - I'm just
                    > > > trying it
                    > > > > > out with the 4K card and the giant board. I've
                    > > > tried different
                    > > > jumper
                    > > > > > combinations on the giant card (including J12)-
                    > > > I remember how
                    > > > kludgey
                    > > > > > that was from my high school days - I have all
                    > > > the jumpers
                    > > > connected
                    > > > > > to a little dipswitch control panel to make it
                    > > > easier. I've
                    > > > traced
                    > > > > > the keyboard signals and input key signals all
                    > > > the way back to
                    > > > the
                    > > > > > 1802. So I replaced it and it still doesn't
                    > > > work- just
                    > > > shows "00" in
                    > > > > > the display. It may be some strange jumper
                    > > > setting/combination -
                    > > > but
                    > > > > > I've followed the old manuals to no avail.
                    > > > > >
                    > > > > > Notes in my manual show I took it out and tested
                    > > > it (and it
                    > > > worked)
                    > > > > > back in 2002. Something must have happened
                    > > > while it was sitting
                    > > > on a
                    > > > > > shelf...
                    > > > > >
                    > > > > > Any ideas on how I can wake the sleeping beauty?
                    > > > > >
                    > > > >
                    > > >
                    > > >
                    > > >
                    > >
                    > >
                    > >
                    > >
                    > >
                    > _____________________________________________________________________
                    > _______________
                    > > Boardwalk for $500? In 2007? Ha! Play Monopoly Here and Now (it's
                    > updated for today's economy) at Yahoo! Games.
                    > > http://get.games.yahoo.com/proddesc?gamekey=monopolyherenow
                    > >
                    >
                  • 64536
                    Lee, Thanks again for the chips. The problem turned out to be a faulty A5 chip on the motherboard. The NAND gates weren t triggering and therefore the
                    Message 9 of 12 , Sep 29 8:04 PM
                    • 0 Attachment
                      Lee,

                      Thanks again for the chips. The problem turned out to be a faulty
                      A5 chip on the motherboard. The NAND gates weren't triggering and
                      therefore the display wasn't getting updated. I tore everything
                      down to the bare bones; now I get to build it back :) Back in the
                      day, I had the Netronics RPN-style BASIC running from ROM. Wish me
                      luck getting back to that point!

                      -John



                      --- In cosmacelf@yahoogroups.com, Lee Hart <leeahart@...> wrote:
                      >
                      > <john_crane_97@> wrote:
                      > > I've found that the motherboard no longer has the 2101 RAM chips,
                      > > so I guess I'm stuck with running on the 4K card. Can't seem to
                      > > find 2101 chips anywhere these days!
                      >
                      > I have lots of them. They are actually 5101; the CMOS version of
                      the
                      > 2101, but are pin-for-pin compatible and take lots less power. I'll
                      > gladly send you two for the cost of postage if you send me a
                      stamped
                      > self-addressed envelope to return them in.
                      >
                      > A useful trick for debugging an 1802 system is to pull the DMA-IN
                      or
                      > DMA-OUT pin low. The 1802 will then sequentially output addresses
                      on the
                      > MA0-8 lines, and either read (DMA-OUT) or write (DMA-IN) to memory
                      in
                      > every cycle. An oscilloscope or even a simple logic probe can
                      check the
                      > various lines to see if they are toggling, and if not, find where
                      the
                      > problem lies.
                      > --
                      > "Never doubt that the work of a small group of thoughtful,
                      committed
                      > citizens can change the world. Indeed, it's the only thing that
                      ever
                      > has!" -- Margaret Mead
                      > --
                      > Lee A. Hart, 814 8th Ave N, Sartell MN 56377,
                      leeahart_at_earthlink.net
                      >
                    • Lee Hart
                      From: 64536 ... Ah! Glad you got it working. Now you have some memory for expansion. :-) -- Excellence does not require perfection.
                      Message 10 of 12 , Sep 29 10:03 PM
                      • 0 Attachment
                        From: 64536 <john_crane_97@...>

                        > Lee,
                        >
                        > Thanks again for the chips. The problem turned out to be
                        > a faulty A5 chip on the motherboard.

                        Ah! Glad you got it working. Now you have some memory for expansion. :-)


                        --
                        "Excellence does not require perfection." -- Henry James
                        --
                        Lee A. Hart, 814 8th Ave N, Sartell MN 56377, leeahart-at-earthlink.net
                      • 64536
                        ... expansion. :-) ... earthlink.net ... Just an update for Lee or anyone else who s following.... 1) Ran a few simple programs via hex keypad. The Monitor
                        Message 11 of 12 , Oct 18, 2007
                        • 0 Attachment
                          --- In cosmacelf@yahoogroups.com, Lee Hart <leeahart@...> wrote:
                          >
                          > From: 64536 <john_crane_97@...>
                          >
                          > > Lee,
                          > >
                          > > Thanks again for the chips. The problem turned out to be
                          > > a faulty A5 chip on the motherboard.
                          >
                          > Ah! Glad you got it working. Now you have some memory for
                          expansion. :-)
                          >
                          >
                          > --
                          > "Excellence does not require perfection." -- Henry James
                          > --
                          > Lee A. Hart, 814 8th Ave N, Sartell MN 56377, leeahart-at-
                          earthlink.net
                          >


                          Just an update for Lee or anyone else who's following....

                          1) Ran a few simple programs via hex keypad. The Monitor on the
                          Giant board works fine. I've been using a Radio Shack CTR-121 Tape
                          Recorder with TDK "normal position" tapes and it works perfectly.
                          Even so, I've been experimenting with more modern storage. I tried
                          an Olympus VN-4100PC digital voice recorder. No good. I have
                          another one with a better frequency range on order. I'll let you
                          know how it works.

                          2) My 4K card was really screwy. I turned out to be a faulty A7
                          chip; a NOR gate wasn't NORing, so the memory chips never got
                          selected properly.

                          3) The Netronics keyboard/video card combo works nice, though slow
                          at 300 baud. I tested by shorting pins 2 & 3 on the RS232 connector
                          and got the "TV Typewriter" mode.

                          4) Currently looking at the Full Basic / EPROM card. I suspect I
                          have a problem with RS-232 input to the Elf II. I'm relearning
                          again how difficult tinkering with the elf was back in '79. Lots of
                          fiddly jumpers everywhere and cryptic instruction manuals. Jumpers
                          11 and 12 on my Giant Board have a switch (to switch between hex
                          keypad and terminal). Though the Full Basic manual says you don't
                          need a switch anymore because the software can compensate. However,
                          you need to use the hex keypad to tell the monitor to branch to
                          0003, the start of BASIC. So I do this and I think that's what's
                          making my RS232 input invisible. I tried switching it back quickly
                          while it's running, but no go. The RS232 input circuit is pretty
                          simple. J4 is used, so the only semiconductor used is Q1 a 2N5232
                          transistor. As luck would have it, I had one in my spares box. I
                          replaced it, thinking it was the most likely failure. But, it
                          didn't change anything. Now I'm beginning to wonder if it isn't
                          some obscure jumper setting somewhere, maybe even on another card,
                          or the motherboard. It wouldn't surprise me now. :)

                          I'll keep you posted....


                          -John
                        • Paul Backhouse
                          John just a thought that you may have covered already - when in run mode to start full Netronics Basic be sure not to press the Input key a second time after
                          Message 12 of 12 , Nov 25, 2009
                          • 0 Attachment
                            John just a thought that you may have covered already - when in run mode to
                            start full Netronics Basic be sure not to press the Input key a second time
                            after entering 3rd byte of the 000003 branch instruction from the Hex
                            keypad. The handshake is established from hitting <Return> once on the VID
                            keyboard or Windows terminal.. This should bring up the Width=32/64?
                            prompt. Pressing anything else seems to halt the sequence. I've never
                            manged to persuade the RS232 to run faster than 110 or 300 baud.

                            Paul.

                            -----Original Message-----
                            From: cosmacelf@yahoogroups.com [mailto:cosmacelf@yahoogroups.com]On
                            Behalf Of 64536
                            Sent: 19 October 2007 07:31
                            To: cosmacelf@yahoogroups.com
                            Subject: [cosmacelf] Re: ELF II in a coma....






                            [Non-text portions of this message have been removed]
                          Your message has been successfully submitted and would be delivered to recipients shortly.