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

Re: [cosmacelf] Is there any hardware junkies in this group

Expand Messages
  • Jim Hall
    Hello: I know I am the new guy on the block, and I enjoyed the digi-key item you attached. I tried doing the same with budget minded thinking of a clip to
    Message 1 of 28 , Jan 31, 2011
    • 0 Attachment
      Hello:
      I know I am the new guy on the block, and I enjoyed the digi-key item you
      attached. I tried doing the same with budget minded thinking of a clip to logic
      board of 2n2222 transistor tied to each line, and it worked very nice till I
      tried bringing all the lines into a cable. It all worked fine but too many
      things could go wrong with my budget minded thinking. I have  to say that a
      logic clip is the best bet and thank you for the idea and site to go to.

       (I did a lot of my tinkering with what I found in a local Radio Shack and what
      money I could get together, and a small bread board and a mind that only thought
      in wire wrapoing)

      Thanks again
      jim

       



      ________________________________
      From: Lee Hart <leeahart@...>
      To: cosmacelf@yahoogroups.com
      Sent: Mon, January 31, 2011 8:32:44 PM
      Subject: Re: [cosmacelf] Is there any hardware junkies in this group

       
      On 1/31/2011 5:26 PM, Jim Hall wrote:
      > Hello:
      > Once more I am thinkful for any answer to my humble questions, in the past I
      > have made a slue of logic analyzers from transistor types to bread board
      jangle
      > of wires that did it. Just that I dont want to fry any of the CMOS that is on
      > the board. In the past they have buffered everything on the viper board, but I
      > wonder if it is necessary. As I said b4 I like blinking lights.

      Modern high-brightness LEDs are easily visible even at currents well
      under 1ma. An LED and 10k series resistor keeps the current under 400
      microamps, which isn't going to hurt any IC.

      It's handy to make a "dip clip" with such an LED+resistor on every pin.
      Clip it onto any chip, and you can see what all its pins are doing.
      A DIP-clip is like a wide alligator clip that has leads to connect to
      every pin on an IC. Here's an example for a 16-pin DIP from Digikey:

      http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail&name=923700-ND
      --
      Lee A. Hart | Ring the bells that still can ring
      814 8th Ave N | Forget the perfect offering
      Sartell MN 56377 | There is a crack in everything
      leeahart earthlink.net | That's how the light gets in -- Leonard Cohen






      [Non-text portions of this message have been removed]
    • Jim Hall
      Hello: I just want to be sure I read my mail and answer everything as much as my old brain can take. In the distant past of the 1802 they user to buffer alot
      Message 2 of 28 , Jan 31, 2011
      • 0 Attachment
        Hello:
        I just want to be sure I read my mail and answer everything as much as my old
        brain can take. In the distant past of the 1802 they user to buffer alot of the
        work with CMOS 4050's and CMOS 4049's so that nothing would go wrong, and now we
        are saying a simple resistor and LED can do the "seeing" of the logic there? I
        just have to get out more and try something now that I have got the Vip out of
        the dust. I am just curious what the 1802 is doing, low state or high state, if
        I can at least understand what is happening as a program is loaded or
        functioning.


        Hope I havent confused you, as a new comer to the group, I have so much to
        learn, but have trouble trying to phrase it all.
        jim




        ________________________________
        From: awasson2001 <awasson@...>
        To: cosmacelf@yahoogroups.com
        Sent: Mon, January 31, 2011 6:42:50 PM
        Subject: [cosmacelf] Re: Is there any hardware junkies in this group

         
        If I read this correctly, you're interested in an add-on for your Cosmac VIP or
        other 1802 based compute so that you can see the contents or states registers
        and/or any other interesting parts of the system.

        So maybe a front panel type of project is what you're looking for. All of the
        ELF based computers I've seen have been fairly minimalistic as far as blinking
        lights go. Have a look at the Quest Super Elf for some enhancements with regard
        to monitoring state (fetch, execute, DMA, INT) and mode (load, reset, run,
        wait). The schematic and notes are in the Super Elf manual starting on pages 19.

        Andrew

        --- In cosmacelf@yahoogroups.com, Jim Hall <jehall1442596@...> wrote:
        >
        > Hello:
        > Once more I am thinkful for any answer to my humble questions, in the past I
        > have made a slue of logic analyzers from transistor types to bread board jangle
        >
        > of wires that did it. Just that I dont want to fry any of the CMOS that is on
        > the board. In the past they have buffered everything on the viper board, but I

        > wonder if it is necessary. As I said b4 I like blinking lights.
        >
        >
        > jim
        >
        >
        >
        >
        > ________________________________
        > From: Lee Hart <leeahart@...>
        > To: cosmacelf@yahoogroups.com
        > Sent: Mon, January 31, 2011 6:09:23 PM
        > Subject: Re: [cosmacelf] Is there any hardware junkies in this group
        >
        >  
        > On 1/31/2011 2:25 PM, Jim Hall wrote:
        > > any hardware junkies for the 1802 machines here?
        >
        > Naa... we're all sane normal rational people, that all believe the Intel
        > x86 CPUs are the epitome of perfection. :-)
        >
        > > Why I ask is that I am thinking of something simple to start
        > > with, say what is the best interface chip to see what the bus
        > > or even the 1802 chip is doing?
        >
        > A nice thing about CMOS in general (and the 1802 in particular) is that
        > you don't need any drivers or special logic to make a logic analyzer.
        > Just connect an LED and series resistor to show the state of the line in
        > question. The 1802 is static, so your "clock" can run as slow as you
        > like, or even be a debounced pushbutton to micro-step it through a bus
        > cycle.
        >
        > --
        > Lee A. Hart | Ring the bells that still can ring
        > 814 8th Ave N | Forget the perfect offering
        > Sartell MN 56377 | There is a crack in everything
        > leeahart earthlink.net | That's how the light gets in -- Leonard Cohen
        >
        >
        >
        >
        >
        >
        > [Non-text portions of this message have been removed]
        >







        [Non-text portions of this message have been removed]
      • Lee Hart
        ... Yeah; adding a cable adds a lot of capacitance, which can mess up a high speed digital circuit. DIP clips are kind of expensive ($10-30 depending on number
        Message 3 of 28 , Jan 31, 2011
        • 0 Attachment
          Jim Hall wrote:
          > I enjoyed the digi-key item you attached [DIP-clip]. I tried doing the same with budget minded thinking of a clip to logic
          > board of 2n2222 transistor tied to each line, and it worked very nice till I
          > tried bringing all the lines into a cable.

          Yeah; adding a cable adds a lot of capacitance, which can mess up a high
          speed digital circuit.

          DIP clips are kind of expensive ($10-30 depending on number of pins),
          but they sure are convenient. Since the power and ground pins of ICs are
          (nearly) standardized, it isn't hard to put the LEDs and resistors all
          right on the DIP clip.

          If you're on a budget, you can make a similar "probe" from a pair of IC
          sockets. The two sockets are stacked, with all pins wired straight
          through (pin 1 to 2, 2, to 2, etc.) The LEDs and resistors also connect
          to the pins, and peek out the sides.

          Unplug the IC from your board, plug this "probe" in its place, and plug
          the chip into the top socket.

          A useful variant of this that I've made has DIP switches between the
          upper and lower sockets. With all switches closed, it has no effect --
          the chip is connected to the circuit normally. But when you open a
          switch, you break a pin. This can let you see whether the chip or
          something else in the circuit is holding a line high or low, or let you
          inject something different (a different clock frequency), or measure
          something (like power supply current).

          One cute variation of this is to open the 8 data lines to a micro. Hard
          wire the 8 micro pins to whatever is the no-op instruction for that
          micro. The micro now marches through memory, incrementing the address,
          "reading" a no-op, and repeating. You can now probe all the other chips
          in the system, to see what memories are being selected, etc.

          Or you can make it an I/O instruction, to repetitively access some port
          address to test for hardware problems.

          --
          Lee A. Hart | Ring the bells that still can ring
          814 8th Ave N | Forget the perfect offering
          Sartell MN 56377 | There is a crack in everything
          leeahart earthlink.net | That's how the light gets in -- Leonard Cohen
        • Lee Hart
          ... Yes. In the 1970 s an LED needed 10ma or so for reasonably brightness. Today s super-bright LEDs are dim but fully visible in normal room light with 0.1ma!
          Message 4 of 28 , Jan 31, 2011
          • 0 Attachment
            Jim Hall wrote:
            > I just want to be sure I read my mail and answer everything as much as my old
            > brain can take. In the distant past of the 1802 they user to buffer alot of the
            > work with CMOS 4050's and CMOS 4049's so that nothing would go wrong, and now we
            > are saying a simple resistor and LED can do the "seeing" of the logic there?

            Yes. In the 1970's an LED needed 10ma or so for reasonably brightness.
            Today's super-bright LEDs are dim but fully visible in normal room light
            with 0.1ma! This is so low that it won't load down the circuit under test.

            --
            Lee A. Hart | Ring the bells that still can ring
            814 8th Ave N | Forget the perfect offering
            Sartell MN 56377 | There is a crack in everything
            leeahart earthlink.net | That's how the light gets in -- Leonard Cohen
          • sbirdasn
            comments inline... ... ... The VIP is expandable, but not with the kind of flexibility of traditional ELF s due to only RESET/RUN capability and
            Message 5 of 28 , Jan 31, 2011
            • 0 Attachment
              comments inline...

              --- In cosmacelf@yahoogroups.com, Jim Hall <jehall1442596@...> wrote:
              <snip>
              > say what is the best interface chip to see what the bus or even the 1802
              > chip is doing?
              > I have a small bread board and like to start learning more about my VIP.
              </snip>

              The VIP is expandable, but not with the kind of flexibility of traditional ELF's due to only RESET/RUN capability and the requirement to access memory/registers via the limited ROM monitor.

              Since the VIP can't single-step/or pause (without modification), there's limited use for a full bus display capability add-on circuit. Note that you can get a dump of the internal register file (excepting R0-R2) via the Monitor, but you must do that as the first thing from entering the Monitor after resetting the VIP from whatever program you were running at the time. In a pinch, it is useful for debugging though.

              Your choices are as follows:

              1) Glomp onto various individual chips/signals, or use the expansion bus and make a plug-in card that does some kind of signal display functionality.

              2) Use test equipment like oscilloscopes or logic probes to examine signals of interest (mostly useful for debugging hardware, not software).

              3) Get a logic analyzer (old ones are relatively cheap on fleabay these days) and use it to give you a bus trace up to the limits of the capture memory. The main issue here is that you probably will not have an 1802 decode pod that gives you mnemonic decoded displays.

              4) Invent something of your own that will do some of the above. If you're not a skilled hardware/software developer, then you'll have a hard time with this one.

              Alas, most of us VIP users simply tough it out and debug the old fashioned way-- study our code, set "break points" (branch to itself), and use the monitor dump capability when possible.

              Oh, and keep a logic probe/scope handy and use the Q output or the output port to latch a signal to give clues as to what is happening for debugging purposes.

              Sbirdasn
            • sbirdasn
              Speaking of which-- I ve been toying with the idea of building a software debugging tool using Parallax s Propeller CPU s. I initially thought that a real-time
              Message 6 of 28 , Jan 31, 2011
              • 0 Attachment
                Speaking of which--

                I've been toying with the idea of building a software debugging tool using Parallax's Propeller CPU's. I initially thought that a real-time hardware ICE might be doable, but I've now come to the conclusion that it isn't possible at full 1802 speed with this chip (maybe 2nd gen *might* barely do it).

                However, I've recently kicked the idea around in my head that I *could* do a software trace bus sniffer at speeds up to about 2MHz (Pixie speed) with the Propeller chip.

                It would essentially be a poor man's logic analyzer with 1802 pod functionality. There would be a trace buffer, and some limited triggering to stop the sniffing. It would fake the register contents by emulating each instruction sniffed and update its own register file/D register in parallel with the real 1802 as the program runs.

                As long as there are no hardware faults or getting out of sync, the register contents should match. The Propeller would always have a video display that would show the present (emulated) contents of the registers (updated 60 times/second). When a trigger condition was met, then the register display updates would stop with the final contents after the trigger event. Then you could scroll through the trace memory to see what happened.

                Once the sniffing stopped, the trace display could get quite sophisticated.

                The total hardware to make such a gadget should be pretty minimal, even if you were careful and did full 5V-3.3V level conversion, the chip count should be less than 10.

                The down side to my little thought experiment is that I'm not a Propeller expert, so I figure there's a lot of development time to get something useful.

                Just a thought.

                Sbirdasn.
              • Jim Hall
                Hello: I have been tryurying to spark interest in this group and I am amazed on the help I am getting. Once more I am looking at all and everyones suggestions
                Message 7 of 28 , Jan 31, 2011
                • 0 Attachment
                  Hello:

                  I have been tryurying to spark interest in this group and I am amazed on the
                  help I am getting. Once more I am looking at all and everyones suggestions on
                  how to understand the VIP I have. I am going to try my best to keep up with it,
                  and I just hope I can look for the LED's you are talking about, and see what I
                  can cook up.


                  Could some kind of a crude latch grab the signal I am looking at and hold it,
                  send it to the LED ? I am also wondering if that crude latch could access the
                  memory lines and see what is there as well ? Forgive me if I am dense or not
                  asking the right question, I am trying to start over again after 25plus years of
                  working with a IBM and not thinking of what electronics do .


                  jim




                  ________________________________
                  From: Lee Hart <leeahart@...>
                  To: cosmacelf@yahoogroups.com
                  Sent: Mon, January 31, 2011 11:49:39 PM
                  Subject: Re: [cosmacelf] Re: Is there any hardware junkies in this group

                   
                  Jim Hall wrote:
                  > I just want to be sure I read my mail and answer everything as much as my old
                  > brain can take. In the distant past of the 1802 they user to buffer alot of
                  the
                  > work with CMOS 4050's and CMOS 4049's so that nothing would go wrong, and now
                  >we
                  > are saying a simple resistor and LED can do the "seeing" of the logic there?

                  Yes. In the 1970's an LED needed 10ma or so for reasonably brightness.
                  Today's super-bright LEDs are dim but fully visible in normal room light
                  with 0.1ma! This is so low that it won't load down the circuit under test.

                  --
                  Lee A. Hart | Ring the bells that still can ring
                  814 8th Ave N | Forget the perfect offering
                  Sartell MN 56377 | There is a crack in everything
                  leeahart earthlink.net | That's how the light gets in -- Leonard Cohen






                  [Non-text portions of this message have been removed]
                • Jim Hall
                  Hello: As I have been out of sink for more than 25yrs doing hardware projects, please explain what a paralax propeller does ? jim (gone looking on the net to
                  Message 8 of 28 , Jan 31, 2011
                  • 0 Attachment
                    Hello:

                    As I have been out of sink for more than 25yrs doing hardware projects, please
                    explain what a 'paralax propeller' does ?


                    jim
                    (gone looking on the net to find out about it right now ! )

                     



                    ________________________________
                    From: sbirdasn <sbirdasn@...>
                    To: cosmacelf@yahoogroups.com
                    Sent: Tue, February 1, 2011 12:17:45 AM
                    Subject: [cosmacelf] Re: Is there any hardware junkies in this group

                     


                    Speaking of which--

                    I've been toying with the idea of building a software debugging tool using
                    Parallax's Propeller CPU's. I initially thought that a real-time hardware ICE
                    might be doable, but I've now come to the conclusion that it isn't possible at
                    full 1802 speed with this chip (maybe 2nd gen *might* barely do it).

                    However, I've recently kicked the idea around in my head that I *could* do a
                    software trace bus sniffer at speeds up to about 2MHz (Pixie speed) with the
                    Propeller chip.

                    It would essentially be a poor man's logic analyzer with 1802 pod functionality.
                    There would be a trace buffer, and some limited triggering to stop the sniffing.
                    It would fake the register contents by emulating each instruction sniffed and
                    update its own register file/D register in parallel with the real 1802 as the
                    program runs.

                    As long as there are no hardware faults or getting out of sync, the register
                    contents should match. The Propeller would always have a video display that
                    would show the present (emulated) contents of the registers (updated 60
                    times/second). When a trigger condition was met, then the register display
                    updates would stop with the final contents after the trigger event. Then you
                    could scroll through the trace memory to see what happened.

                    Once the sniffing stopped, the trace display could get quite sophisticated.

                    The total hardware to make such a gadget should be pretty minimal, even if you
                    were careful and did full 5V-3.3V level conversion, the chip count should be
                    less than 10.

                    The down side to my little thought experiment is that I'm not a Propeller
                    expert, so I figure there's a lot of development time to get something useful.

                    Just a thought.

                    Sbirdasn.







                    [Non-text portions of this message have been removed]
                  • Lee Hart
                    ... Yes, you can add a latch, and trigger it at the right instant to capture (for example), the last high or low byte of the address, data bus, status of the
                    Message 9 of 28 , Feb 1, 2011
                    • 0 Attachment
                      Jim Hall wrote:
                      > Could some kind of a crude latch grab the signal I am looking at and hold it,
                      > send it to the LED? I am also wondering if that crude latch could access the
                      > memory lines and see what is there as well? Forgive me if I am dense or not
                      > asking the right question, I am trying to start over again after 25plus years of
                      > working with a IBM and not thinking of what electronics do.

                      Yes, you can add a latch, and trigger it at the right instant to capture
                      (for example), the last high or low byte of the address, data bus,
                      status of the control signals (read, write, instruction fetch, DMA
                      cycle, etc.) But this tends to take a lot of parts, and requires many
                      connections into the VIP logic. So, a lot of time is involved in hooking
                      it all up.

                      But, the 1802 is *so* simple that this sort of conservative approach
                      used with other microprocessors is overkill. After you work with the
                      1802 a bit, you will have "Aha!" moments when you realized there are
                      much easier ways to do it.

                      First, you can do it all in software. I'm fairly sure there are better
                      monitor programs for the VIP that will single-step a program in
                      software; The program to test is put in RAM. Let's say you want to
                      execute the first byte of the program at 1000h. The monitor program:

                      - looks at the instruction at 1000h to see if it is a 1, 2, or 3-byte
                      instruction.
                      - looks at the address of the *next* instruction after this one.
                      - saves the byte at that location, and writes a "return to monitor"
                      instruction in its place. This is called a "software breakpoint".
                      - The monitor then executes the instruction at 1000h.
                      - Just one instruction executes; then it returns to the monitor.
                      - The monitor can save all the registers, display their contents,
                      etc.
                      - Repeat: The monitor can restore the original instruction at (say)
                      1001h, write the software breakpoint into 1002h, and do it again.

                      In this way, you can single step through a program one instruction at a
                      time, and display the results after each step. All with no additional
                      hardware at all!

                      Or, if the problem is that the 1802 has hardware problems, you can add
                      hardware single-stepping. RCA has the simple circuit for this in MPM-201
                      (User manual for the COSMAC 1802 microprocessor).

                      Or, since the 1802 is static (clock frequency can be zero), it is even
                      simpler to add a circuit to over-ride the clock so it is controlled with
                      a switch. All the signals then "freeze" right where they are, so you can
                      easily view them with a simple LED+resistor.

                      Does this make sense?
                      --
                      Lee A. Hart | Ring the bells that still can ring
                      814 8th Ave N | Forget the perfect offering
                      Sartell MN 56377 | There is a crack in everything
                      leeahart earthlink.net | That's how the light gets in -- Leonard Cohen
                    • ted_rossin
                      If you have about $10 you could build my cheap logic analyzer to get a good look at the 1802 signals:
                      Message 10 of 28 , Feb 1, 2011
                      • 0 Attachment
                        If you have about $10 you could build my cheap logic analyzer to get a good look at the 1802 signals:

                        http://www.tedrossin.0sites.net/Electronics/Pic/Pic.html#LogicAnalyzer

                        This will will run at up to 5M samples/second and also give you 5 analog channels to look at voltages if you want. I used this to help debug my Elf Clone:

                        http://www.tedrossin.0sites.net/Electronics/RCA/RCA.html#ElfClone

                        The only hitch is that you need to program the PIC microcontroller. The programmer costs about $35 but if you send me a part I can program it up for you. Just pay the return postage and I'd be happy to do it.

                        --- In cosmacelf@yahoogroups.com, "sbirdasn" <sbirdasn@...> wrote:
                        >
                        >
                        >
                        > Speaking of which--
                        >
                        > I've been toying with the idea of building a software debugging tool using Parallax's Propeller CPU's. I initially thought that a real-time hardware ICE might be doable, but I've now come to the conclusion that it isn't possible at full 1802 speed with this chip (maybe 2nd gen *might* barely do it).
                        >
                        > However, I've recently kicked the idea around in my head that I *could* do a software trace bus sniffer at speeds up to about 2MHz (Pixie speed) with the Propeller chip.
                        >
                        > It would essentially be a poor man's logic analyzer with 1802 pod functionality. There would be a trace buffer, and some limited triggering to stop the sniffing. It would fake the register contents by emulating each instruction sniffed and update its own register file/D register in parallel with the real 1802 as the program runs.
                        >
                        > As long as there are no hardware faults or getting out of sync, the register contents should match. The Propeller would always have a video display that would show the present (emulated) contents of the registers (updated 60 times/second). When a trigger condition was met, then the register display updates would stop with the final contents after the trigger event. Then you could scroll through the trace memory to see what happened.
                        >
                        > Once the sniffing stopped, the trace display could get quite sophisticated.
                        >
                        > The total hardware to make such a gadget should be pretty minimal, even if you were careful and did full 5V-3.3V level conversion, the chip count should be less than 10.
                        >
                        > The down side to my little thought experiment is that I'm not a Propeller expert, so I figure there's a lot of development time to get something useful.
                        >
                        > Just a thought.
                        >
                        > Sbirdasn.
                        >
                      • Jim Hall
                        Hello: Thanks for the great information on interfacing to the 1802, I will try some of the basic ideas as well as dust off some of my electronic thinking, and
                        Message 11 of 28 , Feb 1, 2011
                        • 0 Attachment
                          Hello:
                          Thanks for the great information on interfacing to the 1802, I will try some of
                          the basic ideas as well as dust off some of my electronic thinking, and hope I
                          have a brief moment of AAHHHHHH when I find what I am loooking for.


                          jim




                          ________________________________
                          From: Lee Hart <leeahart@...>
                          To: cosmacelf@yahoogroups.com
                          Sent: Tue, February 1, 2011 1:08:06 PM
                          Subject: Re: [cosmacelf] Re: Is there any hardware junkies in this group

                           
                          Jim Hall wrote:
                          > Could some kind of a crude latch grab the signal I am looking at and hold it,
                          > send it to the LED? I am also wondering if that crude latch could access the
                          > memory lines and see what is there as well? Forgive me if I am dense or not
                          > asking the right question, I am trying to start over again after 25plus years
                          >of
                          > working with a IBM and not thinking of what electronics do.

                          Yes, you can add a latch, and trigger it at the right instant to capture
                          (for example), the last high or low byte of the address, data bus,
                          status of the control signals (read, write, instruction fetch, DMA
                          cycle, etc.) But this tends to take a lot of parts, and requires many
                          connections into the VIP logic. So, a lot of time is involved in hooking
                          it all up.

                          But, the 1802 is *so* simple that this sort of conservative approach
                          used with other microprocessors is overkill. After you work with the
                          1802 a bit, you will have "Aha!" moments when you realized there are
                          much easier ways to do it.

                          First, you can do it all in software. I'm fairly sure there are better
                          monitor programs for the VIP that will single-step a program in
                          software; The program to test is put in RAM. Let's say you want to
                          execute the first byte of the program at 1000h. The monitor program:

                          - looks at the instruction at 1000h to see if it is a 1, 2, or 3-byte
                          instruction.
                          - looks at the address of the *next* instruction after this one.
                          - saves the byte at that location, and writes a "return to monitor"
                          instruction in its place. This is called a "software breakpoint".
                          - The monitor then executes the instruction at 1000h.
                          - Just one instruction executes; then it returns to the monitor.
                          - The monitor can save all the registers, display their contents,
                          etc.
                          - Repeat: The monitor can restore the original instruction at (say)
                          1001h, write the software breakpoint into 1002h, and do it again.

                          In this way, you can single step through a program one instruction at a
                          time, and display the results after each step. All with no additional
                          hardware at all!

                          Or, if the problem is that the 1802 has hardware problems, you can add
                          hardware single-stepping. RCA has the simple circuit for this in MPM-201
                          (User manual for the COSMAC 1802 microprocessor).

                          Or, since the 1802 is static (clock frequency can be zero), it is even
                          simpler to add a circuit to over-ride the clock so it is controlled with
                          a switch. All the signals then "freeze" right where they are, so you can
                          easily view them with a simple LED+resistor.

                          Does this make sense?
                          --
                          Lee A. Hart | Ring the bells that still can ring
                          814 8th Ave N | Forget the perfect offering
                          Sartell MN 56377 | There is a crack in everything
                          leeahart earthlink.net | That's how the light gets in -- Leonard Cohen






                          [Non-text portions of this message have been removed]
                        • Jim Hall
                          Hello: Im once again thankful for any kind of informaton on this great chip, I have been busy dusting off my electronic smarts and thinking of spending my time
                          Message 12 of 28 , Feb 1, 2011
                          • 0 Attachment
                            Hello:

                            Im once again thankful for any kind of informaton on this great chip, I have
                            been busy dusting off my electronic smarts and thinking of spending my time
                            learning LOGIC and the basic's. Gee Spock would liket hat !


                            jim




                            ________________________________
                            From: ted_rossin <ted_rossin@...>
                            To: cosmacelf@yahoogroups.com
                            Sent: Tue, February 1, 2011 3:54:20 PM
                            Subject: [cosmacelf] Re: Is there any hardware junkies in this group

                             


                            If you have about $10 you could build my cheap logic analyzer to get a good look
                            at the 1802 signals:

                            http://www.tedrossin.0sites.net/Electronics/Pic/Pic.html#LogicAnalyzer

                            This will will run at up to 5M samples/second and also give you 5 analog
                            channels to look at voltages if you want. I used this to help debug my Elf
                            Clone:

                            http://www.tedrossin.0sites.net/Electronics/RCA/RCA.html#ElfClone

                            The only hitch is that you need to program the PIC microcontroller. The
                            programmer costs about $35 but if you send me a part I can program it up for
                            you. Just pay the return postage and I'd be happy to do it.

                            --- In cosmacelf@yahoogroups.com, "sbirdasn" <sbirdasn@...> wrote:
                            >
                            >
                            >
                            > Speaking of which--
                            >
                            > I've been toying with the idea of building a software debugging tool using
                            >Parallax's Propeller CPU's. I initially thought that a real-time hardware ICE
                            >might be doable, but I've now come to the conclusion that it isn't possible at
                            >full 1802 speed with this chip (maybe 2nd gen *might* barely do it).
                            >
                            > However, I've recently kicked the idea around in my head that I *could* do a
                            >software trace bus sniffer at speeds up to about 2MHz (Pixie speed) with the
                            >Propeller chip.
                            >
                            > It would essentially be a poor man's logic analyzer with 1802 pod
                            >functionality. There would be a trace buffer, and some limited triggering to
                            >stop the sniffing. It would fake the register contents by emulating each
                            >instruction sniffed and update its own register file/D register in parallel with
                            >the real 1802 as the program runs.
                            >
                            > As long as there are no hardware faults or getting out of sync, the register
                            >contents should match. The Propeller would always have a video display that
                            >would show the present (emulated) contents of the registers (updated 60
                            >times/second). When a trigger condition was met, then the register display
                            >updates would stop with the final contents after the trigger event. Then you
                            >could scroll through the trace memory to see what happened.
                            >
                            > Once the sniffing stopped, the trace display could get quite sophisticated.
                            >
                            > The total hardware to make such a gadget should be pretty minimal, even if you
                            >were careful and did full 5V-3.3V level conversion, the chip count should be
                            >less than 10.
                            >
                            > The down side to my little thought experiment is that I'm not a Propeller
                            >expert, so I figure there's a lot of development time to get something useful.
                            >
                            > Just a thought.
                            >
                            > Sbirdasn.
                            >







                            [Non-text portions of this message have been removed]
                          • Vince
                            ... I was going to write 25 yrs, I ve got the perfect example until I realized my example is rapidly pushing 40 yrs old... Err I guess its not so much my
                            Message 13 of 28 , Feb 1, 2011
                            • 0 Attachment
                              --- In cosmacelf@yahoogroups.com, Jim Hall <jehall1442596@...> wrote:
                              >
                              > Hello:
                              >
                              > As I have been out of sink for more than 25yrs doing hardware projects, please
                              > explain what a 'paralax propeller' does ?
                              >
                              >
                              > jim

                              I was going to write "25 yrs, I've got the perfect example" until I realized my example is rapidly pushing 40 yrs old... Err I guess its not so much my example getting old as it remains static in the late 60s/70s, but me whos gettin old...

                              Anyway, you know what a CDC 6000 series PPU does? The whole "virtual ten processors" deal? Imagine all that in a single chip with some memory but without the attached supercomputer of course. And at least until very recently only programmable on a windows PC (so of little interest to me). A nifty, crazy, unique and lovable microcontroller architecture.

                              Everything old is new again! Can't wait till they reinvent channels, VTAM, and bus -n- tag, thats gonna be fun.

                              Vince
                            • sbirdasn
                              Though you may not have been on the forum before, we ve discussed this chip once or twice, as it s a chip now familiar to the hobbyist/hardware hacker
                              Message 14 of 28 , Feb 1, 2011
                              • 0 Attachment
                                Though you may not have been on the forum before, we've discussed this chip once or twice, as it's a chip now familiar to the hobbyist/hardware hacker community.

                                Just google with keywords: Parallax Propeller

                                I only throw out this chip since it has some really peculiar features that allow low chip count projects.

                                I really don't *need* such a tool, as I have a Tektronix 338 32-channel x 256 sample logic analyzer, and a collection of oscilloscopes (analog & DSO) (and LP-1 logic probe) that are more than capable for 1802 development/debug.

                                I ponder building Propeller projects like an 1802 sniffer with the idea that I might be able to have a tool that has features never dreamed of back in the old days, while not costing a mint to build, that others might enjoy making for their own use too.

                                I think that having a "live" display of the 1802 register set would be the real key selling point for having such a gadget.

                                I remember using a Z80 ICE (In-Circuit Emulator) back in the day that did show live display, but the output was only the address, data, and maybe one register at a time along with the status byte being monitored.

                                An 1802 with its large register file would have been a real challenge for ICE live register display back when the 1802 was in current development.

                                Also, the best of RCA's 1800 development tools are rare and fetch rather stiff prices on "fleabay", and still don't have the abilities I could put into my own little debugger project.

                                Sbirdasn.
                              • sbirdasn
                                That little PIC logic analyzer is an interesting solution to sniffing a few lines. The main issue here is that for both the PIC and the Propeller MPU s, it
                                Message 15 of 28 , Feb 1, 2011
                                • 0 Attachment
                                  That little PIC logic analyzer is an interesting solution to sniffing a few lines.

                                  The main issue here is that for both the PIC and the Propeller MPU's, it takes an additional development tool set for playing with the new devices, just to do the debugging/learning about the target 1802 system.

                                  For someone just getting started again, that's a pretty messy solution.

                                  I think that the original poster is best served by keeping things simple. The biggest problem is that the VIP is *not* designed for single-stepping and displaying the address/data bus live while the user walks through the program.

                                  Yes, a VIP *could* be modified to single-step, but glomping on full address/data bus sniffing display LED's is a lot of wires for a computer that doesn't benefit much from it unless modified.

                                  It would be easier to instead build a second computer as a basic ELF from scratch (wire-wrap since OP mentioned those skills) and add a bank of discrete LED's to see address/control lines as he sees the need for it.

                                  Also, a few medium size solder-less breadboards would work for a temporary ELF project, and be reusable for other circuits once comfortable enough with 1802 code to go back to the VIP.

                                  The basic ELF's Hex display with the built-in single step & load functionality goes a long way to understanding how the 1802 works without even an address display.

                                  Yes, the Super Elf does have the address display option, but that adds at least 25-35% more wiring to the project if not more. And the power requirements jump up substantially as a result.

                                  If the OP really wants live address/data display on his *VIP*, then he should buy a 22/44 pin edge card proto board large enough to mount the required latch/buffer/display chips on it, and plug it into the expansion bus of the VIP.

                                  The decision to do hex vs discrete LED's will decide how much wiring is needed for the display board project, and the total cost.

                                  Single-stepping modifications don't *have* to be made if branch/idle instructions are used to halt program execution during debugging.

                                  Sbirdasn.
                                • Lee Hart
                                  ... Right. You can *simulate* it in software, by walking a breakpoint through a program in RAM, stopping after each instruction and displaying the address and
                                  Message 16 of 28 , Feb 1, 2011
                                  • 0 Attachment
                                    sbirdasn wrote:
                                    > I think that the original poster is best served by keeping things
                                    > simple. The biggest problem is that the VIP is *not* designed for
                                    > single-stepping and displaying the address/data bus live while the
                                    > user walks through the program.

                                    Right. You can *simulate* it in software, by walking a breakpoint
                                    through a program in RAM, stopping after each instruction and displaying
                                    the address and data on the video screen. (I know I've heard of such a
                                    program for the VIP, but don't have it myself).

                                    > Yes, a VIP *could* be modified to single-step

                                    Yes; that's pretty easy. Basically, you defeat the crystal clock, and
                                    use your own very slow clock. It could even be a debounced pushbutton.

                                    > glomping on full address/data bus sniffing display LED's is a lot of
                                    > wires for a computer that doesn't benefit much from it unless modified.

                                    Absolutely. That's why I suggested just making some DIP clips that show
                                    the state of the address and data buses when clipped onto the
                                    appropriate chips. Address lines A0-A11 are already latched and available.

                                    As built, the VIP wires 1802 /WAIT (pin 2) high. Add a switch or some
                                    gate to pull this pin low, and the 1802 will freeze in place, so you can
                                    examine all the lines with LEDs.

                                    > If the OP really wants live address/data display on his *VIP*, then
                                    > he should buy a 22/44 pin edge card proto board large enough to mount
                                    > the required latch/buffer/display chips on it, and plug it into the
                                    > expansion bus of the VIP.

                                    That could work; but not all the signals needed are in the expansion
                                    connectors.

                                    --
                                    Lee A. Hart | Ring the bells that still can ring
                                    814 8th Ave N | Forget the perfect offering
                                    Sartell MN 56377 | There is a crack in everything
                                    leeahart earthlink.net | That's how the light gets in -- Leonard Cohen
                                  • Jim Hall
                                    Hello: Dont get me wrong in how simple the 1802 for the time, and how it came about, but if I was going to do something special to make the 1802 just a bit
                                    Message 17 of 28 , Feb 1, 2011
                                    • 0 Attachment
                                      Hello:
                                      Dont get me wrong in how simple the 1802 for the time, and how it came about,
                                      but if I was going to do something special to make the 1802 just a bit better, I
                                      would have a bread board area, spots for more memory and a place for more ROM. I
                                      like the idea of having CHIP-8 in ROM, maybe a area for games and something
                                      better than a hex keyboard.


                                      Am I asking for too much here?

                                      I am attaching a picture of something I wish I had,
                                      jim




                                      ________________________________
                                      From: Lee Hart <leeahart@...>
                                      To: cosmacelf@yahoogroups.com
                                      Sent: Tue, February 1, 2011 11:56:10 PM
                                      Subject: Re: [cosmacelf] Re: Is there any hardware junkies in this group

                                       
                                      sbirdasn wrote:
                                      > I think that the original poster is best served by keeping things
                                      > simple. The biggest problem is that the VIP is *not* designed for
                                      > single-stepping and displaying the address/data bus live while the
                                      > user walks through the program.

                                      Right. You can *simulate* it in software, by walking a breakpoint
                                      through a program in RAM, stopping after each instruction and displaying
                                      the address and data on the video screen. (I know I've heard of such a
                                      program for the VIP, but don't have it myself).

                                      > Yes, a VIP *could* be modified to single-step

                                      Yes; that's pretty easy. Basically, you defeat the crystal clock, and
                                      use your own very slow clock. It could even be a debounced pushbutton.

                                      > glomping on full address/data bus sniffing display LED's is a lot of
                                      > wires for a computer that doesn't benefit much from it unless modified.

                                      Absolutely. That's why I suggested just making some DIP clips that show
                                      the state of the address and data buses when clipped onto the
                                      appropriate chips. Address lines A0-A11 are already latched and available.

                                      As built, the VIP wires 1802 /WAIT (pin 2) high. Add a switch or some
                                      gate to pull this pin low, and the 1802 will freeze in place, so you can
                                      examine all the lines with LEDs.

                                      > If the OP really wants live address/data display on his *VIP*, then
                                      > he should buy a 22/44 pin edge card proto board large enough to mount
                                      > the required latch/buffer/display chips on it, and plug it into the
                                      > expansion bus of the VIP.

                                      That could work; but not all the signals needed are in the expansion
                                      connectors.

                                      --
                                      Lee A. Hart | Ring the bells that still can ring
                                      814 8th Ave N | Forget the perfect offering
                                      Sartell MN 56377 | There is a crack in everything
                                      leeahart earthlink.net | That's how the light gets in -- Leonard Cohen






                                      [Non-text portions of this message have been removed]
                                    • Jim Hall
                                      Hello: I dont want to beat this subject to death, but I like simplicity of things, but I like to have some kind smarts in what I am seeing. I will have to look
                                      Message 18 of 28 , Feb 1, 2011
                                      • 0 Attachment
                                        Hello:

                                        I dont want to beat this subject to death, but I like simplicity of things, but
                                        I like to have some kind smarts in what I am seeing. I will have to look into
                                        what could be done and maybe try it.


                                        jim

                                         



                                        ________________________________
                                        From: sbirdasn <sbirdasn@...>
                                        To: cosmacelf@yahoogroups.com
                                        Sent: Tue, February 1, 2011 10:52:21 PM
                                        Subject: [cosmacelf] Re: Is there any hardware junkies in this group

                                         


                                        That little PIC logic analyzer is an interesting solution to sniffing a few
                                        lines.

                                        The main issue here is that for both the PIC and the Propeller MPU's, it takes
                                        an additional development tool set for playing with the new devices, just to do
                                        the debugging/learning about the target 1802 system.

                                        For someone just getting started again, that's a pretty messy solution.

                                        I think that the original poster is best served by keeping things simple. The
                                        biggest problem is that the VIP is *not* designed for single-stepping and
                                        displaying the address/data bus live while the user walks through the program.

                                        Yes, a VIP *could* be modified to single-step, but glomping on full address/data
                                        bus sniffing display LED's is a lot of wires for a computer that doesn't benefit
                                        much from it unless modified.

                                        It would be easier to instead build a second computer as a basic ELF from
                                        scratch (wire-wrap since OP mentioned those skills) and add a bank of discrete
                                        LED's to see address/control lines as he sees the need for it.

                                        Also, a few medium size solder-less breadboards would work for a temporary ELF
                                        project, and be reusable for other circuits once comfortable enough with 1802
                                        code to go back to the VIP.

                                        The basic ELF's Hex display with the built-in single step & load functionality
                                        goes a long way to understanding how the 1802 works without even an address
                                        display.

                                        Yes, the Super Elf does have the address display option, but that adds at least
                                        25-35% more wiring to the project if not more. And the power requirements jump
                                        up substantially as a result.

                                        If the OP really wants live address/data display on his *VIP*, then he should
                                        buy a 22/44 pin edge card proto board large enough to mount the required
                                        latch/buffer/display chips on it, and plug it into the expansion bus of the VIP.

                                        The decision to do hex vs discrete LED's will decide how much wiring is needed
                                        for the display board project, and the total cost.

                                        Single-stepping modifications don't *have* to be made if branch/idle
                                        instructions are used to halt program execution during debugging.

                                        Sbirdasn.







                                        [Non-text portions of this message have been removed]
                                      • Jim Hall
                                        Hello: Sorry I am the new guy on the block, and just dusted of my VIP and started reading the VIPER s that I have found after all the years of sitting there
                                        Message 19 of 28 , Feb 1, 2011
                                        • 0 Attachment
                                          Hello:

                                          Sorry I am the new guy on the block, and just dusted of my VIP and started
                                          reading the VIPER's that I have found after all the years of sitting there
                                          gathering dust. I will have to get back to you and see if I can get the supplies
                                          I need. I like wire wrapping and bread boards as well, and have time to spend on
                                          something that will help me pass the time, being retired for only a few months.

                                          jim




                                          ________________________________
                                          From: Vince <vince@...>
                                          To: cosmacelf@yahoogroups.com
                                          Sent: Tue, February 1, 2011 7:43:59 PM
                                          Subject: [cosmacelf] Re: Is there any hardware junkies in this group

                                           
                                          --- In cosmacelf@yahoogroups.com, Jim Hall <jehall1442596@...> wrote:
                                          >
                                          > Hello:
                                          >
                                          > As I have been out of sink for more than 25yrs doing hardware projects, please

                                          > explain what a 'paralax propeller' does ?
                                          >
                                          >
                                          > jim

                                          I was going to write "25 yrs, I've got the perfect example" until I realized my
                                          example is rapidly pushing 40 yrs old... Err I guess its not so much my example
                                          getting old as it remains static in the late 60s/70s, but me whos gettin old...

                                          Anyway, you know what a CDC 6000 series PPU does? The whole "virtual ten
                                          processors" deal? Imagine all that in a single chip with some memory but without
                                          the attached supercomputer of course. And at least until very recently only
                                          programmable on a windows PC (so of little interest to me). A nifty, crazy,
                                          unique and lovable microcontroller architecture.

                                          Everything old is new again! Can't wait till they reinvent channels, VTAM, and
                                          bus -n- tag, thats gonna be fun.

                                          Vince







                                          [Non-text portions of this message have been removed]
                                        • awasson2001
                                          Unfortunately the attachment didn t make it through (at least not to me). Maybe upload it to the photos or files section. Just to satisfy intellectual
                                          Message 20 of 28 , Feb 1, 2011
                                          • 0 Attachment
                                            Unfortunately the attachment didn't make it through (at least not to me). Maybe upload it to the photos or files section.

                                            Just to satisfy intellectual curiosity I'm keen to try out some clips with LED's that Lee was describing but I'd really like to see something along the lines of what Sbirdasn was describing with hardware and a row of LEDs to select and display the register contents. That would be cool in a blinken lights kind of way. Not that it would be necessary but it's always fun to watch blinking LEDs on our old hardware.

                                            Andrew

                                            --- In cosmacelf@yahoogroups.com, Jim Hall <jehall1442596@...> wrote:
                                            >
                                            > Hello:
                                            > Dont get me wrong in how simple the 1802 for the time, and how it came about,
                                            > but if I was going to do something special to make the 1802 just a bit better, I
                                            > would have a bread board area, spots for more memory and a place for more ROM. I
                                            > like the idea of having CHIP-8 in ROM, maybe a area for games and something
                                            > better than a hex keyboard.
                                            >
                                            >
                                            > Am I asking for too much here?
                                            >
                                            > I am attaching a picture of something I wish I had,
                                            > jim
                                            >
                                            >
                                            >
                                            >
                                            > ________________________________
                                            > From: Lee Hart <leeahart@...>
                                            > To: cosmacelf@yahoogroups.com
                                            > Sent: Tue, February 1, 2011 11:56:10 PM
                                            > Subject: Re: [cosmacelf] Re: Is there any hardware junkies in this group
                                            >
                                            >  
                                            > sbirdasn wrote:
                                            > > I think that the original poster is best served by keeping things
                                            > > simple. The biggest problem is that the VIP is *not* designed for
                                            > > single-stepping and displaying the address/data bus live while the
                                            > > user walks through the program.
                                            >
                                            > Right. You can *simulate* it in software, by walking a breakpoint
                                            > through a program in RAM, stopping after each instruction and displaying
                                            > the address and data on the video screen. (I know I've heard of such a
                                            > program for the VIP, but don't have it myself).
                                            >
                                            > > Yes, a VIP *could* be modified to single-step
                                            >
                                            > Yes; that's pretty easy. Basically, you defeat the crystal clock, and
                                            > use your own very slow clock. It could even be a debounced pushbutton.
                                            >
                                            > > glomping on full address/data bus sniffing display LED's is a lot of
                                            > > wires for a computer that doesn't benefit much from it unless modified.
                                            >
                                            > Absolutely. That's why I suggested just making some DIP clips that show
                                            > the state of the address and data buses when clipped onto the
                                            > appropriate chips. Address lines A0-A11 are already latched and available.
                                            >
                                            > As built, the VIP wires 1802 /WAIT (pin 2) high. Add a switch or some
                                            > gate to pull this pin low, and the 1802 will freeze in place, so you can
                                            > examine all the lines with LEDs.
                                            >
                                            > > If the OP really wants live address/data display on his *VIP*, then
                                            > > he should buy a 22/44 pin edge card proto board large enough to mount
                                            > > the required latch/buffer/display chips on it, and plug it into the
                                            > > expansion bus of the VIP.
                                            >
                                            > That could work; but not all the signals needed are in the expansion
                                            > connectors.
                                            >
                                            > --
                                            > Lee A. Hart | Ring the bells that still can ring
                                            > 814 8th Ave N | Forget the perfect offering
                                            > Sartell MN 56377 | There is a crack in everything
                                            > leeahart earthlink.net | That's how the light gets in -- Leonard Cohen
                                            >
                                            >
                                            >
                                            >
                                            >
                                            >
                                            > [Non-text portions of this message have been removed]
                                            >
                                          • Lee Hart
                                            ... Not to me either. I don t think this list accepts attachments (too high a risk of malware). ... Well, here s how I did it on my very first ELF. Everything
                                            Message 21 of 28 , Feb 2, 2011
                                            • 0 Attachment
                                              awasson2001 wrote:
                                              > Unfortunately the attachment didn't make it through (at least not to
                                              > me). Maybe upload it to the photos or files section.

                                              Not to me either. I don't think this list accepts attachments (too high
                                              a risk of malware).

                                              > Just to satisfy intellectual curiosity I'm keen to try out some clips
                                              > with LED's that Lee was describing but I'd really like to see
                                              > something along the lines of what Sbirdasn was describing with
                                              > hardware and a row of LEDs to select and display the register
                                              > contents. That would be cool in a blinken lights kind of way. Not
                                              > that it would be necessary but it's always fun to watch blinking LEDs
                                              > on our old hardware.

                                              Well, here's how I did it on my very first ELF.

                                              Everything was on separate 4.5" x 6.5" cards; CPU, RAM, ROM, I/O etc.
                                              Each had a 44-pin edge connector. They all plugged into a bus that
                                              connected all pin 1-s together, all pin 2's etc.

                                              One of the cards was a front panel. It had the usual toggle switches and
                                              LEDs for control and data. The data was captured by two 4042 quad
                                              latches and displayed by 8 LEDs. The address was captured by four 4042
                                              4-bit latches, and displayed by 16 LEDs. The high byte address was
                                              latched by TPA, and the low byte by TPB. There were also 4049 buffers
                                              driving LEDs for the various control lines.

                                              The crystal clock was gated by a 4011 so it could be stopped. While
                                              stopped, a bounceless pushbutton could pulse the clock low to single
                                              step one clock pulse at a time (took 8 button pushes for each bus
                                              cycle). Like this (view with a fixed width font like Courier):

                                              +5v _ crystal oscillator (crystal,
                                              | ____|| ||____ resistor, 2 capacitors)
                                              > | ||_|| |
                                              10k > | _____ |
                                              Run __ > |__|U1A \ | _____
                                              S1 \__|_____|4011 )O__|__|U1B \
                                              Stop __ | |_____/ __|4011B )O____1802 clock
                                              | | +5v___ | |_____/
                                              gnd | STEP S2 \___|
                                              | gnd___ _|_
                                              D1__|/|__| ___ 0.01uF
                                              |\| | |
                                              ... | gnd
                                              Dn__|/|__|
                                              |\|

                                              S1 is the Run/Stop toggle switch. When up (Run), the 10k resistor pulls
                                              the lower input of U1A high, and the crystal oscillator runs. When S1 is
                                              down (Stop), it grounds the input of U1A, forcing its output high and
                                              the oscillator is stopped.

                                              S2 is the single-step pushbutton. It is normally high, letting the
                                              oscillator pass through U1B to the 1802. When you push S2, it grounds
                                              the lower input of U1B, forcing its output high. You normally stop the
                                              oscillator with S1 so the 1802 clock is low; then pressing S2 pulses the
                                              1802 clock high.

                                              The 0.01uf capacitor on C2 makes it a "bounceless" pushbutton. When the
                                              switch slowly moves from up to down, its contacts "bounce" open and
                                              closed many times. The capacitor holds the state of the last contact to
                                              have touched (up or down) despite the bouncing, until the switch has
                                              moved enough to actually touch the opposite contact. This instantly
                                              charges or discharges the capacitor, holding the new level until the
                                              switch stops bouncing.

                                              The diodes D1-Dn are a crude logic analyzer. They form a simple OR gate,
                                              where the cathodes of all the diodes must be high to Run. If you want to
                                              stop at the first write cycle, connect a diode cathode to MWR. The 4042
                                              quad latch has both true and inverted outputs, which means you can stop
                                              on any high or low on any address or data line as well.
                                              --
                                              Lee A. Hart | Ring the bells that still can ring
                                              814 8th Ave N | Forget the perfect offering
                                              Sartell MN 56377 | There is a crack in everything
                                              leeahart earthlink.net | That's how the light gets in -- Leonard Cohen
                                            • Jim Hall
                                              Hello: Sorry that I dont know how to upload to the file area, please advise me how , so I can upload the 1802 board I found that I like most. jim
                                              Message 22 of 28 , Feb 2, 2011
                                              • 0 Attachment
                                                Hello:

                                                Sorry that I dont know how to upload to the file area, please advise me how , so
                                                I can upload the 1802 board I found that I like most.
                                                jim




                                                ________________________________
                                                From: Lee Hart <leeahart@...>
                                                To: cosmacelf@yahoogroups.com
                                                Sent: Wed, February 2, 2011 1:18:21 PM
                                                Subject: Re: [cosmacelf] Re: Is there any hardware junkies in this group

                                                 
                                                awasson2001 wrote:
                                                > Unfortunately the attachment didn't make it through (at least not to
                                                > me). Maybe upload it to the photos or files section.

                                                Not to me either. I don't think this list accepts attachments (too high
                                                a risk of malware).

                                                > Just to satisfy intellectual curiosity I'm keen to try out some clips
                                                > with LED's that Lee was describing but I'd really like to see
                                                > something along the lines of what Sbirdasn was describing with
                                                > hardware and a row of LEDs to select and display the register
                                                > contents. That would be cool in a blinken lights kind of way. Not
                                                > that it would be necessary but it's always fun to watch blinking LEDs
                                                > on our old hardware.

                                                Well, here's how I did it on my very first ELF.

                                                Everything was on separate 4.5" x 6.5" cards; CPU, RAM, ROM, I/O etc.
                                                Each had a 44-pin edge connector. They all plugged into a bus that
                                                connected all pin 1-s together, all pin 2's etc.

                                                One of the cards was a front panel. It had the usual toggle switches and
                                                LEDs for control and data. The data was captured by two 4042 quad
                                                latches and displayed by 8 LEDs. The address was captured by four 4042
                                                4-bit latches, and displayed by 16 LEDs. The high byte address was
                                                latched by TPA, and the low byte by TPB. There were also 4049 buffers
                                                driving LEDs for the various control lines.

                                                The crystal clock was gated by a 4011 so it could be stopped. While
                                                stopped, a bounceless pushbutton could pulse the clock low to single
                                                step one clock pulse at a time (took 8 button pushes for each bus
                                                cycle). Like this (view with a fixed width font like Courier):

                                                +5v _ crystal oscillator (crystal,
                                                | ____|| ||____ resistor, 2 capacitors)
                                                > | ||_|| |
                                                10k > | _____ |
                                                Run __ > |__|U1A \ | _____
                                                S1 \__|_____|4011 )O__|__|U1B \
                                                Stop __ | |_____/ __|4011B )O____1802 clock
                                                | | +5v___ | |_____/
                                                gnd | STEP S2 \___|
                                                | gnd___ _|_
                                                D1__|/|__| ___ 0.01uF
                                                |\| | |
                                                ... | gnd
                                                Dn__|/|__|
                                                |\|

                                                S1 is the Run/Stop toggle switch. When up (Run), the 10k resistor pulls
                                                the lower input of U1A high, and the crystal oscillator runs. When S1 is
                                                down (Stop), it grounds the input of U1A, forcing its output high and
                                                the oscillator is stopped.

                                                S2 is the single-step pushbutton. It is normally high, letting the
                                                oscillator pass through U1B to the 1802. When you push S2, it grounds
                                                the lower input of U1B, forcing its output high. You normally stop the
                                                oscillator with S1 so the 1802 clock is low; then pressing S2 pulses the
                                                1802 clock high.

                                                The 0.01uf capacitor on C2 makes it a "bounceless" pushbutton. When the
                                                switch slowly moves from up to down, its contacts "bounce" open and
                                                closed many times. The capacitor holds the state of the last contact to
                                                have touched (up or down) despite the bouncing, until the switch has
                                                moved enough to actually touch the opposite contact. This instantly
                                                charges or discharges the capacitor, holding the new level until the
                                                switch stops bouncing.

                                                The diodes D1-Dn are a crude logic analyzer. They form a simple OR gate,
                                                where the cathodes of all the diodes must be high to Run. If you want to
                                                stop at the first write cycle, connect a diode cathode to MWR. The 4042
                                                quad latch has both true and inverted outputs, which means you can stop
                                                on any high or low on any address or data line as well.
                                                --
                                                Lee A. Hart | Ring the bells that still can ring
                                                814 8th Ave N | Forget the perfect offering
                                                Sartell MN 56377 | There is a crack in everything
                                                leeahart earthlink.net | That's how the light gets in -- Leonard Cohen






                                                [Non-text portions of this message have been removed]
                                              • ted_rossin
                                                That is the reason I offered to program the chip and send it to the orignal poster. No need to learn the Microchip tool set. Just wire it up on a solderless
                                                Message 23 of 28 , Feb 2, 2011
                                                • 0 Attachment
                                                  That is the reason I offered to program the chip and send it to the orignal poster. No need to learn the Microchip tool set. Just wire it up on a solderless breadboard and away you go. You can build it in under 10 minutes.

                                                  --- In cosmacelf@yahoogroups.com, "sbirdasn" <sbirdasn@...> wrote:
                                                  >
                                                  >
                                                  >
                                                  > That little PIC logic analyzer is an interesting solution to sniffing a few lines.
                                                  >
                                                  > The main issue here is that for both the PIC and the Propeller MPU's, it takes an additional development tool set for playing with the new devices, just to do the debugging/learning about the target 1802 system.
                                                  >
                                                  > For someone just getting started again, that's a pretty messy solution.
                                                  >

                                                  ...

                                                  \> Sbirdasn.
                                                  >
                                                Your message has been successfully submitted and would be delivered to recipients shortly.