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

Sol-20 ACI question

Expand Messages
  • corey986
    So I got to work today with my oscilliscope on the MARCH museum s Sol-20 cassette interface. I replaced the PLL and was able to set calibrate the adjustment as
    Message 1 of 3 , Jun 26, 2013
    • 0 Attachment
      So I got to work today with my oscilliscope on the MARCH museum's Sol-20 cassette interface.

      I replaced the PLL and was able to set calibrate the adjustment as per the sol manual.
      Next I replaced the UART as it was bad.

      I can now get a XEQ command to read the name of the file off my iPod, but it has a load error. I checked all the other chips in the input part of the circuit.

      What I am suspecting is ripple on the supply. I put the big caps on my scope and the power was pretty "fuzzy". BTW: on my personal sol that works the powe across the caps is smooth with just a little cycling around the expected voltage.

      So could my suspicion be correct? The noise on the supply from the caps is causing a cassette interface read error.

      Just want to understand if that is a viable place to look next. I'm making progress in that I get the program name and the error instead of "nothing"

      Thanks,
      Corey
    • Bob Grieb
      Corey,     Different circuits can tolerate different amounts of supply noise.  If the supplies are only being used to power op amps, then a lot of ripple
      Message 2 of 3 , Jun 26, 2013
      • 0 Attachment
        Corey,

            Different circuits can tolerate different amounts of supply noise.  If the supplies are
        only being used to power op amps, then a lot of ripple would be OK.  (Op amps have
        a spec called "power supply rejection ratio" which measures their ability to reject noise
        on the supplies.  They are pretty good at it) The amount of expected ripple
        can be calculated based on the frequency (120 Hz, usually) the current, and the supply
        capacitance.  Also, if there are regulators between the voltage with the ripple and the
        circuit you are interested in, then it's really not an issue unless the voltage drops way down,
        which could mean a bad filter cap (or something drawing a lot of current).

            How much ripple are you seeing?  On a 15V supply, I would expect less than 0.25V
        pk-pk, but that's just a rough guess.  In many cases, more would be OK.  Something I
        was working on recently had about 2V of ripple on a 10V supply.  It was only used to
        feed a 5V regulator, so that was not a problem.  The 5V regulator needed about 7V minimum
        to hold its output up, and that was met.

            But some circuits may use a voltage divider with one end tied to one of the supply rails
        to level shift a signal.  In that case, supply noise can couple directly into the signals in the circuit
        and would be a much bigger issue.  So it depends on the circuit.

            Bob



        From: corey986 <no_reply@yahoogroups.com>
        To: altaircomputerclub@yahoogroups.com
        Sent: Wednesday, June 26, 2013 6:55 PM
        Subject: [Altair Computer Club] Sol-20 ACI question

         
        So I got to work today with my oscilliscope on the MARCH museum's Sol-20 cassette interface.

        I replaced the PLL and was able to set calibrate the adjustment as per the sol manual.
        Next I replaced the UART as it was bad.

        I can now get a XEQ command to read the name of the file off my iPod, but it has a load error. I checked all the other chips in the input part of the circuit.

        What I am suspecting is ripple on the supply. I put the big caps on my scope and the power was pretty "fuzzy". BTW: on my personal sol that works the powe across the caps is smooth with just a little cycling around the expected voltage.

        So could my suspicion be correct? The noise on the supply from the caps is causing a cassette interface read error.

        Just want to understand if that is a viable place to look next. I'm making progress in that I get the program name and the error instead of "nothing"

        Thanks,
        Corey



      • corey986
        I pretty much knew it would depend on the circuit, and the ripple is on the regulated side so it would feed into everything. I guess I was just wondering if
        Message 3 of 3 , Jun 26, 2013
        • 0 Attachment
          I pretty much knew it would depend on the circuit, and the ripple is on the regulated side so it would "feed" into everything. I guess I was just wondering if anyone had experience with this on the Sol-20 cassette adapter circuit as I could easily see it being a problem but would really have to look the circuit over to see how tollerant it should be, I'm guessing not, but I might as well ask.

          Cheers,
          Corey

          --- In altaircomputerclub@yahoogroups.com, Bob Grieb <bobgrieb@...> wrote:
          >
          > Corey,
          >
          >     Different circuits can tolerate different amounts of supply noise.  If the supplies are
          > only being used to power op amps, then a lot of ripple would be OK.  (Op amps have
          > a spec called "power supply rejection ratio" which measures their ability to reject noise
          > on the supplies.  They are pretty good at it) The amount of expected ripple
          > can be calculated based on the frequency (120 Hz, usually) the current, and the supply
          > capacitance.  Also, if there are regulators between the voltage with the ripple and the
          > circuit you are interested in, then it's really not an issue unless the voltage drops way down,
          > which could mean a bad filter cap (or something drawing a lot of current).
          >
          >     How much ripple are you seeing?  On a 15V supply, I would expect less than 0.25V
          > pk-pk, but that's just a rough guess.  In many cases, more would be OK.  Something I
          > was working on recently had about 2V of ripple on a 10V supply.  It was only used to
          > feed a 5V regulator, so that was not a problem.  The 5V regulator needed about 7V minimum
          > to hold its output up, and that was met.
          >
          >     But some circuits may use a voltage divider with one end tied to one of the supply rails
          > to level shift a signal.  In that case, supply noise can couple directly into the signals in the circuit
          > and would be a much bigger issue.  So it depends on the circuit.
          >
          >     Bob
          >
          >
          >
          >
          > ________________________________
          > From: corey986 <no_reply@yahoogroups.com>
          > To: altaircomputerclub@yahoogroups.com
          > Sent: Wednesday, June 26, 2013 6:55 PM
          > Subject: [Altair Computer Club] Sol-20 ACI question
          >
          >
          >
          >  
          > So I got to work today with my oscilliscope on the MARCH museum's Sol-20 cassette interface.
          >
          > I replaced the PLL and was able to set calibrate the adjustment as per the sol manual.
          > Next I replaced the UART as it was bad.
          >
          > I can now get a XEQ command to read the name of the file off my iPod, but it has a load error. I checked all the other chips in the input part of the circuit.
          >
          > What I am suspecting is ripple on the supply. I put the big caps on my scope and the power was pretty "fuzzy". BTW: on my personal sol that works the powe across the caps is smooth with just a little cycling around the expected voltage.
          >
          > So could my suspicion be correct? The noise on the supply from the caps is causing a cassette interface read error.
          >
          > Just want to understand if that is a viable place to look next. I'm making progress in that I get the program name and the error instead of "nothing"
          >
          > Thanks,
          > Corey
          >
        Your message has been successfully submitted and would be delivered to recipients shortly.