power supply help...

[Deleted User]
[Deleted User] Posts: 7,658
edited April 2011 in Electronics
Hello,
Does anyone have a computer based electronics modeling capability to look at a fairly simple power supply schematic and determine the output impedance versus frequency? I'm trying to determine what values for a RC decoupling network based upon power supply output impedance.
Or, if someone has a fairly simple approach to measuring PS output impedance, I'd appreciate info. I have the 1964 H-P white paper, but not sure how to do it with out H-P 4301A.
Post edited by [Deleted User] on

Comments

  • megasat16
    megasat16 Posts: 3,521
    edited April 2011
    Linear DC Power Supply? Regulated or non-Regulated?

    Switch Mode DC Power Supply?

    Or Tube Amp AC power Supply?

    It's been a long time since I draw circuits on EWB and Protel but I have MultiSim on my computer so I might give it a shot if you want to upload the Schematics.
    Trying out Different Audio Cables is a Religious Affair. You don't discuss it with anyone. :redface::biggrin:
  • [Deleted User]
    [Deleted User] Posts: 7,658
    edited April 2011
    Hi Mega,
    Thanks for the reply. It is a linear DC regulated supply for the audio sections of a Revox A77. Let me see if I can find it online and post it. I appreciate the assistance!
    Cheers, Ken
  • [Deleted User]
    [Deleted User] Posts: 7,658
    edited April 2011
    Hello,
    Here is the schematic for the power supply, it is the +21VDC that is in the middle of the sheet.
    Thanks, Ken
  • megasat16
    megasat16 Posts: 3,521
    edited April 2011
    Hello Ken,

    Don't thank me yet! I am not even sure I could help but I'll try.

    I found a bunch of manual and I am also looking at the service manual.

    Here is the link to Download them. I will see if I can simulate the power supply section of it. But at what load?

    http://www.analogrules.com/manual_list.html
    Trying out Different Audio Cables is a Religious Affair. You don't discuss it with anyone. :redface::biggrin:
  • [Deleted User]
    [Deleted User] Posts: 7,658
    edited April 2011
    Hi Mega,
    Listen, I appreciate your offer to help. But, its high time I got off my lame duff and learned to use an electronics program. This is as good a time as any, is there one that is Mac compatible that you might recommend? I've sent a message off to my brother-in-law who is a digital engineer and works with Mac.
    Again, your offer was appreciated.
    Cheers, Ken
  • megasat16
    megasat16 Posts: 3,521
    edited April 2011
    Hi Ken,

    No sweat. Actually, trying to help you makes me relearn electronics. It's a win-win but I am sure your bro-in-law with the practice in the field will know the answer easier.

    From what I gather, you need to disconnect the power supply part from the A77 (and also remove the bidge rectifier from the circuit) and measure the output amplitude with various input frequency using a signal generator as the input source.

    It'll plot the filter slopes of the power supply circuit. I will try to simulate that part of the circuit since it's all in good fun. Oh well, some things are worth to relearn again.
    is there one that is Mac compatible that you might recommend?
    Cheers, Ken

    I think there is something called macspice you can use on Mac. I know how well it works or have never used it though.

    http://www.macspice.com/Download.html


    Cheers,
    James
    Trying out Different Audio Cables is a Religious Affair. You don't discuss it with anyone. :redface::biggrin:
  • [Deleted User]
    [Deleted User] Posts: 7,658
    edited April 2011
    Hi James,
    The way to empirically measure the impedance is to connect an oscillator to a power amp to the power supply. The power amp is isolated from the PS with a blocking cap. There is a load resistor and a series resistor. You set the oscillator to low frequency with the PS operating and measure the ACV across the load and the series resistor then Zo=Rs x Vl/Vs. Keep changing frequency and continue measuring Zo. This gets plotted on log-log paper, then this is used to select the value cap and resistor that will provide a localized decoupling network at each point "downstream". The goal is to provide a nice bucket of electrons close to the circuit demand instead of having the AC path go all the way back to the PS. But, rather than just sticking a big cap near the circuit (which can cause instability problems) the ESR and ESL of the cap have to be compared to the PS Zo. Sort of like the damping factor of an amp influences how the wire/speaker load behaves.
    That's what I've gathered, anyway.
    Cheers, Ken