LTSpiceIV - modelling a computer PSU

Discussion in 'The Projects Forum' started by stevendt, Dec 10, 2012.

  1. stevendt

    Thread Starter New Member

    Dec 10, 2012
    Hi Folks,
    As will no doubt become abundantly clear, I am very new to this, but here goes . . . .

    I am trying to model the power circuit for an old 1980s computer (Memotech MTX), just to try and understand it more than anything else.

    The external "PSU" is nothing more than a step down transformer with multiple taps at nominal voltages of 22.3, 18 and 9. The regulation etc. is done on the mother board inside the main computer chassis.

    I am trying to use LTspiceIV to model the circuit. I have tried to cobble together a transformer using some inductors to give me the approximate AC voltages. I have downloaed Spice models for the Diodes, regulators and TIP2955 that are not part of the LTspice standard distribution.

    I have added resistors to the 12VDC, 5VDC and -5VDC to simulate the loads that I expect the computer to run. The model appears to work fine until I add the TIP2955 and then things go very wrong - giving wierd current and voltages. I am sure that the circuit topography is correct, but my modelling skills are suspect.

    I'd appreciate any help/advice that anyone can give please?
    I have attached my model and include files

      File size:
      3.3 KB
  2. Ron H

    AAC Fanatic!

    Apr 14, 2005
    I don't have any answers for you, but I have a few comments.
    Your regulator input voltages are way too high. M78XX regulators need a minimum of about 3V across them for regulation. Anything higher wastes power and causes the regulators to get hotter than necessary. I realize you measured AC voltages, but I'm having trouble believing the designers would have built in that much headroom. I see you are in Scotland. Are you applying 220VAC to a supply that was designed for 110VAC?

    The normal way of boosting the output of a linear regulator is to use a power PNP as shown, but with the collector connected to the output of the regulator. I tried that, and your simulation oscillated. I moved C39 to the other side of R3, and it still oscillated.
    I don't understand the "value" of R2. What does that mean?
  3. stevendt

    Thread Starter New Member

    Dec 10, 2012
    Hi there,

    Thanks a lot for the reply.

    The PSU is rated for 240VAC - it was 240 here in those days before we got "Eurofied".

    The transformer tap voltages are printed on the PSU case, 22.5, 18, 9, it is not clear whether they are p-p or RMS though, but I would think they should be RMS?

    The "value" for R2 was just a way to make the simulated 5v load vary a bit, I just look at one of the AC voltages and manipulate it a bit, then make sure the value of the resistor is within certain limits to keep the current value reasonable,