I'm trying to figure out why the 43/-43V rails in this PSU are actually reading 41/-41V. I'm presuming that it must be due to a single shared component that is out of spec, but I'm not sure what.

I suspected the zenner diode, but the voltage across that is 15v, as it should be.

I've ruled out supply voltage issues, as the voltages remain at 41v regardless of whether supply is slightly too hight or slightly too low.

Attached schematic below. Please ignore the unregulated parts of the supply (coloured).

enter image description here

  • \$\begingroup\$ That's not a full circuit - there are connections via R513 and R505 that go somewhere not shown. Have you tried simulating this under LTSpice. I'd recommend that. \$\endgroup\$
    – Andy aka
    Nov 18, 2014 at 8:41
  • \$\begingroup\$ Hi Andy - this is the full PSU circuit, those connections just go off to other modules that use the 43v. I've never used LTSpice, but will take a look, thanks. \$\endgroup\$
    – UpTheCreek
    Nov 18, 2014 at 9:04
  • \$\begingroup\$ In a circuit rated to supply 42V, to be reading 41V is only 2.4% off. That could easily be due to component tolerance. What are the actual hard specifications of the power supply? It looks pretty old, so I'd be surprised if it is rated tighter than +/-5% \$\endgroup\$
    – user36129
    Nov 18, 2014 at 10:01
  • \$\begingroup\$ Couldn't you measurement device be a little off? \$\endgroup\$ Nov 18, 2014 at 11:11
  • 1
    \$\begingroup\$ Oh - bad circuit diagram technique - never join four wires like that - should stagger them because the "dots" are hard to read. \$\endgroup\$
    – Andy aka
    Nov 18, 2014 at 12:29

1 Answer 1


Looks to me as if some of the feedback divider resistors may have drifted in value. Or maybe the transistors fed by these feedback resistors may have changed characteristics with age.

If you look at the snippet below the area in the red box are resistors involved with setting the level of the +43V. The fact that the design shows the 18K and 180K in parallel implies that these may have intended to be hand selected and soldered in to tweak the supply voltage into the +43V level. You could try tweaking these values to raise the positive rail voltage some. It may even be possible to replace the 180K with a trimpot to allow adjustability.

enter image description here

Once you get the positive rail adjusted the negative rail will likely follow along. In case it is still off then the resistor glob in the yellow box is used to adjust the negative rail. Note how the feedback of the negative rail is developed off the positive rail. This is done to allow the two voltages to track. Once again the parallel combinations of the resistors in the upper and lower parts of the divider seem to imply that one or more of these resistors was meant to be hand selected to set the negative rail voltage. You could also try replacing one of the resistors with a trimpot to allow easy adjustment of the negative rail.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.