I made a 13.8V 10A linear power supply from a transformer took from a computer UPS whose battery was dead. Now the issue is when I touch the output of the regulator leads, or the power supply casing I felt a slight tingling sensation. I measured the voltages using a multimeter with one lead referenced to earth and got these readings,

negative rail and earth = 34 VAC

positive rail and earth = 34VAC

Casing and earth = 15VAC

some parts of the transformer and earth =15VAC.

This stray voltage was interfering with my other circuits I'm testing. Fortunately I was able to solve the problem by earthing the case and the output rails directly, but I still read a 1 to 2 Volts on my DMM. How is this effect happening? How do I get rid of this completely? Also I can read a 700mV on a 1N4148 RF detector probe in the close proximity to my PSU when ever I touch it or attach a 10cm wire to it. Is it just static from all the switching devices in my house or does it have to do anything with the PSU.

Here is the schematic,


simulate this circuit – Schematic created using CircuitLab

  • 2
    \$\begingroup\$ Sounds like inter-winding capacitance but really, you should post a schematic to be sure. \$\endgroup\$ – Andy aka Oct 23 at 7:59
  • \$\begingroup\$ I have added the schematic. Lets assume its inter-winding capacitance, it cant be avoided, so how does one minimize the effects? \$\endgroup\$ – ASWIN VENU Oct 23 at 8:33
  • 1
    \$\begingroup\$ It's a bigger problem in switching power supplies because of the high frequencies involved but, it's also countered by the inter-capacitance being smaller. If "earth" is available and you can earth the DC output that is the easiest fix. If not, you will have to use non-electrolytic capacitors down to earth from the negative (and sometimes positive) DC output. \$\endgroup\$ – Andy aka Oct 23 at 8:39
  • \$\begingroup\$ I'll try both, how much should the non-electrolytics be in value? Also what do you think about the spooky readings on my rf pobe? \$\endgroup\$ – ASWIN VENU Oct 23 at 8:50
  • \$\begingroup\$ I can't comment on the RF probe but I'd start about 1 uF and see what happens. \$\endgroup\$ – Andy aka Oct 23 at 10:17

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.