0
\$\begingroup\$

Recently after having some issues in my gaming PC, i checked the 12v rail under load(benchmark). The 12v rail dropped to 11.6v that's within spec. But when i switched my DMM to AC it was giving me 100mV to 200mV while checking through the 8 pin connector that plugs into the gpu. I am wondering if this is the right way to check for ripple in PSU? And if so how bad are these readings? Also while the 12v rail drops to 11.6v, the 5v rail increases from 5.04v to 5.15v

\$\endgroup\$
  • \$\begingroup\$ Your meter likely shows "RMS" measurements. The peak-peak will be 2.828x larger, if a sinwave. IF a sinwave. Find someone who has a scope. Or borrow one for 1 day. \$\endgroup\$ – analogsystemsrf Apr 18 at 12:34
  • \$\begingroup\$ @analogsystemsrf If the signal is a sine wave, or if the meter isn't true rms, yes? \$\endgroup\$ – Hearth Apr 18 at 13:02
1
\$\begingroup\$

I am wondering if this is the right way to check for ripple in PSU?

I would say no unless you know the behavior of your DMM. On most DMMs the AC range is designed for low frequency (often up to 400 Hz) and sinusoidal signals.

How will your DMM respond to high frequency AC, your PC's power supply will output DC rails with a (for example) 100 kHz ripple due to the switching of the DCDC converter in that power supply?

How will your DMM respond to a DC voltage being present while measuring AC?

If that ripple was indeed 100 mV to 200 mV that is to be expected and nothing to worry about. Most circuits in a PC are designed to allow for so much supply ripple.

Also while the 12v rail drops to 11.6v, the 5v rail increases from 5.04v to 5.15v

Also nothing to worry about. In many PSUs (Power Supply Units) for PCs the 12 V and 5 V outputs do not have their own regulation loop but share one. So it might be that the 5 V isn't regulated by itself but relies on the 12 V rail.

If you load the 12 V rail the PSU has to "work harder" to compensate, that might increase the voltage at the 5 V output which might still have the same load.

The drop from 12 V to 11.6 V can also be (partly) due to IR-drop, meaning voltage drop through the wires and connectors.

\$\endgroup\$
  • \$\begingroup\$ According to intel's specification ripple should be lower than 120mVp-p. I know it's not the right way to measure ripple as my dmm might only measure upto certain frequency but if it is showing correct readings (which i also doubt) then aren't they higher than what intel has allowed as these are rms values. \$\endgroup\$ – BumbleB Apr 19 at 5:23
  • 1
    \$\begingroup\$ @BumbleB The only way to correctly measure the supply ripple is using an oscilloscope and a Spectrum Analyzer (that could also be a function of the scope) if you want to know the frequency content. So if you want to compare to Intel's specs you need better equipment. But what would be the point? To pinpoint the issues you have with your PC, swap components (PSU, GPU etc) to determine what is causing the issue. Measuring the ripple won't tell you much. \$\endgroup\$ – Bimpelrekkie Apr 19 at 7:08

Your Answer

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

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