My passive first order RC circuit is not filtering out the transient response as expected. There's barely any attenuation considering that my cut off frequency is around 2Hz and the frequency of the transient is at around 20MHz. I require a DC voltage with no ripples.

This is the original signal with transients:

This is the filtered output:

This is the circuit diagram used:

I utilized a resistor of 10k and a capacitor of 10uF.

  • 2
    \$\begingroup\$ But... what does your circuit look like? \$\endgroup\$ Jun 5, 2019 at 12:04
  • 1
    \$\begingroup\$ And how did you connect your oscilloscope? \$\endgroup\$
    – Andy aka
    Jun 5, 2019 at 12:05
  • 3
    \$\begingroup\$ Show us a complete schematic of the filter, along with component values. Explain exactly how you made these measurements. Discuss how you constructed the filter to avoid parasitic capacitance. \$\endgroup\$ Jun 5, 2019 at 12:05
  • 2
    \$\begingroup\$ the frequency of the transient is at around 20MHz Then you need to make sure that your capacitor is still behaving as a capacitor at 20 MHz and higher frequencies. Also, how much filtering is theoretically possible with the filter you use. No filter filters out "everything". \$\endgroup\$ Jun 5, 2019 at 12:31
  • \$\begingroup\$ Try to place a smaller ceramic-type capacitor (say 0.1 µF) in parallel to the electrolytic one... \$\endgroup\$
    – aschipfl
    Jun 5, 2019 at 12:44

1 Answer 1


To make circuits at high frequencies you have to model real, physical components with more care than just treating them as ideal lumped elements. Twenty megahertz is not very high, but it’s high enough in this case. Most important here is inductance of the lead wires and the hookup wire you have used. The inductance of the capacitor and its leads means that its impedance is increasing at high frequencies rather than decreasing, like an ideal capacitor would. Also, there is always capacitance from the input to output (across the 10k resistor), which lowers the impedance of this part at high frequencies.

Redo your physical circuit. Cut the leads of the 10ufd cap as short as possible. Physically separate the input side of the circuit from the output side with some care. Make sure the scope probe has a very short ground connection. You should then see the filter working more ideally.


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.