How to design a Sallen-key Low pass Oscillator that gives undamped oscillations? I've grounded the input of the Sallen-key low pass filter but still can't get any oscillation at the output. I've searched on the internet for days and I have not made any progress. Please help.

  • \$\begingroup\$ Sallen key is a filter and if the gain is set too high it will oscillate but there’s no guarantee it will be as good as a standard oscillator circuit. \$\endgroup\$
    – Andy aka
    Commented Jun 23, 2019 at 14:42
  • \$\begingroup\$ @Andyaka do you have any idea on how to build that circuit? \$\endgroup\$
    – Shawn Khoo
    Commented Jun 23, 2019 at 15:07
  • 1
    \$\begingroup\$ Build the Sallen key filter with the gain large than 3. And you will turn Sallen key filter into an oscillator. \$\endgroup\$
    – G36
    Commented Jun 23, 2019 at 16:50

1 Answer 1


G36 gave you good advice. If you set \$\zeta=0\$, you have infinite \$Q\$. And doing so also sets the voltage gain of a Sallen Key LP filter to \$A_v=3\$.

Here's the simple template:


simulate this circuit – Schematic created using CircuitLab

As shown, I've set the values of \$R_1\$, \$R_2\$, \$C_1\$, and \$C_2\$ to provide about \$1\:\text{kHz}\$ for the corner frequency. You can adjust them as you see fit. Just make sure that \$R_4\$ is more than twice the value of \$R_3\$ to achieve oscillation. (Exactly twice the value sets \$A_v=3\$. But you want just a little more than that, I think.) \$C_3\$ is required to compensate the opamp and avoid very high frequency (unwanted) oscillations.

The output won't be a perfect sine wave, though. But you can always filter the output more, if you want. Just paste another two-pole Sallen Key filter at the output with the same corner frequency, perhaps:


simulate this circuit

I'm just shooting from the hip. But that's the basic approach.

  • \$\begingroup\$ I think, it is interesting to note that such a S&K oscillator is nearly identical to the well-known WIEN oscillator. In both cases, we have a RC-bandpass in the feedback path with identical properties (damping factor 1/3 at w=wo). It is really surprising that the WIEN type oscillator can be found in each relevant book or article - opposite to the S&K oscillator which can be rarely found. \$\endgroup\$
    – LvW
    Commented Jun 25, 2019 at 7:19
  • \$\begingroup\$ @LvW See this link for example. \$\endgroup\$
    – jonk
    Commented Jun 25, 2019 at 7:24
  • \$\begingroup\$ @JonkThank you for the link. However, I doubt if the author is right in assuming that the S&K oscillator would have better properties regarding harmonic suppression. I think, the harmonics are created within the amplifier (overdrive, signal clipping) and it does not matter if the feedback path is a WIEN type bandpass or a ladder type bandpass (with identical Q values). What do YOU think? \$\endgroup\$
    – LvW
    Commented Jun 25, 2019 at 9:45
  • \$\begingroup\$ @LvW I've never given it any thought before. (Frankly, I'd never before considered the idea of turning a Sallen Key into an oscillator until this question and seeing G36's correct comment.) But the Sallen Key oscillator produces something of a truncated triangle wave, now that I've looked. This suggests to me odd harmonics. And without the PTC thermistor (lamp) in the Wein, it's also distorted. Still, I've not taken the time to compare the two. [I have spent the time to completely read Sallen & Key's TR-50 paper, though. The 'classified' one about a year prior to their cited publication.] \$\endgroup\$
    – jonk
    Commented Jun 25, 2019 at 15:32

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.