0
\$\begingroup\$

I have been trying to simulated a filter and amplification circuit. In one of the stages I am using a LP Sallen and Key circuit and another I am using a HP Sallen and Key circuit. I can successfully get the shape of the filter in LTspice AC analysis with 307kHz being my center frequency, but when I add gain enhancement the shape completely goes away. The first image is my calculations for my component values. The first image below is my LP filter circuit and the resulting AC analysis. The second image is my circuit when I add gain enhancement and the resulting AC analysis that exits. My calculations are at the bottom.

LP Filter no gainLP filter with 20 gain

I am hoping to get the same or similiar shape to the filter without the gain enhancement, but would just like a higher gain. Looking at the datasheet of the LT6233 the gain bandwidth is 320MHz. Does anybody know what I might be doing wrong?

schematic

simulate this circuit – Schematic created using CircuitLab

With Q=1.5, ωo=307000*2π and H=20 we can find the component values for each:

C3=(2*Q/ωo)*((H-1)/H)=1.477*10-6
C2=1/(2*Q* ωo)=1.728(10-7)
C4=2*Q/ ωo=1.555(10-7) R3=R4=1 R1=R3*H/(H-1)=1.0526 R2=R3H=20

Then, normalizing each component to C4=1nF, we get: C4n=C4*(1nF)/C4=1nF C2n=1.111(10-10)≈0.1nF C3n=9.5nF≈0.01µF R3n=R4n=R3*C4/(1nF)=1555Ω≈1.6kΩ R1=1637Ω≈1.6kΩ R2=31.1kΩ≈30kΩ

I am not 100% confident in the source I got these notes from and after simulations I found that I needed to cut out C3

\$\endgroup\$
  • \$\begingroup\$ Apologies, the first image is not showing up on its own, so you have to click to see it. \$\endgroup\$ – BRogers Aug 10 '17 at 19:34
  • \$\begingroup\$ Provide the details about how you calculated your component values. Also, I recollect gain and damping are coupled in the Sallen-Key. \$\endgroup\$ – jonk Aug 10 '17 at 19:50
  • \$\begingroup\$ Ah, in you saying that I found where my error was for the high pass Sallen and Key. Thank you very much. Let me re-post with my low pass Sallen and Key and I'll post my calculations too. \$\endgroup\$ – BRogers Aug 10 '17 at 19:58
0
\$\begingroup\$

It's quite simple what's going wrong. The filter shape is gain dependent and if you increase gain you start to add a big peak to the bode plot. Somewhere in the range 5 to 10, the gain will turn the circuit into an oscillator.

The sallen key filter characteristic depends entirely on amplifier gain.

\$\endgroup\$
  • \$\begingroup\$ The big peak it what I am after. When I simulate the filter without any gain enhancement I see that 'bulge' at the 307kHz region I am after. However, when I add the gain enhancement, I find that the peak of the output drops to around 4kHz and I have significant attenuation at the 307kHz region that I am after \$\endgroup\$ – BRogers Aug 11 '17 at 12:39

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.