Im starting to think that too.
So, I pulled out the oscilloscope to see if my MFOS state variable filter exhibits the same behavior and, what do you know? It sure does.
I decided to compare it to my MFOS 24db low pass filter. No resonance at all when the resonance is set to zero.
I dont know if this info is helpful, but at least it points to the design of the filter rather than faulty components.
1 Like
Given their position in the circuit, it doesn’t seem to me like those larger resistors should affect the resonance.
They could limit the filter’s maximum cutoff frequency and/or your ability to tune the filter to 1 V/octave. But again, I don’t fully understand the roles of Q1 to Q4 as part of the exponential converter and associated current mirror(s).
2 Likes
I agree. Q1 and 2 appear to be a standard exponential converter and Q3 and 4 look like buffers. They shouldnt play a role in the resonance.
1 Like
Yes, very much so, thank you!
2 Likes
Hi @TBone thanks for joining in.
Do i see in your first screenshot that you have it too? The spike looks the same as mine.
1 Like
So for my understanding, you have both 2 pole 12db and 4 pole 24db mfos filters and the first one has also the reso spike when no reso is dailed in, and you have compared it to the 4 pole 24db filter and does not have this spike?
Yep, I compared the two filters and the 12db state variable filter has that small resonance spike with the resonance completely “off” while the 24db low pass filter has zero resonance withe the resonance control off. Honestly, I prefer the 12db filter and think that the resonance issue is more of a feature rather than a bug. It adds to its character in a good way in my opinion. I recommend the MFOS 24db low pass filter if you want zero resonance with the resonance control all the way off. Its a really good filter. That said, if there is a way to totally squash the resonance on the 12db filter I’m all in for the sluething : )
1 Like
Ok thanks for your effort to test both filters. That means that my filter is ok and free from solder bridges or other rarities. Brings me to the next steps to compare both schematics, see if there are significant differences in component values that might cause the spike. Btw the spike happens between opamps not the ota. I removed the ota to see if the spike takes a hike but unfortunately it sticks😁
With the ota removed the connection between points RA and RB and signal path to the opamp feedback circuit are interrupted. Spike still there so that is interessting to investigate what causes this.
1 Like
So glad that im not the only one having this. Its not a very big deal ofcourse but i find it to much resonance imho
Think i might have something
This is the 2 pole 12db mfos schematic i used to build my filter
This is the 4 pole 24db schematic
When i look at RA and RB, the 2 pole schemo has R22 added. Does not exist on the 4 pole.
Next U3-D opamp to point RA has a 150k while the 4 pole has 100k.
What if i remove R22 (150k) from point RB and change the R33 (150k) to 100k to point RA?
I’m by no means any sort of authority on the matter, but the circuits seem very similar. So, it is worth a shot! Let us know what happens.
1 Like
The equivalent resistor is R53 on the resonance control page of the four pole filter.
The four pole low pass filter doesn’t have any other feedback than the resonance circuit like the state variable filter has, that is likely the reason why it doesn’t show the resonance bump.
2 Likes
Ok thanks for sharing.
Have a nice day and thanks for all the help.
I think for now, i’ll leave the circuit as is as we have checked almost everything on the circuit so lets say that this reso spike is a feature.
If someone is continueing trying, searching for a solution im happy to hear. Untill then again many many thanks
Hello,
What does the square-wave input look like at the output of the input buffer op-amp U3-A? Is it already ringing there?
Cheers
Hi jaradical, thanks for stepping in. I will let you know tomorrow. 
Hi, i hooked up the scope to output U3-A opamp and here it is.
Cutoff fully open ofcourse for full signal, and resonance fully closed (pot ccw)
Spike is there, but because the output is high pass also im not sure if this is correct. Hope your advice will bring clearance.
Other thing i noticed is that when i slowly dail in resonance, it starts on the little 2nd spikey next to the big one (1st big spike). Maybe this brings ideas to your mind?
Yes i did. Output pin ditectly (not after resistor.)
It showed my this spike on my scope
I used this schematic. The 2 pole 2db one.