Designing a diode ladder filter from scratch

Great stuff! How did you make the kickdrum at the end of the video? is it an oscillator going into the filter and then a VCA? Some frequency change with the envelope? Can you explain? :slight_smile:

Super !! Merci pour ton travail toujours aussi rapide et soigné, speedy gonzales ! :wink:

1 Like

Thx, maybe a module for a next case :wink:

1 Like

C’est même certain !! Je le garde bien au chaud pour plus tard mais je compte d’abord essayé le vcf-6c de EFM, une version du diode ladder qui me chatouille depuis un moment, avec contrôle cv de la resonance

1 Like

J’ai fait le Steiner Parker il est vraiment excellent

Trad :
I did the Steiner Parker (@EddyBergman version) and it is really excellent !


Ok cool I go to try it, Thank you !

1 Like

Steiner Parker diode filter :wink:


It sure is one of my personal favorites! :slight_smile:

1 Like

it’s very simple: i just turned the resonance way up, so that the filter goes into self-oscillation mode. then i used an ADSR envelope to control the filter’s cutoff. give it a super short decay with a bit of sustain and release, and you have a kick!


I saw your YT-videos and was so pleased to see yet another enthusiast showing the magic of electrickery.
Thumbs up!


Hi! Has anybody built this vcf using the layout or does “not verified” mean that it does not work?

1 Like

Generally “not verified” or “untested” or something along those lines means that nobody’s built it to check that it works yet.

The Verified Stripboard Layouts thread is a great repository of layouts that are known to be functional.

And a list of all posts in that thread with images/actual layouts to save you scrolling:


Welcome to the forum :slight_smile: and sorry not yet tested
maybe it works but no one knows …

I traced it in case anyone wants a schematic (also plz help verify just in case).
I also took the bandpass example Moritz provided as an invitation to add a switch to alternate LP/BP (S1) and make it a single circuit.

I was thinking about adding a LPF to a microcontroller project, but that means 5V single supply. Any chance I can make this circuit work with a virtual ground (Vref = V/2 with two resistors as voltage dividers) and, where should I replace GND with Vref in this one?
Im thinking R6, R11, IC2D and IC1C non inverting inputs.
Am I missing something?
Would it even work?
Do I need D1, D2 and D3 working with a single 5V rail and a 5V PWM from the microcontroller as the CV?
I also swapped TL074 for the LM324 as it already works in single supply.


PS: The FREQ pot goes from +V to -V, not +V to GND, sorry.


Normally, when going from a bipolar supply design to a single supply, you would replace all the GND connections by connections to the half way voltage.

Since you have an unused opamp, you could use that to buffer the V/2 obtained with a resistor divider to get a low impedance ground replacement without having to use small resistors in your divider.

With a maximum of 5V for both the FREQ pot and the CV, the maximum voltage at IC1B’s non-inverting input will be about 0.83V, way below the conduction threshold of the series arrangement of D1,D2 and D3, so removing those three diodes should not change anything. You might actually have to add some gain around IC1B, or increase R5, to get the filter to fully open.