Tuesday 7 December 2021

Synthersizer Build part-45: STEINER-PARKER DIODE FILTER for EURORACK.

This is the same Yusynth Steiner Parker diode multimode filter I posted in project 26 but with a new layout for Eurorack.

I'm busy setting up a Eurorack system dedicated to live performances, so I want to remake some of the modules I built earlier to make them fit the Eurorack 3U size. So here's the first one I converted, one of my absolute favourite filters, the Steiner Parker multimode diode filter with Lowpass, Highpass, Bandpass and Allpass. This is a Sallen-Key type filter with positive feedback so that you don't loose volume when you increase the resonance like you do with the Moog ladderfilter for instance.
I won't go further into how it works etc. You can go to the previous Steiner Parker article for more details.
I started out matching the diodes I needed by measuring the voltage drop but they all came from the same batch and the measurements were so close that I stopped matching and just put them in (and the filter works absolutely fine). The transistors however must be closely matched otherwise the filter won't be in balance. You can set the right balance with the 1K potmeter but that's only a fine control so make sure the transistors are matched. You can match them by simply measuring the HFE and look for two with the same values.
When you start out building, make the cuts in the copper strips first and then put in the wirebridges. Then you can put in the rest of the components.

About component values:
For the level potmeters I used 10K linear ones because that's what I had. You can use any value from 10K up, it doesn't matter for level potmeters. Keep to the recommended value for the Cut-Off and Resonance though. I used a 100K for the Cut-Off frequency potmeter and I changed R26 to a 100K to make the voltage drop over the potmeter the right value. This works perfectly fine. You can of course use a 47K (50K) potmeter but then use a 47K resistor for R26. (R26 is the 100K resistor in strip A to the right).
For the 1,5nF filter capacitors I would recommend using good quality polystyrene, polyester or silver mica types. These form the heart of the filter so don't use ceramic caps for those.
Btw, I left out the two 10 Ohm resistors in the + and -12V strips because this filter was designed for 15V but running on 12V so I wanted to avoid any further voltage drops. I also left out the bypass capacitors but if you want to include those just put a 100nF capacitor from +12V to ground and one from -12V to ground right above the location of the chip. There's room enough left. (I did put them in later, just to be sure, but they are not on the layout or the bill of materials.)

Below are the layouts I made for this project. They are verified as always. I used these for my own build. I left out the second CV-IN and the second AUDIO-IN potmeters and jacks to keep the layout free from clutter. You just copy the first input if you want two of them (which I strongly advise you to do especially for the CV). The stripboard is 24 by 41 holes. The switch to choose between Lowpass, Bandpass, Highpass and Allpass is a normal 2 pole 4 way rotary switch. 
Instead of using a reverse logarithmic 50K potmeter for Resonance I used a 100K linear type with a 100K resistor soldered onto it to get the reverse logarithmic characteristic. (See layout below. Two 100K resistors in parallel make for one 50K resistor). This is the recommended alternative in the original Yusynth article and it works really well. Of course, if you happen to have a reverse logarithmic 50K potmeter then use that instead of the 100K pot + 100K resistor solution. Should you have problems with resonance coming in too soon, put a 10K resistor in series with pin 3 of the resonance trimmer potmeter to get the throw of the resonance panel potmeter more to the clockwise side. Thanks to Nick in the comments below for the heads up on that one!  Here's an image of the alternative wiring of the Resonance potmeter:

Wiring Diagram:

Stripboard only:

About trimmer T1: 
I changed trimmer T1 from a normal one to a multiturn trimmer which made it much easy to set. You need to set this trimmer so that the Cutoff frequency potmeter has the correct throw with full resonance at about 2/3 clockwise with the resonance potmeter set to almost self oscillation. I measured the resistance of T1 when I was done and it was about 640 Ohm.

This filter works best if it has a 1V/Oct CV permanently connected to it, although you can't play the self oscillation as you can with some other filters where you can use the filter as an oscillator. This filter's resonance is just too agressive for that.

Making the cuts accurately:
Here's a layout of all the cuts you must make and the wirebridges you need to solder in. This is viewed from the component side. Mark the cuts on the component side, with an Edding pen, and then stick a needle through the marked holes and mark them again on the copper side. Then you can cut them with a hand held 7mm dril bit. The cuts are all over the place so concentrate and be accurate otherwise the filter won't work. Don't forget the cut underneath the wirebridge at position S-19:

Bill of materials: Buy a batch of 100 BC547 transistors if you don't have any, so you have enough to choose from when looking for a matched pair. If you want to include de-coupling capacitors then order two extra 100nF caps because these are not included in the BOM. Order good quality polystyrene or silver mica or polyester types for the three 1,5nF filter caps.

Here's the schematic drawing by Yusynth:

Here are some pictures of the build proces and the finished product. Notice I had to put two capacitors in parallel to create a 680nF capacitor. I didn't have one in stock.

I soldered all the wires directly to the copper side of the print and mounted the print with the component side pointing backwards of course, otherwise you can't get at the trimmers. I put some Gaffa tape over the pins of the 4 way rotary switch to avoid accidental contact with the print or wiring.

A look at the finished panel. I managed to fit everything in nicely. I had this piece of powdercoated aluminium left over so that was perfect for this project. I made the 3mm mounting holes wider to give me some room to move the module sideways to fit the rest of the modules (which are yet to come ^___^)

Finally a word about hum. This filter is susceptible to hum especially when using a switching powersupply. I have no problems with mine in my DIY synth because it uses linear powersupplies but my Eurorack filter did suffer from this. I solved it by putting two 1000µF/25V electrolytic capacitors over the powerrails. This provides enough capacitance to suppress the hum. It worked fine for me. See picture below. The overall depth of the module is 45mm with these caps installed. Still below 5 CM.
There's a forum post about the hum problem here -- CLICK HERE --

Okay that's it for this one.  
If you have any questions or remarks please comment below or post them in the Facebook group for this website where we have a great little community willing to help anyone encountering problems with the projects.

If you find these projects helpful and would like to support the website and its upkeep then you can buy me a Coffee. There's a button for that underneath the menu if you're on a PC or Mac. Or you can use this PayPal.Me link to donate directly. All donations go towards the website and projects. Thank you!


  1. Hi Eddy,

    Thanks for this I was going to make my case so that it would take your original boards. These Stripboard layouts will make it much easier, fair play to you.

    May I ask for the Thomas Henry VCO-555 next please I would like to make 3 of them. And also if all the boards will be 24 x 41 cut from the 24x56 ones.

    Ohh and before I forget is this panel 50mm wide?

    Cheers - Jim

    1. Hi Jim! Thanks for your support, that's awesome! But don't get too enthusiastical because I'm not going to convert all the projects I have on this website to Eurorack. I will see if I can do the VCO-555 but I make no promises. All the stripboards I will use will indeed be cut from the 24 by 56 ones into a maximum length of 41 holes. The panel I used for this filter is 87mm wide. It was an off-cut I had lying around. Cheers! Eddy

    2. Actually the VCO555 already has a eurorack layout further down in that article. -___^

  2. Hello Eddy, Thanks for the reply, doing some measurements I'll only save 28mm using 41 hole stripboards, I can just as easily stick to the 56 hole ones and still get two rows within 300mm case depth. Cheers - Jim

    1. Yes I think that's the best approach. Good luck and let me know if you have any questions.

  3. Hi Eddy! I dont rlly understand (Im a new in DIY world) why this filter needs 1V/oct. And I dont know if the CVin could be used like a ADSR input.

    This is my first eurorack project and i want to make a simple MIDI2CV -> VCO -> VCF / ADSR -> VCA /ADSR and I dont know if this project works for me.


    1. Hi Carlos. Those are good questions. The 1V/Oct is for the resonance. The resonance goes up and down in frequency when you play a high or a low note on the keyboard if you use this input. This sounds much better. The ADSR input into a CV-in makes the resonance ring in the first moments you play a note and this produces that typical synthesizer sound where you don't just hear a steady note but it is high in resonance at first and then settles down more after a second of so. This is a typical synthesizer sound and that is the way to achieve that.

  4. Then, cutoff frec cannot be modulated by ADSR envelope in this filter? I want to have this possibily in my filter, but this resonance ring is rlly interesting!

    1. Yes, I'm sorry. I think it IS the actual CutOff that is influenced by the ADSR and not the resonance as I said earlier. It works just like any other filter.

  5. Where on Earth do you get a 2-pole rotary switch?

    1. On ebay of course. I got mine for 89 cents a piece. The cheapo ones work fine.

    2. tayda has the exact switch decent quality for cheap

  6. Hey I got this circuit to run smoothly and everything works all the filter modes and all the controls and pots work. I used the original 50k anti log pot for resonance and ran into a problem with it going into self oscillation way too early. Right around 10-12 oclock with the resonance trimpot all the way down. I slapped a 10k resistor on leg three of the resonance trimpot and now its exactly where I'd want it in a filter after readjusting that resonance throw trimpot. Its an incredible sounding filter and works so no real complaints lol. I just didnt know if anyone else ran into this happening or not

    1. Hi Nick! Thanks for the feedback. I saw your post on Facebook. I'll make a note of your solution in the article. I think this will be useful information for everyone building this filter. Glad you got it working so well. It's my favourite filter at the moment!

  7. Hello. Why is there twice an input for audio and twice for CV? Could someone explain to me what it is for and how it is specifically used in practice? Thank you. MAARFY.

    1. The double inputs are there so you can mix signals right at the inputs to get some cool audio effects. For instance if you have two CV input signals you can get some weird effects if they are mixed together. But you don't have to include it if you don't think you'll need it.


Note: comments are moderated and do not appear straightaway. Your first comment is not allowed to contain any links.