Skip to content
Snippets Groups Projects
Select Git revision
  • 2ccc942d6d28ae2115ad90aa2600abde705b0b17
  • master default
  • tssop protected
  • v1.0
4 results

emes-vcf

  • Clone with SSH
  • Clone with HTTPS
  • David Huss's avatar
    David Huss authored
    maybe fix diode orientation in pick and place correction script (verify
    manually uppon next order)
    2ccc942d
    History

    EMES-VCF

    panel

    A analogue voltage controlled Filter for modular Eurorack Synthesizers based loosely on the Korg MS-20 VCF from 1978 with a few modernizations and extras:

    • Blend between the 12 dB-Lowpass and the 6 dB-Highpass filters
    • Switch between three different types of resonance
      • Symmetric clipping with two anti-parallel diodes (classical MS-20 behaviour)
      • Asymmetric clipping with one diode (weird/wild)
      • No diode clipping (loud, self-oscillating behaviour)
    • CV Input with Attenuverter and a bicolor LED to display the incoming signal. The Attenuverter allows you to e.g. turn positive control voltages into negative control voltages, which is useful for the Highpass mode
    • Drive controls the level, but has a wider range than just a passive control, can saturate the signal and influence the volume and character of the filter and the resonance

    For a sound-demo of the breadboarded prototype click here

    Build Instructions (Through Hole)

    Experience showed that following a certain order of soldering is a feasible method of delivering the most comfort to the person soldering. Namely:

    1. Solder the (usually blue) trimmer on the back side first. Clip the legs very closely with you non-facceted wire cutter (there will be another part on top here)
    2. Solder the top-most potentiometer on the front side next. Straigthen out the mechanical pins and make dead sure that:
      • the part is flush with the surface after you soldered one pin
      • you don't touch the large cylindrical electrolytic capacitors or the small parts in the area with the soldering tip
    3. Solder the power pinheader on the back side next
    4. Straighten out the pins for all the potentiometers and put all the panel parts in (WITHOUT SOLDERING):
      • 3× remaining green potentiometers
      • 1× blue trimmer potentiometer
      • 1× On-Off-On-toggle-switch
      • 4× 3.5mm Mono TS jacks
      • 1× bicolor 3mm LED (long leg goes into +)
    5. Put the panel on top and screw the washer/nut of the soldered, top-most potentiometer onto the thread
    6. Screw all the rest of the parts on and make sure they are straight
    7. Solder all the pins on the back side, again making dead sure never to touch anything other than the solder joint with the tip of your iron. Just pretend you are defusing a bomb and everything will be fine.
    8. Create a power cable
    9. Measure power consumption with a lab power supply before putting it into a rack (15-22mA on positive/negative rails are expected).

    Scripts

    If you change anything on the board, and export new Gerbers make sure to run before commit/production"

    ./scripts/cleanup.py
    ./scripts/bom-to-jlcb

    This cleans up unneeded files and creates new pick and place and BOM files for JLCPCB (essentially maps horizon MPNs and rotations to LCSC parts and rotations)

    Errata

    • revision 1.0:
      • The reverse polarity protection diodes D201 and D202 are 180° wrong in the pick and place file, but not on the silkscreen, resulting in no operation when the power connector is connected the right way and destructive operation when the connector is connected the wrong way. Fix by flipping the two diodes 180° to their true orientation as indicated by the silkscreen (or bridge the pads, which would result into a loss of reverse polarity protection, but would still work).
      • The "universal footprint" for the switch turns out to be not all that useful, you still need to get 1.27 mm lead spaced On-Off-On toggle switches, which are much harder to find. I am not sure if this can be fixed in the next revision because there is not much space for bigger footprints.
      • Some of the TH-solder joints are close to SMD parts and could thus result in destructive action by inexperienced solderers, where possible they will be moved
      • 1.5k for R207 might be too much, maybe a value of 1k would be better suited to allow true 1 V/Oct tracking