Monday, December 16, 2013

quad ADSR simplified yu synth ADSR.

I am an aspiring electrical engineer. Circuit design inspires me in the same way that songwriting once inspired me.  In electronics as in songwriting, one often imitates until he can create on his own. When modifying another person's circuit, it's hard to say exactly when one can claim a circuit.  How much needs to be different?  What kind of changes are considered more than arbitrary and therefore justify claiming ownership (my circuit)?  

A majority of the circuits I've worked on for my synth started as either yu synth designs or MFOS designs.  The designers of these circuits have far more electronics knowledge than myself and clearly execute a higher level of intention in their circuits. Though I've learned a lot since I began my modular project, I still often find myself in the "shit, lets try this resistor" process. The designs available on these sites are sometimes built upon older designs in which the newer design only varies in peripherals (like input/output buffer impedance, trigger method, resistor value, LED indicator), keeping the core function identical.  In all cases, these designs are based on simple electronic processes which were discovered by someone else years ago-- a fact to which these sites openly acknowledge.  As participants, we cannot ignore this chain of influence.  

I take the postmodern view when it comes to circuit design.  No one owns the function in which these circuits are designed to perform. No one owns the function of a filter or the function of an oscillator. These simple electronic functions can be performed in a variety of ways and as we get further away from the tree trunk of an idea, the branches become thinner and closer together. The things I change about these circuits--however arbitrary--make them more suited to my purpose and therefore are as "mine" as  anyone's claim to a recipe for homemade nachos. 

That being said, I would never directly clone a person's circuit or disregard the chain of influence.  I don't believe that is in the spirit of sharing knowledge or creative engineering.  In circuit design as in songwriting, my goal is to participate, not "steal"(whatever that word even means anymore). 

That diatribe was meant to address the fact that "my" quad ADSR circuit is really nothing more than a simplified version of the yu synth ADSR.  By reducing the number of functions it performed and therefore reducing the number of components, I was able to squeeze 2 simple circuits onto a PCB using only 2 chips (quad op amp and LM556).  I have no need for an inverted ADSR output because most of the cv inputs on my synth were designed with attenuverters. This ADSR is a sort of a simpler compromise between the two yu synth designs.

This is likely one of the last PCB I will layout in photoshop.  The circuit repeats 4 times and so it made sense to do it in photoshop as eagle cad will not allow you to simply copy a section of  your PCB layout and stamp it around. 

The transistors are 3904's and the collector is the square pad.  ha ha.  except the very bottom one that controls the led. it's backwards and so the emitter is the square pad.  That kind of mistake is easier to make in photoshop. the arrows point to the negative side of the diodes.  I doubt anyone will try and build this based on my minimal instruction but posted the PCB just in case. comment any questions, I respond pretty fast. 

since I began building modules, I find myself planning more.  I like to print version of the panels and lay the components out on them in order to insure that they will fit correctly. 

a little DEVO joke for you. 

I managed to take a million pictures of my build, except for one showing all the panel wires.

Ms. Twenty module

The MS20 is a excellent beginner filter because it sounds awesome and there is tons of documentation available for those interested in how it works.  Generally when I implement a circuit, I like to find a few different version of it.  The MS20 has been cloned about a thousands times and so there are a variety of schematics in which you can compare and contrast component values.  I primarily used Tim Stinchcombe's page but I printed and consulted tons of other MS20 filter schematics available online. 

There are only a few things which need to be changed in order to switch the filter between lowpass and hi pass functions.  I implemented a switch to move between the two functions.  The lowpass filter output volume is much lower than the hi pass filter volume so I also incorporated a gain circuit which switches with the filter function.  

In retrospect, my method for switching the gain circuit is pretty sloppy and wasn't really seeing the whole circuit when I designed it.  The next version would switch the non-inverting buffer at the audio signal input to an inverting buffer, thus the gain amplifier (located at the end of the circuit) could also be inverting (returning it to the correct phase), then I would simply have it switch between two different "Rin" resistors to adjust the gain of the inverting amplifier.  There is an audible bleep when switching between filter modes. 

also, it is definitely NOT 1v/octave. 

I got the boards made at OSH Park. This was one of my first projects in eagle cad.  There are things I like about it better than photoshop and things I like less.  Unfortunately, when a circuit board has repeating parts, it's next to impossible to copy and past PCB layouts. 

Initially I was going to use a picture of Martin Sheen (MS pun) from Badlands on my panel but I couldn't find one that fit the panel layout.  In the end, this picture of young Steve Martin (Still MS but backwards) won out.  After all of the knobs and jacks are put in, it's hard to tell it Steve but it's still fun for me to know he's on there. 

 In my earlier modules, I've generally been pretty lazy and just done all my panel wiring in one color. My modules have become a lot complex and so I had to implement the multicolored "twist" method. Since doing so, I do enjoy the final "wire up" process a lot more.  less stress.