Ah, panels... I tried cardboard, thin wood, but the best material is aluminum for sure. Make sure to drill the panel before soldering any component that goes through it, so you can solder those components with the panel on (I remember spending hours trying to adjust jack holes to make the panel fit). Great episode as always!
Just a (very late!) note that filing aluminium gums up files, a process called “pinning”, the preventative is to chalk the file, easiest done with those giant chalks for sidewalk scrawling. And it’s fun having multi-coloured files.
Nice video ! But next time use a center punch for your holes. It helps a lot ! Also give yourself a bit wider tolerance at the dimensions. From the other side the rework of the faceplate gives your sequancer its charm
Great way to start the new year! edit: I'm curious how you approached the issue of the first step. Analog sequencers tend to not play the step the sequence starts on as the clock pulse advances to the next step right away. Digital sequencers typically will play the first step and I've always wondered what the trick behind that was.
I noticed I was losing the first step in my first tests and in the next build I changed the initialization so that the "current step" value starts at 16 rather than 1. Seems to work for now - there's also the external "reset" input to set the start instantaneously - my midi-cv doesn't have a reset out but I do want to test with a more standard euro clock (pamela's workout maybe?).
@@extralifedisco Very interesting. Is your sequencer based around a counter? I noticed that my Varigate 4+ doesn't actually respond to the clock gate (especially gate patterns that aren't really clocks like euclidean sequences or drum patterns) the way my SH-101 does. The SH-101 is one gate one step, so I get a straight forward variation on whatever is in the buffer. The Varigate 4+ has ratchets and multiply/divide options so it averages the pulses to determine bpm and then resyncs the step as best it can but the result is often charmingly wonky..
@@robertsyrett1992 Ah, clock multiplication is more art than science. Mine is all in software, so when it gets a change from low to high on the clock input, it stops the internal clock, advances a step, and counts milliseconds between pulses (solely for duration and glide length). The clock input has a transistor buffer to bring higher voltages down and lower voltages up to 5v. My MIDI-CV module has some crazy low impedance clock out, so it doesn't even work when hooked up through a resistor divider - that might be a difference between your devices (input buffer design). My CD4017 counter sequencer triggers on input low instead of high, but otherwise seems to keep time pretty well.
heyy this is really starting to come together, looks like a really fun sequencer to play with. Digging the advanced features you've implemented, switching to a proper IDE and breaking it all out into classes helps so much. I think I got to around 800 lines before i had to fully refactor, but it was totally worth it. re: pattern storage - are you looking at FRAM or flash storage? I was originally going to try FRAM but I kinda want to try leveraging the EEPROM as my sequence encoding is quite compact. the read/write times (especially write) are pretty harsh though :/
Looking at using Flash - had in mind the W25Q80DL because it was through-hole and big on adafruit/sparkfun, but seems to be obsolete. Will try breadboarding with it but may switch to a more current chip - in any event all the new flash stuff is 3.3v only so requires some level shifting with transistors or another IC. Excited about it as it's really the last thing to prototype before spinning up some boards!
Ahh yep yep. I've been trying to keep everything 5v to try and minimize the component count and the overall complexity of the circuit as this is my first proper dive into circuit design. I'd definitely like to have my name put down on some kind of a list if you have plans to sell boards! Did you have any intentions of pushing the code to github? Apologies if you've articulated this in any of your previous videos, I just have a really bad memory :P
@@JarrydNielsenLMAO Yes, planning to sell boards at some point! Will start a mailing list when I'm closer to it. The code is indeed up on github: github.com/matthewcieplak/super-sixteen Still a mess, no classes yet but if you have refactoring ideas I'm all ears!
@@extralifedisco Ahh yes I have actually been to your github before, silly me! I can definitely see some similarities in the way we've chunked our projects out, as well as a few differences. I probably can't offer any solid advice for refactoring as i work on my code in a very incremental way, but feel free to have a dig through the repo for one of the sequencers i started working on last year - github.com/atomsmasha/ModularSequencer Might have to pinch your portamento code for a new project i've started working on, if you don't mind :3
Yep, the filter is a YuSynth minimoog VCF, which is a transistor ladder design (matched BC547s). I don't know much about the 303 filter but have seen people say it's also a diode ladder clone. (I have noticed that the "acid" filter mode on my bass station sounds nothing like either one!) I am running the synth out through a Sansamp Para Driver DI (a guitar amp sim pedal) which has a sweepable mid boost before overdrive - really pulls up that squelchy sound.
@@extralifedisco hmm the Moog filter is indeed a transistor ladder but this is not the same as a diode ladder (which is in the 303). So I guess this specific sound is indeed mostly because of the drive/amp-sim you are describing. Thanks!
I'm in love with jam at the end.
Loved the DYI Play More Music sticker
This is so refreshing, next time predrill bigger holes with a smaller drillbit
Ah, panels... I tried cardboard, thin wood, but the best material is aluminum for sure. Make sure to drill the panel before soldering any component that goes through it, so you can solder those components with the panel on (I remember spending hours trying to adjust jack holes to make the panel fit).
Great episode as always!
Just a (very late!) note that filing aluminium gums up files, a process called “pinning”, the preventative is to chalk the file, easiest done with those giant chalks for sidewalk scrawling. And it’s fun having multi-coloured files.
this is dope!! congrats
wow a rellay great serie!!!
That acieed sounded sweeet!
It's back!
whatmultimeter is used ofr the calibration? needs to be one with a high accuracy.
Nice video ! But next time use a center punch for your holes. It helps a lot ! Also give yourself a bit wider tolerance at the dimensions. From the other side the rework of the faceplate gives your sequancer its charm
Great way to start the new year!
edit: I'm curious how you approached the issue of the first step. Analog sequencers tend to not play the step the sequence starts on as the clock pulse advances to the next step right away. Digital sequencers typically will play the first step and I've always wondered what the trick behind that was.
I noticed I was losing the first step in my first tests and in the next build I changed the initialization so that the "current step" value starts at 16 rather than 1. Seems to work for now - there's also the external "reset" input to set the start instantaneously - my midi-cv doesn't have a reset out but I do want to test with a more standard euro clock (pamela's workout maybe?).
@@extralifedisco Very interesting. Is your sequencer based around a counter? I noticed that my Varigate 4+ doesn't actually respond to the clock gate (especially gate patterns that aren't really clocks like euclidean sequences or drum patterns) the way my SH-101 does. The SH-101 is one gate one step, so I get a straight forward variation on whatever is in the buffer. The Varigate 4+ has ratchets and multiply/divide options so it averages the pulses to determine bpm and then resyncs the step as best it can but the result is often charmingly wonky..
@@robertsyrett1992 Ah, clock multiplication is more art than science. Mine is all in software, so when it gets a change from low to high on the clock input, it stops the internal clock, advances a step, and counts milliseconds between pulses (solely for duration and glide length).
The clock input has a transistor buffer to bring higher voltages down and lower voltages up to 5v. My MIDI-CV module has some crazy low impedance clock out, so it doesn't even work when hooked up through a resistor divider - that might be a difference between your devices (input buffer design). My CD4017 counter sequencer triggers on input low instead of high, but otherwise seems to keep time pretty well.
@@extralifedisco Thanks for answering some extra-geeky questions :)
Wow looks amazing, i might try to make on of my own. What king of VCO are you using?
respect
heyy this is really starting to come together, looks like a really fun sequencer to play with. Digging the advanced features you've implemented, switching to a proper IDE and breaking it all out into classes helps so much. I think I got to around 800 lines before i had to fully refactor, but it was totally worth it.
re: pattern storage - are you looking at FRAM or flash storage? I was originally going to try FRAM but I kinda want to try leveraging the EEPROM as my sequence encoding is quite compact. the read/write times (especially write) are pretty harsh though :/
Looking at using Flash - had in mind the W25Q80DL because it was through-hole and big on adafruit/sparkfun, but seems to be obsolete. Will try breadboarding with it but may switch to a more current chip - in any event all the new flash stuff is 3.3v only so requires some level shifting with transistors or another IC. Excited about it as it's really the last thing to prototype before spinning up some boards!
Ahh yep yep. I've been trying to keep everything 5v to try and minimize the component count and the overall complexity of the circuit as this is my first proper dive into circuit design. I'd definitely like to have my name put down on some kind of a list if you have plans to sell boards!
Did you have any intentions of pushing the code to github? Apologies if you've articulated this in any of your previous videos, I just have a really bad memory :P
@@JarrydNielsenLMAO Yes, planning to sell boards at some point! Will start a mailing list when I'm closer to it. The code is indeed up on github: github.com/matthewcieplak/super-sixteen Still a mess, no classes yet but if you have refactoring ideas I'm all ears!
@@extralifedisco Ahh yes I have actually been to your github before, silly me!
I can definitely see some similarities in the way we've chunked our projects out, as well as a few differences. I probably can't offer any solid advice for refactoring as i work on my code in a very incremental way, but feel free to have a dig through the repo for one of the sequencers i started working on last year - github.com/atomsmasha/ModularSequencer
Might have to pinch your portamento code for a new project i've started working on, if you don't mind :3
Wow this sounds pretty 303-like. What other parts contribute to this and how? Is the filter explicitly a diode-ladder for example?
Yep, the filter is a YuSynth minimoog VCF, which is a transistor ladder design (matched BC547s). I don't know much about the 303 filter but have seen people say it's also a diode ladder clone. (I have noticed that the "acid" filter mode on my bass station sounds nothing like either one!) I am running the synth out through a Sansamp Para Driver DI (a guitar amp sim pedal) which has a sweepable mid boost before overdrive - really pulls up that squelchy sound.
@@extralifedisco hmm the Moog filter is indeed a transistor ladder but this is not the same as a diode ladder (which is in the 303). So I guess this specific sound is indeed mostly because of the drive/amp-sim you are describing.
Thanks!