Interested in dipping my toe in the water with DIY analog synthesis,

you already have a computer which can do more than 1000 synth circuits without once burning your fingers on a soldering iron. it's not 1950.

>>1020568
this board is called /diy/ - Do It Yourself, not /ddiy/ - Don't Do It Yourself

>>1020569

Exactly this

the most active section of the ''analog synth'' discussion groups is the obituary section. this is an obviously fruitless course, and it's our duty to at least point out more fruitful approaches. hitting your head against a brick wall is not DIY.

>>1020593
Yeah, look at muff-wiggler and gearslutz, not like they have literally thousands of active users. The eurorack market is growing not declining

File: 20160620_231744.jpg (2MB, 2448x3264px) Image search: [Google]

2MB, 2448x3264px

>>1020564
i*m getting into eurorack as well.
built some pcbs form synthrotek and other places.

I scratchedbuilt a few other like a barebones vco and a dual lfo.. also a sequencer but im not that happy with it.

1v/octave is the standard and for good reason.
makes modulation very simple and voltage control easy to understand in a musical context,
even if its a bit of a hassle to set up.

the synth DIY group on facebook is my top recommendation.
people of all skill levels post, and they are all very helpful.

im not all that synth savvey yet, but ive been an electronics technician for almost 6 years, and hobbyist for even longer.
I help out where i can.

I'm currently working on a quad VCA, a basic ADSR and maybe a CV keyboard instead of the sequencer.. not sure.

pic is the dual lfo.
its got rate(duh) symmetry and level knob for each.
the symmetry skews the triangle wave so you can shape it into saw, ramp or anything in between. very cool for the VCF.
the level just cuts down the output level. simple and useful, but missing on may lfo's for som reason.
my synthrotek VCLFO has lots of stuff, but not level adjustment... wierd.

>>1020629
>>1020629

Very nice. I'm not on Facebook, never will be.

When you say scratch built you mean from an existing design or literally nothing?

File: 20160411_175250.jpg (3MB, 3264x2448px) Image search: [Google]

3MB, 3264x2448px

>>1020634
>not on facebook.
I understand.
though i feel being in specialinterest groups makes it worth while.
the daily jibber-jabber of friends and family is pretty tedious, i admit.

bu scratchbuilt, I mean no pcb or kit, just protoboard and parts i got.
mostly from some design, but the lfo was pretty much my design, though probably not all that original.
its a pretty simple thing, so there is not all that much room for original, i suppose.

the vco is from a design.
http://schmitzbits.de/vco4069.html

pic is the vco.

I'll check that out. Thanks

>>1020648
Did you build your own case, PSU and power strip?

File: 20160414_233746.jpg (3MB, 2448x3264px) Image search: [Google]

3MB, 2448x3264px

>>1021015
yes, yes and yes.

The psu i made myself form a multi tap transformer from ebay.
The usual synth power supplies regulate 5v from the already regulated +12v.
Seeing as the +12v rail is the one seeing the most load i though this wasnt all that great.
On my psu the 5v has its own transformer winding, rectifier,caps and of course regulator.

the power strip is just stripboard cut in three and soldered together with a bunch of 2x8 pinheader sockets soldered on.

the 'case' is for now just a single set of rails with some thin wood on each side to hold it together.

I i got two other sets of rails, so i can make a 3x3U case someday.
I'm not sure what setup i want though. console setup perhaps?

I got into this stuff around 2010 where i choosed the 5U/MOTM format because i've heard they were easier to diy due to the size, i had a lot of components suitable for that (jacks, cables etc.) due to my guitar effect DIYing and Ikea RAST cases were cheap (i did not knew anything about woodworking back then)

i used to visit these:

muffwiggler
http://electro-music.com/forum/forum-112.html
cgs synth
yusynth
bridechamber

i think some of these stopped operating but synthcube and thonk is still working and i used to deal with them.

i have never finished my modular btw but once i finish my 3 TH555VCOs it will at least start to work.

>>1021106
5v , 1vprr octave, limited to 5 octaves?


Got a schematic for the PSU you built?

File: 40106 vco.jpg (43KB, 956x554px) Image search: [Google]

43KB, 956x554px

>>1021162
well its eurorack, so it has +-12v and a 5v rail.

i can pump the full 12 volts into my vco, but my simple vco doesn't track well beyond 5 octaves.

since it starts at 0, five volts would be 6 octaves.
mind you, 5 octaves is quite a lot.
if you start at A3 whitch is 220Hz, five octaves up you got A8
A8 is 7040hz.

the vco goes from subsonic to supersonic, but its accuracy tapers off beyond 4-5 octaves.

i also got a VCO form MFOS which is much better and has more waveforms, but the pcb is to big for the format so I've shelved it for some other project.
I might cannibalize it and build a better VCO some other time.

>Got a schematic
I never drew one, but its this with a separate 5v winding.
i did it that way because i felt like it, not becuase its necceary.
http://www.modularsynth.ru/wp-content/uploads/2013/10/psu_bipolar_schematic_v1.2.png
i think this is pretty much the standard for eurorack.

pic: interesting barebones design i found

>>1020568
I'm not a faggot that brings a lap top to a gig

>>1021519
Yeah, this wasn't me (op)

>>1021261
Thanks.

Is it common for the output of modules to be between 0 - 5v, or does the standard mean that they should all have to deal with input voltages (via patch cord) over 5v?

>>1021543
>Is it common for the output of modules to be between 0 - 5v,
not really.

its usually more since its mostly 12v.
that example is just something i threw in, not really eurorack at all.

the standard used to not even have a 5v rail, but it got included at some time.

>>1021552
Awesome. Thanks for the info

>>1020593
Not all /diy/ has to have a practical application you know. Sometimes it's just for fun, and building an analog synth is much more fun than using your computer.

>>1022729
I like to think learning analog electronics isn't fruitless at all.

its all arduinos and ARM's these days.
which isn't really a bad thing, but it seems most reach for the MCU for even the most trivial of tasks.
I'm no analog guru yet, but if you cant set up an op amp comparator, you are missing a big piece of electronics knowlege in my opinion.

I've built meany guitar effects and even a few amps, but I must say I've learned quite a bit more these last months messing with modular synthesis.
all the parameters and modulation inputs and outputs demand a bit of understanding, and i enjoy learning about it.. mostly:D

>>1020569
Learning is also "Do It Yourself".
Even if it is learning about DAWs.

Quick question, right now I know nothing about building electronics, but how reasonable would building a theremin be for a first "big project". Schematics in general all look about the same to me so it's hard to judge difficulty.

>>1022978
Shouldn't be terribly hard if you are building it from an existing design, no more than any other circuit with a comparable number of components. Pay very close attention to actual builds though, schematics are only half the story: You might not get the expected performance or have some bugs you might need to track down and that could be non-trivial with little/no experience (or lacking the test equipment, eg: no oscilloscope). RF circuits like that are very sensitive to the actual construction, real life circuits things like conductor length adding in resistance, capacitance and inductance can have noticeable effects on something like a theremin and tracking down and correcting those would be hard for a novice.

>>1022998
A real, functional, musical theremin is not easy.

>>1023067
Not op, you know this from experience?

>>1023598
looked into building one off and on.

the principle is that you have two oscillators running at high frequency and you detune one of them by the capacitance of yourself and the antenna, the resulting beat-frequency is the pitch.

that is the controlled by another antenna-oscillator that runs an amplifier, so you can control the volume with your other hand.

not impossible, but you should have a decent understanding of electronics and signal behavior.

Check out 'Music From Outer Space' OP.

>>1023067
>>1023598
>>1023619
Theremin guy here. Designing one once you understand the basic principles is pretty simple. The main hurdle is calibration. The oscillators generally run up in the lower RF range, and it's pretty crucial to tweak them with the aid of a good scope that stays accurate at those frequencies.

Other factors to look at are timbre, which can be affected by several factors integral to the design, and response on the volume side, which can get slightly tricky, as the amp is usually controlled from a detector tuned to peak at whatever frequency the volume oscillator(s) are centered on, and the width of that envelope can be a little hard to control for with a simple R-C-diode detector.

I'm not a total expert, but I've studied and built quite a few and play one in a band, so ask me anything and I'll answer to the best of my ability.

http://tinypic.com/player.php?v=332nqx3%3E&s=9
phone quality.

working on this.
this is a trigger sequencer, well its really two.

4 first bits of a binary counter and the 4 bits of a decade counter that counts to 4.
this way i can make some interesting beats imho.

some ideas:
-switch to set the clock for the decade counter to Q1 of the binary counter, halving the clock rate for it.
Resulting in 8ths instead of 4ths
I tried this on the breadboard, seemed useful.

-switchable inverter for the decade counter making offsetting the triggers by "half" a clocksycle.
i haven't tried this, but it could be interesting as the triggers would be between the triggers of Q1 on the binary counter.

-internal LFO for clocks that you can switch of with the pot.
I got pots that have an inbuilt switch that disconnects when CCW.
This way i can have the clock in addition to the internal clock to make sudden jumps in the sequence with external stuff.
A straight clock out seems handy as well,
maybe trigger so i have one trigger even faster than Q1.

the board I'm sequencing is a CB55, a boss DR55 clone with no sequencer.
no trigger buffering as of yet, just ordered some 4050's today.

File: Screenshot 2016-07-17 17.24.17.png (10KB, 616x204px) Image search: [Google]

10KB, 616x204px

>>1023813
made this this to illustrate the inverted clock idea.
ill try it later tonight.

File: 20160717_230458.jpg (2MB, 2448x3264px) Image search: [Google]

2MB, 2448x3264px

fixed up my mfos VCO.

replaced the fine/coarse pots with a single 10 turn pot.
the pot was 10k instead of 100k so i needed to replace the range limiting resistors on each side of it by a factor of 1/10th.


the old panel was some scrap pcb i drilled out.
this blank was from some cabinet at the old job.
same for the 10 turn pot and indication knob.

File: 20160717_231326.jpg (2MB, 3264x2448px) Image search: [Google]

2MB, 3264x2448px

>>1024337

File: DSCN4257.jpg (538KB, 1600x1200px) Image search: [Google]

538KB, 1600x1200px

I spent today building a rack/box for my video synth project.

File: DSCN4259.jpg (542KB, 1600x1200px) Image search: [Google]

542KB, 1600x1200px

>>1024405
It fits two monitors in the base and holds the synth electronics in the lid. It'll have table legs and wheels when done.

This is turning into a great thread

>>1024405
>>1024406
cool stuff.
looks really sturdy, is this a road/tour case?

what are the monitors for?
are you the anon with the weird/awesome nes zapper screen synth stuff from way back?

>>1024683
>video synth project.
I need to learn to read.

File: 20160718_141014.jpg (2MB, 3264x2448px) Image search: [Google]

2MB, 3264x2448px

>>1024338
small update.
but my synthrotek sample and hold working.
silly as designers useing a p-channel jfet instead of an n-channel.

had to wait for an order to complete it.
sample and hold works well and is very fun, but the noise generators are kinda lame.

synthrotek expects you to have a scope with a fuggen spectrum analyzer to calibrage the noise generators. lel.

sounds great with just ramp/saw/tri lfo at varying speads as a sample source though.
still, i feel like making my own sample and hold, with internal lfo and gate triggering... maybe even slew.

but I'm getting ahead of myself. i think next i'll see if i cant finish my trigger sequencer(s) and put it on a panel.

>>1024694
FFT not sufficient? Even many budget DSO's these days have FFT functionality.

>>1024702
I got an old-ass analog 10mhz scope.

File: mf2.jpg (112KB, 960x960px) Image search: [Google]

112KB, 960x960px

>>1024683
>are you the anon with the weird/awesome nes zapper screen synth stuff

That's me, but I moved up to a five sensor glove system a while ago and now I'm working on a two monitor two hand system.

> is this a road/tour case
That's the plan. I've got some gig invitations already but I'm putting them off until I've put a lot more work into the project.

>>1024869
fucking sweet man.

video synthesis is tempting stuff, but its awfully hard to find proper info on.
could you point me in the right direction for a simple start?

I'll be sticking to audio for now but its pretty darn interesting.
i assume you work with vga?

>>1024871
> point me in the right direction for a simple start

Here's my project log for the zapper project. You can do it without the zapper and just make a colorful VGA signal.
https://hackaday.io/project/9782/logs

>>1024875
damn, this is great.

question:
do you really feed the 5volt form your 'rainbow' oscillator to the vga connector.

i understand it goes through the porching switces, but is there no attenuation.
i thought vga had 0-07v rgb signals.

anyway. I'm working on fitting the whole sequencer on a 5hp panel.
I worked out i want 1pot, 2 switches and 10 jacks.
i will be tight, but i have faith.

>>1024894
>do you really feed the 5volt form your 'rainbow' oscillator to the vga connector.

Yah. The VGA spec says it's 0-0.7v but every monitor manufacturer realized they needed to auto adjust to whatever the graphics card was giving them before the first models hit the shelves. I've done several VGA projects over the years and never had a problem with 0-5v. I've discovered graphics cards doing this too.

File: 4313041456040082087[1].png (18KB, 1026x1029px) Image search: [Google]

18KB, 1026x1029px

>>1024976
I can just use 4000series equivalents at 5v for this, right?
I dont have all that much 74 series lying about anymore, except for a load of 595 sipo's.

could be cool to just make a box(or even a panel?) for this circuit and just modulate the RGB's with my synth setup. just make sure to clip everything above 5v and i should be fine.
what do you think?

>>1025099
>I can just use 4000series equivalents at 5v for this, right?

Yes.

File: 20160720_031349.jpg (2MB, 3264x2448px) Image search: [Google]

2MB, 3264x2448px

made some progress on the panel for the trigger sequencer.

File: 20160720_031304.jpg (2MB, 3264x2448px) Image search: [Google]

2MB, 3264x2448px

>>1025502
by total accident, the protoboard fits exactly between the jacks. so i just blued them all on.

>>1025504
Why no LEDs?

>>1025519
it might be a bit tight, but i might at leds later when the rest is done.
a led for clock at the very least.

i already use yellow leds for rate/clock indication. so red for steps?

>>1025536
I would probably use same color LEDs but I guess green would look better with yellow, RYO logic modules use that combination.

>>1025545
I really like yellow and amber leds.
looks nice with the black blanks I'm using, I feel that I should vary it up and have different colours for different things.

reds are less intense than green imho. though I could always just have them a little dim.

>>1025614
Where did you get those black blanks btw?

>>1025653
gmsn.co.uk for 1 £ each.
some "bad" stock they had, sold for cheap.
dunno if they still have em.

they got 5 and 2.5 cm or 5 and 10 HP...

>>1025667
Oh, they are indeed cheap, too bad only two sizes and questionable availability for the future. I like to have a visual correspondence within the modules.

>>1023688
Not OP

Reading this now. The first third of the book is waffle/filler

>>1025962
Well, he meant the website and not the book probably. The book is useless since all that information and much more is already freely available on his website.

File: 20160721_073554.jpg (2MB, 3264x2448px) Image search: [Google]

2MB, 3264x2448px

>>1026001
got i bit further yesterday. still not there.
still don't have the 4050 hex buffers, but it will work with nonbuffered outputs in the meantime if i finish before I get them.

>>1026315
Looking good, hopefully those top row of leds have enough place under the toggle switches.

I see you started with the cables for power, what are you doing rest of the connections? I started using enameled wire lately and they seem to do the trick but don't know how reliable they will be in the long run.

>>1022766

Right now I would say there is a massive lack of people doing really powerful diy synths using modern arm chips with DSP.

I'm amazed no one has made a decent synth using the teensy 3 and audio library.

File: 20160722_162605.jpg (2MB, 2448x3264px) Image search: [Google]

2MB, 2448x3264px

>>1026315
enameled wire is fine for most onboard stuff,
as long as they don't see much movement, the tire and break easily.
the nice thing about enameled wire is that its so thin, you can "sew" in tinto the protoboard and hav multiple wires running in the same hole without shorts.

its pretty much done at this point. still no buffers, but it works fine without them.
still have boardspace should i need it later.

the lfo's top range is too fast for the counters so i need to increase the series resistance a bit.
the clock inn doesnt "AND" with the internal clock properly.
when one is high and the other has a rising edge, it doesn't clock.
i feel silly for not seeing this problem before.
a resistor and cap on each of those pulses should fix that.

I'm considering replacing the clock divider switch for the dacade counter with an on-off-on switch so i can divide it down another step, as in 16ths.
no rush though.

File: 20160722_162539.jpg (2MB, 3264x2448px) Image search: [Google]

2MB, 3264x2448px

>>1026552
I'm homeless in about a weeks time.
so i need to rig down the electronics lab and focus on finding places to store my shit.
meaning no electronics for a while i suppose :S
ironic as I'm going back to school in 2 weeks time, technical school, dunno what its called in english.

>>1026554
I also had a cool idea yesterday.

when using the dac0800, how about adjusting the voltage reference so that the pics correspond with musical intervals?

1/12 = 0.08333+V per note
.
if you feed the dac 10.58V you get half a musical interval per Increment in the dac.
meaning every even value on the dac is at a correct musical interval.

if you just ground(by switch?) bit 1 instead of hooking it up to the shift register, every possible output is already "quantized".. sort of, no scales, just chromatic.

the DAC would then have 128 states each at a 0.08333V increment.
0,08333v * 127 = 10.58v
(remember that 0 is a state too :D)

I've wanted to scratchbuild a turing machine for a while, adding my own ideas and testing stuff out.

making a decent adjustable voltage reference shouldnt be too hard, an op amp and a using a diode and zener diode could work fine.
(zeners and regular diodes have opposing temp coefficiencies AFAIK)
something like the lm10 could be even better.

I don't think it would ever be spot on, but how close does it need to be anyway?

I'm planning to build a midibox sid.
I went out and bought 8 SID 6581 chips and the realised that the PCBs for building a midibox sid mb6582 have been out of stock for the last 8 years.
I can't really be arsed to go laying everything out on veroboard.
The midibox site has a PCB layout but it's a PDF.
Is there any way a PCB could be manufactured from this instead of a gerber?

>>1026719
>sids
each chip has 3 voices, what do you need 28 voices for?

I have a sid project too, but its been on the shelf for very long.
I planned on using a single sid chip, breaking out most variables to pots or CV inputs for my modular, and midi for the notes.

>>1026537
>I'm amazed no one has made a decent synth using the teensy 3 and audio library.

Do it then.

>>1026537
Mutable use arm. There are teensy projects iirc

>>1026756
>3 X 8 ≠ 28

Technically, I'll have 12 voices as the 8 SIDS will be in 4 stereo modules

Some rumors about that Ray Wilson has died.
can anyone confirm?

damn shame if true.

>>1027034
I don't see anything other than some chatter on twitter.

Hopefully he is ok, he has even been working on new stuff lately.

>>1027056
>he has even been working on new stuff lately.
he has?
i thought he was out of the game, dealing with cancer. hence no sales on his page.

>>1027056
nah man, he's really dead :S

>>1027166
He closed the shop but still published new designs for people to etch themselves. Nothing groundbreaking but still. VC Spring Reverb and Quad mixers for example.

RIP Ray, like most people, I too moved from building guitar pedals to synths thanks to his site.

F

>>1027166
There's a thread over at muff wiggler, he has died

>>1027335
yeah, my first build was his VCO.
seen here
>>1024337
>>1024338

>>1027384
Mine was Sound Lab Mini-Synth, had to do all the wiring twice to get it to work.

Right now, I am working on a reduced version of his vari-clock sequencer with 8 steps in total and with 8 clock counts max pro step.

>>1027443
cool

oh man, RIP ray

>>1028146
Yeah. Good guy