My polysynth is finally finished :D

Discussions about anything analog, digital, MIDI, synth technology, techniques, theories and more.
User avatar
D-Collector
Expert Member
Expert Member
Posts: 1233
Joined: Sat Dec 16, 2006 3:01 pm
Location: Norway

Re: My polysynth is finally finished :D

Post by D-Collector » Wed Jun 08, 2011 1:41 pm

adamstan wrote:Yet another recording. Time for some 80's synthpop :-)
I'm speechless, it's so nice! Not only do you build your own polysynth, you also make music like a champion!

Super work! :shock:

User avatar
crystalmsc
Synth Explorer
Synth Explorer
Posts: 2862
Joined: Thu Jul 13, 2006 7:15 am
Gear: uWaveXT,uQ,Pulse+;Ion;MiniBrt,Karma,Mntrb,Volcas,01R;Venom,NordModular,L2X;Evolver;JP8080,MKS70,JD990,D50,MC909,MC80ex,Microcon2;TG77,RS7k,AN200,AN1x
Band: Crystaline
Location: Synth Garden
Contact:

Re: My polysynth is finally finished :D

Post by crystalmsc » Wed Jun 08, 2011 4:01 pm

Lovely 80's synth song and great synth.
Kaossilatron - Voicillator
http://crystaline.bandcamp.com
Station: Ableton Live 10 Suite, Obscurium, Push 2, Ultranova, MS-20m, Wavedrums

User avatar
gordonmerrick
Newbie
Newbie
Posts: 14
Joined: Thu Apr 28, 2011 4:05 pm
Gear: Telharmonium, Novachord, Clavivox, Thereminocello

Re: My polysynth is finally finished :D

Post by gordonmerrick » Fri Jun 10, 2011 12:46 am

Truly amazing work.

User avatar
adamstan
Active Member
Active Member
Posts: 511
Joined: Tue Aug 22, 2006 2:56 pm
Real name: Adam
Gear: Yamaha 2xDX7II|QY-70|PSR-S750|PSR-2000|TYROS|Electone D85|Electone E50
B5 DIY polysynth
KORG Poly61
Vermona DRM
Solton Programmer 24
LEMA EDD-5
Location: Poland
Contact:

Re: My polysynth is finally finished :D

Post by adamstan » Tue Feb 19, 2013 8:22 am

Hello again :-)

Recently, I've started to notice, that pitch CV in my synth suffers from 50Hz mains hum bleeding into it :-( This results in higher notes sounding "dirty", as they have subtle FM applied to them. It really annoys me. It is probably caused by some grounding problems - the wiring is true rats nest... And CVs are travellling via unshielded ribbon cables, so they can pick up all sorts of interference. I suppose, that almost complete rewiring may be needed :-(

Meanwhile, I've designed another version, with different approach. In polysynth with patch memory CVs are generated digitally, especially when one uses software EGs. So, considering that digital parts are currently dirt cheap as opposed to 70's or 80's, my new voicecards would have their own CPU and DACs, so each card would be complete, self-contained monosynth, with parameters controlled via serial interface. I think that it's better than generating all CVs on mainboard and then distributing them using cables that pick up interference (see paragraph above). And it makes multi-timbrality possible :-)

Another design change is to replace VCAs with DCAs made with DACs. Why? Let me explain:
- When all CVs are generated with software, there is almost no benefit in having VCAs. It is just another multiplying stage. First, DAC multiplies constant reference voltage by its digital input, thus making CV, and then VCA multiplies audio signal by this CV. So really what we do, is multiplying signal by digital value, but using intermediate CV. Let's do it directly then! Let's just feed signal into Vref input of DAC, and we get simple and elegant digitally-controlled attenuator.

I don't know if and when such version will be built, but I wanted to share the idea with you.
Man with a tape recorder | Living in the '80s ;-)

User avatar
StepLogik
Expert Member
Expert Member
Posts: 1493
Joined: Fri Dec 29, 2006 4:16 am
Gear: Gauntlets of Ogre Power, Vorpal Sword, Chain Mail +3, Ring of Invisibility, Staff of the Magi, Boots of Speed
Location: Boston, MA

Re: My polysynth is finally finished :D

Post by StepLogik » Tue Feb 19, 2013 3:40 pm

I like the new voice card idea! What about LFO's? Will each voice card have its own LFO that syncs with the "main" CPU or will the main CPU just send parameter offsets for values that are modulated by a master LFO calculated in the "main" CPU?

I assume the "main" CPU will handle keyboard scanning, panel scanning, MIDI, patch storage, and voice allocation? I suppose you would use a chip-2-chip serial bus (such as SPI) to allow the "main" CPU to direct the slave voice cards? Would you use MIDI as the protocol for the internal serial data from main CPU to the slave voice cards?

Mooger5
Expert Member
Expert Member
Posts: 1459
Joined: Tue May 06, 2008 5:11 pm
Location: Lisbon

Re: My polysynth is finally finished :D

Post by Mooger5 » Tue Feb 19, 2013 4:56 pm

adamstan wrote:Hello again :-)

Recently, I've started to notice, that pitch CV in my synth suffers from 50Hz mains hum bleeding into it :-( This results in higher notes sounding "dirty", as they have subtle FM applied to them. It really annoys me. It is probably caused by some grounding problems - the wiring is true rats nest... And CVs are travellling via unshielded ribbon cables, so they can pick up all sorts of interference. I suppose, that almost complete rewiring may be needed :-(
Hi! Is the PSU well filtered and regulated? If so, is everything CV-related (wires and components) well away of the mains transformer and internal mains wiring?
Also, use star-grounding scheme, and separate signal ground from mains earth. If the signal wires have to cross PSU wires, rearrange them so they cross at 90º if possible.

Hope this helps!

About the improvements, careful not to overengineer or your polysynth may turn into a VA :mrgreen: kidding...
Herrare umanum est.

User avatar
adamstan
Active Member
Active Member
Posts: 511
Joined: Tue Aug 22, 2006 2:56 pm
Real name: Adam
Gear: Yamaha 2xDX7II|QY-70|PSR-S750|PSR-2000|TYROS|Electone D85|Electone E50
B5 DIY polysynth
KORG Poly61
Vermona DRM
Solton Programmer 24
LEMA EDD-5
Location: Poland
Contact:

Re: My polysynth is finally finished :D

Post by adamstan » Tue Feb 19, 2013 6:03 pm

@Mooger5
When I connect scope to PSU output (with everything connected of course), there is almost no ripple or noise - below 10mV p-p. CPU is regulated, it uses 78xx/79xx regulators and 6800uF filter capacitors before and after the regulators. However, it seems that I have very noisy ground due to some ground loops. I've just connected the scope between ground of the main CPU board and ground of DAC board, and there was around 80mV p-p of noise, hum, spikes etc.

Unfortunately, my mainboard has HUGE design flaw - the LS154 chip used as address decoder isn't strobed while switching, and I suppose it is main source of noise in this circuit. Sadly, I can't correct this without rebuliding and redesigning the whole thing as I've run out of CPU pins :-(.

I will try to find and break any ground loops, and bring the grounding as close to star as it can be. I too hope this helps :-)

@StepLogik

I don't have these LFO things sorted out yet. But yes - the mainboard would do just what you've described.
Man with a tape recorder | Living in the '80s ;-)

Mooger5
Expert Member
Expert Member
Posts: 1459
Joined: Tue May 06, 2008 5:11 pm
Location: Lisbon

Re: My polysynth is finally finished :D

Post by Mooger5 » Tue Feb 19, 2013 7:05 pm

OK, there is no ripple, but the physical presence of the mains transformer might be inducing the hum, that´s why the sensitive circuits should be located away from it.

I´m not familiar with the ls154 and the switching process, but I guess it´s in the order of the MHz?? More decoupling could be of some use. Try to place the decoupling caps across the chip´s power rails as close as possible. These are supposed to eliminate spikes and high noise, but the interference will still go down to 50 Hz, I believe.

When star-grounding one should also separate digital and analogue grounds. Connect all digital ground wires to a single point and all analogue ground wires to another single point. Then join both points with a thin solid core wire. The middle point of the wire is then connected to the PSU ground.

The earth wire from the outlet should connect to another point in the chassis, near the IEC inlet or where the power cord goes to. You could try to plug the cord into an earthless outlet to see if the hum goes away, but it´s nof a safe procedure. The secondary (or dual secondaries, depending on your PSU design) of the transformer shouldn´t be in contact with mains earth. Anyone in the know correct me if I´m wrong, as this not a matter to be taken lightly.

A good solution would be separate PSUs for analogue and digital. Either dedicated transformers or a single transformer with dedicated secodnaries.

Can´t think of any other solutions atm. I guess you know all this, but well, just trying to help. Good luck!
Herrare umanum est.

User avatar
adamstan
Active Member
Active Member
Posts: 511
Joined: Tue Aug 22, 2006 2:56 pm
Real name: Adam
Gear: Yamaha 2xDX7II|QY-70|PSR-S750|PSR-2000|TYROS|Electone D85|Electone E50
B5 DIY polysynth
KORG Poly61
Vermona DRM
Solton Programmer 24
LEMA EDD-5
Location: Poland
Contact:

Re: My polysynth is finally finished :D

Post by adamstan » Tue Feb 19, 2013 7:17 pm

Well, I've found the culprit of the mains hum induced in oscillator CV. It was so simple that it almost made me laugh at myself. Master tuning pot! It was a 50k pot, that had additional 150k resistors at both ends to limit its range. So it made quite poor voltage source, due to high impedance. And it was connected with quite long, unshielded wire directly to the modulation bus. The wires from the PSU were also running through the whole length of the instrument. No wonder it picked up everything it could. When I disconnected it from mainboard, almost all unwanted modulation was gone. So I just need to rearrange this - probably I should add some attenuator with filter between the pot and the CV bus, instead of current direct connection.
Man with a tape recorder | Living in the '80s ;-)

Mooger5
Expert Member
Expert Member
Posts: 1459
Joined: Tue May 06, 2008 5:11 pm
Location: Lisbon

Re: My polysynth is finally finished :D

Post by Mooger5 » Tue Feb 19, 2013 7:20 pm

Good to know it was so simple to sort out! Cheers!
Herrare umanum est.

User avatar
bochelli
Supporting Member!
Supporting Member!
Posts: 482
Joined: Wed Jun 28, 2006 11:57 pm
Gear: minimoog polymoog memorymoog multimoog micromoog minitmoog and tons more.
Location: UK

Re: My polysynth is finally finished :D

Post by bochelli » Wed Feb 20, 2013 12:06 am

Well done Sir , truly amazing Moog take note you might learn something.
yamoto773

cgren72
Active Member
Active Member
Posts: 587
Joined: Wed Mar 17, 2010 11:41 pm

Re: My polysynth is finally finished :D

Post by cgren72 » Wed Feb 20, 2013 1:32 am

adamstan wrote:Hello again :-)
...

Another design change is to replace VCAs with DCAs made with DACs. Why? Let me explain:
- When all CVs are generated with software, there is almost no benefit in having VCAs. It is just another multiplying stage. First, DAC multiplies constant reference voltage by its digital input, thus making CV, and then VCA multiplies audio signal by this CV. So really what we do, is multiplying signal by digital value, but using intermediate CV. Let's do it directly then! Let's just feed signal into Vref input of DAC, and we get simple and elegant digitally-controlled attenuator.

I don't know if and when such version will be built, but I wanted to share the idea with you.
That is interesting, I'd like to see that.

thehighesttree
Newbie
Newbie
Posts: 84
Joined: Mon Sep 12, 2011 10:42 pm

Re: My polysynth is finally finished :D

Post by thehighesttree » Fri Feb 22, 2013 9:52 pm

Just signed in to say I'm mindboggled. Outstanding work; genius! Glad to hear that you've diagnosed and are resolving the interference issue.

User avatar
adamstan
Active Member
Active Member
Posts: 511
Joined: Tue Aug 22, 2006 2:56 pm
Real name: Adam
Gear: Yamaha 2xDX7II|QY-70|PSR-S750|PSR-2000|TYROS|Electone D85|Electone E50
B5 DIY polysynth
KORG Poly61
Vermona DRM
Solton Programmer 24
LEMA EDD-5
Location: Poland
Contact:

Re: My polysynth is finally finished :D

Post by adamstan » Thu Jul 02, 2015 8:21 pm

Today I've managed to greatly improve S/N ratio - the instrument is MUCH quieter now - as it should be. During last five or six years, I used very bizarre output stage - output from each voice was fed into SSM2164 VCA controlled from volume pot, then outputs of these VCAs were mixed, and sent to output jack. This board was leftover from earliest design stage, where every module was meant to be on the separate board, and these were going to be final voice VCAs. I thought "well, there are SSMs on it, I must use it somehow" ;-) That was the worst thing I could do, and the source of annoying hum, noise and buzz, especially at full volume. Why? Well, the volume pot was powered directly from +5V line, the same that was powering all digital circuitry. Of course it was very noisy - and all that noise has been amplified by every VCA. Today I replaced it with simplest unity gain summing amp, passive volume control and buffer:
voices -> mixer -> volume pot -> buffer -> output.
And now it is as quiet as a mouse :D

I hope to finally make proper demonstration video this weekend.
Man with a tape recorder | Living in the '80s ;-)

User avatar
adamstan
Active Member
Active Member
Posts: 511
Joined: Tue Aug 22, 2006 2:56 pm
Real name: Adam
Gear: Yamaha 2xDX7II|QY-70|PSR-S750|PSR-2000|TYROS|Electone D85|Electone E50
B5 DIY polysynth
KORG Poly61
Vermona DRM
Solton Programmer 24
LEMA EDD-5
Location: Poland
Contact:

Re: My polysynth is finally finished :D

Post by adamstan » Sun Jul 05, 2015 5:05 pm

And the bloody thing broke down again. Just as I was recording demonstration. Second voice keeps on droning - it reacts to envelope, but doesn't shut off completely. Probably the transistor on the VCA CV input went belly up. Damn...
Man with a tape recorder | Living in the '80s ;-)

Post Reply