My Mini Moog Voyager. Click to go to my web site...

Saturday, May 29, 2010

General Progress

The prototype EG boards have shipped and the prototype VCO boards are in production! I have further refined the layout for the EG and VCO, and have a prototype layout ready to go for the VCA/mixer and VCF. As soon as I get the EG boards in, I'll send these ones in.

I've started the sequencer layout, and it is going to be a mess. Fortunately, there is a lot of redundancy, so a lot of the discrete diodes, leds, resistors, and pots should be easy. My dilemma is whether or not to try to put the pots on PCBs. It is so much easier to wire them this way, but it also limits the panel design. I think I will just space the pads to allow for the 1 mil headers and will use flying wires.

Even though I may never be able to share my work on this project, I'm nevertheless trying to make it user friendly. I'm adding construction note to the silkscreen, I'm laying it out so that monolithic matched transistor pairs can be used or discrete parts, etc. Especially for the VCO. I laid it out so I can use a THAT340 with NPNs and PNPs on the same chip, SSM2210s or LS319s and a discrete PNP that should be close enough to touch the NPNs and thermistors, or all discrete, hand matched parts. I also labeled the emitter, base, collector, source, drain, and gate on all the transistors so one can easily use different small signal transistors. I'm also doing it to minimize errors stuffing the boards when I build them...

Tuesday, May 25, 2010

VCO PCB ready to order

The envelope generator boards are still in production, but they are etched, and screened.

The big news is I have the prototype VCO board ready to go. It can be stuffed as a Steiner Type A, B, or C, but I will just build the Type A which has sin, triangle, saw, and pulse waveforms. Here is the Eagle layout with traces not shown for clarity.

While I think it is cleaner to have PCB mounted pots and/or jacks. I think that for the Steiner modules I'll use flying wires. This allows the most flexibility for panel format and design.

Next, I think I'll tackle either the Type 3 envelope generator or a VCA. I'm also refining the Type 1/2 envelope generator layout. I've hit on a system for placing the parts which seemed to work really well on the VCO card. I'll try it on the envelope generator and will make the board a bit more compact and efficient.

Sunday, May 23, 2010

Which modules and general progress

What modules to build?

One decision I am facing right now is which modules and how many. The basic studio system included:

  • 1 - VCO-A - 4 wave forms, sine, triangle, saw, pulse
  • 1 - VCO-B - 2 wave forms, sine and saw
  • 1 - VCO-C - 1 waveform user specified to be square, triangle, or saw
  • 1 - VCF - selectable HP, LP, or BP, classic...
  • 1 - Dual Random Voltage Generator
  • 1 - VC Trigger generator
  • 1 - Triple Envelope Generator
  • 1 - VCA/Mixer
  • 1 - Noise Generator
  • 1 - Ring Modulator
  • 1- Tuner/Monitor
  • 1- Input pre-amp
The Synthasystem added:
  • Three Octave Keyboard
  • 1 - Phaser
  • 1 - Frequency Divider
  • 1 - Sequencer
Optional modules included:
  • Peak Selector - trigger based on the signal input level
  • Dual Voltage Follower - "The voltage follower does the same thing for control voltages that the Portamento doe to keyboard voltages."
  • Voltage Processor - 6 input, two output mixer. DC/AC signal compatible. Outputs A and B are the same except for the phase.
  • Selective Inverter - I have no documentation on this module.
I don't have all the schematics I need for a complete Synthasystem. I'm missing the Phaser, Frequency divider, Ring modulator, Pre-amp, and Tuner/Monitor. I have the others as either original Synthasysem modules or from the Synthacon.

I'm thinking the Tuner/Monitor and Frequency divider aren't really important to the sound of the system so I could cobble something together if needed. The Phaser and Ring Modulator, however, would be unique to the Synthasystem. These ones will require some thought on what to do. When I was first building this in the 70's, my thought was to mimic a Moog system 55. The 55 did not have a Phaser or Ring Modulator, so I probably would not have built them anyway, if I do build these, I'll find something "vintage" and simple and add it.

For my resurrection, and to match the general function of the Moog 55 I wanted to mimic, I've decided to build:
  • 7 VCOs. All type A to give all the waveforms.
  • 5 VCAs
  • 3 VCFs. This way, I can have one LP, one HP, and one BP
  • 1 Noise
  • 2 Triple EGs. This gives me 6 total EGs
  • 1 or 2 Sequencers
I'll probably also add:
  • 1 Dual Sample and Hold
  • 1 or 2 Dual Voltage Followers. I don't have a Steiner schematic for this, but I think I can find something "vintage" to use. The Synthacon keyboard has a portamento control I might be able to use.
  • 1 VC Trigger Generator. I only have the Synthacon schematic which does not include any voltage control. I will have to add this, so this won't be strictly a Steiner design.
  • Voltage Processor. I'm pretty sure Nyle would have used his mixer core with some Op-amps on the inputs/outputs to provide the normal and inverted signals.
  • A keyboard controller. I'm not really sure on this as I don't have the Synthasystem schematic with the legato play mode. I have the Synthacon schematic. What I'll probably do is build a module that has a keyboard CV in, and standard voltage gate in and then use the Synthasystem schematic to add portamento, voltage gate to S-trigger, and keyboard S-type gate, and maybe the pitch bend. Nyle had a keyboard interface module, and this will be similar and will let me use a modern keyboard controller like the .com one.
I'm thinking of adding some modules for which I don't have specific Steiner schematics, so I'll have to roll my own:
  • Tuner/Monitor. The Tellun/Dave Brown A440 tuner and monitor would be useful here. I have an extra A440 chip, and I might even add an O-Scope like I did to my other modular.
  • Phaser. If I can find a simple vintage transistor based phaser, maybe with a BBD delay, I'll build it. It won't be Nyle's, but, in keeping with my original desire, I may not have done a 100% Steiner system. In fact, I know I built a cross fader because Walter (at the time) Carlos used that effect in one of his SOB renditions.
As for the Ring Modulator, I'm not too keen on them, but if I can find a design which incorporates the same controls as Nyle's did, I might build it to fill out the box.

Panels will be Euro size. Nyle used a 5.75 inch tall and 1.75 inch width multiple panels with 1/8 inch jacks, so a Euro, or Frac for that matter, size is in keeping with the original scale and I have the advantage of being able to buy ready made power supplies and racks. I'm not sure, but I expect the finished modular will probably take three or four racks.

I want to try to keep the same layout as the Synthasystem, but will depart from it if it means substantial space saving. I will use similar, but probably smaller knobs so I can be a bit more compact.

VCO layout

I have the VCO schematic entered and some of it partially simulated. The core simply isn't working and I'm convinced it is the transistor "diode" component that is the problem. So, I'm going to try to bread board it, but I'm going to lay it out so I can use discrete components, a monolithic NPN pair like the SSM2210, or use a monolithic NPN and PNP pairs like the THAT340. I'm pretty excited for that.

Envelope Generators

I'm thinking I'll get the EG boards sometime this week. I'm excited to get them and build one as a ENVDL-1 and ENVDL-2 type. I will build a few for my other modular for fun if they work well.

Tuesday, May 18, 2010

EG Working

Thanks to some great help on the SDIY list, I know what the problem was with the EG simulation. There is a transistor with the base wired to the emitter and then wired backwards to a diode which resets the Flip-Flop. SPICE did not like this and did not properly model it. Oren, from the SDIY list was the first to recognize this transistor was acting like a zener diode. I merely replaced the transistor in the simulation with a zener diode and the circuit works perfectly. The zener voltage determines the peak voltage for the envelope, for a transistor that will be about 5 volts.

Here is a screen shot of the EG. The red trace is the "key down" and the green is the output. The first trace shows that in normal operation, even a short duration key down causes the EG to peak then decay. Next, I held the key down past the peak to show the EG will peak, then decay to the duration level, then decay to 0 with key up. Next is a trace with the Damp switch on. The EG decays immediately and quickly to 0 after a key up. Last is the Quathin switch on. It, too decays immediately on key up, but decays with the decay setting rather than quickly as with the Damp.

Monday, May 17, 2010

Envelope generator PCBs ordered

I ordered some prototype envelope generator PCBs this weekend. I'm excited to get them in to try.If they work fine, I'll buy several more for the modular.

The PCB is labeled for BC550 and BC560 transistors, but I will probably use the original 2N5172s and 2N5138s to stay authentic. For the second run, I might change the package to reflect the pinout for the vintage parts.

Sunday, May 9, 2010


The latest development was a switch in SPICE simulation software. I bought a copy of Multisim to replace the B2 SPICE I had been using. What is comforting, but also aggravating is the EG circuit works exactly the same in  both versions of SPICE. I got a reply on the B2 SPICE forum which was quite helpful, but there must be something unique about the design. Both the Synthasystem and Synthacon EG's are similar enough to tell me I don't think there is a schematic error unless it was made twice. I just really wish I could get the SPICE model working. I'd really like to start layout of PCB's and panels.

This is actually a module I'd like to build for use in my other modular as well as in the "tribute" modular which will stand alone.

In the interim, I got the PCB's in for the Quad VCA I put together. It is derived from work that Marc Barielle and David Dixon did, with some of my own tweaking to make them what I was wanting. I did the layout in Eagle. I also got the parts in, and some panels ordered. I might take a break from the Steiner system and get these built. I also have to calibrate my recently finished JH Varislope VCF.