Jupiter One – Update

Panel Mockup

Overview Back in 2018 I bought an original Jupiter 4 Voice Card with the plan to turn it into a complete single voice synthesizer in a Moog 60 HP case. I added a replica Controller PCB and a new Pot PCB that holds the switches, LED’s and potentiometers. In 2020 I got as far as a prototype but with some problems:

  • Voice Card too close to Controller PCB, causing shorts
  • The 4053 in the LFO was wired to +/-15V
  • There were errors in my schematics
  • It sounded rather thin and weak, with only one VCO
  • BA6110 signals are too hot (+3dB over BA662)
  • The Controller board is too complex for a synth with no presets
  • I could not find suitable 4-way slide switches

Pro Mars Tune

Revised Plan I decided to simplify the design and strip back the Controller PCB, remove the Bend Modulation section, and align the design to be used with an Arturia Key Step which has a mod output (and arpeggiator). The 4-way slide switches were replaced with a rotary switch (LFO waveform), or potentiometers (Key Follow and PW).

To solve the thin sound I took a look at the Roland Pro Mars and realised it uses a second VCO card and not a full voice card. I could do the same, add a second VCO, especially as I had already proven my VCO deign in the AM8400. It requires a small modification to the Roland voice card to enable the second VCO to be patched in before the VCF. I have added the TUNE-A and TUNE-B  potentiometers for VCO2 and the slide switch.

I reworked both the Panel and Controller PCB’s in the Spring of 2021 and 2022, and added a VCO card. Here are the changes:

  • The Voice Card mounting has been reversed with component side facing out
  • Implemented the JP-8 LFO with CMOS switch waveform selection, and a S&H output
  • Implemented the JP-8 LFO Delay circuit (VCA based)
  • S&H modulates the VCF via a dedicated slider
  • PW is a slider rather than 4-way switch
  • LFO waveform selection is on a rotary switch
  • Key Follow is a slider rather than a switch
  • Key Follow circuit removed
  • VCF Envelope output patched to KF0 and then onto PWM
  • Removed the voltage control from the VCO, VCF and VCA modulation
  • Implemented SMD AS662 VCA’s to eliminate the +3dB boost of the BA6110
  • Moved to PTL30 slide potentiometers with LED’s
  • Added a Gate Boost circuit
  • Added VCO2 controls to the front panel
  • Added front panel audio output (as well as rear)

This involved quite a lot of work and the release of the AS662 chip inspired me to get the job done. The synth is really now a Pro-Mars without presets, or at least a Jupiter Two!

Outcomes I ordered the VCO2 PCB in March 2021 and the revised PANEL and CONTROLLER PCB’s in May 2022.

Behringer 1050 Mix Sequencer

B1050 module

Introduction During 2019 I worked with Behringer to develop their range of 2500 modules, and one of the most challenging was the 1050 Mix Sequencer. This post describes how it works and how I designed it, ably assisted by the Behringer Manchester team to achieve production status in late 2021.

The original ARP design dates back to 1969 and uses a large number of TTL chips. They draw a lot of current (500mA in total) and only the original 7474 chips from 1970 will work in the circuit (which are impossible to source). I built my own 1050 replica in 2017 and knew of the various issues in trying to recreate it. The CMOS version consumes only 130mA of power, the majority to run the CMOS logic and LED’s.

B1050 rear

The Design Rather than using TTL logic or a micro-processor I decided to use CMOS logic. I have written PIC assembler code to replicate the 1050 but Behringer use ARM chips, so it was quicker to go down a hardware route. The CMOS logic requires an inversion of the negative logic signals used in the ARP design to positive logic. It all requires a lot of IC’s! Over 30.

The original analog clock circuit has been retained, and miniaturized into SMD, with a BJT pair replacing the unijunction transistor. There is a small user accessible trimmer to adjust the base frequency of the clock, this is factory set and should not need adjustment. The clock runs from 0.5Hz to 150Hz and  generates a square wave, which is divided down by a 4024 binary counter and drives the heart of the circuit a 4555 Dual Binary to 1-of-4 Decoder/Demultiplexer. The limitation of not using a UJT is that the lowest clock frequency of the original (1 pulse every 30 seconds) cannot be achieved.

The analog circuit consists of eight inputs which are controlled individually by level pots and mixed into either a 4:2 or 8; 1 configuration with two outputs. 4580 dual op amps are used for the mixing of the analog signals and 4053 analog switches used to select whether the signal is ON or OFF. A toggle switch controls the 4:2 or 8:1 setting and sets the DG403 DPDT analog switch to the correct mixing positions. I originally specified high quality analog switches in both locations, but Behringer have gone with the 4053 for the input channel switching (lower cost).

B1050 Module

The digital circuit is a bit more complex. The clock divides down via a 4024 binary counter (only 3 stages are needed) to create the 8 channel positions as BCD. The 4555 chip is used to create a sequence of 4:2 or 8:1, with the individual channels switched ON/OFF and latched by the dual 4013 flip flops. These eight signals are than combined (OR gates) with the 4555 outputs and the external 12-pin IDC inputs from a B1027.

An eight position switch selects the number of positions before the sequence is repeated. The manual incorrectly states the rotary switch as 9-way. The LED’s in the white button switches are controlled by these 8 channels with transistor drivers, and the red buttons enable a position/channel to be activated, typically used for auditioning.

The EXTERNAL GATE toggle switch digitally enables the 10 gate positions fed in via the 12-pin IDC socket from a B1027 sequencer. A 12 pin cable is used to avoid confusion with the 10-way power cable. The clock has a toggle switch to select ON/OFF, and a single position advance. An external advance jack socket also enable step advance.

B1050 Switch

There is a 3P3T toggle switch (especially sourced by Behringer) for selecting 8:1 or 4:2 for the analog mixing and the digital sequence, so all 3 combinations are possible. The original used an even bigger 4P3T switch but the CMOS logic does some of the switching. The analog path in the B1050 is not AC coupled, so it can be used for mixing bi-polar CV signals up to 20v peak to peak.

There are three tests points on the rear of the PCB:

  • TP1
  • TP2
  • TP3

Conclusion The CMOS version of the ARP 1050 has turned out really well, with a reliable module at a low price with acceptable power consumption. A microprocessor version would be cheaper to manufacture, use less power but require software development and debugging.

Integrated Circuits Used There are over 30 IC’s in the B1050; ranging from Op Amps to analog switches and CMOS logic. The majority of the chips are on the rear of the PCB with only two on the front. Here is a list of the chips used and the functions:

  • U1 HCF4053 – analog switch
  • U2 HEF4013 – dual flip flop
  • U3 DG403 – analog switch
  • U4 HEF4072 – dual 4 input OR
  • U5 HEF4013 – dual flip flop
  • U6 HEF4013 – dual flip flop
  • U7 CD4071 – quad 2 input OR
  • U8 MC14081 – quad 2 input AND
  • U10 CD4071 – quad 2 input OR
  • U11 RC4580 – dual op amp
  • U12 HEF4013 – dual flip flop
  • U13 HC4024 – 7-stage binary counter
  • U14 HEF4069 – hex inverter
  • U15 HEF4069 – hex inverter
  • U16 HEF4011 – quad 2 input NAND
  • U17 HEF4013 – dual flip flop
  • U18 CD4555 – dual binary to 1-of-4 decoder/demultiplexer
  • U19 CD4071 – quad 2 input OR
  • U20 HEF4013 – dual flip flop
  • U21 HEF4011 – quad 2 input NAND
  • U22 RC4580 – dual op amp
  • U23 – not used
  • U24 – not used
  • U25 HEF4071 – quad 2 input OR
  • U26 HEF4071 – quad 2 input OR
  • U27 – Dual NPN transistor (LED driver)
  • U28- Dual NPN transistor (LED driver)
  • U29 – Dual NPN transistor (LED driver)
  • U30 – Dual NPN transistor (LED driver)
  • U31 HCF4053 – analog switch

 

Jupiter 6 Waveshaper

Original Roland Chip

Introduction The Roland Jupiter 6 (and MKS-80) use a custom hybrid chip (EHM-S226W83S ) for transforming the CEM3340 oscillator waveforms to a consistent level of 5v centered around 0V (-2,5V to +2.5V). The CEM3340 creates waveforms that are at different levels (Pulse 0 to +12V, Triangle 0 to +5V, Sawtooth 0 to +10V) which need level shifting for mixing into the VCF. This is implemented by a mixture of SMD resistors and transistors.

The chip also contains a transistor based sync circuit, as Roland chose not to use the inbuilt CEM3340 sync (weak and strong), possibly because they wanted to control the sync on/off with a +5V programmable signal from the micro-controller and didn’t want use an analog switch.

Roland took this approach to reduce the footprint of the waveform translation to a minimum using a 12-pin SIP package. This enabled the Jupiter 6 PCB’s to be more compact than the Jupiter 8. An alternative approach is to use resistors and Op Amps but this is 5x the PCB size in the days before SMD components could be used.

Roland EHM-S226W83S

Pin Out The EHM-S226W83S pin out can be easily determined from the schematic in the Jupiter 6 Service Manual:

  1.  +15V
  2. -15V
  3. Pulse Output
  4. Pulse Input
  5. Sync Input from second VCO sawtooth
  6. Sync ON/OFF from microcontroller
  7. Sync Output to CEM3340
  8. Triangle Input
  9. Sawtooth Input
  10. Triangle Output
  11. Sawtooth Output
  12. GND

Tightly Packed!

How it works There are four separate circuits; one for each waveform and one for the sync. The triangle wave generated by the CEM3340 is the 0V to +5V and is level shifted to -2.5V and +2.5V by a PNP transistor (Q1) and a resistor network (R2, R5, R6).

The 0 to +10V sawtooth wave is reduced and shifted to -2.5V to +2.5V using three resistors (R1, R3, R4).

The 0 to +12V pulse wave is reduced and shifted to -2.5V to +2.5V using a more complex circuit of six resistors, an external capacitor and a NPN transistor (R13 – R17 and Q4.). The SH-101 service manual uses a similar circuit without the final level shifting. So it is easy to determine what most of the values are without measurement.

The sync circuit is also more complex. The sawtooth wave from the source VCO is switched off by grounding the input to a NPN transistors (Q2), or going high to enable the Sync. The sawtooth is then pulse shaped by a NPN transistor, 2 diodes, a capacitor and 3 resistors, The narrow pulse resets the VCO integrator via its capacitor (Q2, Q3, C1, R7-R12), creating hard sync.

Tear Down For my Jupiter 6 VCO clone I wanted to replicate the EHM-S226W83S circuit, along with the RA3A resistor array connected to the CEM3340. The array resistor values are easy to determine from the schematics and values in the MKS-80 service manual. The Jupiter 6 makes use of all 6 resistors in the array, whilst the MKS-80 only uses five of them.

The EHM-S226W83S was harder to replicate, as although the schematics are in the manual, the values are not printed and are a mystery. I bought one in April 2018 and removed the plastic encapsulation with acetone (took 2 days of soaking), so that I could then measure the various printed on resistors using a multi meter, some components had to be desoldered to be measured.

Prototyping Once the values had been obtained the full VCO schematics were laid out using Eagle CAD. Roland has implemented the VCO with some changes to the original CEM3340 datasheet  which have also been captured.

A prototype PCB with THD components was ordered in April 2022. The AMSynths Jupiter 6 VCO will operate with 12V power rails, so the resistor values will need changing anyway, as the original uses 15V power rails. I plan to keep with THD components rather than building SMD replicas.

 

Synthanorma SQ312

Prototype SQ312

Overview The first Synthanorma analog sequencer was famously used by Klaus Schulze in the Spring of 1975 in Berlin, during the recording of his Timewind album, which was released in the August. The track  Bayreuth Return uses the sequencer in a live recording to 2-track tape. Klaus first saw the SQ312 at the Frankfurt Music fair on February 1975, and he purchased the prototype.

The Synthanorma SQ312 was the first product made by the Synthesizerstudio Bonn, founded by Matten & Wiechers in the early 1970’s. They went on to build a 16 step sequencer for Tangerine Dream – the 316.

In early 1975 there was a limited choice of sequencers for musicians to buy in Europe; only the expensive Moog 960, or the basic sequencer in the EMS AKS. During the next 18 months the options improved with the ARP 1601 (at DM2,800) the PPG 213 (clone of the 960), PPG 314 and the Roland 104 and 717. But at the Frankfurt Fair in 1975 the Synthanorma was a new innovation.

Research Approach The Synthanorma SQ312 has been tricky to research and document, as there is not much information about this very rare innovation from 1975. None of the 20 or so built seemed to have survived into modern times, and pictures of them are rare. The front panels have no or minimal lettering, so I have used the later 316 model as a way to guessing the functionality. An article published in the German Keyboards magazine in October 2004 provides a rare overview of the Synthanorma SQ312 and the image is their copyright.

Production SQ312

Sequencer Features The SQ312 has the 12x step controls on the left, each in a column, and a control section on the right.  There are three rows of potentiometers which set the control voltages for the three channels, below this a multi-way rotary switch, a push button and a toggle switch.  Looking at the later Synthanorma 316 I am guessing that:

  • Rotary switch – 3 way (END, SHIFT, SKIP)
  • Momentary push button – SET, activates the step
  • Toggle switch – ON = STOP the sequence

There is a red LED at the top of each step column to indicate when the step is active.

The control section is harder to work out and changed over the production run. The prototype that Klaus used has less controls than the ones used by Edgar Froese. The first part of the control section is the CV row controls and indicators, a sequential switch. For each row there is a red indicator LED, a rotary switch and a momentary push button. I am guessing that:

  • LED- indicates when the row is active
  • Rotary switch – 3 way (SKIP, SHIFT, STOP) controls the row
  • Push button – SET, activates the selected row when used with the step SET

To the right of the sequential switch is the voltage controlled clock. In the prototype there are potentiometers from clock rate and pulse width. Below this is a group of three push buttons and a LED. I am guessing that they are STOP, START and STEP. Below the clock are four toggle switches and one push button. I don’t know what these are for.

In the prototype there is a set of 12 jack sockets below these toggle switches and a push button. At least three of these jack sockets are for the row CV outputs (upper left and maybe duplicated in lower left) and one for the clock output. On production models the number of jack sockets was reduced to ten, and clock section expanded with more potentiometers.

The SQ312 could be synchronized to tape, so some of the jack sockets must be for this feature.  And finally at the top of the control section are two toggle switches mounted sideways. I am guessing they are for transposing the sequence.

If you have any more detailed knowledge of the controls or technical design please use the comment section below. Much appreciated!

Versions The SQ312 sequencer sold in small numbers in Germany (less than 20?) and the design evolved from the prototype to the production model. Edgar Froese had a model which looks like it has two VU meters at the top right. Edgar bought his immediately he saw the prototype at Klaus’s house (which was next door to his in Berlin). I am assuming the sequencer was based on low power CMOS logic, which was becoming available in the mid 1970’s due to its use in high volume digital calculators.

Klaus used his Synthanorma to sequence his ARP 2600 from 1975 to 1977, and used the sync feature when recording to his early 8-track tape recorder. The 2600 and Synthanorma were replaced by a set of PPG modules and dual sequencers.

 

Oberheim Xpander – 2

Pot and Display PCBs

Overview In 2021 I acquired a set of original Oberheim Xpander PCB’s – the Voice and Processor boards, so that I could build my own Xpander at around £2000. This is an expensive and crazy project that requires lots of difficult to source parts.

In 2021 I bought the CEM3372 and CEM3374 chips, and waited until the “Elephant in the Room” had been addressed. I also bought wooden ends so I could get some of the panel dimensions, and panel switches and buttons. I have not located all the correct caps.

Membrane Display

Membrane Panel This is the “Elephant in the Room”, as I cant make a membrane and have to wait until one is remanufactured. In late 2021 Sweet Discrete announced they would make the panel if they got 200 orders (see mock up picture).

I jumped in at order 196 in February 2022 and there are only 4 more orders need to get the membrane into production. I am confident this will happen, so I can start planning the rest of the build. Sweet Discrete make a wide range of replacement membrane panels and are highly recommended.

Pot PCB A smaller “elephant in the room” is that I don’t have a Pot PCB, and I will have to make one, using the membrane panel as a template. This will take a few weeks to lay out and check before ordering one. A one off PCB this size is around £300, so its a major expense and something I have to get right. Whilst a NOS Display PCB is available, the new VFD displays mean I don’t need to use one.

Metal Chassis I need to design a metal top panel, which I will do in CAD software. This will be expensive to have made as a “one off”. The bottom panel may be based on a Roland D-50 base plate. I doubt anyone has an old Xpander chassis but I will keep searching!

Xpander Encoders

Next Big Purchases The next step, in parallel to the Pot PCB design, is to gradually buy the major components needed for an initial build and test;

  • New VFD displays £400
  • Switched power supply £400
  • Processor logic chips and OS chips £200
  • New set of encoders £100

The Xpander Mockup Panel was created by Sweet Discrete and the image is their copyright – More in Episode 3!

Yamaha SPX900 Refurb

Yamaha SPX900

Introduction I bought a used Yamaha SPX900 multi-effects unit in December 2021 to complete the FX rack for my main analog console, the Soundcraft 22MTK which has five AUX sends. The plan is to refurbish the SPX, and use it for delays with pitch shifting and the Symphonic preset. Launched in 1988 (at over£700) the SPX900 was Yamahas first full bandwidth effects processor with 16-bits at 44.1Khz.

With only a mono analog input and multiplexed stereo output it was eclipsed by the more powerful SPX1000 in 1989, which has 4x the sound memory and digital I/O but at twice the second hand price. The SPX990 is a 20-bit version of the SPX900, with the same DSP chip but more memory and effect types.

The SPX900 arrived without much packaging and the rack ears were bent back. It is always important to protect rack ears when packing a unit to avoid this.  I managed to straighten them out but they have cracks in the metal now.

Technical Design The SPX900 uses an 8-bit microprocessor the HD6306, which is a derivative of the Motorola 6800 running at 4Mhz and with a 32kbyte ROM  chip holding the main OS. There is a large main custom DSP chip, as well as custom chips for EQ and modulation, this is when Yamaha were designed their own chips. The main sound memory is 7x 256k bit DRAM chips, with equates to 1.3 seconds of stereo audio signal at 16-bits and 44.1KHz. Delays times of up to 1.48s are possible, with Stereo Delay providing 740ms. In Freeze mode up to 1.35 seconds of mono audio can be sampled and replayed.

The ADC chip is the PCM78 and the DAC the often used PCM56, which deliver the 16-bit performance. The Op Amps are medium quality bi-polar NJM4556/4558’s, and the ones in the audio path could be upgraded with care to faster and lower noise versions.

Effect Types The SPX900 has 13 different effect types, with the spatial effect generated by the DSP chip, whilst others involve the EQ and Modulation chips.

  • Basic Reverbs, Detailed Reverbs, Early Reflection Reverbs
  • Delay, Stereo Echo, Modulation
  • Noise Gate, Pitch Change, Freeze, Pan
  • Compressor, Distortion, Aural Exciter

The majority of the 50 factory presets are SINGLE effects (all with digital EQ), but there are of course some MULTI effects where the different types are combined in series, and DUAL effects where two are combined in parallel (limited to reverbs and delays). There are 49 user memory slots which I will soon fill up!

Custom Early Reflections Preset 17 (Programmable ER) enables access to four sets of user modified early reflection parameters, so you can design your own reverb space. They are edited using the USER ED EDIT function. The parameters are; room size, liveness, diffusion, initial delay, high and low pass filters. This feature is often overlooked as it buried in menus but it is covered in the user manual and is a great way to create new spaces.

New LED Display

New Display The original LCD backlight had faded, and I was keen to replace or even upgrade to an OLED. However the LCD is a narrow version, which reduces the potential replacements, and rules out an OLED. I have used a LED display with white text on blue (the Midas MC21605J6W-BNMLW-V2), just £11 from Farnell.  The height and width of the new display are the same as the original, and the new display fits snuggly against the front panel bezel.

The LCD mounting is a simple snap fit into the false panel, with Yamaha deciding not to use any machine screws. The LCD is connected to the SPX900 main PCB with a 14-way ribbon cable with a captive IDC header on the LCD and a plug/socket on the main PCB.

The new display uses the same cable and pin out but with 2 extra pins (15 and 16) for the power to the LED backlight. I desoldered the header from the old LCD and soldered it into the new display leaving Pin 15 and 16 unconnected. I then wired a 220R resistor from Pin 2 to Pin 15 (+5V) and connected Pin 1 (GND) to Pin 16.

Contrast Trimmer

The contrast circuit at R226/R227 needs replacing as it does not provide the right voltage to VO at Pin 3, the LCD shows no characters. I desoldered the resistors and replaced them with a small 10K trimmer that adjusts the contrast for the display.

The contrast can now be set to have a clear blue background with no ghosting of characters. A perfect blue display with light white lettering. Much better! The contrast adjustment means you can set it up for different positions in your rack, an external trimmer pot fitted at the rear would be a great upgrade.

Recapped PSU

Power Supply Recap The SPX900 has a standard linear power supply rather than a switched variety (as used in the SPX90). I replaced the main reservoir capacitors at R503/R504 (2,200uF 35V) and R505/R511 (2200uf 16v), along with the three 7XXX voltage regulators and all the 100nF bypass capacitors which did look worn.

  • Nichicon 2,200uf 35V
  • Nichicon 2,200uF 16V

I also replaced all the local 10uF/35V capacitors near the Op Amps, 1/2.2/3.3uF near the ADC/DAC chips and 10/33uF BP capacitors in the audio signal path. 100nF ceramic capacitors were also replaced through out the main and power supply boards.

Yamaha SPX990 I bought a cheap SPX990 soon after the SPX900 purchase, which needed no refurbishment as it was in better condition and had less “miles”. The SPX990 lives in the left of my studio in a 12U 19″ rack where it is the fifth FX unit connected to the Soundcraft 22MTK, this enables the 18-bit quality and wide range of effects to be used by my collection of poly synths.

The SPX900 is mounted in a 19″ rack to the right of my studio, where it is used on the FX sends of my Behringer RX1602 mixer. This provides a synth sub mix into the Soundcraft for my Guin Guin MME, Model D, Pro-One and Korg Odyssey (Mk3).

 

Guin Guin MME

Guin Guin MME

Introduction Part of my “Big Moog” setup is a Behringer Model D, which occupies the space where the third bank of 921A/B VCO’s are in the replica Moog IIIP. This added the Minimoog sound which Klaus has used for many years and it avoids the setup issues in the Behringer VCO’s. In 2022 the Model D is being moved and replaced with a bank of AM901 VCO’s, with the MME acting as main lead Moog, whilst the Model D sits on top of the “Big Moog”.

The plan is to add a SCI 700 Programmer (3U) replica above the “Big Moog” to enable preset patches with the Model D. and traditional ADSR envelopes, to broaden the sound. However this approach needs separate VCO CV inputs, which the Model D does not have. I did consider adding a patch bay to the right of the Model D but then I came across the 60 HP Guin Guin MME.

The MME replicates the Minimoog circuit with THD components, except the power regulator chips and 3046 which are SMD. It has CV inputs for the VCO’s, VCF and VCA and is a perfect match for the 700 Programmer. It also enables me to make component changes. The Model D sounds very good and uses SMD PPS capacitors in the filter, so it will be interesting to compare.

I ordered a panel and PCB set in October 2021, and watched them travel across to the UK from France. Very excited to get this project started!

The Build Buying most of the components looked simple initially, especially as Tayda had nearly everything in stock. However the SMD LDO’s were not in stock anywhere, and with 1 year delivery timeframes. However I did find them at UTSource (which I have success with in the past), so I ordered them. The required 38x 2N3392 transistors were also hard to locate, I found 15x at UTSource and the rest from eBay. I  later found that Futerlec have loads at a low price!

The Micro Match connectors, which connect the Control Board to the Jack Board proved difficult to source, with very long lead times for some parts. In the end I used a pre-built cable from Farnell to finish the build. The connectors are small and the plugs are difficult to close over the cable, I used a vise.

Original Moog VCF

Component Choices The Minimoog went through a number of revisions during its manufacturing life. The MME PCB’s are designed with the 3046 based VCO’s which makes it an “old” version produced from 1972 to 1978 and I am making my replica as a 1972 version. I have used 5% carbon resistors in the build, except for where 1% metal resistors are needed for precision. Some resistors would have been carbon composite in 1972, but there is no space on the board for them.

I have matched all the transistor pairs in the ladder filter and VCA, which was typical of an original 1972 Minimoog. Later versions reduced the hand matching in the filter to save production costs. The capacitors in the VCF and VCA audio path were polyester in the Minimoog; either Tropical Fish in the early 1970’s or polyester box in the late 1970’s.

  • C62, C65, C66, C67 = 68nF – VCF ladder
  • C63, C64 = 220nf – VCF buffer
  • C51 = 330nF – VCA input

I fitted 68nF Tropical Fish capacitors in the VCF ladder and upgraded some of the electrolytic capacitors. The Minimoog and MME make extensive use of the 741 Op Amp in the VCO’s to gain stability over the previous all transistor design. Today we have options about whether to upgrade some of these to further improve frequency stability, as the wave shaping remains all discrete. The VCO Buffer Op Amps at IC4, IC8, IC15 are obvious candidates.

Powering On By the end of November I had received all the components and populated the boards, it was about 5 days work to solder all the components in. The Micro Molex connectors and control knobs proved hard to find, but success in the end. I powered on the MME with all boards connected and set the operating voltages. All the SMD regulators worked perfectly.

Initial power on faults, with resolutions:

  • VCO3 dead – Dual JFET pins swapped
  • Broken VCO2 frequency pot – bought new one
  • VCO2 frequency too high due to above
  • Noise level low – trimmed louder
  • VCO1 PW goes to 0% and 100% – incorrect pot fitted
  • No sound through VCF and VCA – reflow of solder joint
  • No glide – incorrect pot fitted
  • VCO3 scaling fails – cleaned under Dual JFET

The schematics do not come with the kit, but can be requested. I used the original Moog schematics to help fault find. The components are very tightly packed on the Analog Board and its very easy to make an unwanted solder bridge. I checked the top of the VCF ladder for -10V and with some re-soldering got the VCF and VCA working. I broke the VCO2 pot when fitting the panel, so I ordered a replacement. VCO3 is not oscillating, so I checked all the transistors and came across the problem – the dual FET had its legs crossed!

The noise level was be adjusted with the trimmer, if this isn’t sufficient I will replace the modern 2N3904 with an old 2SC828R which provides a loud and wide noise spectrum (it was used in the TR808). Sorting out the PW is more complicated, once I had all 3 VCO’s working I compared the wave shapes and checked the resistor values – all ok. Turns out I had put 5K pot in Glide and 1MA in VCO1 PW! Easily corrected! The glide pot should be really 5MA rather than 1MA, so I may switch this over to give longer glide times.

Calibration This is carefully explained in the build notes and the VCO’s scaled up accurately with only a few cents error over 5 octaves. I set my analog synths at C1 =32Hz = 0V. There are additional trimmers for the VCA which need setting. The MME uses the “old” Moog oscillator design with 3046 transistor arrays but I noticed no temperature or scaling drift once warmed up.

Outcome A very nice Minimoog clone with added features and a fantastic filter sound thanks to the Tropical Fish capacitors. The use of precision resistors has helped create an accurate set of VCO’s and I don’t think I will upgrade the buffer Op Amps. The build documentation is very good but some explanation of the new features would be useful (see below). It is a very large and challenging build, which requires lots of care in soldering, but delivers a very accurate Minimoog in 60HP!

MME Extra Features The MME has a very useful set of additional features over the Minimoog (and Model D) plus some omissions;

  • VCO2 and VCO3 Sync to VCO1
  • PWM on all VCO’s with external PWM inputs
  • Modulation has a level pot
  • VCO Mod and Wheel CV inputs
  • Individual CV inputs for the VCO’s
  • VCF audio input with level pot
  • VCF and VCF CV inputs
  • Envelope CV outputs
  • No VCF input switches (use the level pots)
  • No power on LED indication
  • No headphone output or 440Hz reference tone

 

Moog 901A/B VCO – Part I

Moog 901A/B

Overview In December 2021 one of my customers asked me if I could build the 901A/B VCO’s as an AMSynths product. In the past I had rejected the idea as they are relatively unstable and I cant get anywhere close to a Behringer 921A/B price point. However I have a small stock of CA3019 diode array chips (and they are still available new), and there is space in my “Big Moog” to fit a set of 901A/B/B/B, so let’s have a go! I will also add the rare 901C module which creates a LFO from a 901B.

The early 901’s used diodes whilst the later versions used enclosed diode array chips. The stability issue is caused by the uncompensated exponential converter, which was never corrected (MOS-LAB have  produced 901 VCO’s with a modern expo). The limited pitch range is caused by using a unijunction transistor as the sawtooth core of the module, 3 octaves of reasonable scaling appears the limit. I will be measuring this once my design is completed.

Research and Design The 901A/B went through many schematic changes in its life to make it more reliable/stable, so I will base my replica on the later 1970 version, to hopefully capture the best ideas from Bob Moog. This later version is the one in Klaus Schulze’s 1969 IIIP “Big Moog”. The modules have been successfully cloned by various 5U/MU manufacturers, and of course Moog have re-issued them as exact replicas. I haven’t seen one squeezed into EuroRack but the clones do seem relatively easy to get working.

My design criteria is to follow the original Moog schematics and use the same components, except where I cant find them or I can chose a better part. I will improve the power supply design as that is clearly what Bob was trying to achieve but without access to modern components in 1965! I have squeezed the designs into 8HP EuroRack modules with 3 PCB layers. No SMD components will be used (except possibly in the power supply), all THD, so I can solder them up myself.

Moog 901A – 1969

Stable Power The power supplies of the original are -6V, -10V, +12V and +11.5V, with the 901A powering the tracking circuit of the 901B from its +12V supply. Precision power rails are very important to the front end of a VCO and I have used modern LDO regulators across the modules.

The -6V is generated by LM2991 LDO regulators from the EuroRack -12V supply and a LM2941 LDO is used to generate the +11.5V in the 901B. The +12V power supply is interconnected across the modules using the BUS connector – see discussion below. The -10V is delivered from a LM337 adjustable linear regulator.

Voltage Divider The 901A uses a resistor ladder to generate the initial frequency control voltages, with a set of 12 voltages from -5V to +6V. I have used a precision voltage source (REF02)  to provide +5.00V, which is then buffered and inverted by a LT1013 Op Amp. VCO transposition is therefore limited to 5 rather than 6 octaves, but I doubt I would ever transpose up 6 octaves!

Moog 901B – 1969

Semiconductors The transistors used in the later versions are 2N3392 and 2N4058, which I will use as they are easy to obtain. Switching to modern transistors with higher HFE needs changes in resistor values which I wish to avoid. The transistors need to be hand matched in some of the circuits, for example the exponential generator.

There are two selected transistors; the 2N2646 UJT at Q10 and the 2N4058 at Q7. These will be hand selected using the Moog calibration process of the correct voltage at Test Point A and the sine waveform symmetry. The 2N2646 is surprisingly easy to obtain new, 50 years after it was used in the Moog 901B’s!

The later 901A modules use two CA3019 diode array chips as temperature compensation (heated like the CA3046 in the Prodigy) and are an improvement over the early versions which had no temperature compensation at all. I have located a source of NOS CA3019 chips and ordered a batch. There are five 1N34A germanium diodes, which are fortunately still available. The unobtainable CL-1 regulator diode is replaced with a contemporary 1N458A, which is what Moog used in the Minimoog re-issue.

Resistors Whilst the original made use of carbon composite and carbon resistors, with precision resistors in the voltage ladder. I have used 1% metal resistors throughout, with 0.1% in the voltage ladder of the 901A and the tracking circuit in the 901B’s (to match the tracking across the three 901B’s).

Trimmers All trimmers have been upgraded from single turn in the Moog to multiturn Bourns. The later versions of the 901A have Scale, High and Low compensation trimmers, with the Mid compensation trimmer removed (as Moog did). I have added a Pulse Width trimmer to the 901A to get accurate pulse widths. The 901B has a Tracking trimmer, as well as Sawtooth offset, Triangle symmetry and Sine offset and symmetry.

Capacitors The timing capacitor values are critical to the octave matching of the 901B and have the following values;

  • 10uF – low mode, not critical
  • 200nF – use x2 100nF in parallel
  • 100nF – use 100nF
  • 50nF – use 2x 100nF in series
  • 25nF – use 22nF and 3n3F in parallel
  • 12nF – use 10nF and 2n2F in parallel

The first three octaves will be accurate and use hand matched Wima MKP2 polypropylene capacitors, although the upper two octaves will be out a bit, This can be corrected manually by the frequency pot and its unlikely I will be switching octaves live. These capacitors have 5mm lead spacing to get them to fit on my 901A panel PCB.

There is a second set of capacitors on the  2P6W rotary switch that go into the Triangle output (to help symmetry?). These were independent of the timing capacitors in early 901A’s but later versions only added a capacitor at low and a high range points, and reused the timing capacitors for the first 4 octaves.

The rest of the capacitors are either ceramic (pF) or polyester (nF).

Mechanical Design The 901A and 901B are each built from three PCB’s, with the circuit spread across all available board areas.

  • POT PCB holds the Alps rotary switch
  • JACKS PCB holds the jacks sockets and pots.
  • MAIN PCB holds the power supplies and 10-pin connector.

Traditional 1P12W and 2P6W rotary switches are used for the frequency ranges, they require more space between the PCB and panel than my usual designs. This created the 3 layer PCB approach which also provides more room for the circuits. The 901A and three 901B’s are inter-connected using a 6-pin IDC cable, this replicates the Moog design but with a more direct dedicated wiring system.

901C Output Stage This is a rare Moog module with only 15 made on perf board in 1967. It provides complementary outputs (one normal, one inverted), for each of the four waveforms and further 2 inversion modes for the pulse waveform. The circuit is transistor based and provides waveform amplification as well, with a dual gang gain potentiometer.

The circuit requires a 3P6W rotary switch which is expensive and simply too big to fit in an 8HP module. I have used CMOS 4051 switches to replicate the functionality, with a modern Alps rotary switch controlling the six signal paths on three layers (input, output+, output-). 2N3392 and 2N4058 transistors are used, along with metal film resistors. There is one ceramic capacitor and single turn trimmer to adjust the initial gain.

 

 

 

 

Minisonik Synth – Part I

Minisonic 2

Overview  Back in 1974 G.D.Shaw designed the Practical Electronics Minisonic 2 analog synthesizer. It was published as a magazine project to build, and complete kits were available from a company called Phonosonics. I was lucky enough to get this kit as a Xmas present in 1974 and I gradually built the whole synthesizer with the keyboard and controls in a Roland SH3 style casing. My first synth!

The Minisonic 2 synthesizer proved disappointing, as the VCO’s drifted and I could not get close to the keyboard playing octaves accurately. I recorded a few songs with this synth but sold it in 1979 to help pay for a visit to the USA, as I never managed to sort out the two VCO’s which drifted and had hopeless keyboard tracking. An experience shared by many UK DIY synth enthusiasts!

Original Design The Minisonic 2 came with a 4 octave keyboard, two VCO’s, a diode filter, two basic AR envelope generators, two VCA’s with manual output panning. There was no LFO but VCO2 could be set as a drone, and there was a ring modulator. Listening to the music I wrote with the synth in 1979, you can hear the low drones and the VCS3 style filter.

AM8050 Prototype

Minisonik Replica I toyed with the idea of a replica for some years and built a copy of the diode filter as the AM8050. I was put off the idea of a full replica, as the VCA’s uses rare 6-pin MFC6040 chips which are hard to find and expensive. There is no LFO, the two envelopes are AR and  only a sawtooth waveform from each VCO.

On the positive side there is lots of modulation routing via over 20x latching push button switches, the VCO’s can be set to drone and detached from the keyboard, there is bi-directional variable cross mod, variable oscillator sync, and VCO1 can FM modulate the filter as a LFO. There is also a ring modulator, are two outputs with level and panning.

Then in 2021 I found by chance, an 8-pin DIL version of the VCA chip; it was cheap and available as NOS. I knew I could sort out the tracking and stability of the VCO’s with an amended exponential converter design, so the project was on!

Minisonic 2 Routing

Features My Minisonik synth is based on the original design, with a some subtle changes and improvements:

  • New exponential converters in the VCO’s and VCF
  • Dual matched transistor pairs in the expo converters
  • Temperature compensation in the VCO’s.
  • Revised noise source using a transistor rather than a diode
  • Removed the +/-6V power supply
  • Retained the +/-9V power using LDO regulators
  • LED indicator for power ON/OFF
  • Tune control but no Span or Portamento
  • Eurorack Pitch and Gate inputs
  • External input to Ring Mod and added one to the VCF.
  • Color code push button caps
  • 60 HP black anodised panel

Minisonik Panel

Development The synth was laid out across two PCB’s, Panel and Main, in Autumn 2021. The front panel design was done first, so I could determine where the pots and switches would be. I have kept to the original left to right layout with the VCF, ES/VCA’s lined up vertically. The pan controls and envelope levels use 9mm trimmers, so I could fit the synth into 60 HP, with large knobs for the main controls.

The noise circuit is a traditional transistor design (ARP, Oberheim) using the 2SC828 which is particularly good at generating a wide noise spectrum (as used in the Roland TR808). The SG3402 based ring modulator circuit has been retained, as it has a slightly different internal schematic than the easier to find MC1496. The headphone feature has been omitted, which means we don’t need the rare MFC400B chips. The Eurorack standard +5V gate signal is converter to -9V for the envelope shapers using a comparator chip and a transistor inverter.

Outcome  The two PCB’s were ordered in December 2021. Read Part II for how they turn out!

Alesis Midiverb II Repair

Overview I bought a couple of MidiVerb II’s in May 2020, both were listed for repair and at a super low cost of £30. Although there is no service manual or schematics online, some information can be gleaned from the MidiVerb III service manual, and I was hopeful that they would be simple repairs. Typical issues are voltage regulators going out of spec, or the crystal oscillators fail. The rest of the analog and digital circuits seem reliable and robust.

Midiverb II – S/N 91606 The first one I bought has seen a lot of use, the rack ears have been cut off and the lettering was rubbing off. Powering it up initially proved it was working and the effects were really quite good and the noise floor low (its uses TL084 chips, the LF354’s in early units are noisy). I went ahead and recapped the power supply and replaced the voltage regulators. The PCB is a later version and has separate oscillators for the 8031 microprocessor and the DSP chip. The OS is MVII.OBJ dated 4/10/87.

Midiverb II – S/N 39303 The second one I bought was sold as broken and the seller stated it was non functional with all LED’s on at power up. Once in the studio I switched it on and it was working ok, so I went ahead and recapped the power supply and replaced the voltage regulators (as they were below specification). I noticed there were some PCB cuts and the timing capacitor for the 16 MHz oscillator was not connected. Powering back on resulted in failure, the most common error code was all LED’s on, sometimes OVLD off. This error code means the microprocessor has not started.

Further investigation shows a 12 MHz crystal and timing capacitor added onto the rear of the PCB for the 8031 microprocessor and the PCB cuts were to disable the divide down clock from the 16 MHz oscillator. This looks like a factory modification to increase the clock speed for the microprocessor from 8 to 12 MHz, as newer models had two clocks. I checked that both clocks were working and found the RESET signal for the 8031 was high. This is why it was bricked, it should be zero.

The reset circuit for the MV2 is very simple; a resistor and capacitor charge up after power on. This is not 100% reliable and the Midiverb III changed it during production to a more reliable transistor based circuit. When I replaced the 4.7uF capacitor I had shorted it out on top of the PCB, so hidden from view, this shorted the RESET pin to +5V. I re-soldered a new cap and added in a missing 10k resistor. Power back on and the MV2 now works perfectly.

Power On Patch One of the differences I noticed between the two MidVerbs is the patch that they initialized to on power up is different. With no battery backed RAM it wasn’t until I looked at the Xicor X2444P chip near the 8031 that I realised there was a very small amount of on board NVRAM. Very small is just 32 bytes, but it does last 100 years! This is where the MIDI program change message (01-32) mapping to the patch number (00-99) is stored and retained during power off. It also stores the MIDI control channel number (0-16) and at power on the first program (01) in the table is loaded.

This explains why the different MidiVerbs power on with different patches. It is because the mapping tables are different, they can of course be configured to get any patch (00-99) at power on. Mystery solved!

Mixer Pots The front panel potentiometers get a lot of use over the years especially the MIX. This is a dual 10kB PCB mounted potentiometer, looks like an easy to find Alpha pot.

Outcomes  I sold the Midiverb II’s in 2021 as I have too many FX!  I bought some Behringer 100M replica modules instead.

 

Copyright AMSynths 2022