AM1027 Clocked Sequential Control

ARP 1027 Module

Overview This is a replica of the 1027 Analog Sequencer from the ARP 2500 Analog Modular Synthesizer, which has 10x step rows and 3x control voltage vertical columns. The ARP module is double width and has an analog clock which drives the 3 rows of control voltages on the right. The left hand side of the panel has the clock controls; stop, start, rest, step and frequency. There is also a useful pulse width control.

The 1027 is a basic sequencer in today’s terms, but probably the first sequencer to use a logic chip and the only one with 10 steps! Don’t be worried about having to always use 10 steps in your sequence, their are position gate outputs for each step, which can be patched into the reset input for shorter sequence lengths.

BIP2611-1 Chip

My AM1027 is installed with 3x AM1045 Synthesizer Voice Modules into an Oak 6U Cabinet made in France, with the panels taking the top 5U. The bottom 1U is the 15V power supply with gate, trigger, cv sockets along with a Roland 100M 8-pin keyboard connector. I use a Roland 181 keyboard to control the AM2500 racks, although most of the time the 1027 is driving the 1045’s in this particular rack.

AM1027 Module My design uses the original ARP schematics and the core is laid out over two PCB’s; a main PCB which is 150 x 150 mm and mounted onto a metal sub assembly, and a smaller 150 x 100 mm PCB for the analog clock, mounted on top. Metal film resistors and modern capacitors have been used, with no attempt to use legacy carbon composition resistors. The voltage controlled clock uses an AM 4001 Expo Generator which contains a matched Linear Systems transistor pair and 100 ohm SMD Tempco.

High quality (and expensive) 100k Spectrol potentiometers with 1/8″ shafts have been used for the control voltages, they are PCB mounted which means the 3x potentiometer PCB’s are mounted straight to the panel. The original Switchcraft square push button switches with incandescent bulbs have been retained, along with the correct round “red” push buttons and black slide switches. The clock and pulse width potentiometers are TT precision pots with a very precise feel.

Keystone 8558 Knob

The correct green coloured row indicator lens have been used with green 5mm diffused LED’s. I experimented with 12V incandescent bulbs but they went dim at higher clock rates. I had to use non ARP like silver control knobs, as I don’t have enough coloured knobs at the moment. I have used control knobs similar to the original that I bought in quantity in the 2000’s as part of my parts sourcing. The large amount of hand wiring in the original 1027 has been partially replaced with ribbon cables, but the left hand panel is still hand wired.

BIP2611-1 The core of the sequencer is a very early decade counter the Burroughs BIP2611-1 chip. This was launched in the late 1960’s for providing the decimal counting signals for Nixie tubes, although the 2611 version could not drive them directly. It can work at 300 kHz, which is rather too high a rate for a sequencer! Understandably these are nearly impossible to find and are rare collector items. I have used a modern CMOS 4017 decade counter, which has the same functionality but inverts the logic state and therefore requires many changes.

Controls The module has the following front panel controls:

  • Clock Rate potentiometer
  • Clock Pulse Width potentiometer
  • Clock ON and OFF Illuminated Push Buttons
  • Clock Rate LED
  • Clock STEP and RESET Push Buttons
  • Clock Frequency HIGH/LOW switch
  • Clock Source INT/EXT switch
  • Clock Control GATE/TRIGGER switch
  • 10x potentiometers for Control Voltage A steps
  • 10x potentiometers for Control Voltage B steps
  • 10x potentiometers for Control Voltage C steps
  • 10x LED lamps for each step

Connections There are 20x 3.5mm jack sockets mounted on the panel:

  • GATE1 to GATE10, position step outputs on front panel
  • STEP and RESET inputs
  • ON and OFF trigger inputs
  • CLOCK OUT output
  • OUTA, OUTB, OUTC – the control voltages
  • OUT1 to OUT10, position step outputs on rear mounted DIL10 socket

Module Build & Outcome I have a set of original 1027 blueprints from 2004 but it wasn’t spring 2017 that I laid the circuits out onto 5 PCB’s; a MAIN PCB for the clock and output drivers, a CLOCK PCB for the analog voltage controlled clock and its controls, and three POT PCB’s to hold the control voltage potentiometers. I ordered the PCB’s and populated them later in 2017 but the project was paused in 2018 whilst I moved into a new workshop.

I restarted the build of the AM1027 in early March 2019, and used a test 4001 expo generator with non-matched 2N390X transistors to start up the prototype. After correcting a couple of schematic errors (which took a few hours to locate) I had the clock oscillating nicely with a pulse waveform, along with push button start/stop control and indicators, as well as the clock LED.

I spent another 20 minutes working out what the three curiously unmarked trimmers were for, and how to adjust the clock to its factory settings.

  • R9 Low Frequency – trim to set lower clock speed
  • R12 High Frequency – trim to set upper clock speed of 400Hz
  • R19 Pulse Width – trim to set Pulse Width at 10%
  • R22 Offset – trim to 0 mV

The clock has two ranges;

  • LOW – 20 pulse per minute to 30Hz
  • HIGH – 30Hz to 400Hz

With the clock working just like the ARP factory, I could now put the CD4017 counter chip into its socket and see what happens. There are a number of changes needed to the ARP circuit to switch from the BIP2611 chip to the modern 4017. The voltage reduction circuit around the BIP2611 chip can be removed, the pull up resistors go to ground rather than +15V, and they are larger values than the 470R resistors in the original.

The CD4017 can drive circuits directly without buffers, unlike the BIP2611 which can not and therefore has inverted outputs. NPN transistors are needed in the output circuits, rather than PNP, and I used 2N3904’s and took out a set of pull up resistors. The pull up resistor on the STEP button needs to be reduced to 330R from 1k to get the right sized voltage trigger into the CD4017. The external step and reset inputs come straight into the CD4017, so I may buffer these with Schmidt triggers to tidy up the waveforms.

The position gate outputs and the CV mixed outputs needed adjusting to +10V with another set of resistor changes. Once the circuit was working well I designed a front panel and did a test fitting using plastic mounting card, and then when everything was correct I ordered an aluminium front panel. This was fitted to the sequencer in April and the module mounted in the oak cabinet. Job done!



Copyright AMSynths 2017