AM1005 Mod Amp


ARP 1005 Prototype

Overview This is a replica of the unusual ARP 1005 module from the ARP 2500 analog modular synthesizer. The module consists of a discrete transistor based Ring Modulator, whose output is then connected to a Voltage Controlled Amplifier. The Ring Modulator from 1971 onward was the 4014 encapsulated sub module, which was used in the ARP 2600. It is based on two Wilson current mirrors and a LM301 Op Amp.

In the 1005 there is an extra feature for generating in-harmonic preset voltages, which was not seen in the subsequent 2600. Two accurate control voltages can be adjusted using the Ratio and Tune pots. They are designed to be fed into the source VCO’s to control the frequencies and ring modulation effect from the 1005 module.

The modulation effect and the precision voltages can be turned ON and OFF by the front panel illuminated push buttons or by a gating the inputs. The VCA has both linear and exponential mode and uses a standard ARP transistor OTA design, similar to the 4019 sub module. The modulation controls drive a simple transistor flip-flop circuit to switch from ON to OFF states. The precision voltages are generated from the main power rails with two LM301 Op Amps.

Hybrid 107C

In the early days ARP used a third party 107C ring modulator module from Hybrid Systems Corp to speed up the development of the 1005 module. It was later replaced by the ARP 4014 module in 1971 as part of the 2600 project. The 4014 is functionally equivalent but unlikely to be the same circuit. The pin out positions changed and external X and Y trimmers were added, which resulted in a PCB revision to the 1005.

Overall this is a fantastic modulator module with its ability to switch the ring modulation ON and OFF by control voltages, and the the additional VCA for controlling the output level.

AM 1005 Module

AM1005 Module The AMSynths design uses the same schematic and similar PCB layout as the original, and is built as a 5U high MOTM compatible design. The PCB is 150 x 150 mm with 40 mil (fat) traces.

The original component choices have been retained in many locations; LM301 Op Amp in the CV buffer, carbon composition resistors, ceramic disc capacitors, Switchcraft illuminated push buttons, high quality PRV6 rotary pots and ARP 2500 style knobs. Improved Op Amps have been used; the LF412 for the precision voltages and LF356 for the audio path (rather than MC1439).

Linear Systems matched transistor pairs are used in the Wilson current mirrors and a 10V precision power rail in the CV control (removing some resistor selection at setup). There are 1% metal film resistors in the expo generator and precision voltage circuit, as in the original.

Controls  The module has the following front panel controls:

  • Unmod Gain
  • Amplifier Gain
  • VCA CV1 Level
  • VCA CV2 Level
  • Ratio and Tune for the Preset Voltages
  • Unmod and Mod Push Buttons
  • Slide switch for VCA response – exponential or linear

Connections There are 10x 3.5mm jack sockets mounted at the base of the front panel, these are:

  • Mod/Unmod Triggers
  • Mod Gate
  • Audio Signal Inputs (x2)
  • VCA CV Inputs (x2)
  • Preset Voltage Outputs (x2)
  • Signal Output

AM1005 PCB

Build Outcomes I bought the ARP detailed service notes back in 2004 and made a couple of attempts to design a PCB over the next few years, and even tried to fit the design into EuroRack. The MOTM format for my ARP 2500 Rack modules was finally decided upon in June 2015, and in July the large 150 x 150 mm PCB was laid out, and a prototype ordered on 18 July.

I populated the PCB in August and mounted it to a nice 5U aluminium panel. However the buttons were not controlling the ON/OFF nature of the precision output voltages or the lamps. The transistor flip flop worked but somehow the signal was failing to drive through the transistor inverters correctly. Many long hours passed!

By March 2016 I gave up trying to make the original ARP circuit work, and designed a new circuit to drive the switching. The transistor flip flop was retained (so that external signals could be used), and it now drove a CMOS Schmitt trigger inverters (40106), followed by high quality CMOS switches controlling the CV and audio paths (DG418 and DG419). The lamp drive circuit was retained and the new “digital core” worked like a dream with 0 to 15V on/off voltages and with the benefits of CMOS analog switches over the old FET’s.

I moved all the components from the old to the new PCB (which took me 3 hours!), as I wanted to re-use the expensive carbon composite resistors and transistor arrays. Then I fired up the new PCB in mid April 2016 and connected it up the panel. At Last…

AM1005 PCB Close Up

2019 Update A couple of mistakes in the schematics were identified in the REV02 board in 2019, which were corrected with kludges. The DG419 analog switch that controls Input B needs to be driven from the MOD OFF signal and the transistors in the 4014 current mirrors were missing an inter connection between collectors. I had also wired the main GAIN pot to +15V rather than the correct +10V precision rail. With these corrections I was able to trim the balanced modulator and amplifier to perfection and the module was finally commissioned on 26 October 2019.

REV01 Update After talking with MOS-LAB about his successful 1005 clone in November 2019, I identified that R79 in the ARP schematics is 4K7 and not the 47K I had used. There is also an error in the ARP service notes, the emitter of Q7 should go to GND and not -15V. I confirmed the error by looking at a real ARP 1005 PCB where its clear that the connection is to GND. I corrected this along with another mistake I had made around Q1 and Q2, and bingo! It works! It took me four years to find the error…

With a spare REV1 PCB in stock I went ahead and built a second AM1005 using the same component choice as the REV02, except going back to the BJT and JFET transistors in the switching circuit. A new front panel was purchase in December and the buttons caps are RED and BLUE.

 

 

Copyright AMSynths 2019