Skip to the content
2 layers60.5 × 43.8 mmNeuromorphic Soma Node, the printed circuit board in 3D

Neuromorphic Soma Node

aliihsan1913/Analog-HH-Neuron

Instead of utilizing microprocessors or algorithmic software to approximate the differential equations, this circuit solves the time and voltage-dependent variables ($V, m, h, n$) in real-time using the inherent…

Size
60.5 × 43.8 mm
Copper layers
2
Thickness
1.6 mm
Parts
24
Nets
57
Pads
164
Tracks, vias
283, 47
Pours
1
Routed
Yes
KiCad
KiCad 9.0 (file 20241229)
Created
9 Jul 2026
Updated
9 Jul 2026

Overview

Neuromorphic Soma Node is an open-source printed circuit board by aliihsan1913, from the aliihsan1913/Analog-HH-Neuron repository on GitHub, published under the MIT licence. It is a 2-layer board of 60.5 × 43.8 mm with 24 parts on 57 nets, routed with 283 tracks and 47 vias.

It is built around the MPY634KU and TL072. Its parts include 9 ICs, 7 resistors, 4 capacitors and 2 connectors.

Main components

2

README

README.md

Analog Hardware Simulation of the Hodgkin-Huxley Neuron Model

3D PCB Render

1. Project Overview and Purpose

This repository contains the design, simulation, and manufacturing files for an analog computational hardware that physically models the Hodgkin-Huxley (HH) biological neuron. The primary objective is to transition neuro-biophysical concepts from theoretical mathematics into a tangible engineering prototype.

Instead of utilizing microprocessors or algorithmic software to approximate the differential equations, this circuit solves the time and voltage-dependent variables ($V, m, h, n$) in real-time using the inherent physical properties of electrons, capacitors, and non-linear analog multipliers.

2. Biological Reference and Parameter Extraction

The electronic component values in this design are strictly derived from the morphological data of a real biological cell, utilizing foundational physics principles rather than hypothetical values.

  • Reference Neuron: 1-10-8 (Rat, Neocortex)
  • Soma Surface Area: 1147.99 µm² ($A \approx 1.15 \times 10^{-5}\ cm^{2}$)
  • Data Source: NeuroMorpho Database

By multiplying the universal specific membrane capacitance ($1\ \mu F / cm^{2}$) and specific channel conductances by the actual surface area of the 1-10-8 neuron, the following hardware parameters were calculated and implemented in the circuit:

  • Soma Membrane Capacitance: $\approx 11.5\ pF$
  • Sodium (Na) Channel Resistance: $\approx 724\ k\Omega$
  • Potassium (K) Channel Resistance: $\approx 2.41\ M\Omega$

3. Hardware Architecture

The system is designed on a 2-layer Printed Circuit Board (PCB) using KiCad, featuring an unbroken ground plane on the bottom layer to ensure signal integrity for the analog pathways. Trace widths are constrained to 0.25 mm with 0.20 mm clearances to accommodate the dense IC placement.

KiCad Schematic

The architecture is divided into three core functional blocks:

  • Soma: Represented by the central capacitor. The accumulation and discharge of charge on this component physically represent the resting potential and action potentials (spikes).
  • Voltage-Controlled Channels: Built using TL072 operational amplifiers (acting as integrators for gating variables) and MPY634 precision analog multipliers. The multipliers physically execute the non-linear conductance equations (e.g., $m^3h$ for Na and $n^4$ for K).
  • Stimulus (STIM) and Threshold (THRESH): External input ports for injecting nanoampere-level synaptic currents and a leakage resistor network to maintain the resting potential at approximately -65 mV.

2D PCB Layout

4. Simulation and Ground Truth

Prior to hardware manufacturing, the extracted parameters were simulated in two distinct environments to verify the mathematical model against the electronic implementation.

Mathematical Ground Truth (Python/SciPy)

The standard HH differential equations were solved using numerical integration to establish the baseline biological response. Under a 0.2 nA stimulus, the mathematical model exhibits precise spiking behavior.

Python Ground Truth

Analog Circuit Simulation (LTspice)

The physical hardware schematic was simulated in LTspice. The resulting transient analysis waveforms confirm that the analog components successfully replicate the action potentials, matching the mathematical ground truth and proving the viability of the physical circuit.

LTspice Simulation

5. Repository Structure

  • /hardware: Contains KiCad PCB design files and LTspice simulation models.
  • /simulation: Contains the Python script (hh_model.py) for the numerical ground truth.
  • /images: Renderings and plots used in this documentation.
  • 1-10-8_Neuromorpho.md: Detailed mathematical proofs and parameter conversion calculations.

Parts

24
QtyValueFootprintReferences
5MPY634KUPackage_SO:SOIC-16W_7.5x10.3mm_P1.27mmU2, U3, U4, U5, U9
4100 nFCapacitor_SMD:C_0805_2012MetricC4, C5, C6, C7
4TL072Package_SO:SOIC-8-1EP_3.9x4.9mm_P1.27mm_EP2.62x3.51mmU1, U6, U7, U8
110 kΩPotentiometer_SMD:Potentiometer_Bourns_TC33X_VerticalRV1
110 kΩResistor_SMD:R_0805_2012MetricR2
1100 kΩResistor_SMD:R_0805_2012MetricR3
122 kΩResistor_SMD:R_0805_2012MetricR8
133 kΩResistor_SMD:R_0805_2012MetricR4
14 MΩResistor_SMD:R_0805_2012MetricR6
14.7 kΩResistor_SMD:R_0805_2012MetricR7
147 kΩResistor_SMD:R_0805_2012MetricR5
1Conn_01x02Connector_PinHeader_2.54mm:PinHeader_1x02_P2.54mm_HorizontalJ2
1Conn_01x03Connector_PinHeader_2.54mm:PinHeader_1x03_P2.54mm_HorizontalJ1
1RResistor_SMD:R_0805_2012Metric3.9M1

Similar boards

  • 2 layers148 × 84.5 mmPanel daisy, the printed circuit board in 3D

    Panel daisy

    ohmtech-rdi/eurorack-blocks

    Software to Hardware Prototyping for Eurorack using C++, Max/Gen~ or Faust

    259 parts410
  • 2 layers39.6 × 99.1 mmESP32-Eurorack Audio, the printed circuit board in 3D

    ESP32-Eurorack Audio

    ctag-fh-kiel/ctag-tbd

    CTAG TBD >>to be determined<< an extendible open source Eurorack sound module

    128 parts260
  • 4 layers121.9 × 109.2 mmPlinkyblack hwmidi jacks, the printed circuit board in 3D

    Plinkyblack hwmidi jacks

    plinkysynth/plinky_public

    a public release of the plinky sw and hw

    300 parts227
  • 2 layers58 × 110 mmVco core, the printed circuit board in 3D

    Vco core

    shmoergh/moduleur

    Fully analog DIY-friendly modular synth

    146 parts222
  • 2 layers37 × 113.9 mmVCO, the printed circuit board in 3D

    VCO

    nickajeglin/Eurorack-pcbs

    A collection of easy to solder DIY eurorack PCB's

    104 parts208
  • 2 layers111.3 × 106.4 mmSuper Sixteen v1, the printed circuit board in 3D

    Super Sixteen v1

    matthewcieplak/super-sixteen

    Code and schematics for the Super Sixteen Eurorack sequencer

    141 parts171
  • 2 layers85.1 × 62.2 mmMic extension board, the printed circuit board in 3D

    Mic extension board

    StoneRose35/cortexguitarfx

    guitar effects processor implemented on various arm cortex platforms

    75 parts146

Questions

How big is the Neuromorphic Soma Node PCB?

It measures 60.5 × 43.8 mm and has 2 copper layers, 1.6 mm thick.

What parts does the Neuromorphic Soma Node use?

24 parts, built around the MPY634KU and TL072. The full list of its parts is on this page, with links to each part's symbol and footprint where PCBKit's library has them.

Where can I download the Neuromorphic Soma Node KiCad files?

From the aliihsan1913/Analog-HH-Neuron repository on GitHub, at the commit this page was made from: the board is Hardware/Kicad/Neuromorphic_Soma_Node.kicad_pcb and its schematic Hardware/Kicad/Neuromorphic_Soma_Node.kicad_sch. PCBKit opens the board in the browser, with nothing to install.

Can I use the Neuromorphic Soma Node design in my own project?

Its repository publishes it under the MIT licence: you may reuse it on that licence's terms (keep the author's notice, and for a copyleft licence share your changes alike).