An open source Jetson Nano baseboard and tools to design your own.

Related tags

Deep Learninghardware
Overview

My Jetson Nano Baseboard

Render of My Jetson Nano Baseboard

Picture of My Jetson Nano Baseboard

This basic baseboard gives the user the foundation and the flexibility to design their own baseboard for the Jetson Nano. It also repurposes some of the Jetson Nano’s interface signals for simple beginner projects.

This baseboard, as designed, contains:

  • A 5V, 4A DC barrel jack
  • 4 USB 2.0 connectors
  • An HDMI connector
  • A UART-to-USB bridge
  • A debug USB
  • A 40-pin GPIO
  • A servo header
  • Three user-interactive buttons (power, reset, and force recovery)
  • A flex connector for an OLED display

Use these files to kickstart your own application-specific baseboard or implement some quick and easy projects!

Quick Start

You only need a computer to get started right now! For a more in-depth setup guide, check out docs/setup.md.

  1. Install the appropriate version of KiCAD, an open source schematic and layout design program, for your operating system here.

  2. Download this GitHub repository either as a ZIP or on the command line.

  3. Save the following symbol and footprint libraries to the “Design Files/Libraries” folder (you may have to make an account – alternatively, if you want the practice, you could try making them yourself):

    1. B3SL-1002P
    2. TPD4E02B04DQAR
    3. 3-1734592-0
    4. DC-005-2.0A
    5. ACM2012-201-2P-T001
    6. 10029449-111RLF
    7. 2309413-1
  4. Open the project (.pro) file in KiCAD.

  5. Add the symbol and footprint libraries as project-specific to your KiCAD program, following this guide. You will know that the libraries are correctly loaded when there are no more boxes with question marks.

  6. You are now set up to tinker with the files and launch your own hardware designs!

Be sure to download the Jetson Nano Product Design Guide here (NVIDIA Developer account required) to help you with your design.

If you want to make your boards and test them, you’ll need the Jetson Nano module (not included, can be bought as part of the developer kit here). It is also helpful to have electronics equipment like an oscilloscope, a multimeter, and a soldering iron.

Questions and Improvements

If you have a suggestion, please open an issue on GitHub.

Please share your projects with us on the Jetson Developer Forums.

Comments
  • Servo PWM signal does not switch logic levels, remains high

    Servo PWM signal does not switch logic levels, remains high

    Problem When running the sampleproj/servo_pwm.py script, SERVO_PWM remains high, even as the GPIO07 pin outputs a PWM signal.

    Release A00, no modifications

    Observed Behavior Servo PWM remains high even when PWM script is run.

    GPIO07 at 40-pin header: gpio7 configured0

    SERVO_PWM at servo header: gpio 7

    Root Cause Theory The pullup could be too strong; the transistor connection may not be consistent across boards. This is another issue that switching to a larger package transistor should fix.

    Suggested Next Steps

    • Switch to larger package transistor (as in issue #5 & 2)
    • Test additional boards to ensure transistor connection remains consistent
    bug A01 Fix 
    opened by wolframalexa 1
  • Fails sleep/wake software cycle

    Fails sleep/wake software cycle

    Problem Board cannot be wake from software; the physical button must be pushed.

    Release A00, no modifications

    Observed Behavior When running the validation/sleep_func.sh script, the device does not wake until the power button is pushed.

    Root Cause Theory An issue with how L4T interacts with the baseboard.

    Suggested Next Steps

    • Probe power logic signals to ensure sequencing is correct
    • Read L4T documentation for power design
    bug 
    opened by wolframalexa 1
  • OLED display does not fit in specified connector

    OLED display does not fit in specified connector

    Problem The specified OLED display does not fit in the specified connector. Users cannot use the display.

    Release A00, no modifications

    Observed Behavior The connector is too small for the display.

    Root Cause Theory Mistakes were made.

    Suggested Next Steps Choose new connector for display.

    bug A01 Fix 
    opened by wolframalexa 1
  • Fan PWM signal not in compliance

    Fan PWM signal not in compliance

    Problem Users cannot use a fan, because the fan PWM signal is not in electrical compliance.

    Observed Behavior GPIO14 provides a nice PWM signal, but the signal becomes less crisp as it goes through the level shifter. In both images below, the blue line is the FAN_PWM_LS node. Yellow: left = GPIO14, right = FAN_PWM_INV.

    image image

    The transistor Q3 should invert the PWM signal, but does not appear to do so.

    Root Cause Theory At 20kHz, it is unlikely the fan signal is switching too fast for the transistor. It may be due to a transistor misalignment; switch from the DMN26D0UFB4-7 to the DMN26D0UT-7, which is in a larger SOT-523 package, to avoid misalignments.

    Suggested Next Steps Ask a more experienced engineer. May need a larger pullup, or more power control.

    bug A01 Fix 
    opened by wolframalexa 1
  • Some USB 2.0 type A ports are not functional

    Some USB 2.0 type A ports are not functional

    Problem Some USB ports on some boards do not respond when a USB mouse or keyboard is plugged into them.

    Observed Behavior When a USB mouse or keyboard is plugged into one of the four type A ports, it occasionally does not work (the pointer does not move, no text appears on screen). This hinders the ability for the user to interact with the display, and to use the USB devices they need.

    Root Cause Theory This is only present on some ports on some boards; it could be a manufacturing or hub chip error.

    Suggested Next Steps

    • Visually check all USB components to ensure there is no damage
    • Ensure hub chip is strapped correctly
    • Ensure USB layout guidelines are followed for signal integrity
    bug A01 Fix 
    opened by wolframalexa 1
  • Low-resolution HDMI (1280x720)

    Low-resolution HDMI (1280x720)

    Problem The maximum HDMI resolution seems to be 1280x720, whereas the Jetson Nano achieves a resolution of 2560x1440. As a result, the display appears zoomed in.

    Release A00, no modifications

    Observed Behavior

    • This is present on all boards, with the same module as on the official baseboard - probably a hardware design issue, and not a software issue
    • The resolution on the official Jetson Nano is twice that of "My Jetson Nano Baseboard"

    Root Cause Theory Resistor values may need to be tuned for better resolutions. Additionally, an EEPROM may be read to confirm HDMI resolution.

    Suggested Next Steps Investigate L4T behavior with regard to HDMI and resistor tuning on CEC line.

    bug A01 Fix 
    opened by wolframalexa 1
  • Power LED does not light up on some boards

    Power LED does not light up on some boards

    Problem On some boards, the power LED D6 does not light, even though the board completes its power-on sequence and has booted normally.

    Release A00, no modifications

    Observed Behavior

    • The LED is the correct direction
    • The board completes its power on sequence and the software functions as expected
    • GPIO04 remains LOW at 0.6V, even though it should be driven HIGH upon power-up
    • There is no measured voltage drop across R32
    • There is an insufficient voltage drop across D6

    Root Cause Theory The gate threshold voltage varies depending on the individual transistor. It could be that this transistor has a higher V_GS and does not turn on when GPIO04 is at 0.6V. Additionally, the transistor Q7 could be misaligned.

    Suggested Next Steps

    • Investigate behavior of GPIO04, which should be HIGH upon power-on. Remove R31 to ensure no loading effects from transistor.
    • Change all transistors DMN26D0UFB4-7 to the DMN26D0UT-7, which is in a larger SOT-523 package, to avoid misalignments.
    bug A01 Fix 
    opened by wolframalexa 1
  • Make silkscreen more readable

    Make silkscreen more readable

    • Increase silkscreen size from 0.5x0.5mm to 0.438x0.7mm to make the text more readable.
    • Add polarity for all ICs and polarized components to aid in soldering
    • Add silkscreen on front for 40-pin header
    enhancement A01 Fix 
    opened by wolframalexa 0
Releases(A01)
  • A01(Aug 12, 2021)

    After having manufactured the boards and validated them, we're fixing some functionality. Here are the changes, which you can read about in our issues:

    • Fix footprints (#11):
      • DC Jack flipped
      • GPIO header flipped (pin 1 should be pin 2)
      • USB footprint with soldermask (#4)
      • New OLED display connector & display (#6)
    • Usability enhancements, with more readable silkscreen (#10)
    • BOM errors:
      • SODIMM connector (#9)
      • HDMI current limiting resistor, which allows for correct resolution (#3)
      • Larger transistor footprint to avoid misalignments (#8, #5)
    • Fix pullup
      • Add pullup to GPIO04 (#2)
    Source code(tar.gz)
    Source code(zip)
  • A00(Jul 28, 2021)

Owner
NVIDIA AI IOT
NVIDIA AI IOT
Multi-atlas segmentation (MAS) is a promising framework for medical image segmentation

Multi-atlas segmentation (MAS) is a promising framework for medical image segmentation. Generally, MAS methods register multiple atlases, i.e., medical images with corresponding labels, to a target i

NanYoMy 13 Oct 09, 2022
Official TensorFlow code for the forthcoming paper

~ Efficient-CapsNet ~ Are you tired of over inflated and overused convolutional neural networks? You're right! It's time for CAPSULES :)

Vittorio Mazzia 203 Jan 08, 2023
Official Keras Implementation for UNet++ in IEEE Transactions on Medical Imaging and DLMIA 2018

UNet++: A Nested U-Net Architecture for Medical Image Segmentation UNet++ is a new general purpose image segmentation architecture for more accurate i

Zongwei Zhou 1.8k Dec 27, 2022
A forwarding MPI implementation that can use any other MPI implementation via an MPI ABI

MPItrampoline MPI wrapper library: MPI trampoline library: MPI integration tests: MPI is the de-facto standard for inter-node communication on HPC sys

Erik Schnetter 31 Dec 22, 2022
In this project we investigate the performance of the SetCon model on realistic video footage. Therefore, we implemented the model in PyTorch and tested the model on two example videos.

Contrastive Learning of Object Representations Supervisor: Prof. Dr. Gemma Roig Institutions: Goethe University CVAI - Computational Vision & Artifici

Dirk Neuhäuser 6 Dec 08, 2022
Python scripts to detect faces in Python with the BlazeFace Tensorflow Lite models

Python scripts to detect faces using Python with the BlazeFace Tensorflow Lite models. Tested on Windows 10, Tensorflow 2.4.0 (Python 3.8).

Ibai Gorordo 46 Nov 17, 2022
Pytorch Implementation of LNSNet for Superpixel Segmentation

LNSNet Overview Official implementation of Learning the Superpixel in a Non-iterative and Lifelong Manner (CVPR'21) Learning Strategy The proposed LNS

42 Oct 11, 2022
tsai is an open-source deep learning package built on top of Pytorch & fastai focused on state-of-the-art techniques for time series classification, regression and forecasting.

Time series Timeseries Deep Learning Pytorch fastai - State-of-the-art Deep Learning with Time Series and Sequences in Pytorch / fastai

timeseriesAI 2.8k Jan 08, 2023
Differentiable molecular simulation of proteins with a coarse-grained potential

Differentiable molecular simulation of proteins with a coarse-grained potential This repository contains the learned potential, simulation scripts and

UCL Bioinformatics Group 44 Dec 10, 2022
.NET bindings for the Pytorch engine

TorchSharp TorchSharp is a .NET library that provides access to the library that powers PyTorch. It is a work in progress, but already provides a .NET

Matteo Interlandi 17 Aug 30, 2021
A learning-based data collection tool for human segmentation

FullBodyFilter A Learning-Based Data Collection Tool For Human Segmentation Contents Documentation Source Code and Scripts Overview of Project Usage O

Robert Jiang 4 Jun 24, 2022
Replication package for the manuscript "Using Personality Detection Tools for Software Engineering Research: How Far Can We Go?" submitted to TOSEM

tosem2021-personality-rep-package Replication package for the manuscript "Using Personality Detection Tools for Software Engineering Research: How Far

Collaborative Development Group 1 Dec 13, 2021
Easy Parallel Library (EPL) is a general and efficient deep learning framework for distributed model training.

English | 简体中文 Easy Parallel Library Overview Easy Parallel Library (EPL) is a general and efficient library for distributed model training. Usability

Alibaba 185 Dec 21, 2022
This is the official code for the paper "Tracker Meets Night: A Transformer Enhancer for UAV Tracking".

SCT This is the official code for the paper "Tracker Meets Night: A Transformer Enhancer for UAV Tracking" The spatial-channel Transformer (SCT) enhan

Intelligent Vision for Robotics in Complex Environment 27 Nov 23, 2022
Virtual Dance Reality Stage: a feature that offers you to share a stage with another user virtually

Portrait Segmentation using Tensorflow This script removes the background from an input image. You can read more about segmentation here Setup The scr

291 Dec 24, 2022
Solution to the Weather4cast 2021 challenge

This code was used for the entry by the team "antfugue" for the Weather4cast 2021 Challenge. Below, you can find the instructions for generating predi

Jussi Leinonen 13 Jan 03, 2023
Weakly Supervised Learning of Instance Segmentation with Inter-pixel Relations, CVPR 2019 (Oral)

Weakly Supervised Learning of Instance Segmentation with Inter-pixel Relations The code of: Weakly Supervised Learning of Instance Segmentation with I

Jiwoon Ahn 472 Dec 29, 2022
Parameter Efficient Deep Probabilistic Forecasting

PEDPF Parameter Efficient Deep Probabilistic Forecasting (PEDPF) is a repository containing code to run experiments for several deep learning based pr

Olivier Sprangers 10 Jun 13, 2022
An implementation of Video Frame Interpolation via Adaptive Separable Convolution using PyTorch

This work has now been superseded by: https://github.com/sniklaus/revisiting-sepconv sepconv-slomo This is a reference implementation of Video Frame I

Simon Niklaus 984 Dec 16, 2022
Train CNNs for the fruits360 data set in NTOU CS「Machine Vision」class.

CNNs fruits360 Train CNNs for the fruits360 data set in NTOU CS「Machine Vision」class. CNN on a pretrained model Build a CNN on a pretrained model, Res

Ricky Chuang 1 Mar 07, 2022