Table of Contents

Introduction

• Context

  • Tracking Antenna Long Term Vision

• What it will not have

  • support ardupilot as a backup

  • support Bluetooth

  • design around nucleo

  • similar to pixracer

  • 6S or 3S battery cell monitor

• What it will have

  • Fix servo backfeed issue

    • https://discord.com/channels/776618956638388305/1236736566223962183/1247278642271227929

  • seeduino xiao

    • main proccessor EFS is already devving around

    • needs to have access to pins for EFS debugging w/ logic analyzer

  • gps connector for external gps

    • contains compass as well

    • Neo M9N or M9P (see arch doc for link)

    • same as we use on the drone

    • this device has the CMOS battery inside of it

  • wifi module on the board or external, EE designer choice

  • barometer on the board

    • Doing it offboard since efs doesn’t really need it right now.

  • a few neopixels on the board

  • low battery indicator (need to have input voltage sense) → done already?

  • power supplies support 6S 5S 4S and 3S battery input

  • maximize mechanical packaging benefits

    • support 30.5x30.5mm mounting pattern

  • just do connectors to sensors instead of on the same pcba

  • sd card on board for logging → done already? Not in this rev for the sake of pin space

  • Display screen on board or external (TFT variant)

  • support 1:1 harnesses for all external modules

  • JST connectors for all off board stuff

  • Add diode for Neopixel supply voltage, drop 0.7V to meet data sheet supply voltage to logic voltage ratio

  • If possible usb port for debugging mavlink? - Not mandatory

• Why

  • improve functionality for ease of development and connectivity

  • integrated in one thing for flight tests, easy to verify it looks good

  • Justified in Tracking Antenna Long Term Vision & 2024-06-01 EFS Director Sync

• Who

  • Ryan Scomazzon

• When

  • For Tracking Ant M3

Engineering

• Ryan Scomazzon

• Siddharth Muralikrishnan

Architecture

Very straight forward

To-do

Here is an action list to help with collaboration with Siddharth. Set status to “In Progress” when you start something so I don't take it. Put “Finished” when done. I can start with the pcb after the schematic is done. I will base it off of the nucleo shield file to make life easy.

Changes to existing board

I am basing the schematic off of the current nucleo shield board. I made a new schematic for the new onboard devices and removed the interface sheet for the nucleo interconnectors. The Seeeduino has way less pins than the nucleo so things have changed. I have removed the additional servo pins that do not have overcurrent protection since we only need three pins. I have also removed the SD card reader/writer since debug data is sent over WiFi. I am also removing the protocol connectors that we don’t need and putting everything that uses the same bussed protocols on the same bus.

Connectors

“For now the sensors we are using on the tracking antenna: GPS(UART), IMU(I2C/SPI), WIFI(SPI).” “the imu we were using I believe only has i2c.“ https://discord.com/channels/776618956638388305/1113893873891160074/1259663283191943198

GPS

The GPS uses UART. The Pixhawk standard connectors for UART are either the telemetry connector or the standard GPS connector which has UART and I2C in one. Right now our GPS (SAM-M8Q) only uses the UART but we plan to switch use a GPS+Compass module (M9N), which uses this GPS connector. Therefore, I will use that connector standard. When using another GPS, we might need a harness but this would be needed anyways since our other GPS uses header pins for UART.

I2C

• One I2C connector is needed for the IMU (when separate from the GPS)

• Adding one more for the barometer (for if it is chosen to be added in the future.

• I am adding another spare one because it adds zero extra pin usage to put another device on the same I2C bus.

SPI

• Not adding SPI anymore since it is only needed for the WIFI module and it takes too many pins. Since the WIFI will be part of the MCU breakout, this would just be a spare anyways.

WIFI Module

We will no longer be adding the WIFI module since it takes too many pins from the seeeduino. Instead, we are choosing a different processor board which has WIFI built in.

Deciding to keep this on board (removable via headers) to make integration easy and reliable. The board we use (Overview | Adafruit AirLift - ESP32 WiFi Co-Processor Breakout | Adafruit Learning System) needs SPI and two more pins, so it won’t fit a standard Pixhawk connector anyways.

Barometer

EFS doesn’t seem to care too much about it. I think the play is to not add it on board and just use the extra I2C connector if we need one and connect it to an external baro breakout board. This also allows us to defer the job of breaking out the IC to someone else so this board is easier to solder and more reliable.

ICP-20100 TDK InvenSense | Sensors, Transducers | DigiKey

Plopping this here for now since it is what the pixhawk uses but it is probably too small and I have to check with EFS to make sure they are cool with it using I2C (since we are out of pins on the seeeduino and I2C is the only one where we can just throw it in the existing bus).

USB debug port

Not doing this since it is just a nice to have and we do not have more pins for another serial port.

Seeeduino

Turns out the Seeduino XIAO SAMD21 (our current one) does not even have enough pins for the bare minimum of what we need. I will be selecting a different board instead. We will use the seeeduino XIAO ESP32C3 because it has WIFI built in so we can free up the SPI pins.

113991054 Seeed Technology Co., Ltd | Development Boards, Kits, Programmers | DigiKey

The datasheet isn’t great so I will be using these ones instead that I found elsewhere

Seeed-Studio-XIAO-Series-SOM-Datasheet.pdf (seeedstudio.com)

Seeed-Studio-XIAO-Series-Package-and-PCB-Design.pdf (seeedstudio.com) for the pad layout.

XIAO ESP32C3.sch (seeedstudio.com) for the schematic

The seeeduino with headers has 7 adjacent 2.54mm headers on either side. For the header pins, this should do PH1-07-UA Adam Tech | Connectors, Interconnects | DigiKey. For the female header sockets, these should do PPPC071LFBN-RC Sullins Connector Solutions | Connectors, Interconnects | DigiKey. I plan to mount them in the correct place for the seeeduino and then have another set just mounted outside of that for EFS to use for debugging. Female headers should be best for that anyways since they are more protected. The footprint for the seeeduino will just be two of these headers in the right spot.

The battery power pins are on the bottom of the board and must be soldered, so when connected with the header pins, we can not access that. To fix this, we can just connect our 5V buck to the USB 5V power rail. The seeeduino has a fair bit of protection on the inputs, but it will still short the USB to the battery if both are connected at once. I will place a diode from the 5V buck to the input/USB power so the USB power does not get fed into the battery. There is internally a regulator to drop the voltage to 3.3V so I will use a Shottkey diode so the input voltage to this regulator is high enough.

ESP32 datasheet for power reference esp32-c3_datasheet_en.pdf (espressif.com) - this suggests 1A of output current could theoretically be pulled. The schematic shows the regulator only can withstand 750mA anyways so I will design towards that spec when choosing the diode. I will choose something at least 1A for safety factor. Since diodes are cheap, I will give lots of overhead.

Nathan’s seeeduino tracking antenna board is here for reference hardware/Projects/Arduino_AAT at master · UWARG/hardware (github.com).