Skip to the content
4 layers64 × 60 mmDronemagnav, the printed circuit board in 3D

Magnetic-anomaly navigation (MagNav) board + firmware for drones: ESP32-S3, BMI088, LIS3MDL, BMP280, u-blox GNSS, WDMAM map matching, particle filter, MAVLink, phone route planner. Proprietary - all rights reserved.

Size
64 × 60 mm
Copper layers
4
Thickness
1.6 mm
Parts
60
Nets
67
Pads
280
Tracks, vias
787, 145
Pours
7
Routed
Yes
KiCad
KiCad 10.0 (file 20260206)
Created
12 Sep 2026
Updated
13 Sep 2026

Overview

Dronemagnav is an open-source printed circuit board by ajisrawi, from the ajisrawi/dronemagnav repository on GitHub. It is a 4-layer board of 64 × 60 mm with 60 parts on 67 nets, routed with 787 tracks and 145 vias.

It is built around the ESP32-S3-WROOM-1-N16R8, BMP280 and LIS3MDL, with the AMS1117-3.3, AP2112K-3.3 and BMI088 among its other parts. Its parts include 18 capacitors, 17 resistors, 8 ICs, 6 diodes and 5 connectors. It belongs with the radio and wireless boards and the ESP32 boards of this library.

Main components

9

README

README.md

DroneMagNav — magnetic-anomaly navigation board for drones

GNSS-denied positioning from the Earth's crustal magnetic field, plus a phone-operated great-circle route planner.

text
  LIS3MDL magnetometer ─┐                          ┌─ WDMAM anomaly map (SD)
  BMI088 IMU ───────────┼─ ESP32-S3 ─ particle ────┤  WMM core field model
  BMP280 baro ──────────┘   filter    matching     └─ Tolles-Lawson compensation
        │                        │
   u-blox GNSS (init/truth)      ├─ MAVLink GPS_INPUT → flight controller
                                 └─ WiFi AP → phone: offline map, route, upload

Deliverables

ItemWhereStatus
Schematics (5 sheets)hardware/dronemagnav/*.kicad_sch, dronemagnav-schematic.pdfERC 0 errors (1 intended warning)
PCB 64 × 60 mm, 4-layer, 30.5 mm Pixhawk hole patternhardware/dronemagnav/dronemagnav.kicad_pcbrouted, 0 unconnected, 0 electrical DRC
Fab packagehardware/dronemagnav/fab/ (Gerber ZIP, drill, pick-and-place)ready to upload
BOMdocs/BOM.csv28 line items, 56 parts, MPNs
Firmware (ESP32-S3, PlatformIO)firmware/, images in firmware/release/compiles with 0 warnings (Arduino-ESP32 3.3.11); WMM core-field model validated 100/100 against NOAA test vectors (0.001 nT)
Phone UIfirmware/sdcard/www/index.htmloffline map, tap-to-route, upload
SD-card preparationtools/prepare_sd.pyWDMAM v2 (1.5 GB text) → 51.9 MB binary grid, verified by tools/wdmam_check.py; WMM, coastline
Design documentDroneMagNav-Design-Document.pdfstudy, diagrams, PCB, BOM, verification

Hardware

BlockPartWhy
MCU / radioESP32-S3-WROOM-1-N16R8WiFi AP for the phone page, BLE, native USB, 8 MB PSRAM for map tiles + particle filter
MagnetometerLIS3MDL (SPI)16-bit, ~40 nT after averaging; RM3100 upgrade via mast connector J5
IMUBMI088 (SPI)drone-grade accel + gyro, vibration-robust
BarometerBMP280 (SPI)altitude for the field model
GNSSu-blox MAX-M10S + U.FLinitialisation and truth reference
StoragemicroSD (SDMMC 4-bit)world anomaly map, WMM, coastline, web page
InterfacesUSB-C, JST-GH TELEM (MAVLink, 5 V in), JST-GH sensor mast (I2C)Pixhawk-standard
PowerUSB/FC 5 V → AMS1117-3.3 (digital) + AP2112K-3.3 (sensors)separate clean sensor rail

How it navigates

  1. Compensate the scalar field for the drone's own magnetism (18-term Tolles-Lawson, fitted once in flight with GNSS).
  2. Predict position with the flight controller's velocity (or IMU dead reckoning) in a 600-particle filter.
  3. Update each particle against WMM core field + WDMAM anomaly at the particle's location; resample; output mean and 1-σ.
  4. Feed the estimate to the autopilot as GPS_INPUT when GNSS is lost.

Expectations, honestly: the WDMAM world grid is 3 arc-min (~5 km) and upward-continued to 5 km, so with it alone you get drift-free kilometre-class positioning — right for oceanic/jammed cruise, not landing. Load a local aeromagnetic survey (/maps/local.bin, same format) for tens of metres at drone altitude.

Routing

The phone page shows an offline world coastline, you tap START and DEST, the board computes the great-circle (shortest) path as evenly spaced waypoints and uploads them to the flight controller as a MAVLink mission.

Build / flash / prepare

See firmware/README.md. Short version: pio run -t upload over USB-C, then python tools/prepare_sd.py --wdmam <file> --wmm WMM.COF --coast <geojson> and copy firmware/sdcard/* to the card.

Status / caveats

  • Prototype; not certified for any regulated use.
  • Firmware builds cleanly (PlatformIO, pioarduino platform, Arduino-ESP32 core 3.3.11: 0 warnings, 1.08 MB flash of 4 MB app partition, 94 KB RAM). Ready-to-flash images are in firmware/release/. It has not yet run on real hardware, since the board has not been built.
  • Vias in pads: 74 power vias were moved out of their pads into dog-bone stubs. Three remain where the routing leaves no room (BMI088 pads 2 and 4, microSD VDD pad) and must be filled and capped, see hardware/dronemagnav/fab/FABRICATION-NOTES.md (included in the Gerber ZIP).
  • MAX-M10S straps checked against the u-blox integration manual (UBX-20053088 R05): VIO_SEL left open selects 3.3 V I/O, which is what the board supplies, so no change was needed. TIMEPULSE is tied inside the module to SAFEBOOT_N, so the firmware keeps GPIO16 input-only.
  • Map data: WDMAM from the link you supplied; WMM.COF (NOAA) and Natural Earth coastline are free downloads you run through prepare_sd.py.

Parts

60
QtyValueFootprintReferences
9100nFC_0603_1608MetricC3, C6, C9, C10, C11, C12, C13, C14, C17
910KR_0603_1608MetricR2, R3, R9, R10, R11, R12, R13, R14, R15
710uFC_0805_2012MetricC1, C2, C5, C8, C15, C16, C18
4M3MountingHole_3.2mm_M3H1, H2, H3, H4
3330RR_0603_1608MetricR4, R5, R6
21uFC_0603_1608MetricC4, C7
24.7KR_0603_1608MetricR7, R8
25.1KR_0603_1608MetricR16, R17
2SS14D_SMAD1, D2
11KR_0603_1608MetricR1
1AMS1117-3.3SOT-223-3_TabPin2U6
1AP2112K-3.3SOT-23-5U7
1BMI088Bosch_LGA-16_4.5x3mm_P0.5mm_LayoutBorder7x1y_ClockwisePinNumberingU2
1BMP280Bosch_LGA-8_2x2.5mm_P0.65mm_ClockwisePinNumberingU4
1BOOTSW_SPST_PTS645Sx43SMTR92SW2
1ESP32-S3-WROOM-1-N16R8ESP32-S3-WROOM-1U1
1FC_TELEMJST_GH_SM06B-GHS-TB_1x06-1MP_P1.25mm_HorizontalJ2
1FIXLED_0603_1608MetricD5
1GNSS_ANTU.FL_Hirose_U.FL-R-SMT-1_VerticalJ4
1LINKLED_0603_1608MetricD6
1LIS3MDLLGA-12_2x2mm_P0.5mmU3
1MAX-M10Sublox_MAXU5
1NAVLED_0603_1608MetricD4
1PWRLED_0603_1608MetricD3
1RESETSW_SPST_PTS645Sx43SMTR92SW1
1SENSOR_MASTJST_GH_SM06B-GHS-TB_1x06-1MP_P1.25mm_HorizontalJ5
1USB-CUSB_C_Receptacle_HRO_TYPE-C-31-M-12J1
1USBLC6-2SC6SOT-23-6U8
1microSDmicroSD_HC_Hirose_DM3AT-SF-PEJM5J3

Similar boards

  • 4 layers128 × 84 mmDronemagnav2, the printed circuit board in 3D

    Dronemagnav2

    ajisrawi/DroneMagNav2.0

    DroneMagNav 2.0 - survey-fluxgate magnetic-anomaly navigation board for drones (IGRF-14 + EMAG2v3 continued to flight altitude)

    222 parts
  • 4 layers81 × 53 mmPratham, the printed circuit board in 3D

    Pratham

    PeriwalPratham/FlightController

    A basic PCB design which contains all components for a solo yet affordable basic drone controller

    102 partsMIT
  • 2 layers21 × 54 mmSensorBoardNoMCU, the printed circuit board in 3D

    SensorBoardNoMCU

    Terrapin-Rocket-Team/SRAD-Avionics

    Code and PCBs for the avionics subteam for the SRAD flight computer and side projects

    29 parts17
  • 2 layers94.5 × 64 mmAir monitor, the printed circuit board in 3D

    Air monitor

    kaushalch1/air-monitor

    this is a portable air quality monitoring device designed to measure particulate matter and environmental conditions in real time.

    42 parts
  • 4 layers55 × 55 mmSTM32-FC, the printed circuit board in 3D

    STM32-FC

    thecasterian/STM32-FC

    Flight controller for a quadcopter with STM32F405

    59 parts3
  • 4 layers60 × 85 mmEsp32s3-camera, the printed circuit board in 3D

    Esp32s3-camera

    LCQIANN/esp32s3-camera

    ESP32-S3 camera board v1.2: OV2640 DVP camera, microSD, USB-C, LiPo, KiCad 10 project for JLCPCB PCBA

    64 parts
  • 2 layers141.8 × 50 mmCwl, the printed circuit board in 3D

    Cwl

    philipparndt/opentherm-wolf-cwl

    Decoder tool for analyzing OpenTherm protocol communication between a BM remote control and a Wolf CWL ventilation/heat-recovery unit. Built to reverse-engineer the protocol for building a custom ESP32-based controller.

    63 parts
  • 4 layers73 × 98 mmALBERT FLIGHTCOMPUTER, the printed circuit board in 3D

    ALBERT FLIGHTCOMPUTER

    felixznr/ALBERT

    Albert I was a rhesus macaque monkey and the first mammal launched on a rocket (V-2 Rocket "Blossom No. 3") on June 18, 1948.

    161 partsMIT

Questions

How big is the Dronemagnav PCB?

It measures 64 × 60 mm and has 4 copper layers, 1.6 mm thick.

What parts does the Dronemagnav use?

60 parts, built around the ESP32-S3-WROOM-1-N16R8, BMP280, LIS3MDL and AMS1117-3.3. 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 Dronemagnav KiCad files?

From the ajisrawi/dronemagnav repository on GitHub, at the commit this page was made from: the board is hardware/dronemagnav/dronemagnav.kicad_pcb and its schematic hardware/dronemagnav/dronemagnav.kicad_sch. PCBKit opens the board in the browser, with nothing to install.

Can I use the Dronemagnav design in my own project?

Its repository states no licence, so its author keeps all rights: ask ajisrawi before reusing the design.