I2C slave not detected on Crazyflie 2.1 expansion header (MP6-XOEM / MAX14521E)

[romaindgc]

Hello,

This is my first time attempting I2C communication using the Crazyflie 2.1 expansion header (I2C1 bus).

I am trying to interface a piezoelectric micropump driver (Bartels MP6-XOEM / MAX14521E).
According to its datasheet, the default 8-bit write address is 0x78. Since the Crazyflie firmware uses standard 7-bit addressing, I target 0x3C (0x78 >> 1).

However, direct writes to 0x3C return NACK. To troubleshoot, I adapted the hello deck driver example to scan the I2C bus across the full range (skipping 0x50 to 0x57 to protect the onboard deck memory).

Here is my deck driver implementation:

#define DEBUG_MODULE "PUMP_DECK"

#include <stdbool.h>
#include <stdint.h>
#include "FreeRTOS.h"
#include "task.h"
#include "debug.h"
#include "deck.h"
#include "i2cdev.h"
#include "system.h"

// MP6-XOEM / MAX14521E registers
#define REG_POWERMODE         0x01
#define REG_FREQUENCY         0x02
#define REG_SHAPE             0x03
#define REG_P1VOLTAGE         0x06
#define REG_P2VOLTAGE         0x07
#define REG_P3VOLTAGE         0x08
#define REG_P4VOLTAGE         0x09
#define REG_UPDATEVOLTAGE     0x0A

#define POWERMODE_NORMAL      0x01
#define FREQ_300HZ            0xA0
#define SHAPE_SRS             0x01
#define VOLTAGE_MAX_250VPP    0x1F
#define UPDATE_VOLTAGE_CMD    0x01

#define MAX_ATTEMPTS          5

static bool writeReg(uint8_t addr, uint8_t reg, uint8_t val)
{
    bool ack = i2cdevWriteByte(I2C1_DEV, addr, reg, val);
    if (!ack) {
        DEBUG_PRINT("Write error on reg 0x%02X at addr 0x%02X\n", reg, addr);
    }
    return ack;
}

static bool configurePump(uint8_t addr)
{
    if (!writeReg(addr, REG_POWERMODE, POWERMODE_NORMAL)) return false;
    vTaskDelay(M2T(10));
    if (!writeReg(addr, REG_FREQUENCY, FREQ_300HZ)) return false;
    vTaskDelay(M2T(10));
    if (!writeReg(addr, REG_SHAPE, SHAPE_SRS)) return false;
    vTaskDelay(M2T(10));

    if (!writeReg(addr, REG_P1VOLTAGE, VOLTAGE_MAX_250VPP)) return false;
    if (!writeReg(addr, REG_P2VOLTAGE, VOLTAGE_MAX_250VPP)) return false;
    if (!writeReg(addr, REG_P3VOLTAGE, VOLTAGE_MAX_250VPP)) return false;
    if (!writeReg(addr, REG_P4VOLTAGE, VOLTAGE_MAX_250VPP)) return false;
    vTaskDelay(M2T(10));

    if (!writeReg(addr, REG_UPDATEVOLTAGE, UPDATE_VOLTAGE_CMD)) return false;
    return true;
}

static void helloTask(void *param)
{
    systemWaitStart();
    vTaskDelay(M2T(2000));

    DEBUG_PRINT("\n========================================\n");
    DEBUG_PRINT("      STARTING AUTOMATIC I2C SCAN       \n");
    DEBUG_PRINT("========================================\n");

    uint8_t targetAddr = 0;

    for (uint8_t addr = 0x08; addr <= 0x77; addr++) {
        if (addr >= 0x50 && addr <= 0x57) {
            continue;
        }

        DEBUG_PRINT("[SCAN] Test 0x%02X... ", addr);
        bool ack = i2cdevWrite(I2C1_DEV, addr, 0, NULL);

        if (ack) {
            DEBUG_PRINT("-> [ACK DETECTED !]\n");
            if (targetAddr == 0) {
                targetAddr = addr;
            }
        } else {
            DEBUG_PRINT("-> [NACK]\n");
        }

        vTaskDelay(M2T(30));
    }

    DEBUG_PRINT("========================================\n");

    if (targetAddr == 0) {
        DEBUG_PRINT("ERROR: No device detected on I2C bus.\n");
    } else {
        DEBUG_PRINT("Device found at address: 0x%02X\n", targetAddr);
        DEBUG_PRINT("Applying configuration...\n");

        bool success = false;
        for (int attempt = 1; attempt <= MAX_ATTEMPTS; attempt++) {
            DEBUG_PRINT("Attempt %d/%d...\n", attempt, MAX_ATTEMPTS);
            if (configurePump(targetAddr)) {
                DEBUG_PRINT(">>> Pump successfully activated! <<<\n");
                success = true;
                break;
            }
            vTaskDelay(M2T(500));
        }

        if (!success) {
            DEBUG_PRINT("Configuration failed after %d attempts.\n", MAX_ATTEMPTS);
        }
    }

    while (1) {
        vTaskDelay(M2T(5000));
    }
}

static void helloInit()
{
    xTaskCreate(helloTask, "helloTask", configMINIMAL_STACK_SIZE + 1500, NULL, 2, NULL);
}

static bool helloTest()
{
    return true;
}

static const DeckDriver helloDriver = {
    .name = "hello",
    .usedPeriph = DECK_USING_I2C,
    .init = helloInit,
    .test = helloTest,
};

DECK_DRIVER(helloDriver);

Hardware setup:

• Power : I use the LiPo 1S to power the driver via the VCOM PIN
• Connections: SCL -> PB6, SDA -> PB7, with a shared common GND.
• Driver configuration: Pin 15 (/RESET) is tied to GND (Slave I2C mode). Pins 4, 5, and 6 are left floating as specified by the manufacturer.
• Pull-up resistors:
  • When I added external 4.7k pull-up resistors to the bus, the Crazyflie frequently failed during boot on the internal EEPROM test (EEPROM: I2C connection [FAIL]).
  • When relying solely on the Crazyflie 2.1 internal pull-up resistors (without external pull-ups), the self-test passes consistently, but the scan still fails to detect the slave:

SYS: ----------------------------
SYS: Crazyflie 2.1 is up and running!
SYS: Build 0:252b41341a07 (2025.12.1) MODIFIED
SYS: I am 0x323038313533510F004F0031 and I have 1024KB of flash!
CFGBLK: v1, verification [OK]
DECK_DISCOVERY: Backend found: onewire
DECK_DISCOVERY: Backend found: forced
DECK_DISCOVERY: Backend found: deckctrl
DECK_FORCED: CONFIG_DECK_FORCE=hello found
DECK_CORE: 1 deck(s) found
DECK_CORE: Calling INIT on driver hello for deck 0
IMU: BMI088: Using I2C interface.
IMU: BMI088 Gyro connection [OK].
IMU: BMI088 Accel connection [OK]
IMU: BMP388 I2C connection [OK]
ESTIMATOR: Using Complementary (1) estimator
CONTROLLER: Using PID (1) controller
MTR-DRV: Using brushed motor driver
SYS: About to run tests in system.c.
SYS: NRF51 version: 2024.2 (CF21)
EEPROM: I2C connection [OK].
STORAGE: Storage check [OK].
IMU: BMI088 gyro self-test [OK]
DECK_CORE: Deck 0 test [OK].
SYS: Self test passed!
STAB: Wait for sensor calibration...
SYS: Free heap: 11224 bytes
STAB: Starting stabilizer loop
PUMP_DECK:
========================================
PUMP_DECK:       STARTING AUTOMATIC I2C SCAN       
PUMP_DECK: ========================================
PUMP_DECK: [SCAN] Test 0x08... PUMP_DECK: -> [NACK]
PUMP_DECK: [SCAN] Test 0x09... PUMP_DECK: -> [NACK]
...
[All addresses 0x0A through 0x77 return NACK]
...
PUMP_DECK: [SCAN] Test 0x76... PUMP_DECK: -> [NACK]
PUMP_DECK: [SCAN] Test 0x77... PUMP_DECK: -> [NACK]
PUMP_DECK: ========================================
PUMP_DECK: ERROR: No device detected on I2C bus.

The oscilloscope trace shows valid 3.0V clock and data transitions, but the 9th clock cycle is never pulled down by the slave (NACK).

Questions:

1. Is a 0-byte write (i2cdevWrite(I2C1_DEV, addr, 0, NULL)) the correct way to ping slaves on I2C1_DEV, or does the driver expect at least one register byte?
2. Does powering a slave from VCOM while the bus logic is at 3.0V require a specific level shifter on the Crazyflie 2.1 expansion header?

Thank you in advance for your help!

[ataffanel]

Hi

For your questions:

1. Yes it should, since you see the signal on the oscilloscope it proves that it is working.
2. You must add a voltage regulator if you use VCOM: the Crazyflie I/O are 3.0V, so you must make sure your I/Os are also 3.0V.

From what I see of your experiment, I would make sure that the address is correct and that the signal makes it to the sensors. Passing the question through Codex seems to show that you are not scanning enough address to reach the device address, I did not double check that though.

[Esmail-sarhadi]

0x3C is the correct 7-bit form of 0x78, so the address maths is right.

A NACK there usually means the device is not answering yet rather than not existing: the MAX14521E needs its supply up and a settle time before it responds, and some breakout boards tie an enable pin. Two more things specific to the Crazyflie expansion header:

• the I2C pull-ups are not always populated on the deck side — check before assuming the bus is idle-high;
• scan the bus first (i2cdetect-style) to see whether anything answers, which separates "wrong address" from "no bus".