b'FeatureDIY geophysicistcontrol. The Arduino system has also adopted this IC and theirparallel and share data (basic support in MicroPython for explicit extensive daughter board sensor and peripheral systems will bemulti-threading). Best of all you can buy these devices and get available (as used by Dean etal., 2018(b)). The microcontrollerunderway immediately. Least costly by far is the Raspberry Pi systems can be inexpensive, currently costing less than $10-50Pico (Figure 3, and Pico W), which outside of the MicroPython for each item. interpreter also needs a special programming environment for Windows- and Mac-based programmers to download code/The various embedded microcontroller systems, often basedscripts to run (e.g. Thonny, https://thonny.org/). The Pico is very upon Advanced RISC Machine (ARM) architectures (similar tominimal in mass storage support. Thus, I feel the Arduino and microprocessors used in mobile phones) are far more capablePyboard systems are the quicker path to satisfaction if you do in directly controlling signal inputs and outputs. Withoutnot have many electronic skills and tools.sophisticated human interface devices and high overhead peripherals these little systems can respond to external inputSensing and measurementin in less than a microsecond even with MicroPython code via hardware interrupts, or the use of timers running with lessTraditional sensors such as geophones may be directly wired than 10 ns of resolution. Many modern ARM microcontrollersto pins on the embedded processors, but while simple it often have internal bus switching to allow MicroPython to directlyproduces very poor performance. The typical 12-bit resolution command Serial Peripheral Interface (SPI) and I2C streams intofound in most microcontrollers may seem primitive and too on-chip memory while executing instructions and data frompoor in performance for most geophysical applications. This isnt other areas, known as direct memory access; a technologytrue, despite modern production geophysical systems typically developed for mainframe computers in the 1970s. These littleusing 24-bit conversion. Adding an instrumentation amplifier, or microprocessors using less 100 mW of power can be faster thantwo, will result in better utilisation of dynamic range and signal the original supercomputers, even in 32-bit floating point. buffering (Dupuis etal., 2020). Recording a single sensor with two measurements at different gains, or even different types of My current recommendations for embedded microcontrollerssensors, will provide very good performance in many cases after are: some post-processing of signals (e.g. gain merging of channels Pyboard v1.1versatile and extensive peripheral set and theas we once did with TEM systems). For example, a conventional foundation of MicroPython support geophone sensor recorded with multiple gains (e.g. gain =2 Arduino Portenta H7relatively costly but extremely capableand gain =100, for later gain merging) allows high dynamic and well supported range (90-100dB) even with limited 12-bit ADC resources. Raspberry Pi Pico/Pico W inexpensive, well supported, Wi-FiMy recommendation for an amplifier for most purposes is an version, and widely available instrumentation amplifier. For example, Analog Devices AD623 Arduino Nano RP2040less costly with Wi-Fi and wellfor use with 3 to 5V systems and directly compatible with supported Nano accessory shields 3.3V powered microcontroller ADCs. Another very fine and easy to use instrumentation amplifier is the Analog Devices The PyBoard systems with a small daughter add-on board for(Linear Technology) LT1920, or if you want to measure down external programable amplifiers are capable of performingthe fundamental thermal noise of geophones then use the as simple engineering 12 channel seismographs storing 4SSM2019 or INA103/163/849. These are all relatively easy to second records at 1 ms sample rate, then streaming the datause and implement, but all analog circuits are trickier to work to to a FAT file on a microSD card. The data on the mSD card canspecification than digital devices.be later harvested or read directly by a PC connecting to theAs a basic datalogger the Pyboard v1.1 appears to be well suited board via USB (as the mSD card presents as a flash drive). Afor many basic data logger geophysical measurements. The STM related microcontroller, STM32H7 series processor, has withinNucleao-144 STM32H743/753 evaluation board is inexpensive the Integrated Circuit 16-bit ADC conversion and pin access toand has greater access to various pins and function than the eight channels, or three differential channels on the evaluationPortenta H7 but requires more work and expertise to interface board when not using external buffer amplifiers (Figure 2).to measurement and control. Like the Pyboard it has both ADC The PWM waveforms and external signals can be controlledand DAC channels, with native 16-bit ADC x 9 and 12-bit DAC to a precision of less than 20 ns. There are many tips and tricksx 2. However, beware the specs: the ADCs will barely achieve on various forums, such as GitHub (https://github.com/) andbetter than 12-bit Signal-to-noise without modification of the the IC manufacturers website www.stm.com. Thus, simplereference circuit (removing a resistor on the board and feeding seismographs, EM or electrical resistivity acquisition anda stable low impedance reference voltage). Many on-chip ADCs transmitter control may be made from these boards. The analogcannot perform better than 12 bits because of noise from the signals (with or without amplifications and buffering) may bedigital circuits nearby. An issue challenging even experienced directly soldered/wired to the terminals with equipment costingengineers. The STM Nucleo is also unfortunately currently out of hundreds of dollars, or as little as $50 for a decent low voltagestock everywhere and it is not looking likely to be in stock until soldering iron. mid-2023.The Arduino Potenta H7 lite with the breakout board (Figure 1,For those wanting more resolution and instantaneous dynamic order codes ABX00042 and ASX00031) provides an affordablerange out of the box then connecting the embedded ready-to-go acquisition system and data logger with microSDmicrocontroller board to an external dedicated analog-to-card plus relatively easy to solder and wire to fast 16-bit ADC,digital convertor via SPI or I2C serial interface is a good option. 12-bit DAC, multiple peripheral interfaces (I2C, SPI, and UART).An example of a capable 16 bit converter not requiring much Its very capable and a newly maturing system with respect toelectrical know-how (or soldering) to interface are small software support and hardware add-on boards (shields). Thedaughter boards based upon the ADS1115 (eg manufactured microcontroller in the Potenta H7 (and the RP2040) has twoby Adafruit) and distributed by Core Electronics (in Australia), separate processing units that can execute separate programs inDigikey (US) and Mouser (US). Core electronics also have a DECEMBER 2022 PREVIEW 52'