Technology must transfer data to the central system in real time, otherwise it may have negative consequences. If the sensor battery power runs out, a machine failure may stop production for one day or lead to direct danger. If battery life is unbelievable and short, IoT applications will become useless, causing more interference rather than making life easier for its intended purpose. Therefore battery powered IoT devices come as a standard in up-to-date IoT installations

Wireless sensors and sensor networks are one of the elements of the Internet of Things systems and intelligent factories. Replacing the standard sensors and data collection devices with versions that communicate wirelessly gives many benefits, but also enforces a highly thought-out system design that will minimize energy consumption. This is important because these systems must work for many years without servicing. In the article we present the issues regarding the design of systems and forecasting of energy consumption in IoT systems.

Wireless communication vs Battery power

The idea of wireless sensor networks has been around for at least two decades, while the IEEE subgroup working on personal wireless networks defined the 802.15.4 standard in 2003, a year later the first versions of ZigBee appeared. Since then, many varieties of wireless communication have been developed, such as LoRa & NarrowBand-IoT and additional functions introduced, as a result of which designers now have a choice of various open or proprietary protocols. What significantly affects the way the entire project is implemented is energy consumption.

Battery powered IoT installation. Source: https://modberry.techbase.eu/

The basic elements of these systems are sensors that measure physical quantities. Some signal and data processing capabilities are also important. After all, the communication interface is important, which will allow you to pass the measured data on. Such a sensor node should wake up from time to time, make contact with its superordinate controller, transfer data and fall back to sleep again. Battery life depends on the total charge collected. Minimizing this consumption in the long run means that you need to minimize energy consumption during each work cycle. In many cases, the sensor will only work for a small fraction of the time. A measurement that lasts a few milliseconds can be triggered once per second, once per minute, or even less frequently. Therefore, the energy consumed in sleep mode may dominate the total energy consumption.

Battery powered sensors market growth

The lifetime of IoT sensors varies greatly: some last a year years, others 10, the first being the most realistic. When organizations need to deploy engineers to install new batteries in sensors and employ staff to monitor them, the benefits of technology itself are beginning to run out.

Battery powered IoT devices crucial to 2020+ standards

It is estimated that in 2020 nearly 31 billion devices will be connected to the Internet of Things. Such forecasts provide ample opportunities, especially for producers associated with the products that make up it, and they are intensified by the developing IoT technology.

Source: https://globenergia.pl/co-laczy-internet-rzeczy-i-perowskity-fotowoltaika-do-zastosowan-wewnetrznych/

Battery-ready IoT devices based on ESP32

Battery / SuperCap power support allows the processes and data to be securely executed, saved or transferred, and the operating system to be safely shutdown or reboot, if the power source has been restored. The power failure alert can also be sent to cloud service, to perform custom task, specified by user or self-learning AI algorithm.

The Moduino device is a comprehensive end-point controller for variety of sensors located throughout any installation. It fully supports temperature and humidity sensors and new ones are currently developed, e.g. accelerometer, gyroscope, magnetometer, etc.

Battery powered Moduino ESP32
Battery powered IoT installation. Source: https://modberry.techbase.eu/

ModuinoModBerry symbiosis allows wide range of wake-up/sleep schedule customization, in order to perform best and save energy accordingly to power supply state. Arduino and MicroPython environments provide libraries to control different scenarios of data and power management.

With built-in algorithms and the possibility to program on your own, the TECHBASE’s sleep/wake addon module can wake the device using schedule/timer. Another option is wake on external trigger, e.g. change of input, etc. All the options for sleep, shutdown and wake can be configured for various scenarios to ensure constant operation of devices, safety of data and continuity of work in case of power failure in any installation.

Sequent Microsystems has introduced the Raspberry Pi „4-relay” add-on to the Industrial IoT market. Four relays switch lines for loads up to 10 A and 250 V, can be stacked up to 8 times with max. 32 switched relays.

It can be used to control loads of up to 10A and 250V. You can control directly all your lights and appliances. Pluggable connectors accept 18 to 22 AWG wires and simplify the installation, specially if you are planning to use more than one card. The cards are stacked with 18mm male-female standoffs. All the hardware required for the installation is provided. Relays are controlled through I2C IO Expanders, using only 2 pins on the GPIO interface and leaving all the other pins available for the user.

Source: https://www.kickstarter.com/projects/279405789/4-relays-for-raspberry-pi-8-level-stackable-10a-250v-each/

Some industrial IoT devices, based on Raspberry Pi platform already offer Relay expansion modules. One of them is ModBerry 500 CM4 series, introduced in 2014 – later upgraded with latest Raspberry Pi’s Compute Module 4+. ModBerry standard configurator offers up to 12x Relay expansion, with additional customizable options available for larger orders.

ExCard modules to peak the performance

Every TECHBASE’s industrial computer is supported by ExCard add-on modules for extra RS-232/485 serial ports, Ethernet ports, PCIe slots, analog input and output, digital I/Os, relays, M-Bus interface, opto-isolation, accelerometer, etc. To provide specific communication paths, ModBerry can be rigged with additional Wi-Fi/Bluetooth module, 3G/LTE, NarrowBand-IoT, LoRa, ZigBee, GPS and Wireless M-Bus.

The latest options for ModBerry series are:

  • SuperCap expansion, to provide constant power supply as a UPS option
  • OLED 0.96” & new OLED 1,3″ screen, allowing the control without the need of connecting into the device
  • ESP32 module as a security chip, to add a firewall into control installation and ensure constant operation of the device, even with power drops and random events
  • Aluminum case, to grant much higher durability for extra harsh industrial conditions
  • M.2 NVMe SSD controller for extra data storage
ModBerry 500 with Raspberry Pi Compute Module 4

ESP-MESH is a network protocol based on the Wi-Fi protocol. ESP-MESH enables the connection of a large number of devices (hereinafter nodes) covering a large physical area (both inside and outside) over a single WLAN (wireless local area network). ESP-MESH self-organizes and self-heals, allowing you to build and maintain networks autonomously.

Traditional Wi-Fi Network Architecture. Source: espressif.com

Traditional Infrastructure of Wi-Fi network is a point-to-multipoint network in which a single central node called an Access Point (AP) is directly connected to all other nodes called stations. The AP is responsible for arbitrating and forwarding transmissions between stations. Some access points relay transmissions to and from external IP networks through routers.

Traditional Infrastructure Wi-Fi networks have the downside of having a limited coverage area as all stations need to be in range to connect directly to the access point. In addition, the maximum number of stations allowed in the network is limited by the bandwidth of the access point, making traditional Wi-Fi networks prone to overload.

ESP-MESH vs traditional Wi-Fi infrastucture

ESP-MESH differs from traditional Wi-Fi infrastructure networks in that the nodes do not have to connect to a central node. Instead, the node can connect with its neighbors. Nodes are responsible for relaying transmissions to each other. This allows for achieving interconnections without the need for nodes to be within the range of the central node, which significantly extends the coverage area of the ESP-MESH network. Likewise, ESP-MESH is less prone to congestion as the number of allowed nodes in the network is no longer limited by a single central node.

ESP-MESH Network Architecture. Source: espressif.com

Industrial use of ESP32-based solutions

One of industrial IoT devices, supporting Espressif’s ESP32 technology is eModGATE from TECHBASE. Economical, ESP32-based solution can serve as an end-point in any installation or works well as a gateway, gathering data from scattered sensor mesh across the installation. For more information check Industrial IoT Shop with all the configuration options for eModGATE.

eModGATE with ESP32

Hyperautomation is a process in which businesses automate as numerous commerce and IT forms as conceivable utilizing apparatuses like AI, machine learning, event-driven computer program, mechanical process automation, and other sorts of choice prepare and task automation instruments.

It is the key to both computerized operational greatness and operational resiliency for organizations. To empower this, organizations had to digitize their documents/artifacts and guarantee their trade and IT process workflows were advanced. They got to mechanize tasks, processes and coordinate computerization over utilitarian zones.

Hyperautomation is irreversible and inevitable. Everything that can and should be automated will be automated.

Brian Burke, Research Vice President, Gartner

Gartner prepared a Tech Trends 2021 summary with key features of the constantly changing market. Read more at: https://www.gartner.com/en/information-technology/trends/top-strategic-technology-trends-iot-gb-pd

Industrial IoT market evolution

Data generated over the Internet of Things is growing exponentially faster than the traditional cloud environment where data is stored, so just the amount of data can justify the acceleration. In addition, in the cloud as the destination, problems related to data transfer (delay and bandwidth) occur, so travel speed is the main issue. This edge is necessary as a solution to the inefficiency of IIoT to Cloud architecture.

Fast data processing of Industrial IoT devices

When industrial IoT devices and edge processing work together, digital information becomes more powerful. Especially in contexts where you need to collect data in a traditional edge context, such as a smart meter, a parking meter or a connected trash can in a street apartment. The installation of sensors with internet access in metropolitan garbage containers is becoming increasingly common for smart urban engineers. You can then remotely monitor the container using the sensor. When it is full, the city sanitation service receives a notification and can register an order and empty the container.

By introducing AI (artificial intelligence) into the device itself, edge computing can also make more context-sensitive, quick decisions at the edge. Data gathered from the sensors can be transferred to the cloud at any time after local work has been completed, contributing to a more global AI process, or archived. With the combination of industrial IoT devices and advanced technology, high quality analysis and small footprint will become the AI standard in 2020.

ModBerry AI GATEWAY with Raspberry Pi CM4 and Google Coral

Latest innovations used in industrial solutions

One of many uses of IoT can be edge devices, dedicated to data management, process control (e.g. with MQTT protocol) and monitoring. Latest ESP32-based eModGATE controller from TECHBASE company is a series utilizing MicroPython environment to provide data management solutions for end-points applications. The eModGATE has built-in Wi-Fi/BT modem and can be equipped with additional NarrowBand-IoT, LoRa, ZigBee, etc.

For example eModGATE eqipped with wireless NB-IoT modem are perfect for industrial automation solutions, e.g. data logging, metering, telemetrics, remote monitoring, security and data management through all Industrial IoT applications.

Silicon Labs recently announced two hardware modules based on its BG22 Secure Bluetooth 5.2 SoC: 6x6mm BGM220S system bundled (SiP) and slightly optimized for wireless performance with a better link budget. BGM220P introduced, large PCB variant, wider range.

Both modules can be integrated into products with a battery life of up to 10 years using a single coin cell battery. All variants of BGM220S/P can support Bluetooth directional discovery, and some components can also support Bluetooth mesh low power protocol.

Main features

  • Silicon Labs EFR32BG22 Arm Cortex-M33 with DSP instructions and floating-point unit, up to 512 kB Flash, 32 kB RAM, 2.4 GHz radio with TX power up to 8 dBm, and Embedded Trace Macrocell (ETM) for advanced debugging
  • Supported Protocols
    • Bluetooth Low Energy (Bluetooth 5.2)
    • Direction-finding
    • Bluetooth mesh Low Power Node

Source: https://www.cnx-software.com/2020/09/14/silicon-labs-bluetooth-5-2-bgm220s-sip-and-bgm220p-pcb-module/

Industrial use of ESP32-based solutions

One of industrial IoT devices, supporting Espressif’s ESP32 and Bluetooth technology is eModGATE from TECHBASE. Economical, ESP32-based solution can serve as an end-point in any installation or works well as a gateway, gathering data from scattered sensor mesh across the installation. For more information check Industrial IoT Shop with all the configuration options for eModGATE.

eModGATE with ESP32

UPDATE 22.10.20: ModBerry 500 with Compute Module 4 available for pre-order

TECHBASE’s ModBerry industrial computer series has received an update to Compute Module 4 and is available for pre-orders. TECHBASE is leading manufacturer of Industrial Raspberry Pi and Industrial Compute Module solutions. ModBerry 500 series is fully compatible with all releases of Compute Module from Rasbperry Pi foundation.

Main features of updated device are:

  • up to 4x faster eMMC Flash with up to 32GB storage
  • up to 2x faster performance of CPU apllications than previous CM3 version
  • up to 8x more RAM (8GB LPDDR4)
  • optional 1Gbit Ethernet interface
  • optional PCIe card support for NVMe SSD drive (via M.2)
  • optional second PCIe support for wireless modem solutions

First orders will be ready with subject to the availability of the CM4 module itself.

A day ago, Raspberry Foundation announced new member of its family, a Rasbperry Pi Compute Module 4. It’s quite obvious, even from the first look, that the new module is very different from its predecessors. Main difference is a new form factor, leaving DDR2 SODIMM in the past.

The same 64-bit quad-core BCM2711 application processor as in Raspberry Pi 4B, the Compute Module 4 brings higher performance: faster CPU cores, better multimedia, more interfacing capabilities, and, for the first time, a choice of RAM densities and a wireless Wi-Fi and Bluetooth connectivity options.

Compute Module 4 comes in 32 variants. Lite, as always, offers no eMMC memory, a and standard versions come with up to 8GB RAM, 32 eMMC Flash and wireless modem.

New Raspberry Pi Compute Module 4 in new form factor

New features of Compute Module 4

  • 1.5GHz quad-core 64-bit ARM Cortex-A72 CPU as in Raspberry Pi 4 version B
  • 1GB, 2GB, 4GB or 8GB LPDDR4-3200 SDRAM
  • 8GB, 16GB or 32GB eMMC Flash storage for Standard version, Lite version without eMMC
  • Optional 2.4GHz and 5GHz IEEE 802.11b/g/n/ac wireless LAN and Bluetooth 5.0
  • Single-lane PCI Express 2.0 interface
  • Gigabit Ethernet PHY with IEEE 1588 support
  • Dual HDMI interfaces, at resolutions up to 4K
  • 28 GPIO pins, with up to 6 × UART, 6 × I2C and 5 × SPI

Source: https://www.raspberrypi.org/blog/raspberry-pi-compute-module-4/

Compute Module 4 cutting edge in Industrial IoT

Few months ago IoT Industrial Devices predicted a possible release date for Compute Module 4 in Standard and Lite version:

Upcoming Raspberry Pi Compute Module 4 possible release date
Raspberry Pi release timeline with probable Compute Module 4 release date

First Rasbperry Pi 1B model had it’s analogy in industrial Compute Module 1 after almost 2 years from it’s premiere. Compute Module 2 was probably omitted because the change from RPi1 to RPI2 mainly involved a minor change of the processor (Cortex-A7 900MHz), which was almost immediately replaced with Cortex-A53 1.2GHz in Raspberry Pi 3.

The premiere of Compute Module 3 occured a year after RPI 3 announcement, providing a significant boost of industrial market solutions. Since Raspberry Pi 4 was a great success in 2019, we might see it’s equivalent in industrial series of Raspberry Pi – Compute Module 4. A possible release date of Raspberry Pi’s Compute Module 4 is mid-2020.

Raspberry Pi Compute Module 4 high-density connector

Raspberry Pi is gaining recognition in Industry

Almost a year ago, in the beginning of 2019, Raspberry Pi Foundation presented Raspberry Pi Compute Module 3+, a successor to previous CM3 version of development board, aimed at businesses and industrial users. The Compute Module uses a standard DDR2 SODIMM (small outline dual in-line memory module) form factor. GPIO and other I/O functions are routed through the 200 pins on the board.

Only a few months later, in June 2019, came big premiere of Raspberry Pi 4 Model B, the long-awaited successor of customer RPi3+. With new processor, larger RAM options and many input/output changes, became new standard in small, embedded PC world.

Raspberry Pi Compute Module 3+
Raspberry Pi Compute Module 3+

It seems a matter of time before the Raspberry Pi Compute Module 3+ will get its own successor, probably called Compute Module 4, a new milestone of professional embedded IoT module. What might be the specification of this highly expected development board?

Industrial use of Compute Module

With Compute Module 3+ options from Raspberry Pi, TECHBASE upgraded their ModBerry 500/9500 industrial computers. From now on the ModBerry 500/9500 can be supported with extended eMMC, up to 32GB. Higher memory volume brings new features available for ModBerry series.

 ModBerry 500 with Compute Module 3+
ModBerry 500 with Compute Module 3+

Higher performance of ModBerry 500/9500 with extended eMMC flash memory, up to 32GB , powered by quad-core Cortex A53 processor allows the device to smoothly run Windows 10 IoT Core system, opening up many possibilities for data management, remote control and visualisation.

UPDATE 22.10.20: ModBerry 500 with Compute Module 4 available for pre-order

TECHBASE’s ModBerry industrial computer series has received an update to Compute Module 4 and is available for pre-orders. TECHBASE is leading manufacturer of Industrial Raspberry Pi and Industrial Compute Module solutions. ModBerry 500 series is fully compatible with all releases of Compute Module from Rasbperry Pi foundation.

Main features of updated device are:

  • up to 4x faster eMMC Flash with up to 32GB storage
  • up to 2x faster performance of CPU apllications than previous CM3 version
  • up to 8x more RAM (8GB LPDDR4)
  • optional 1Gbit Ethernet interface
  • optional PCIe card support for NVMe SSD drive (via M.2)
  • optional second PCIe support for wireless modem solutions

First orders will be ready with subject to the availability of the CM4 module itself.

According to latest leaks about Compute Module 4 specifiaction and features we can be more than sure that:

  • New Compute Module will feature Wi-Fi and Bluetooth on-board! Raspberry Pi Compute Module series will probably include versions with and without these modems to provide modules for variety of industrial applications.
  • PCI-Express line will be available externally to enable extension support via PCIe
  • Ethernet support will be enabled, most probably 1Gbps, since it is a standard in latest Raspberry Pi 4B.
  • 5x UART will be available to Compute Module 4 users

Official Raspberry Pi’s information about upcoming Compute Module 4

In recent interview with Eben Upton, the CEO of Raspberry Pi Trading, we finally had Raspberry Pi Compute Module 4 release confirmation, probably in 2021. He shared some details about the upcoming CM4 features, such as single-lane NVMe support.

The Raspberry Pi Compute Module, CM4, we will support NVMe to some degree on that,  because of course, it [Raspberry Pi 4] has a PCI Express channel. (…) We have a single lane Gen 2 which is used to supply USB 3.0 on the Raspberry Pi [4]. On the [Compute] Module that would be exposed to the edge connector and we’re likely to support NVMe over that.

Eben Upton, CEO of Raspberry Pi Trading

Raspberry Pi is gaining recognition in Industry

Almost a year ago, in the beginning of 2019, Raspberry Pi Foundation presented Raspberry Pi Compute Module 3+, a successor to previous CM3 version of development board, aimed at businesses and industrial users. The Compute Module uses a standard DDR2 SODIMM (small outline dual in-line memory module) form factor. GPIO and other I/O functions are routed through the 200 pins on the board.

Only a few months later, in June 2019, came big premiere of Raspberry Pi 4 Model B, the long-awaited successor of customer RPi3+. With new processor, larger RAM options and PCIe/NVMe support, CM4 might be a black horse of industrial automation in 2021.

It seems a matter of time before the Raspberry Pi Compute Module 3+ will get its own successor, called Compute Module 4, a new milestone of professional embedded IoT module. What might be the specification of this highly expected development board?

Raspberry Pi Compute Module 3+
Raspberry Pi Compute Module 3+

Raspberry Pi Compute Module 4 probable specification

Compute Module 4 specifications probably will look like these:

  • Broadcom BCM2711, Quad core Cortex-A72 @ 1.5GHz will highly plausible replace previous Broadcom BCM2837B0, Cortex-A53 64-bit SoC @ 1.2GHz,
  • 1GB, 2GB or 4GB LPDDR4-3200 SDRAM will become a standard options, instead of fixed 1GB LPDDR2 SDRAM,
  • PCIe/NVMe support via single lane
  • Current flash memory (eMMC) options: 8GB / 16GB / 32GB from CM3+ will probably stay the same,
  • weight and factor will stay the same, to provide a possibility to upgrade current IoT applications of CM3 and CM3+

With much higher performance, the new Raspberry Pi Compute Module 4 will, for sure, support Gigabit Ethernet, USB 3.0 expansions with PCIe/NVMe single lane. We might even see wider working temperature range, if Raspberry Pi Foundation decides to make some hardware changes, to follow, for example, ESP32 – used in end-point IoT automation.

Industrial use of Compute Module

With Compute Module 3+ options from Raspberry Pi, TECHBASE upgraded their ModBerry 500/9500 industrial computers. From now on the ModBerry 500/9500 can be supported with extended eMMC, up to 32GB. Higher memory volume brings new features available for ModBerry series. Upcoming Raspberry Pi’s Compute Module 4 will be fully compatible with TECHBASE’s ModBerry 500/9500 controllers, oferring extended features.

 ModBerry 500 with Compute Module 3+
ModBerry 500 with Compute Module 3+

Higher performance of ModBerry 500/9500 with extended eMMC flash memory, up to 32GB , powered by quad-core Cortex A53 processor allows the device to smoothly run Windows 10 IoT Core system, opening up many possibilities for data management, remote control and visualisation.

EB (Elektrobit) announced new features and functions in the EB GUIDE. This makes the Advanced Human Machine Interface (HMI) more accessible and convenient than ever to a wider range of developers.

EB provides the EB GUIDE Development Toolkit (SDK) specifically for the Raspberry Pi OS. This allows users of Raspberry Pi devices, one of the most accessible and popular embedded systems development platforms worldwide, to take advantage of the user-friendly features of EB GUIDE to make them extremely simple and efficient. The HMI can be developed. How to do it. This EB GUIDE SDK for Raspberry Pi OS is available to users for free and gives you the opportunity to see how easy it is to model HMI with EB GUIDE.

We are excited to make our unique HMI development toolchain even more capable and available to a broader group of designers and developers,” said Bruno Grasset, Head of Product Management User Experience, Elektrobit. “There are more than 30 million Raspberry Pi devices in use around the world. Pairing our advanced software with the versatile, budget-friendly Raspberry Pi development platform will accelerate innovation, allowing pros and students alike to easily create the world’s most advanced user interfaces.

Source: https://www.elektrobit.com/ebguide/blog/enhanced-flexibility-capability-eou-of-hmi-development-software/
 ModBerry 500 with Compute Module 3+
ModBerry 500 with Compute Module 3+

Industrial use of Raspberry Pi Compute Module 3+

With Compute Module 3+ options from Raspberry Pi, TECHBASE upgraded their ModBerry 500/9500 industrial computers. From now on the ModBerry 500/9500 can be supported with extended eMMC, up to 32GB. Higher memory volume brings new features available for ModBerry series. One of the options is SuperCap power support.

Higher performance of ModBerry 500/9500 with extended eMMC flash memory, up to 32GB , powered by quad-core Cortex A53 processor allows the device to smoothly run Windows 10 IoT Core system, opening up many possibilities for data management, remote control and visualisation.

SinoVoip has released specifications for the Banana Pi BPI-M5 SBC, which has the same Amlogic S905X3 SoC and many of the same features as Hardkernel’s Odroid-C4. SBC updates the Banana Pi flagship design, including the latest Banana Pi BPI-M4 based on the A53 Realtek RTD1395 quad. The latest Banana Pi was the April BPI-F2P built around the A7 SunPlus SP7021 quad.

Banana Pi BPI-M5

Characteristics of BPI-M5

  • Processor — Amlogic S905X3 (4x Cortex-A55 @ up to 2GHz); 12nm fab; Mali-G31 GPU @ up to 650MHz
  • Memory/storage:
    • 4GB LPDDR4 RAM
    • 16GB eMMC with optional up to 64GB
    • MicroSD slot
  • Networking — Gigabit Ethernet port
  • Media I/O:
    • HDMI 2.0 port for up to 4K@60Hz with HDR, CEC, EDID
    • 3.5mm audio jack
  • Other I/O:
    • 4x USB 3.0 host ports
    • USB Type-C port for power
    • Serial debug header
    • 40-pin GPIO header (UART, I2C, SPI or PWM, 5V, 3.3V, GND, 28x GPIO)
  • Other features — IR receiver; 2x LEDs
  • Power — 5V DC via Type-C; power, reset, and boot switches
  • Dimensions — 92 x 60mm
  • Weight — 48 g
  • Operating system — Android, Linux

Source: http://linuxgizmos.com/fifth-gen-banana-pi-goes-amlogic/

Industrial use of market Banana Pi-like SBCs

A year ago, TECHBASE released an updated version of the ModBerry M500 industrial IoT computer, replacing the aging Raspberry Pi 3 with a 3B+, giving it better performance. With the recent launch of the Raspberry Pi 4, TECHBASE has yet again, announced another upgrade to the M500, which now packs the latest single-board computer.

ModBerry M500 with Raspberry Pi’s 4

ModBerry M500 also utilizes many more SBC platforms, such as Orange Pi, NanoPi and Intel-based UpBoard. Find more information here: https://iiot-shop.com/product/modberry-m-series/

UPDATE 22.10.20: ModBerry 500 with Compute Module 4 available for pre-order

TECHBASE’s ModBerry industrial computer series has received an update to Compute Module 4 and is available for pre-orders. TECHBASE is leading manufacturer of Industrial Raspberry Pi and Industrial Compute Module solutions. ModBerry 500 series is fully compatible with all releases of Compute Module from Rasbperry Pi foundation.

Main features of updated device are:

  • up to 4x faster eMMC Flash with up to 32GB storage
  • up to 2x faster performance of CPU apllications than previous CM3 version
  • up to 8x more RAM (8GB LPDDR4)
  • optional 1Gbit Ethernet interface
  • optional PCIe card support for NVMe SSD drive (via M.2)
  • optional second PCIe support for wireless modem solutions

First orders will be ready with subject to the availability of the CM4 module itself.

Update on Raspberry Pi’s Compute Module 4 features [15.10.2020]

According to latest leaks about Compute Module 4 specifiaction and features we can be more than sure that:

  • New Compute Module will feature Wi-Fi and Bluetooth on-board! Raspberry Pi Compute Module series will probably include versions with and without these modems to provide modules for variety of industrial applications.
  • PCI-Express line will be available externally to enable extension support via PCIe
  • Ethernet support will be enabled, most probably 1Gbps, since it is a standard in latest Raspberry Pi 4B.
  • 5x UART will be available to Compute Module 4 users

Official Raspberry Pi’s information about upcoming Compute Module 4

In recent interview with Eben Upton, the CEO of Raspberry Pi Trading, we finally had Raspberry Pi Compute Module 4 release confirmation, probably in 2021. He shared some details about the upcoming CM4 features, such as single-lane NVMe support.

The Raspberry Pi Compute Module, CM4, we will support NVMe to some degree on that,  because of course, it [Raspberry Pi 4] has a PCI Express channel. (…) We have a single lane Gen 2 which is used to supply USB 3.0 on the Raspberry Pi [4]. On the [Compute] Module that would be exposed to the edge connector and we’re likely to support NVMe over that.

Eben Upton, CEO of Raspberry Pi Trading

First Rasbperry Pi 1B model had it’s analogy in industrial Compute Module 1 after almost 2 years from it’s premiere. Compute Module 2 was probably omitted because the change from RPi1 to RPI2 mainly involved a minor change of the processor (Cortex-A7 900MHz), which was almost immediately replaced with Cortex-A53 1.2GHz in Raspberry Pi 3.

The premiere of Compute Module 3 occured a year after RPI 3 announcement, providing a significant boost of industrial market solutions. Since Raspberry Pi 4 was a great success in 2019, we might see it’s equivalent in industrial series of Raspberry Pi – Compute Module 4. A possible release date of Raspberry Pi’s Compute Module 4 is somewhere inbetween 2020/2021.

Raspberry Pi is gaining recognition in Industry

Almost a year ago, in the beginning of 2019, Raspberry Pi Foundation presented Raspberry Pi Compute Module 3+, a successor to previous CM3 version of development board, aimed at businesses and industrial users. The Compute Module uses a standard DDR2 SODIMM (small outline dual in-line memory module) form factor. GPIO and other I/O functions are routed through the 200 pins on the board.

Only a few months later, in June 2019, came big premiere of Raspberry Pi 4 Model B, the long-awaited successor of customer RPi3+. With new processor, larger RAM options and PCIe/NVMe support, CM4 might be a black horse of industrial automation in 2021.

It seems a matter of time before the Raspberry Pi Compute Module 3+ will get its own successor, called Compute Module 4, a new milestone of professional embedded IoT module. What might be the specification of this highly expected development board?

Raspberry Pi Compute Module 3+
Raspberry Pi Compute Module 3+

Raspberry Pi Compute Module 4 probable specification

Compute Module 4 specifications probably will look like these:

  • Broadcom BCM2711, Quad core Cortex-A72 @ 1.5GHz will highly plausible replace previous Broadcom BCM2837B0, Cortex-A53 64-bit SoC @ 1.2GHz,
  • 1GB, 2GB or 4GB LPDDR4-3200 SDRAM will become a standard options, instead of fixed 1GB LPDDR2 SDRAM,
  • PCIe/NVMe support via single lane
  • Current flash memory (eMMC) options: 8GB / 16GB / 32GB from CM3+ will probably stay the same,
  • weight and factor will stay the same, to provide a possibility to upgrade current IoT applications of CM3 and CM3+

With much higher performance, the new Raspberry Pi Compute Module 4 will, for sure, support Gigabit Ethernet, USB 3.0 expansions with PCIe/NVMe single lane. We might even see wider working temperature range, if Raspberry Pi Foundation decides to make some hardware changes, to follow, for example, ESP32 – used in end-point IoT automation.

Industrial use of Compute Module

With Compute Module 3+ options from Raspberry Pi, TECHBASE upgraded their ModBerry 500/9500 industrial computers. From now on the ModBerry 500/9500 can be supported with extended eMMC, up to 32GB. Higher memory volume brings new features available for ModBerry series. Upcoming Raspberry Pi’s Compute Module 4 will be fully compatible with TECHBASE’s ModBerry 500/9500 controllers, oferring extended features.

 ModBerry 500 with Compute Module 3+
ModBerry 500 with Compute Module 3+

Higher performance of ModBerry 500/9500 with extended eMMC flash memory, up to 32GB , powered by quad-core Cortex A53 processor allows the device to smoothly run Windows 10 IoT Core system, opening up many possibilities for data management, remote control and visualisation.