Box-level embedded computers, also known as Embedded PCs, have emerged as a key technology for a variety of embedded systems markets. By using a complete, turn-key system-level solution, system developers no longer have to spend time on integrating computing sub-systems and instead can focus on their application and end-user software.
What is happening in Embedded PCs?
BOXER-8230AI from Aaeon
ADLINK Technology’s DLAP-211- Nano
Advantech’s UNO-238
DeviceEdge Mini Series-M1 from Aetina
WEBS-21G0 from American Portwell Technology
eBOX560-52R-FL from Axiomtek
GP-3000 from Cincoze
DFI’s ES220-CS
ASB200-919 from Ibase Technology
Kontron’s KBox A-203
Lanner’s LEC-2290
Helix 600 from OnLogic
Sintrones’ VBOX-3611-4L-D5G
PL-50040 from WIN Enterprises
WINSYSTEMS’ SYS-427
Box-level embedded systems are known by a number of names, including industrial PCs, rugged stand-alone box systems and embedded PCs. These systems are attractive to system developers that want a high-level of integration to from the beginning. A complete embedded PC system means they don’t have to deal with making different SBCs and I/O boards work together, for example.
Because embedded box-level computing is such a broad category, it was a challenge to give this Datasheet roundup some kind of focus. With that in mind, the representative products included in this article’s product gallery is restricted to embedded PCs in the literal sense of systems with GPU or PC-architecture computing (Intel or AMD processing) with Windows OS support. The gallery also excludes box-level systems that are marketed as application-specific systems—aimed at vehicles, digital signage, railway systems, vision system and so forth. This distinction gets somewhat muddy because many of the systems in this article can and are used in vehicles, for example, so there is overlap.
An example of an embedded PC application is an upgraded robotic delivery vehicle designed for BMW’s production lines. BMW’s production of build-to-order cars requires a logistics process which guarantees that the right parts arrive at the right assembly line at the right time. Every day, 230,000 different types of parts are organized into trays to produce 10,000 cars. BMW had been using automated-guided vehicles (AGVs) to move these parts around. But those robots lacked robotic arms, intelligence or flexibility, so the AGVs were unable to load themselves or plan and navigate routes because they cannot perceive their surroundings.
BMW decided to replace the AGVs with a fleet of autonomous mobile robots (AMRs) capable of handling and transporting production material without human intervention. The company’s wholly-owned subsidiary, Idealworks, was tasked with using artificial intelligence (AI) in the logistics processes, starting with the design of autonomous robots (Figure 1). Idealworks developed an AMR called iw.hub and the fleet management software for it called AnyFleet. The AI-based robots required high-performance computing platforms that could operate reliably in stringent industrial conditions, like high vibration and physical impact. With that in mind, Idealworks decided to build a custom solution and selected ADLINK Technology’s DLAP-401 edge AI platform for the design. The compact DLAP-401 computing platform addressed the size, weight and power (SWaP) constraints of transport robots, like running on a single battery charge for at least a full shift.
FIGURE 1 The iw.hub autonomous robot is shown speeding across a BMW automotive production floor to collect its next set of BMW auto parts. The mobile robot embeds a ADLINK DLAP-401 embedded PC to enable the robot to transport auto production materials without human intervention.
The BOXER-8230AI from Aaeon combines the Nvidia Jetson TX2 NX with a loadout designed to power intelligent applications and edge networks. The system offers AI processing speeds up to 1.33TFLOPS. The system also features 16GB of onboard eMMC storage that’s expandable through the microSD card slot. I/O includes four USB 3.2 Gen 1 ports, five Gbit LAN ports and two COM ports.
• Nvidia Jetson TX2 NX
• Arm Cortex-A57 MPCore (Quad-Core)
• 8GB LPDDR4
• Storage options: 16GB eMMC / MicroSD slot / 2.5 drive bay
• HDMI Type A x1 for HDMI 2.0
• Gbit Ethernet LAN x 5
• USB 3.2 Gen 1 x4
• Wide voltage DC-in 10-24VDC
• Operating temperature range -15°C to 65°C
ADLINK Technology’s DLAP-211-Nano is part of the company’s deep learning acceleration platform (DLAP) family. It offers scalable, power-efficient AI inference with support for the full lineup of Jetson family. The DLAP series provides rich I/O interfaces and flexible expansion and support software development kits to reduce design time and efforts to accelerate edge AI deployment.
• Deep learning acceleration with Nvidia Jetson Nano
• 4GB DRAM, 16GB eMMC
• Compact fanless system
• 2x GbE, 4x Type A USB 3.0, 1x USB OTG
• SATA/NVMEx2, SD Card
• 4x SMA antenna holes
• Dimensions:
148mm × 120mm × 49.1mm
• Temperature range: -20°C to 70°C
Advantech’s UNO-238 is a compact IoT edge computer powered by the 8th gen Intel Core i processor. The UNO‑238 offers an up to 50% improvement in performance compared to the previous model. The unit features its 3.5″ compact HD size with swappable RAM for increased flexibility and easy maintenance, while the built-in onboard 32GB eMMC storage with dual M.2 expansion enables rapid deployment for IoT applications.
• 8th gen Intel Core i processor
• 2x GbE, 4x USB 3.2, 2x RS-232/422/485
• 1x HDMI, 1x DP, 1x GPIO (8 bit)
• Compact, fanless design
• Rubber stopper for stand mount
• Optional kit for DIN-rail mounting
• Threaded DC jack for reliable power supply
• Optimized mechanical design for easy RAM swapping
The DeviceEdge Mini Series-M1 from Aetina is a palm-size embedded PC featuring high performance and low power consumption in a limited space. With fanless design, it runs silently and saves the maintenance cost. Users can choose various Nvidia Jetson modules based on different performance demands without changing mechanism design.
• Nvidia Jetson Nano, Xavier NX or TX2 NX module
• 0.5TFLOPS (FP16) at 5W-10W
• Expand storage to 128GB
• 4GB LPDDR4 DRAM
• Front I/O: 2x USB 3.2 Gen1, 1x power button, 1x RJ45
• Rear I/O: 1x USB Type-C, 1x DB15 male connector, 1x Gbit Ethernet, 1x HDMI 1.4, 1x DC-in
• Operating temperature: -20°C to +50°C
The WEBS-21G0 from American Portwell Technology builds on Intel Whiskey Lake-U Core i SoC and takes advantages of Intel Core mobile technologies that can support superior processing power and capability. Support two Gigabit Ethernet ports, two mini-DP ports and M.2 2280 key M for storage. The WEBS-21G0 is well suited for applications in harsh environments such as digital signage, transportation and more.
• 8th Gen Intel Core i7/i5/i3 processor
• 1x DDR4 2400 non-ECC SO-DIMM, up to 32GB
• 1x M.2 key M 2280 for storage
• 3x USB 3.0, 2x Gbit Ethernet
• Dual mini-DP ports
• 1x Antenna hole for Wi-Fi module
• IP40 Rating
• Operation temperature: 0°C to 50°C
• Dimensions: 150mm × 150mm × 62.6mm
The eBOX560-52R-FL from Axiomtek is powered by the Intel Core i5-8365UE or Intel Celeron 4305UE processors. The fanless yet low-power embedded system can operate under wide temperatures ranging from -10°C to 50°C with 0.7m/s airflows and endure up to 50G shock and 3Grms vibration. Moreover, it features two independent 4K Ultra HD displays through its DisplayPort++ and HDMI for multi-display applications.
• 8th gen Intel Core and Celeron processor onboard (Whisky Lake-U)
• One M.2 Key E 2230 for Wi-Fi module
• DDR4-2400 SO-DIMM up to 32GB
• 2x GbE LAN and 2x USB 3.1 Gen2
• Supports DisplayPort++ and HDMI for dual independent displays
• Mounting options: Wall mount,
DIN-rail and VESA mount
• Dimensions:
141.6mm × 106mm × 55 mm
The GP-3000 from Cincoze is powered by a 9th gen Intel Xeon/Core processor and supports up to two 250W high-end graphics cards adding fast performance for AI, machine learning, machine vision and big data analysis. The GP-3000 features a GPU Expansion Box (GEB) that supports the expansion of up to two 250W 328mm full-length graphics cards.
• Intel Xeon/Core (Coffee Lake-R / Coffee Lake) processor
• Up to 2x DDR4-2666 ECC/non-ECC SO-DIMMs, max. of 64GB
• GPU Expansion Box (GEB)
• GEB supports up to 2x 250W full-length GPU cards
• 2x COM, 6x USB 3.2, 1x HDMI/DP/VGA
• Multiple PCIe expansion slots for expansion and
• 5x GbE LAN ports
• Dimensions: 105mm × 190mm × 360mm
DFI’s ES220-CS is a high-performance miniature computer designed to improve computing performance while maintaining compact dimensions. It is equipped with 8th/9th generation Intel Core processors with hardware acceleration of video codes not limited to H.265, and supports up to 8-cores supporting up to 64GB DRAM capacity.
• 8th/9th Gen Intel Core processors
• 2x DDR4 SODIMMs up to 64GB
• Storage: 1x M.2 SSD, 2x 2.5″ SSD (option) (RAID supported)
• 2x GbE, 4x USB 3.1, 2x USB 2.0, 2x COM
• 2x HDMI 2.0 supports resolution of 4K at 60Hz
• Acoustic level supports full load 36dB at operating temperature 45°C
• Dimensions: 17.5cm x 4.8cm x 12cm
The slim-type ASB200-919 from Ibase Technology is a fanless system based on the IB919 3.5″ disk-size SBC. The system is designed for both industrial and commercial applications, including factory automation, production line equipment, POS computers and gaming applications. The box is equipped with the 8th Gen Intel Core i7/i5/i3 / Celeron processors and two DDR4 SO-DIMM slots from 8GB to 32GB.
• Fanless system with Ibase IB919 3.5’’ SBC
• 8th Gen Intel Core™ U-Series Processor
• Supports 2x M.2 sockets (E-Key, CNVi / M-Key, NVMe)
• 12V (-10%) to 24V (+10%) DC-in Power Input
• 2x DDR4 SO-DIMM, Max.32GB
• 4x USB 3.1, 2x Gbit LAN, 1x COM, External GPIO
• 2x DP (1x by Type-C), TPM (2.0)
• Optional VESA mount bracket
• Dimensions:
180mm × 150mm × 66mm
Kontron’s KBox A-203 is based on the Google Coral Edge TPU in combination with the Intel Atom x7‑E3950 processor. The maintenance-free, fanless and battery-free system was developed as an AI inference gateway for visual AI-applications. The KBox comes with a pre-installed Google Coral Edge TPU and Kontron provides a Debian Linux image with pre-installed software packages from Google.
• Intel Atom processor and Google Coral Edge TPU
• Well suited for AI and gateway applications
• TensorFlow Lite compatible
• Maintenance-free: no moving parts, battery-free
• Up to 2x serial ports,
Optional Fieldbus, Wi-Fi, LTE
• Trusted Platform Module 2.0
• Kontron KEAPI middleware
• 24VDC (+/- 20 %) input voltage
• Operating temperature:
-20°C to +60°C
Lanner’s LEC-2290 edge box pairs an x86 CPU with the Hailo-8 AI module to produce a solution capable of optimizing demanding edge workloads and accelerating time-to-market for AI deployments. Lanner’s low-footprint LEC-7242 industrial wireless gateway integrates the Hailo-8 into a fanless, efficient appliance for real-time applications powered by AI at the edge.
Comet Lake Processor-Based System for Industrial Edge
The Helix 600 from OnLogic lets users choose the mix of performance and features that best suit your project. Powered by Intel 10th gen Comet Lake processing, including the latest Core i9 CPU, Helix 600 can accommodate up to 64GB of RAM, has room for dual SSDs and features available VPU accelerators, including Intel Movidius and Google Coral.
• Intel 10th gen Comet Lake Core i3/i5/i7/i9 processing
• 2x Gbit LAN
• DDR4 SO-DIMM (non-ECC)
up to 64GB
• 12VDC to 24VDC power, triple display capable
• PCIe x16 expansion for graphics cards, storage or additional I/O
• 6x antenna holes
• DIN-mount and wall-mount options
• Operating temperature: 0°C to 50°C
• Dimensions:
303mm × 65mm × 210mm
Sintrones’ VBOX-3611-4L-D5G is an in-vehicle computer powered by an Intel Gen 6 Core i7-6600U CPU. The system features wide range 9VDC to 48VDC power inputs, smart vehicle power ignition and adjustable power management. The unit is E13 and EN50155 certified, and can be adapted to unstable vehicle power sources with ease. Optional CAN bus and GPS dead reckoning provide interfaces for real time status monitoring.
• Intel Core i7-6600U
• Wireless support dual 5G w/ Dual SIM Cards
• WLAN, GPS Dead Reckoning, Wi-Fi, Bluetooth (optional)
• 9VDC to 48VDC Power Input
• Dual hot swappable SATA storage, RAID 0, 1, 5
• 1x CAN bus 2.0B (optional)
• E13 and EN50155 certified
• Dimensions:
250mm × 150mm × 55mm
Fanless Systems Serves IoT, Call Center and Medical Apps
PL-50040 from WIN Enterprises is a fanless, high-performance networking system based on the Intel 8th/9th gen Core processors. This platform provides high CPU computing power for environments requiring quiet operation, such as counter-based medical applications. Wi-Fi can be supported. With 2x GbE LANs and 4x USB 3.0 and 2x USB 2.0 ports, it is also well suited for small call centers that demand quiet, but high performance.
• 8th/9th gen Intel Core processors
• Supports DDR4 SODIMM up to 32GB
• Three independent displays: 1x VGA, 1x DVI/HDMI, 1x HDMI/DP
• Compact, cost-effective solution
• 2x GbE, 4x COM, 4x USB 3.0,
2x USB 2.0
• 2x Mini PCIe slots support mSATA and Wi-Fi modules
• Wall-mountable, silent-running
WINSYSTEMS’ SYS-427 is a rugged, embedded computer featuring the latest generation Intel Apollo Lake-I SOC processor. Its small size, low power and extended operational temperature make it well suited for rugged embedded systems in the industrial control, transportation, energy and industrial IoT markets.
• Intel Apollo Lake-I E3900 Processor (dual or quad-core)
• Up to 2.0GHZ CPU clock rate
• Up to 8GB DDR3L ECC system memory
• Supports up to three independent displays
• 2x DisplayPort with 4K resolution
• LVDS with backlight and touch control
• Time-coordinated computing
• Soldered down TPM 2.0 hardware security
• -40°C to +85°C operating temperature Range
• Power input: 10VDC to 50VDC
Note: We’ve made the Dec 2022 issue of Circuit Cellar available as a free sample issue. In it, you’ll find a rich variety of the kinds of articles and information that exemplify a typical issue of the current magazine.
Jeff served as Editor-in-Chief for both LinuxGizmos.com and its sister publication, Circuit Cellar magazine 6/2017—3/2022. In nearly three decades of covering the embedded electronics and computing industry, Jeff has also held senior editorial positions at EE Times, Computer Design, Electronic Design, Embedded Systems Development, and COTS Journal. His knowledge spans a broad range of electronics and computing topics, including CPUs, MCUs, memory, storage, graphics, power supplies, software development, and real-time OSes.
This website uses cookies to improve your experience while you navigate through the website. Out of these cookies, the cookies that are categorized as necessary are stored on your browser as they are essential for the working of basic functionalities of the website. We also use third-party cookies that help us analyze and understand how you use this website. These cookies will be stored in your browser only with your consent. You also have the option to opt-out of these cookies. But opting out of some of these cookies may have an effect on your browsing experience.
Necessary cookies are absolutely essential for the website to function properly. This category only includes cookies that ensures basic functionalities and security features of the website. These cookies do not store any personal information.