Forlinx Embedded has recently released the FCU3011, a compact, fanless industrial AI edge computer built around the NVIDIA Jetson Orin Nano, designed for 24/7 operations in manufacturing, smart cities, robotics, and machine vision systems, where real-time processing is needed.
The fanless system supports NVIDIA Jetson Orin Nano 4GB (34 TOPS) or 8GB (up to 67 TOPS) configurations, with 4GB/8GB LPDDR5 memory and a 128GB PCIe x4 NVMe SSD. Connectivity options include up to four Gigabit Ethernet ports, USB 3.0/2.0, HDMI 2.0 (4K), an SD card slot, optional 4G/5G and dual-band Wi-Fi via M.2 modules, along with industrial interfaces such as isolated RS-485, CAN, opto-isolated inputs, relay outputs, and an RTC. The system takes a wide 9–24V DC power input, features ESD-protected interfaces, and can be used for AGVs, visual inspection, smart factories, intelligent traffic analysis, medical devices, and small commercial robots.
Forlinx FCU3011 specifications:
- SoM options
- NVIDIA Jetson Orin Nano 4GB/8GB
- CPU – 6-core Arm Cortex-A78AE v8.2 64-bit CPU @ 1.5 GHz with 1.5MB L2 + 4MB L3
- GPU – 625 MHz 512-core/1024-core NVIDIA Ampere GPU with 16 tensor cores
- Video Encode – 1080p30 supported by 1-2 CPU cores
- Video Decode
- 1x 4K60 (H.265)
- 2x 4K30 (H.265)
- 5x 1080p60 (H.265)
- 11x 1080p30 (H.265)
- Memory – 4/8GB 128-bit LPDDR5
- AI Performance
- 20 TOPS (sparse) for 4GB model, 40 TOPS (sparse) for 8GB model,
- Super mode – 34 TOPS for 4GB model, 67 TOPS for 8GB model
- NVIDIA Jetson Orin Nano 4GB/8GB
- Storage
- 128 GB PCIe Gen4 NVMe SSD connected to M.2 M-key socket
- MicroSD card slot
- Display interface – HDMI 2.1
- Networking
- 4× Gigabit Ethernet (10/100/1000 Mbps) RJ45 ports with ESD Level 3 protection
- Optional Wi-Fi and Bluetotoh module via M.2 E-Key PCIe slot
- Optional 4G LTE (Quectel EM05) / 5G (Quectel RM500U) cellular module via M.2 slot + SIM card slot
- USB
- 2x USB 3.0 ports
- 2x USB 2.0 ports
- USB Type-C OTG port
- Expansion
- 1x M.2 Key-M socket (for SSD)
- 1x M.2 Key-E socket (for Wi-Fi module)
- 1x M.2 socket + SIM card slot for 4G LTE/5G (not clear specs, but likely Key-B)
- RS-485 (1.5 kV isolation, ESD Level 3)
- CAN bus (1.5 kV isolation, ESD Level 3)
- 2x relay outputs (5 A @ 30 VDC / 250 VAC)
- 2x opto-isolated digital inputs (1.5 kV)
- Misc
- Recovery button
- Real-time clock with CR2032 battery
- Power Supply
-  9–24 V DC via 3-pin terminal block with reverse-polarity and over-current protection
- Max load 22 W
- Dimensions – 178 × 110 × 55 mm
The Forlinx FCU3011 AI Edge Computing Box ships with ForlinxDesktop 22.04, a Linux-based distribution derived from Ubuntu 22.04, and supporting the standard NVIDIA JetPack SDK, including CUDA, cuDNN, and TensorRT for accelerated AI inferencing and development. Built-in OpenSSH support enables remote access and system management. The platform supports system updates via OTG under Linux or USB flash drive under Windows, simplifying deployment and field maintenance. Forlinx also lists OS images, testing demos, and source code resources as part of the software deliverables; however, at the time of writing, these materials are not publicly available for download.
The company mentions that the FCU3011 includes a product datasheet and user guide, along with hardware design files such as the carrier board schematic and PCB layout, and SoM pinmux details. Firmware support includes OS images, testing demos, source code, and manuals, enabling system bring-up, application development, and customization. But you have to contact the company to get access to those.
Other NVIDIA Jetson Orin Nano-powered fanless industrial computers include the Portwell PJAI-100-ON, the Seeed reComputer J3010B, and Compulab EdgeAI-ORN, all of which target industrial edge AI applications like theForlinx FCU301.
At the time of writing, I cannot find any availability and pricing information about the product. Additional details canbe found on the product page.
Debashis Das is a technical content writer and embedded engineer with over five years of experience in the industry. With expertise in Embedded C, PCB Design, and SEO optimization, he effectively blends difficult technical topics with clear communication
Support CNX Software! Donate via cryptocurrencies, become a Patron on Patreon, or purchase goods on Amazon or Aliexpress. We also use affiliate links in articles to earn commissions if you make a purchase after clicking on those links.
