Copperhill Technologies

Name: Copperhill Technologies
Address: 21 Mohawk Trail, Greenfield, MA, 01301, US
Telephone: +14132132500
Price range: $

Home
United States
Greenfield, MA
Copperhill Technologies

Development and sales of embedded systems for Controller Area Network and SAE J1939, which includes popular systems such as the Arduino and Raspberry Pi.

Copperhill Technologies Corproation was established as an electronics engineering and consulting business in 1993. Our main focus is on the development and sales of embedded systems for Controller Area Network and SAE J1939, which includes popular systems such as the Arduino and Raspberry Pi but also our own embedded solutions such as the jCOM board and gateway series. Our skills also include rapid prototyping; we can deliver prototype solutions within shortest time and assist with the manufacturing process of the final product. We deliver not only schematics, bill of material, placement files, Eagle/Gerber files but also the application source code inclduing documentation.

10/14/2026

Planning to connect a Raspberry Pi to your boat’s NMEA 2000 network? Getting the power right deserves as much attention as getting the data flowing.

The PiCAN-M HAT provides a useful bridge between the Raspberry Pi and marine electronics. Its optional onboard power supply can simplify wiring, but there’s an important distinction: the converter’s output capacity and the permitted current draw from the NMEA 2000 backbone are two separate limits.

And because they operate at different voltages, comparing the amp ratings alone can be misleading.

In my latest article, I explain:

• How NMEA 2000 network power limits work
• What changes between Raspberry Pi 3, 4, and 5
• When powering through PiCAN-M can be practical
• Why I recommend a dedicated power supply for Raspberry Pi 5 installations

Whether you’re building a Signal K dashboard, an OpenPlotter system, or an onboard data logger, a suitable power arrangement is the foundation for dependable operation.

Read the full article:
https://www.wilfriedvoss.com/p/pican-m-and-raspberry-pi-getting

10/07/2026

Looking to monitor LIN bus traffic on a Windows computer? There are several tools available, but choosing the right one involves more than finding a USB adapter.

Some packages display and record raw frames. Others interpret signals using a LIN Description File (LDF), simulate master and slave nodes, or execute automated test sequences. Standalone loggers provide another option when you need to capture traffic over long periods without leaving a computer connected.

My latest article compares offerings from PEAK, Lipowsky, Kvaser, Microchip, Vector, Ixxat, NI, and others, explaining their capabilities and where each approach fits.

Before purchasing, a few questions are worth asking:

• Does the included Windows software provide LIN monitoring, or is an additional license required?
• Can it import your LDF and display named signals with meaningful values?
• Does it support passive monitoring, node simulation, or both?
• Which communication errors can it detect and record?
• What cables and power connections are needed?

For straightforward live monitoring, PEAK PLIN-USB is a useful starting point. Lipowsky Baby-LIN deserves attention for simulation and repeatable bench testing, while combined CAN/LIN interfaces can be useful when a project involves both networks.

The article includes a comparison table and practical guidance to help narrow the choices.

Read it here:
https://www.wilfriedvoss.com/p/monitoring-lin-bus-traffic-with-windows

10/04/2026

Could SAE J1939 be an industrial network protocol?

We usually associate J1939 with diesel engines, trucks, construction equipment, and agricultural machinery. But technically, there is little preventing us from using its architecture for general industrial embedded networks.

J1939 already provides network management, address claiming, device discovery, diagnostics, request mechanisms, peer-to-peer communication, and efficient PGN-based data exchange. Protocol stacks can be small, memory requirements modest, and the learning curve is relatively manageable.

Compared with CANopen, I also see a significant advantage in usable bandwidth for process data. CANopen’s Object Dictionary is powerful, but it comes with additional complexity and protocol machinery. J1939 takes a much more direct approach: transmit the PGN, embed the data, and let the receiving application interpret the defined parameters.

And there is no reason a proprietary J1939-based industrial network has to stop at 250 or 500 kbit/s. Classical CAN can reach 1 Mbit/s—and J1939-22 takes the concept into CAN FD.

Maybe we should stop thinking of J1939 simply as a truck protocol.

[Read the full article on wilfriedvoss.com](https://www.wilfriedvoss.com/p/could-sae-j1939-be-an-industrial)

10/03/2026

CAN Bus arbitration is one of the most ingenious features of the CAN protocol.

Multiple nodes can attempt to access the bus at virtually the same time, yet no message is destroyed in the process.

CAN accomplishes this through bit-by-bit, non-destructive arbitration. The message with the highest priority continues transmitting, while the other nodes switch to listening mode and try again when the bus becomes available.

No collision recovery. No lost messages. Predictable communication.

In my latest article, I explain how CAN Bus arbitration works, including dominant and recessive bits and a complete arbitration example.

https://www.wilfriedvoss.com/p/can-bus-arbitration-ingenious-by

10/01/2026

SAE J1939 and NMEA 2000 are close relatives—but they live in very different worlds.

I’ve spent many years working with SAE J1939, developing hardware, software, and protocol stacks around it. NMEA 2000 has always been a little different for me. I’m familiar with it and have worked with it, but I never went as deeply into NMEA 2000 as I did with J1939.

What makes the comparison interesting is just how much the two protocols have in common. Both are based on CAN Bus, both use 29-bit identifiers and PGNs, and many of the underlying concepts will look immediately familiar to a J1939 developer.

But the ecosystems surrounding them—and some important technical details—are quite different.

In my latest article, I take a personal look at the similarities and differences between SAE J1939 and NMEA 2000:

https://www.wilfriedvoss.com/p/sae-j1939-and-nmea-2000-close-relatives

09/26/2026

I developed ARD1939 because I wanted an SAE J1939 protocol stack that I understood completely—and, more importantly, one that was not tied to a particular microcontroller or development environment.

I deliberately started with one of the most resource-constrained platforms available to me: the Arduino Uno with a CAN Bus shield. My thinking was simple: if I could make a J1939 protocol stack work reliably within the Uno’s limited memory and processing power, and write it in portable C/C++, moving it to more capable processors should be relatively straightforward.

That approach worked.

ARD1939 was subsequently ported to the Arduino Mega 2560, Arduino Due, ESP32, and an ARM Cortex platform using NXP MCUXpresso. The hardware-dependent functions—primarily CAN Bus access and timer control—are isolated in a separate hardware layer, while the actual J1939 protocol implementation remains essentially unchanged.

In my latest post on wilfriedvoss.com, I explain how ARD1939 evolved, why I chose this architecture, and why I believe most embedded developers should concentrate on building their application rather than spending their time developing and maintaining the protocol stack underneath it.

The current precompiled ARD1939 protocol stack can be downloaded free of charge from ard1939.com, and I eventually plan to release the complete source code through wilfriedvoss.com.

[Read the full article: ARD1939 — Why I Developed My Own SAE J1939 Protocol Stack](https://www.wilfriedvoss.com/p/ard1939-why-i-developed-my-own-sae?utm_source=chatgpt.com)

09/22/2026

For years, the Arduino Due has been my preferred platform for embedded CAN Bus development. Its two CAN controllers, combined with the simplicity of the Arduino IDE, made it an excellent choice for many of my projects.

But CAN technology is evolving, and CAN FD has changed my hardware requirements.

For new development work, I have been looking more closely at the ESP32-S3. Combined with an MCP2518FD CAN FD controller, it gives me something particularly useful: Classical CAN and CAN FD on the same development platform—while allowing me to continue working with the Arduino IDE.

In my latest post, I discuss why I still appreciate the Arduino Due, what prompted the move toward the ESP32-S3, and why I think this combination is an interesting platform for modern embedded CAN development.

[Embedded CAN Bus Development with the ESP32-S3](https://www.wilfriedvoss.com/p/embedded-can-bus-development-with?utm_source=chatgpt.com)

09/20/2026

Are we overengineering CAN Bus?

After roughly three decades of working with CAN, I still consider Classical CAN one of the most ingenious communication technologies for embedded systems.

But I increasingly question the strategy of continually extending CAN to handle more data and higher speeds. CAN FD provides more bandwidth and larger payloads. CAN XL takes that considerably further. But at what point should we simply use Ethernet for high-bandwidth applications—and let CAN continue doing what it does exceptionally well?

In my latest post, I share my personal perspective on CANopen, J1939, CAN FD, CAN XL, NMEA 2000/OneNet, and the potential of mixed CAN/Ethernet architectures.

Sometimes the right answer isn't CAN or Ethernet. It's both.

https://wilfriedvoss.substack.com/p/are-we-overengineering-can-bus-canopen

09/17/2026

CAN has been around for decades, yet its approach to reliability remains impressive.

My latest post looks at one of the reasons Controller Area Network became so successful far beyond its original automotive application: Classical CAN was designed from the beginning to expect errors—and deal with them.

Bit monitoring, stuff error detection, CRC checking, form errors, acknowledgment errors, automatic retransmission, transmit and receive error counters, and fault confinement all work together to protect communication. A persistently malfunctioning node can ultimately remove itself from the network through the Bus Off mechanism.

That level of built-in reliability helped CAN expand from automobiles into industrial automation, medical equipment, aerospace systems, and even satellites.

I also briefly look at how CAN FD inherited these mechanisms and what changed when the faster data phase was introduced.

Read the full article:

https://wilfriedvoss.substack.com/p/classical-can-designed-for-maximum

09/16/2026

I’ve started a new project: CAN Bus Embedded Development.

Rather than spending a year or more writing a technical book before anyone gets to read it, I’m taking a different approach. I’m publishing it online as a book in progress.

I’ll add new topics regularly, expand and update existing sections, and adjust the table of contents as the project develops. Readers can follow along as the book grows—and, importantly, subscribers can comment, ask questions, and suggest topics that deserve more attention.

The goal is to build a practical resource covering the hardware, software, development tools, testing, and techniques involved in creating CAN Bus embedded applications.

The first version of the table of contents is now online. More to come.

https://wilfriedvoss.substack.com/p/can-bus-embedded-development

Address

21 Mohawk Trail
Greenfield, MA
01301

Telephone

+14132132500

Website

https://copperhilltech.com/, https://jcom1939.com/, https://espberry.com/

Alerts

Be the first to know and let us send you an email when Copperhill Technologies posts news and promotions. Your email address will not be used for any other purpose, and you can unsubscribe at any time.