Expectation vs. Execution: Our Journey Adopting MeshCore for IoT

Developing and deploying Internet of Things (IoT) devices is always a journey—moving from what you hope will happen to realizing how the hardware actually behaves in the wild. At Deviceworx, we recently set out to integrate MeshCore repeaters into our ecosystem to bridge the connectivity gap between our roaming sensors and asset tags and our cloud-connected xGATEWAY devices (IoT gateways). Along the way, we experienced some incredible successes, ran into a few frustrating hardware headaches, and ultimately figured out a plan to build a rock-solid, commercial-grade asset tracking solution.

Here is the story of our journey so far—and what we learned about the real-world limits and possibilities of MeshCore technology.

The Facilitators

It would be completely unfair to discuss MeshCore technology without acknowledging the original project founders. If I am ever fortunate enough to meet Scott Powell or Liam Cottle, I will gladly buy them a beer. Online resources indicate that Scott set up the protocol and Liam has been key in developing the required tools, including the MeshCore app.

Their efforts in developing this standard are massive, and the value of their contributions cannot be understated. If you haven't already, check out their site at meshcore.io. It includes very comprehensive documentation, along with online tools like a web-based flashing tool and a live network map showing active MeshCore devices.

Why MeshCore?

At Deviceworx, we design and manufacture high-performance IoT sensors and asset tags. However, we faced a persistent challenge: we needed a reliable, off-premises or "out-of-band" connectivity option for both our roaming sensors and asset tags. We realized a mesh network could act as the perfect bridge, allowing our sensors and tags to communicate with our xGATEWAY devices, which then pass the data on to the cloud.

When evaluating how to build this connection, we looked at several technologies:

• Legacy Cellular: We considered traditional cellular connectivity, but ruled it out because it is too costly and far too power-hungry for compact, roaming devices. Commonly, cellular-connected devices need a recharge every few weeks or months—which is not ideal when users typically only think of a tag when an asset goes missing.

• Long Range Wireless (LoRa): This emerged as the best fit, offering long-range connections at the lowest part cost, with dramatically reduced power draw that supports years of device operation without needing a battery recharge or replacement.

• Amazon Sidewalk™ & Others: We evaluated other LoRa alternatives like Amazon Sidewalk, but ultimately decided a MeshCore network was the safest, most cost-effective option and offered the best coverage.

Under this architecture, our Deviceworx sensors, tags, and xGATEWAYs function as MeshCore "companion devices". They communicate through MeshCore repeaters (supporting up to 64 repeater hops) to reach an xGATEWAY, which then communicates directly with the cloud.

Current MeshCore Repeater Hardware & Use Cases

If you want to deploy a MeshCore repeater today, you cannot simply buy one off the shelf (to our knowledge). Instead, the standard practice is to purchase older Meshtastic mesh repeaters and replace their internal software (we "reflash" or reprogram them).

Fortunately, this process is highly accessible:

• Cost-Effective LoRa Radio Modules: There are several very budget-friendly Meshtastic radio module options on the market that can be easily reflashed to support MeshCore using straightforward online instructions. These modules handle all MeshCore LoRa messaging within each repeater.

• Pre-Packaged Hardware: You can also buy complete, pre-packaged repeaters that include a Meshtastic radio module, a battery, and a solar panel for recharging, complete with pole-mounting support to make installation incredibly simple.

• Low CapEx: Because these radio modules and packaged repeater units are so inexpensive, the low capital expenditure (CapEx) has fueled massive, prolific growth for the MeshCore ecosystem.

While MeshCore is great for sensor data, its growth is actually being driven by its ability to transmit text messages and GPS location data. Being able to send messages and coordinates over a network that functions entirely without a working power grid or cellular towers is incredibly valuable. Because of this, many HAM radio operators and Search and Rescue (SAR) groups have actively installed MeshCore repeaters to prepare for emergencies when public infrastructure is completely knocked out.

The Reality Check: Repeater Challenges

Our expectations met reality when we began looking closely at commercially available hardware. Through our own online research and hands-on testing, we discovered several critical limitations.

1. The Power Struggle & "Flooded Message" Overload

Most pre-built, consumer-grade repeaters suffer from limited power support, featuring very small batteries and low-power solar panels. We believe that this design choice is partly due to the low battery charge currents supported by the radio module hardware itself, as well as a desire to keep retail costs incredibly low.

However, this creates a major vulnerability in a mesh network:

• The 64-Hop Burden: Because MeshCore supports up to 64 hops, a single flooded message is sent through up to 64 hops from devices that may be thousands of kilometers away. Often, these flooded messages are sent by operators during testing. They place a significant power burden on potentially thousands of repeaters that must wake up and process them, rapidly draining their batteries.

• Grid Exhaustion: We understand that some local user groups found that when their networks were hit with frequent, flooded messages, their inexpensive pre-built repeaters simply ran out of battery because their small solar panels couldn't keep up with the load.

• The Workaround: To avoid this constant power drain, at least one user group resorted to using a non-standard MeshCore frequency within their repeaters so they could communicate without the burden of processing far-away flooded messages. Unfortunately, this meant standard MeshCore users were unable to utilize these reconfigured repeaters. In effect, this group had created a private MeshCore network, only available to users who knew about their frequency change. While these frequencies are publicly available on MeshCore maps, users would have to manually audit local maps to notice the change—and their reach through this non-standard mesh would still be limited to that local area.

2. The Deviceworx Custom Power Solution

To solve the issue of limited power, we built a custom repeater using existing MeshCore-capable LoRa radio module hardware but paired it with heavy-duty power infrastructure.

We successfully tested this repeater featuring a proprietary 20W solar charging system, a proprietary 72 Watt-Hour battery (6 AHr at 12V), and a high-gain, high-efficiency antenna.

The results were outstanding:

• With adequate sunlight, our solar setup can fully recharge a completely depleted battery in under 4 hours.

• In normal operations, the daily battery drain is recovered in mere minutes.

• Even with an automated ping message sent to a repeater every 10 seconds, the power draw on our larger battery is relatively small, allowing the repeater to run for a week or more without any solar recharge at all.

By investing in robust power, the test-message battery drain issue is dramatically mitigated.

3. Fragile Hardware Design

While evaluating multiple LoRa MeshCore radio modules, we noticed some concerning design flaws that prioritized low cost over physical durability:

• Power Input Risks: One manufacturer supported multiple power connections requiring a jumper selection, but only one option supported full transmit power. Selecting the wrong jumper, plugging into the wrong port, or connecting a USB cable while powering the module from another input easily ruined the hardware.

• Antenna Disconnection Failures: None of the tested modules featured protection on their antenna connections. If a module is running at full transmit power and the antenna is unplugged (or never plugged in), the Low Noise Amplifier (LNA) on these modules can easily burn out, rendering the entire device useless.

While these cheap boards are great for hobbyists, replacing damaged hardware during commercial deployments quickly becomes frustrating. Sometimes, traveling to service a remote repeater can cost many times the price of the hardware itself.

4. Poor Remote Telemetry

To run a reliable commercial network with remote radio modules, you need to know how your hardware is doing. Unfortunately, the repeaters we tested provided very limited telemetry—often only reporting basic battery voltage and device uptime.

To remotely monitor repeater systems properly, engineers need critical metrics like battery discharge cycles (which degrade battery capacity over time), charging currents (which indicate solar panel health), and RF performance. Without all of these metrics, it's difficult to know if solar panels need cleaning, if antennas are failing, or if a battery replacement is due.

Real-World Success: Put to the Test

Despite these hardware hurdles, we took our custom repeater setup into the field to test its limits, and the results were highly encouraging.

• Over-Water Range Testing: We tested our repeater across the Georgia Strait, successfully transmitting messages between North Vancouver and Vancouver Island—a distance of approximately 45 kilometers. This test was conducted with our repeater at an elevation of about 10 meters and a companion device at just 2 meters. (For context, some local MeshCore groups have achieved over 100 kilometers of range between repeaters when both are mounted at elevations of 10 meters or higher).

• Urban Canyon Penetration: We also tested the system in a dense urban environment. Our repeater, positioned at a 10-meter elevation in North Vancouver, successfully communicated with companion devices (at 2-meter elevation) located in the middle of downtown Vancouver's high-rise commercial district—covering a distance of over 7.5 kilometers.

What's Next?

Our testing proved that we can successfully use MeshCore repeaters and companion devices to handle our messaging needs. This validation paves the way for our xTRAx Puck Asset Tags to reliably route tracking data back to our cloud-connected xGATEWAYs via a MeshCore network.

However, to scale this into a rugged, enterprise-grade solution, Deviceworx is taking things a step further. We are actively exploring the opportunity to design and manufacture our own commercial-grade MeshCore radio module. Rather than a Meshtastic board requiring a reflash, this will be a native MeshCore solution.

A Deviceworx repeater with our newly designed radio module will directly solve the pain points we identified, featuring:

• Antenna Disconnect Protection: Built-in protection to prevent radio LNA failure if an antenna is disconnected.

• Advanced Remote Metrics: Deep telemetry to monitor solar panel health, antenna performance, and battery degradation remotely.

• Robust Power Delivery: Built-in support for charging much larger batteries and managing high-wattage solar panels to handle heavy network traffic without breaking a sweat.

By bridging the gap between hobbyist-level hardware and industrial reliability, we are excited to unlock the full, uninterrupted potential of MeshCore for our customers.

Interested in learning more about our xTRAx Puck Asset Tags or how we are deploying robust mesh networks for industrial IoT? Head over to Deviceworx.com or leave a comment below!

Amazon Sidewalk™ is a trademark of Amazon.com, Inc. or its affiliates.