A few months back I got pulled into a conversation about instrumenting a wellfield, and it reminded me how different "remote monitoring" looks once there's no mains power or Wi-Fi anywhere nearby. Boreholes are a good example of just how hard that gets. They're usually out in a field somewhere, or scattered across a wellfield, with nothing to plug into and no network to join. And the thing you actually care about, the water level, is happening tens of meters underground where nobody can see it.
Most sites still handle this the old fashioned way: someone drives out, drops a dip meter down the hole, and writes a number on a clipboard. It works, sort of, but it only tells you what's happening at that exact moment. If a pump has been running dry for three days before the next site visit, nobody finds out until it's already a problem.
So I wanted to write up how a battery powered, cellular connected logger actually solves this, and what the real engineering trade offs look like: sensor choice, power budgeting, and picking the right radio.
What's actually in the system
Strip it down and a remote borehole node is four things:
- A hydrostatic pressure sensor sitting in or above the water column
- A sealed, battery powered logger at the wellhead
- A cellular radio, NB-IoT, 2G, or 4G LTE depending on the region
- A dashboard and alerting layer, either your own or a vendor's
Sensor reads the level, logger stores and transmits it, network carries it back, and it lands on a dashboard or an SMS. NORVI's EC-M12 borehole and groundwater monitoring solution is a good real world example of this, mostly because the interesting decisions are all in how it handles power and connectivity rather than the sensor itself.
Sensor choice matters more than I expected
There are basically two ways to take a hydrostatic reading. You can lower a sealed 316L stainless sensor into the water on a cable, which is the classic direct immersion approach and usually holds up for around two years on battery. Or you can use a tube type sensor, where the sensing head stays up at the wellhead and reads the column through a tube run down the hole. Since the electronics aren't sitting in the water constantly, those tend to run past three years.
If you're only doing one or two sites it probably doesn't matter much which you pick. But if you're rolling out a fleet of nodes across a wellfield, that extra year or so adds up fast in fewer truck rolls and lower long term cost, and it's worth thinking about upfront rather than after the first round of battery swaps.
The real trick to multi year battery life
Getting years of battery life from a cellular device is not just about using a bigger battery. It is about reducing how often the cellular radio needs to transmit.
A well-designed data logger wakes up at regular intervals, records sensor readings locally, and only sends data when a meaningful change is detected. It also sends periodic heartbeat messages so you know the device is still online.
Cellular transmission uses far more power than taking sensor readings or storing data locally. By transmitting only when needed, increasing logging frequency when conditions change, and staying quiet when everything is stable, a battery-powered data logger can operate for years instead of months.
The table below shows the power consumption of the battery-powered data logger under different operating conditions.
Source: EC-M12-BC-C6-C LOW POWER APPLICATIONS – SLEEP MODE
One honest caveat here: signal strength matters a lot. A node reaching for a weak signal burns more power per transmission than one sitting under good coverage. So the choice between NB-IoT, which is lower power and reaches deeper into weak coverage areas, and 4G LTE, which draws more power but is available where NB-IoT isn't, really should be made per site rather than as a blanket default.
Alerts that don't depend on your cloud staying up
Since the cellular modem can send SMS directly, the alerting logic can live right on the device instead of in the cloud. It compares readings against thresholds and fires off a text the moment something crosses a line, whether or not your dashboard backend happens to be reachable at that moment. If you've ever had a cloud outage quietly swallow a critical alert, you'll know why that matters more than it sounds like on paper.
If you're building or speccing something similar
If you're a system integrator or just evaluating hardware for a groundwater project, it's worth reading through the full EC-M12 breakdown, which covers sensor options, cellular variants, and actual battery numbers from live deployments. Even if you end up building your own logger, the same constraints apply: no mains power, no local network, and a multi year battery target you have to design around from day one.
Curious what others have run into instrumenting off grid infrastructure like this. Sensor choice, radio choice, or something else entirely that bit you in the field? Have a remote monitoring project in mind? Contact the NORVI engineering team for expert guidance or request a free consultation.
Top comments (0)