Solar · 9 min read
Best Energy Monitor for Solar Systems

What to actually check before specifying a monitor for a solar site — bidirectional measurement, accuracy class, channel count, commissioning and firmware.
Published · Updated
We make energy monitors, so treat this as a specification checklist from someone with a position rather than a neutral roundup. The criteria below are the ones that decide whether a device works on a solar site, and they are worth putting to any supplier, us included.
It has to measure both directions
This is the first filter and it eliminates a surprising number of products. A monitor built for consumption assumes power flows one way: into the building. On a solar site it does not.
A device that measures magnitude but not direction will happily add your exported units to your imported ones and report the sum as consumption. The specification to look for is forward and reverse energy measured separately per channel — import and export distinguished rather than netted.
Accuracy class, and what it actually means
Accuracy is expressed as a class. Class 1 means measurement within one percent under defined test conditions; Class 2 within two percent. Anything that does not state a class has not been characterised, which is itself the answer.
For understanding behaviour, finding waste and verifying that a change worked, Class 2 is entirely adequate. Where the data will be used for anything resembling settlement — submetering between tenants, or apportioning cost between units — Class 1 is worth specifying. Be clear that neither is a utility revenue meter, and any supplier implying otherwise should be pressed on it.
Enough channels, and the right ones
A solar site needs at minimum generation and grid flow. A single channel device forces a choice between them, which means either you cannot see production or you cannot see export.
Three channels covers a single phase home properly. A three phase site needs six to take both grid and generation per phase, which is the only way an inverter running unevenly across phases becomes visible. The full reasoning is in single phase vs three phase monitoring.
Commissioning that fits in one visit
For an installer this is a commercial question, not a technical one. A device that takes forty minutes to get online, or needs a laptop and a cable, or fails silently when the homeowner later changes their router, turns a fitted product into a support liability.
- Wireless commissioning from a phone, on site, without additional equipment
- 2.4 GHz Wi-Fi rather than 5 GHz only — range and wall penetration matter inside a metal distribution board
- Reported signal strength, so a marginal installation is identified at the time rather than after handover
Firmware you can update remotely
Any connected device you install will need updating. The question is whether that update requires a site visit.
Over-the-air updates change the economics for anyone maintaining more than a handful of sites. Without them, a fleet of a hundred monitors is a hundred visits every time something needs fixing, which quietly destroys the margin on the original sale.
Data that survives an outage
Power cuts are routine, and a monitor that loses track of time during one will corrupt the day it comes back to. Look for NTP synchronisation with a real-time clock backed up locally, so readings land on the correct timestamp regardless of what the supply did overnight.
Electrical robustness for the actual grid
A device specified for a stable supply will not last on an Indian one. The figures worth checking are the operational voltage range, the frequency range, and surge immunity.
- Wide operational range — 90 to 264 V AC accommodates the sags and swells a real connection sees
- 47 to 63 Hz, rather than an assumption of a rock-steady 50
- Surge immunity tested to IEC 61000-4-5, which is what stands between the device and a nearby lightning strike or switching transient
The app the customer actually opens
For an installer this determines whether monitoring reduces support calls or generates them. A homeowner who can see why their bill looks the way it does calls less than one who cannot — the question why solar sometimes fails to cut a bill answers itself when the export split is visible on a phone.
Check that both iOS and Android are supported and actively maintained, and that history goes back far enough to compare seasons rather than just days.
Questions worth putting to any supplier
- 1Is forward and reverse energy measured separately on every channel?
- 2What accuracy class, and measured under which standard?
- 3How many CT channels, and what is the maximum on a three phase supply?
- 4How is the device commissioned, and how long does it take on site?
- 5Can firmware be updated over the air across a fleet?
- 6What is the surge immunity rating and the operational voltage range?
- 7What happens to data during a power cut?
- 8Is there an API, if the data needs to reach your own systems?
Common questions
What should I look for in an energy monitor for a solar system?
Separate forward and reverse energy measurement so import and export are distinguished, a stated accuracy class, enough CT channels for both generation and grid flow, wireless commissioning, over-the-air firmware updates, and surge immunity appropriate to the local supply.
What is the difference between Class 1 and Class 2 accuracy?
Class 1 measures within one percent under defined test conditions, Class 2 within two percent. Class 2 is adequate for understanding usage and verifying changes; Class 1 is worth specifying where the data apportions cost between parties, such as submetering.
Why do over-the-air firmware updates matter?
Because without them every firmware change means a site visit. For an installer maintaining a fleet, that turns a routine update into a hundred separate journeys and erodes the margin on the original installation.

