Aug 27, 2026Solar Knowledge

A Deye Three-Phase Zero Export Case: Meter, CTs, and Firmware Troubleshooting

A real Deye three-phase zero export case covering CT wiring, meter compatibility, Eastron communication, and firmware update troubleshooting.

deye-zero-export-meter-troubleshooting-cover

A Deye Three-Phase Zero Export Case: Meter, CTs, and Firmware Troubleshooting

We recently helped a customer troubleshoot a zero export problem on a Deye three-phase hybrid inverter.
It wasn’t a dramatic failure, and it wasn’t something you could fix by swapping one part. Getting to the bottom of it meant checking the inverter, the CTs, the external meter, the communication protocol, the settings, and eventually the firmware — all together.
This is the kind of issue where the customer does not just need a part number. They need someone to help separate the zero export meter problem from the inverter settings, CT wiring, meter compatibility, and firmware response.
The system was built around a Deye SUN-12K-SG04LP3-EU-AM2-LV inverter, with zero export already configured on site. Under a normal, balanced load, everything looked fine. But as soon as roughly 1000W of load landed on one phase, the other two phases started feeding power back to the grid.
The customer was specific about what “zero export” needed to mean here: it wasn’t enough for the combined three-phase power to sit near 0W. Each phase — L1, L2, L3 — had to stay out of the grid individually.
That distinction matters more than people usually assume. Total zero export and phase-by-phase zero export are not the same requirement, and a system can satisfy one while failing the other.

First instinct: blame the meter

There was an external meter on site, so naturally the first question was whether the zero export meter itself was the problem.
That is a fair place to start. In most anti-backflow complaints, the meter is one of the first suspects.
We did not jump to replacing it, though.
The anti-backflow meter only measures. The inverter is what actually controls the output. If the inverter is not reading the meter correctly — or if something is off with CT phase sequence, CT direction, CT placement, or the firmware response logic — a new meter will not fix the root cause.
So before touching hardware, we laid out the problem clearly and checked with both Deye support and Eastron.

Deye’s take: test with CTs first

Deye’s answer was straightforward.
For this class of energy storage inverter, zero export can normally be handled through CTs alone. An external meter is usually recommended only when the CT cable run is too long or the CT signal is not stable enough.
They also flagged something worth remembering: if the external meter was not sourced through Deye, compatibility with the inverter is not guaranteed.
So our first question to the customer was simple:
Were the earlier test photos and videos taken with the external meter still connected?
If yes, disconnect it and re-test using CT control only.
This was not about pointing fingers. It was about cutting variables. We needed to know whether the inverter could hold zero export correctly on CTs alone before saying anything meaningful about the meter’s role.
For customers searching for a Deye zero export meter or Deye anti backflow meter, this point matters. The meter must not only measure correctly; it must also communicate correctly with the inverter.

Eastron’s take: their meter is not automatically “the” meter for Deye

We checked with Eastron too.
They confirmed their three-phase meters can measure import and export on L1, L2, and L3 separately. No issue there.
But here is the catch:
Even though Eastron manufactures meters that Deye sells, a standard Eastron unit bought on the open market may not talk to a Deye inverter the same way.
The Deye-supplied version often uses a specific communication protocol or parameter configuration. A generic Eastron meter can display correct numbers on its own screen and still be unreadable — or misread — by the inverter.
Practical takeaway:
If a project genuinely needs an external meter, source it through Deye or an authorized distributor rather than buying a “compatible-looking” meter separately.
This is why terms like Eastron meter for Deye inverter and Deye compatible Eastron meter are not just SEO phrases. In real projects, they point to a real compatibility issue.

What we had the customer check on site

We put together a short troubleshooting sequence:
  1. Confirm whether prior test data was taken with the external meter connected. If yes, disconnect it.
  1. Switch the setting to Zero Export to CT.
  1. Verify CT1, CT2, and CT3 actually line up with L1, L2, and L3.
  1. Check CT direction. This trips people up constantly.
  1. Confirm the CTs sit at the main grid connection point — not at the inverter output, and not on a single load branch.
  1. Leave functions like Asymmetric Phase Feeding off for now. It is not designed to solve phase-by-phase zero export and just adds another variable.
  1. Keep the inverter online so Deye support can check the firmware version remotely.
None of this is exotic. But in a three-phase system, one wrong CT direction or one mismatched phase sequence is enough to send the whole diagnosis down the wrong path.

The result: a firmware update fixed the response time

After going through the checklist, Deye pushed a firmware update.
The customer’s own words afterward:
“We have received firmware update on our Deye models, reaction for 0 feed-in is way better now. We are testing what it will do on site.”
In other words, the zero feed-in response became noticeably faster, and they were still validating it on site as of that message.
That result told us this was never purely a zero export meter hardware issue.
Zero export is a real-time control loop. The inverter reads CT or meter data continuously and has to adjust output fast enough to keep up with sudden load changes. Older firmware that reacts too slowly can show brief export spikes exactly when a single-phase load jumps — which matches what the customer originally saw.

What we would take from this case

A few things are worth carrying into similar projects.
Total power near 0W does not mean each phase is clean. Check L1, L2, and L3 independently before calling the system stable.
Do not assume the meter is faulty first. The meter only reports data. The inverter decides what to do with it. Confirm the inverter is reading correctly before replacing anything.
A compatible brand is not the same as a compatible unit. Eastron makes meters for Deye, but a standard retail Eastron meter is not guaranteed to work the same way.
CT installation errors quietly break everything downstream. Wrong direction, wrong phase mapping, wrong location — any of these makes later settings meaningless.
Firmware is not a footnote. In this case, the firmware update made a clear difference. Control response speed lives in firmware, not just in hardware specs.For a more technical explanation of why one phase may still export under zero export mode, see our guide on three-phase zero export troubleshooting.

A rough order of operations for similar issues

If you are chasing a similar problem on a Deye three-phase setup, this is the order we would follow:
  1. Pin down whether the requirement is total zero export or phase-by-phase zero export.
  1. Identify whether the system relies on CTs, an external anti-backflow meter, or both.
  1. If there is an external meter, verify it is actually Deye-compatible.
  1. When in doubt about the meter’s origin, disconnect it and test CTs alone first.
  1. Check CT phase sequence, direction, and physical placement.
  1. Set the inverter to Zero Export to CT.
  1. Keep secondary balancing functions off during the initial test.
  1. Keep the inverter online so Deye can check firmware remotely.
  1. After any firmware update, re-test specifically with sudden single-phase load changes.
For solar and energy storage projects, choosing a solar zero export meter or zero export device for solar inverter should not be based only on the product name. It should be based on the inverter model, supported communication protocol, CT layout, and the actual zero export requirement on site.

Final thought

This one did not get solved by swapping a meter.
It got solved by working through the chain: where the data comes from, whether the inverter is actually reading it, whether the CTs are installed correctly, whether the meter is genuinely compatible, and whether the firmware is current.
For us, this case is a good reminder that supplying a zero export meter or inverter accessory is only one part of the job. In real projects, the value is often in helping the customer ask the right questions, check the right points, and avoid replacing parts before the root cause is clear.
Buying the right hardware is just the starting point. Commissioning and troubleshooting are usually what determine how the system actually performs.
Phase-by-phase zero export is not solved by one component. Measurement, communication, configuration, and firmware all have to work together — and in three-phase systems, it usually only takes one of them being slightly off to throw the whole result.