# \[APP\]\[Pro\] EcoFlow STREAM Series - solar, battery, smart meter and Homey Energy automation

**URL:** <https://community.homey.app/t/app-pro-ecoflow-stream-series-solar-battery-smart-meter-and-homey-energy-automation/157399>\
**Category:** Apps\
**Created:** [July 20, 2026, 6:45pm UTC](https://community.homey.app/t/app-pro-ecoflow-stream-series-solar-battery-smart-meter-and-homey-energy-automation/157399 "2026-07-20T18:45:14Z")\
**Posts on this page:** 2\
**Page:** 6

<div class="post-metadata">

**Author:** ![Ronald3](https://sea1.discourse-cdn.com/flex025/user_avatar/community.homey.app/ronald3/32/80106_2.png) [@Ronald3](https://community.homey.app/u/Ronald3)\
**Post date:** [November 11, 2026, 10:35am UTC](https://community.homey.app/t/app-pro-ecoflow-stream-series-solar-battery-smart-meter-and-homey-energy-automation/157399/101 "2026-11-11T10:35:20Z")

</div>

I noticed btw that some guys over at Home Assistant are working on it as well. And if I read it correctly they figured out how to set the charge/discharge level for the AC 5000. For me that would be a very interesting command to give as Homey at that point would be able to control the battery.

> <https://github.com/shuette42/ecoflow-energy-ha/issues/393#issuecomment-5647636137>
>
> \### Home Assistant Version
> 
> 2026.9.1
> 
> \### Integration Version
> 
> 1.20.0
> 
> \### Devic…e Type
> 
> Stream
> 
> \### Device Firmware Version
> 
> 1.1.4.46
> 
> \### Connection Mode
> 
> Enhanced (MQTT Protobuf)
> 
> \### Bug Description
> 
> \<html\>
> \<body\>
> \<h1 dir="ltr"\>STREAM AC 5000 (ES22) – Battery charge power reports 0 W\</h1\>
> 
> \<h2 dir="ltr"\>Environment\</h2\>
> 
> \<ul dir="ltr"\>
> \<li\>Integration: \<code\>ecoflow\_energy\</code\> (shuette42/ecoflow-energy-ha)\</li\>
> \<li\>Mode: Enhanced (MQTT, \<code\>mqtt-e.ecoflow.com:8084\</code\>)\</li\>
> \<li\>Device: STREAM AC 5000, serial prefix \<code\>ES22...\</code\> (serial redacted)\</li\>
> \<li\>Work mode per \<code\>select.ecoflow\_stream\_ac\_5000\_0340\_betriebsmodus\</code\>: \<code\>custom\</code\>\</li\>
> \</ul\>
> 
> \<h2 dir="ltr"\>Symptom\</h2\>
> 
> \<p dir="ltr"\>\<code\>sensor.ecoflow\_stream\_ac\_5000\_0340\_batterieladeleistung\</code\> (battery
> charge power) and \<code\>sensor.ecoflow\_stream\_ac\_5000\_0340\_batterieentladeleistung\</code\>
> (battery discharge power) stayed at a constant \<code\>0 W\</code\> for extended
> periods, even though the EcoFlow app showed active charging/discharging
> at the same time. Other readings from the same device (grid import
> power, home consumption, SoC, battery voltage) updated correctly during
> that same window — the MQTT connection was not the problem
> (\<code\>mqtt\_status.data\_receiving: true\</code\>, LQI/RSSI unremarkable).\</p\>
> 
> \<p dir="ltr"\>At a later point, the \<code\>Batterieladeleistung\</code\> (battery charge power)
> sensor correctly showed \<code\>47 W\</code\> while the device was actively charging
> per the app. So the field isn't permanently broken — it just reads \<code\>0\</code\>
> in some moments while the device is actively charging/discharging.\</p\>
> 
> \<h2 dir="ltr"\>Approach\</h2\>
> 
> \<p dir="ltr"\>Using the \<strong\>"Record extra diagnostic data for 24 hours"\</strong\>
> (\<code\>raw\_capture\</code\>) option, raw MQTT frames of the \<code\>property\</code\> message
> \<code\>cmd\_func=32 / cmd\_id=50\</code\> (the "flow-matrix" telemetry) were captured.
> This message was decoded generically as protobuf (without knowledge of
> the project's internal \<code\>.proto\</code\> definition) to verify two reference
> values:\</p\>
> 
> \<ul dir="ltr"\>
> \<li\>Field \<code\>1.1.6 = 82\</code\> → matches SoC 82 % exactly\</li\>
> \<li\>Field \<code\>1.1.7 = 16635\</code\> → matches battery voltage 16.635 V exactly\</li\>
> \</ul\>
> 
> \<p dir="ltr"\>This confirms the decoding is structurally correct.\</p\>
> 
> \<h2 dir="ltr"\>Raw data – two comparison frames\</h2\>
> 
> \<p dir="ltr"\>\<strong\>Frame A – idle\</strong\> (SoC frozen at 82 %, app shows no activity),
> \<code\>2026-09-10T19:31:45Z\</code\>:\</p\>
> 
> \<div role="group" aria-label="Code" tabindex="0"\>\<div\>\<div\>\</div\>\</div\>\<div\>\<pre style="color: rgb(234, 236, 240); background: transparent; font-family: var(--font-mono);"\>\<code style="color: rgb(234, 236, 240); background: transparent; font-family: var(--font-mono); white-space: pre-wrap;"\>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\</code\>\</pre\>\</div\>\</div\>
> 
> \<p dir="ltr"\>\<strong\>Frame B – actively charging, ~47–58 W per HA/app\</strong\>,
> \<code\>2026-09-10T19:43:55Z\</code\>:\</p\>
> 
> \<div role="group" aria-label="Code" tabindex="0"\>\<div\>\<div\>\</div\>\</div\>\<div\>\<pre style="color: rgb(234, 236, 240); background: transparent; font-family: var(--font-mono);"\>\<code style="color: rgb(234, 236, 240); background: transparent; font-family: var(--font-mono); white-space: pre-wrap;"\>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\</code\>\</pre\>\</div\>\</div\>
> 
> \<h2 dir="ltr"\>Field diff (generic protobuf decoding)\</h2\>
> 
> \<div dir="ltr"\>
> Field path | Idle | Active (charging) | Note
> \-- | -- | -- | --
> 1.1.7 | 16635 | 16637 | Battery voltage (mV) – reference value, matches
> 1.1.8 | 18446744073709548240 (as signed int64: -3376) | 777 | Candidate for signed battery power – the only field that flips from negative to positive
> 1.1.12 | 258958 | 258311 | drops by ~647
> 1.1.17 | 3326 | 3327 | Cell voltage (mV)? minor change
> 1.1.25 / 1.1.42 / 1.1.44 | 1118105890 / 1118105900 / 1118105900 | 1118078878 / 1118078886 / 1118078886 | fixed32, possibly timestamp/counter
> 1.1.26 | 0 | 12 |  
> 1.1.27 | 56 | 0 |  
> 1.1.28 | 5580 | 24247 | notable jump
> 1.1.31 | 2 | 1 | possibly a status flag (charge/discharge state?)
> 1.1.47 | 1 | 2 | possibly a status flag
> 1.1.50 | 304928 | 304931 | increases by 3
> 1.1.51 | 138993 | 139642 | increases by ~649 (complements 1.1.12, possibly an energy counter)
> 1.1.80 | 2282 | 2293 | increases by 11
> 1.10 | 338 | 331 | message length (not a payload field)
> 
> \</div\>
> 
> \<p dir="ltr" style=""\>\<strong\>Most likely candidate for battery power: field \<code\>1.1.8\</code\>\</strong\> (the only
> sign flip). \<code\>1.1.31\</code\> and/or \<code\>1.1.47\</code\> could be the charge/discharge
> status flag (\<code\>batt\_charge\_discharge\_state\</code\>).\</p\>
> 
> \<h2 dir="ltr" style=""\>Request for feedback\</h2\>
> 
> \<p dir="ltr" style=""\>If anyone with access to the project's \<code\>.proto\</code\> definition can confirm
> which field actually carries \<code\>batt\_charge\_power\_w\</code\> /
> \<code\>batt\_discharge\_power\_w\</code\>, the ES22 parser could likely be fixed with
> little effort. Happy to provide further comparison frames (e.g. at a
> higher discharge power) if that would help.\</p\>
> \</body\>
> \</html\>
> 
> \### Steps to Reproduce
> 
> 1. Open the EcoFlow app and start charging or discharging the STREAM AC 5000
> 2. In Home Assistant, check sensor.\<device\>\_battery\_charge\_power (Batterieladeleistung) / battery\_discharge\_power (Batterieentladeleistung)
> 3. Observe that the sensor reads 0 W even though the app shows an active, non-zero charge/discharge power at the same moment
> 4. Other sensors on the same device (grid import power, home consumption, SoC, battery voltage) update correctly in the same time window, ruling out a connection/MQTT issue
> 
> \### Relevant Log Entries
> 
> \`\`\`text
> 
> \`\`\`
> 
> \### Diagnostics
> 
> \`\`\`json
> 
> \`\`\`
> 
> \### Additional Context
> 
> \_No response\_

---

<div class="post-metadata">

**Author:** ![EdNL](https://sea1.discourse-cdn.com/flex025/user_avatar/community.homey.app/ednl/32/90603_2.png) [@EdNL](https://community.homey.app/u/EdNL)\
**Post date:** [November 12, 2026, 2:44pm UTC](https://community.homey.app/t/app-pro-ecoflow-stream-series-solar-battery-smart-meter-and-homey-energy-automation/157399/102 "2026-11-12T14:44:51Z")

</div>

Sorry, sorry, sorry!  
I haven’t had much time over the past week to work on this at home or do any experimenting. I’ll likely be able to set aside some time again after the weekend.

By the way, I also reached out to EcoFlow and received the following response: “Thank you for your patience. I am Neon from the EcoFlow Technical Team. I have compiled your product requests—specifically the addition of a ‘Standby’ mode and support for the STREAM AC5000 on the Homey Pro platform—and forwarded them to our R&D department. I will notify you as soon as I receive an update from them.”

I really hope that EcoFlow takes action to get this sorted out and that they make it easy for app developers to implement.

[Previous page](https://community.homey.app/t/app-pro-ecoflow-stream-series-solar-battery-smart-meter-and-homey-energy-automation/157399.md?page=5)
