I had a drop 3 days ago. And none of the inverters lifetime values have reset. There is one inverter that is at 1197310 and crossed the magic number.
My envoy lifetime value dropped from 25910553 Wh to 24718000 Wh, a 1192553 Wh step down.
So indeed it’s not in the inverter, but somewhere in the Envoy.
Hi @catsmanac ,
Could this open the possebility to calculate the total lifetime value of the system with the sum of the livetime values of inverters in stead of taking ot from the Envoy?
Yes, If you add all inverter lifetime values in a group helper with sum method you will get the lifetime value as well. Or if you prefer templates add the states of each.
I was also thinking of that option for the Envoy integration itself.
I was also thinking of that option for the Envoy integration itself.
So the buggy lifetime value is replaced by the sum of inverters.
The value wil then be correct, and even my (complex) blueprint will not be necessary anymore …
Well the lifetime saga may still be ongoing
On march 11, all of my 24 inverter lifetime values all showed a step change of ~20 kWh down (total -414 kWh) and at the same time the Envoy lifetime value stepped up 778KWh.
As a result, my summed inverter lifetime and the Envoy lifetime values are equal again. Before they were offset by the 1.2MWh step down.
When comparing inverter lifetime to values shown in the Enphase portal, they now differ with the step down value of ~20kWh…
Interestingly, the 778 and 414 steps together are the 1.2 MWh magic number… As if the step down ‘repaired’ the earlier (but might come back again when thresholds are reached again)
At the time of this happening there was an error and a data gap of 6 minutes (envoy restart!?) and right after it the warning that ever increasing values went down.
2026-03-11 01:30:31.831 ERROR (MainThread) [homeassistant.components.enphase_envoy.coordinator] Error fetching Envoy 122302045041 data: Error communicating with Envoy API on 192.168.3.112: HTTP status error https://192.168.3.112/ivp/pdm/device_data 502
2026-03-11 01:40:10.246 WARNING (Recorder) [homeassistant.components.sensor.recorder] Entity sensor.envoy_no_lifetime_productie from integration group has state class total_increasing, but its state is not strictly increasing. Triggered by state 12033.17 (12217.405) with last_updated set to 2026-03-11T00:36:57.778234+00:00. Please create a bug report at https://github.com/home-assistant/core/issues?q=is%3Aopen+is%3Aissue+label%3A%22integration%3A+group%22
2026-03-11 01:40:10.246 WARNING (Recorder) [homeassistant.components.sensor.recorder] Entity sensor.envoy_lifetime_productie from integration group has state class total_increasing, but its state is not strictly increasing. Triggered by state 25650.383 (26065.089) with last_updated set to 2026-03-11T00:36:57.813779+00:00. Please create a bug report at https://github.com/home-assistant/core/issues?q=is%3Aopen+is%3Aissue+label%3A%22integration%3A+group%22
Anyone with similar observations?
1 inverter and envoy lifetime step change:
All inverters step changing:
The step changes in the sum of inverters (-418) and envoy lifetime (+778)
Maybe a firmware change?
I will check my inverter totals, and let you now.
Firmware didn’t change, 8.3.5268 since Feb 4, 9pm, when it changed from 8.3.5167.
There’s also issue 165351 reported which may be the same thing happening. Similar 502 outage in /ivp/pdm/device_data and step changes afterwards. Some hours later as mine.
Weird, behavior here as well:
Drop by 1.4 MWh.
FW: 8.3.5286
Dropped yesterday, but my inverters are no-where near the max number.
joulesProduced": 2643335775
joulesProduced": 2669799150
joulesProduced": 2658092850
joulesProduced": 2646070200
joulesProduced": 2677694175
joulesProduced": 2672346600
joulesProduced": 2679144075
joulesProduced": 2674440450
joulesProduced": 2664396450
joulesProduced": 2678846400
joulesProduced": 2674494000
joulesProduced": 2685932775
joulesProduced": 2671938675
joulesProduced": 2681880975
joulesProduced": 2681094375
joulesProduced": 2664816075
joulesProduced": 2681562150
joulesProduced": 2657138175
joulesProduced": 2672709975
joulesProduced": 2685364875
joulesProduced": 2681519625
joulesProduced": 2670228900
joulesProduced": 2666667375
joulesProduced": 2651329575
joulesProduced": 2659998825
Enphase app still shows the correct number though.
Can’t explain why it started showing “gaps” in its measurements.
Null measurements seem to be nightly:
Not sure if firmware was updated recently. Any way to check?
Enphase app shows a GW reboot, 11:25PM last night. That’s when the drop occurred.
Gaps shortly after 11 pm are caused by Envoy “internal cleanup”. It does some cleaning or cycling of tasks and is not responding for 2-3 minutes. Depending on actual time this happens and when HA sends the requests, this may result in a timeout or not.
In December we bumped the the timeout up to deal with this. But on my Envoy that has FW 8.3.5286 since early February, it became worse with that version. I now have 3 out of 4 days these brief outages.
Again working on a fix to increase timeout gap between 11:00 PM and 11:20 PM to handle this better.
Found a bug in the blueprint:
Inline:
{% set new_overflows = -(delta / drop_amount) | round(0, 'ceil') if (delta < 0 or delta >= drop_amount) else 0 %}
… the round(x,'ceil') was used, taking the higher integer value of x.
For negative values this is wrong. Neede is the next integer value ‘away from zero’.
So the new lines are:
{% set value = -(delta / drop_amount) %}
{% set new_overflows = (value | abs | round(0, 'ceil')) * ((1 if (value | float) >= 0 else -1) ) if (delta < 0 or delta >= drop_amount) else 0 %}
fixing this in the gist now !
I found this thread whilst googling for similar issues to what I’m seeing.
It looks like my envoy has an issue that is getting worse over time. Has anyone else seen similar?
I understand that exceeding the capability of a variable would cause the value to roll-over, but that doesn’t seem to be what’s happening here - it doesn’t look like that, and I don’t understand why the steps appear to have become smaller and more frequent over time. This doesn’t look like a simple over-flow.
Any ideas?
Luke
Also experiencing the same issue: occasional jumps of up to 14MWh per day and sometimes negative numbers. Did anyone find a fix?
To make it even weider: find below the graph of my “Envoy lifetime production” sensor. This is probably caused by the frequent corrections that I had to make in order to keep my energy dashboard tidy (via developer tools → statistics-> outliers)
The step in 2025 may be unrelated to the steps in 2026. Was it same firmware then and now?
The 2026 pattern resembles what the blueprint is trying to handle. Check the size of the individual steps in 2026 (see values below). I’ve now had multiple steps, each time an inverter crossed the 1.19…MWh lifetime value.
The value roll-over of some internal value is an assumption based on the value. Could be coincidence of the values, we don’t know for sure. I’ve reported it to the (Dutch) Enphase support with multiple screenshots, they replied they would hand it to technical support…
This is my lifetime since Jan1, 2025, steps downs started in march of this 2026.
Half om my panels are facing south-west, other half north-east and all installed at the same time. The south-west ones are after 3 year of production now one-by-one reaching the 1.19 MWh mark.
April showed 7 steps:
This weeks statistics corrections (see the matching values):
Think the statistics correction updates the sum field used by the energy dashboard in the statistics table and not the actual last value. I would expect to show actual value as it was received, see my screenshot in my previous reply. I did multiple corrections as well and it only shows correct in the energy dashboard, not in actual history.
I experienced a step up one time also (screenshot in previous reply) at the same time the lifetime production of the inverters stepped down. Do you see something similar?
Interesting. I didn’t have individual inverter lifetime energy production enabled for my 12 inverters, but I’ve enabled them today. All 12 are well over 1.19 MWh.
The most recent step down was just over two weeks ago, yet the inverter with the lowest energy production is currently at 2.4 MWh.
Looking back, it looks like all 12 of my inverters have reset/overflowed. I’ve found all 12 steps. Does this mean I’m good for another 9 years or so?
Well, if firmware doesn’t change, maybe only 4.5 year. If your lowest inverter lifetime value is 2.4MWh it means this time it was the second series of reset/overflow at 1.19 MWh that you just observed.
And if firmware changes, who knows… Don’t think I’m gonna bet on this one…
The most recent step down was just over two weeks ago, yet the inverter with the lowest energy production is currently at 2.4 MWh
The reported inverter lifetime does not reset/overflow. So its probably some internal counter that has the issue.
Of course! This is the second time!
I may not have had HA running approx. 4.5 years ago.
We’ll see…
Thanks,
Luke
