Inaccurate consumption measurements since API change in March
Hi All,
Since the API change in end of March (~March 26), I'm getting very inaccurate consumption values for my NLP type devices (Legrand Netatmo 412015 CX3 energy meters). For NLPM (Legrand Netatmo 064840 - Mobile socket) it seems to be fine.
I've been waiting for a fix, but it's still broken for months... I'm getting huge consumption spikes, which clearly indicate some calculation error. I'm recording my power usage as well from the same devices, those values seem to be fine. It must be some problem with consumption calculation:
I'm getting random 200-400 kWh consumption spikes for almost every day at some point...
There's no way I could consume 400kWh in 5 minutes...
Firmware versions for Gateway and NLP:
{
"id": "00:04:74:3a:39:44",
"type": "NLG",
"offload": false,
"firmware_revision": 702,
"last_seen": 1717413782,
"wifi_strength": 54,
"reachable": true,
"use_pincode": false
},
{
"id": "00:04:74:00:00:15:d8:60",
"type": "NLP",
"on": true,
"offload": false,
"firmware_revision": 83,
"last_seen": 1717414193,
"power": 2,
"reachable": true,
"bridge": "00:04:74:3a:39:44"
},
API response (30 min resolution for today):
{
"body": [
{
"beg_time": 1717366500,
"step_time": 1800,
"value": [
[
100
],
[
55
],
[
406749
],
[
28
],
[
37
],
[
46
],
[
23
],
[
37
],
[
45
],
[
24
],
[
35
],
[
44
],
[
24
],
[
34
],
[
44
],
[
79
],
[
39
],
[
42
],
[
41
],
[
35
],
[
35
],
[
57
],
[
107
],
[
41
],
[
33
],
[
45
]
]
}
],
"status": "ok",
"time_exec": 0.03305816650390625,
"time_server": 1717414552
}
Why am I getting 406749 [Wh] for 30 min?
I have not changed any setup of my devices and before the API change everything was accurate.
Please advise!
Thanks and Regards,
Gergely
Comments
7 comments
Hello Gergely,
Indeed, it's a know problem on our side. I added your post to the opened tickets about this so that they can have usecases to investigate. For now, they don't have an immediate solution to bring
Have a good day,
Leslie - Community Manager
Hi Leslie,
It's good to know that it is still a known issue and it's not due to a problem with my equipment.
On the other hand, it's quite puzzling why it could not be solved for months, or at least making the previous, correctly working API available.
Anyway, thank you for the update!
Regards,
Gergely
Hi Leslie,
I noticed there was a firmware upgrade from
to
for all device types.
Is there a possibility that this revision fixes the issues seen previously?
Thank you :)
Hello Gergely,
Unfortunately no, teams are still investigating by currently doing some tests with a beta gateway version. It seems to impact a very few users, so it's a bit hard for them to fully understand the issue
Have a good day,
Leslie - Community Manager
I see, thanks for the update.
Let me know in case I can provide any more information or help :)
Seems like only NLPC devices are affected, measurements from NLPM and NLP remained accurate. Also, power measurements seem to be accurate for all types.
Hi again,
Indeed we already have 3/4 well-documented cases and it only impacts NLPC measures. Hoping they will quickly find the problem and fix it !
Have a good day,
Leslie - Community Manager
Hi,
Looks like the issue is not present any more, having these versions:
Gateway:
Line meter:
Have not observed any inaccurate spikes for few weeks now.
Thanks! :)
Regards,
Gergely
Please sign in to leave a comment.