Smappee local integration - consumption and production entities reporting same value

Hi:
The smappee integration was auto discovered and has created the right 3 entities but I am getting the wrong value for the solar production. It appears to be returning the same value as the local consumption. Image shows results in a short period of time, The offset between the two is maybe 30 seonds so I assume that is just a query difference.

HA-smappee

This is a smappeeS in Canada
dual phase 180d + solar
Acqisition Software version 4.29
4 CT’s set as in image

I am a HA noob so was wondering if there is somewhere I need to give some configuration information to set the solar production to the appropriate inputs or CTs.

Thanks to anyone that can provide some assistance.

Hi Paul

In case the Smappee Solar is detected on the local network we provide some basic entities like grid and solar consumption. The first one is calculated as the sum of the CT inputs 1, 2 and 3, while the solar consumption is being taken from CT inputs 4, 5 and 6. Based on the configuration you’ve shared, that should work.

Would it be possible to share some results of the instantaneous api calls that are being done while seeing the live values on the config page (Google Chrome > inspect > network)?

Hi bsmappee
Thank you for your assistance.
I think this is what you requested, let me know if you would like to see more or if there is other data that would help.

Yes, thank you for sharing. That should be enough information to get us going. We’re seeing a dual phase grid and solar setup, which is quite rare in our region. We have started working on a patch to support this configuration as well in Home Assistant (keep in mind that the local option only communicates with the Smappee device and not directly with the cloud). Will keep you posted.

1 Like

Again Thanks for your help.
Let me know if I can assist in any way.

Yes aware of the local only. My thinking on local was if my internet goes down I can still run any automations locally. I have not developed any at this point but was thinking of things like if my solar production is greater than consumption then that would be the time to turn on the HRV full to become more efficient.
I also think the Home assistant integration will be useful to find things like peak power, which can provide insight into things like home battery design.

We have opened a pull request to review the code changes (https://github.com/home-assistant/core/pull/45699). If all goes well, the consumption and solar values should be correct within one of the next Home Assistant versions.

1 Like

Thanks, I will watch for it.
Also thanks for your support of home asistant.

I believe the code changes have made it to Home Assistant version 2021.2. Would it be possible for you to update your Home Assistant version and see if the values are correct now?

1 Like

Yes, all looks good.
It is tracking the AP systems consistently. A bit different but the smappee always reports less than the solar system for production consumption a bit different than the cities meter.

Thank you, it is much appreciated.
Paul

Slightly off topic… I’m interested on getting a Smappee, but I’m wondering that is it mandatory to have internet connection always on or can it be run with enough functionality locally? Does appliance detection require internet connection? What functionalities would be missing if ran without cloud connection? Thinking on only running via Home Assistant, so no need of app.

1 Like

Hi Sebu,

did u get an answer on that, did u invest in smappee?
tx

Yes, I actually ended up by buying Smappee Solar but just a week ago. I just installed few days ago and so far the appliance detection has found anything, but I guess that is still expected. So now I’m waiting for the detection to recognize appliances. I have currently the Smappee connected to the cloud/internet.

Based on the MQTT documentation I found, it seems that with local integration you can only access power consumption at phase level. The cloud connection would be needed for getting at appliance level.

tx Sebu

which devices did u get overall? Also some EV charging station ?
migube

The device is Smappee Solar. That device is no longer in current Smappee product range, but I happened to find one from Finnish webstore OmaVahti in sales. I didn’t want to choose the modular Smappee Genius as it would have become too expensive.

I don’t have an EV charging station. Also I dont have solar panels, so I would have chosen the Smappee Energy if it would have been available.

I’ve been now running Smappee Solar for about 3 weeks. The total electricity measurement seems to match quite well, but the appliance detection so far hasn’t been a hit. Over these 3 weeks it has found only 3 appliances: dishwasher, clothes dryer and washing machine. The hit rate on detecting the electricity usage of those is unusable poor. Also I would have expected it to detect at least the ground-source heat pump and the ventilator.

If I try to add manually it still indicates to wait few weeks. How many weeks should I still wait?

What is the experience people in the forum have with Smappee and appliance detection?

Hi
Sorry this is quite delayed.
I have been running the same system as you (smappee solar) for a number of years, since about 2017 with a complete reset and fresh install in 2019. I am on Canadian split phase power. Here are my observations.
I collect my HA energy data from the smappee local integration rather than the cloud.
I find mine reports about 5-6% lower than both my utility companies meter and my solar companies production metering. Cloud and local smappee agree with each other (of course they are a single measurement source).
For appliances I have given up long ago. In the older app I could train it to recognize appliances but never got it to be useful.
First the small use appliances (LED lights) could be trained with other things turned off but they always got masked by bigger use items. It didn’t have fine enough resolution to see the led lights coming on and off reliably.
In the winter I have electric heat so a 10w light coming on would be masked by the fluctuation smappee observed in a 1000 watt baseboard.
It always had trouble with the split phase, you could manually combine the 2 from each phase but they weren’t reliable enough and got confused.
It was not reliable in discerning sources and would report stoves 1500w elements as 1500w baseboard sometimes and visa versa. This of course makes sense because it is just measuring the resistive load which is the same.
In the end I gave up and strictly use it as a whole home measurement device.