Hi,
I’ve a Ecodan Heat pump system with mix of UFH and rads, mainly UFH, working on one stat and running it all open. I’ve got Mel Pump and love the data it gives. I’m going to give Auto Adaptive a go, and wondered if anyone had any view on this v Havenwise?
Dave
Hi. I monitored what havenwise got up to with DHW only. At the time I used HW it would trigger the fast DHW Boost to heat DHW. There’s no option for eco or normal mode. As I have a cascade system, data was very limited via HW app.
I have my system data transmitted via CN105 via MQTT to home assistant. Overall HW worked great tracking Agile prices and treated DHW as a battery. Storing heat when tarrif was exspensive.
I can see HW being a excellent addition if people just want plug and play. I’ve spoken to the chaps at HW and have given my feedback to the team.
For me, it was the lack of data that didn’t sit comfortably . How would I know what HW was up to without home assistant.
Sorry going a little off topic. Other options if you like to tinker and have solar and battery.
I have a solaredge system. I used predbat to control the battery’s status automatically. Having a cascade system working with Predbat has been great. It’s well worth the effort to install. Give it ago if you don’t mind messing about with home assistant. I’ve found Claude Ai work best with Yaml code.
Predbat has learnt the extra loads of the ASHP and keeps battery available for the exspensive peak over the morning. Then solar takes over. Predbat calculates every 10min for the best cost/return.
In home assistant I can monitor where the power is coming from for the ASHP. Also cost.
I’m in the process of adding my 22.4kw mitsubishi cascade system to the open energy monitor league table. I’ll add the combined system data once I’ve worked out the best way to go about it.
I would use havenwise again. Yes.
As I’ll be logging my system for a few years on here I’ll be able to compare with or without havenwise if I decide to use it again.
Good luck
Alan
How do user want the system to anticipate for dynamic energy prices? Mitsi auto adaptive works fine, it just needs one of their thermostat to get it working.
(I am interested since I’m building my own auto adaptive, and currently support setpoint bias to anticipate for this. It can also be used to anticipate for weather. esphome-ecodan-hp/docs/auto-adaptive.md at main · gekkekoe/esphome-ecodan-hp · GitHub)
Has anyone compared how accurate the internal Mitsubishi heat meter is compared to an external meter?
I understand that internal electric meter isn’t the most accurate.
Hi.
I have two utility meters for each ASHP. Optical Utility Meter LED Pulse Sensor - Shop | OpenEnergyMonitor
I’m sending data via esp32 to home assistant.
Power from the CN105 control board does track closely to my live feed. CN105 is normally a little lower. With local data feed on ecomcms of 60s you seen a higher resolution, Emoncms online is 5 min refresh so power looks smoother.
First pic is live. Power spikes was me messing with sensors.
It’s not the electric meter accuracy I’m interested in
It’s the accuracy of the heat meter within the Ecodan compared to an external heat meter for example an Axioma. If anyone has a CN105 to MQTT feeding into Home Assistant but also has an external heat meter feeding heatpumpmonitor?
yes I have a certified mid energy meter on the heatpump and compared with
the reported numbers by the ecodan, it can be pretty good, but sometimes it can differ up to 10%.
yellow: reporting numbers from ecodan (not available via melcloud, but via my esp)
blue: certified meter. You can see it’s pretty close. (I probably re-flashed a couple of times this morning, counter probably got reset, that’s why its off in this case, but normally its really close)
If you don’t have a pulse wired to the ecodan, then it estimates the consumption. It does a pretty decent job.
If you wire your energy meter (pulse) then the consumption of the heatpump is very accurate.
