Here’s the basic YAML, I tested it but haven’t put a trigger/condition on it yet; I was planning to use it to adjust the offset before turning the pump back on to reduce the initial power load:
alias: Reset ASHP Offset to 0
description: “”
trigger:
condition:
action:
You can test it manually.
I can’t see it mentioned in this thread, but mine isn’t connected to the X10A connector of the main unit, I’ve connected it to the X12A connector in the wiring centre in the house.
My entry in the “Show your ESPAltherma!” post:
https://github.com/raomin/ESPAltherma/issues/17#issuecomment-1256862726
I do apologise at hijacking this thread and the newbie question.
This is the closest help I’ve come to understanding how to get some good monitoring done on my heat pump.
I have a Daikin altherma HT 3 16kw heap pump. This has the outdoor and indoor unit. There is a separate Daikin DHW tank too.
As my unit is fairly modern I think it can communicate directly with some types of SW and HW to send all the required info vs getting external temp sensors and heat monitors.
The dumb question I have, what do I need?
Once set up I’d love to share my stats with your all. I’m in a large Victorian uninsulated house with Radiators as the only heat emitters. 250sqm plus. As you can probably imagine, the HP loves to eat up electricity, averages around 60 to 80 kWh per day in winter. (Very greedy HP!)
I am having some defrost issues, Daikin coming to check refrigerant levels. It’s set to 60o fixed flow temp, essentially on 24/7 now. Bit of a story behind WC vs fixed. Probably could do with the odd radiator upgrade but I’d like to get some decent monitoring set up first before I go charging off into changes.
Really appreciate a link to a how to to set up my HP on the relevant SW.
I also know my HP has the LAN adaptor. Ideally, is there a way to use this to get all the data but sent to another App vs the Daikin App on my phone?
Thanks,
Terry
Hi @Haggistrap Terry, if you want to DIY and save some money - 1st of all I would start with Home Assistant, spin up a small server VM, old PC, Home Assistant appliance (Yellow/Green) or Raspberry Pi, then check out some of the instruction within this thread…
or
with a bit of plumbing and electrical work - buy and install one of the following emonHP -
Many thanks. Didn’t know there was a wee box with HA already on it.
My biggest unknown currently. When I have HA on a server can I simply point it to my Daikin wireless LAN IP and it starts connecting and pulling data. Or do I need another device to do this. I’m thinking as I have wireless adapter already I don’t need that?
Think this one may help me, seems to use the wireless lan adaptor
you can use the wifi card and this integration (fork), jwillemsen/daikin_residential_altherma
you can ‘control’ the HP via HA just like the Onecta app, but you cannot calculate the COP as Daikin do not expose enough information and you need the ESPAltherma GitHub - raomin/ESPAltherma: Monitor your Daikin Altherma / ROTEX heat pump with ESP32 (ESP32) directly connected to the Monoblock.
Check out post no. 6 within this thread
Hi,
Would you be kind enough to detail how you setup ESPAltherma to send data straight into EmonCMS please? I really don’t want to set up HA if I don’t need to.
Thanks.
Sure. Assuming you have configured the PlatformIO or equivalent facility for recompiling/loading the software… the outline of actions to take is
You should see these new data items appearing in EmonCMS Input screen (Setup/Inputs). Now you have to configure them into Feeds, along with an appropriate source of energy consumption information e.g. EmonTX + CT clamp on HP feed. Calculating the heat energy produced requires some transformations of the data - screenshot shows my entries for the appropriate flow temp thermistor reading, R2T in my case.
That should get you going. Note also
Bear in mind that you should
Finally, if you want to use emoncms.org, you will have to
0 * * * * php /opt/emoncms/modules/sync/sync_upload.php
To edit crontab you’ll need to connect via SSH then run ‘crontab -e’
Thanks for the detail on this. I went with the stick inside the Daikin unit and then had to move one of my Orbi’s in to my garage, even with line of sight from my corridor (through glass) it wouldn’t connect. My electric still needs a bit of tweaking so will show some errors whenever the Eddi drawing energy, my plan is to add another CT and offset the electricity going through the immersion on my heat pump feed.
A couple of annoying things happened which meant it took longer than it should have - how typical!
It’s worth nothing that in the template file that’s been supplied, I put checks in whether the feeds where available - I am running HA & MQTT in docker containers. I’m not sure whether that causes issues on startup for missing conversions - i.e. from undefined to a float:
availability: >
{{ state_attr('sensor.althermasensors', 'Pressure') is number }}
Also, in the configuration file this has the _2, whereas the template file doesn’t:
- binary_sensor.3way_valve_on_dhw_off_space
Hi John,
Many thanks for the details - I have now got about 20 new inputs in EmonCMS to look over!
It took me a while to figure out why it wasn’t working initially - the dupont connectors to the esp board really are naff! Raomin’s suggestion of using headers is a good plan.
Thanks again,
Mark.
Has anybody got MID monitoring alongside the ESPAltherma.
I’d be interested in knowing how far off the figures are? Surely, that offset would be similar for most models?
That would be marvellous, but… We’d need to get over to @matt-drummer Matt’s and install it for him alongside his MID setup 
, I’ve been suggesting he should do it for months .
However, it wouldn’t necessarily follow that a single comparison would add a lot - it’s quite clear to me that the performance of the thermistors varies enormously, within specific heat pumps and across all instances. We’d probably find that the flow meter is cr*p as well! IMNSHO, here’s no way you can really avoid having to check the relative performance of your individual thermistors.
I need to get Daikin on the phone to confirm/deny the various sources of data for the MMI etc. and their experience of thermistor performance. Maybe mine are particularly rubbish…
I was pondering if the figures are wildly out wouldn’t the HP not work to a certain degree?
Unless, the figures exposed aren’t used for controlling the running of the system.
I can’t really justify to the OH installing MID for however much it would cost. I’m happy with the ESPAltherma monitoring if I can factor in a % in my mind for how much it’s wrong.
This is why I want to ask Daikin about how the readings are used. The MMI calculates Energy used/produced, and to do the latter it must be using dT (Flow-Return) and Flow rate. From my observations, the LWT is R2T (on a BUH-equipped Altherma) i.e after BUH as you’d hope, but CoP is calculated from R1T (before BUH), which is bonkers. However, unless the R2T and R4T (Return) are wildly out and thus dT is miles off, the actual pump would work whatever, because it’s the effect on the house you’re looking for, and it doesn’t matter that you label that. And they don’t really care about CoP unless you pin them down on something they pay for!
I have the P1P2Alterrma and the OEM l3 kit, I will look into setting up a second app with the daikin data later on and will provide links for both.
What values are you guys using for the power and the heat Output (and how have you configured the inputs / feeds) in the app so I can be sure to use the correct values.
I’m using the ESPAltherma package, which interrogates a serial port on the main board. The P1P2 interface connects to the control bus for MMI/Madoka, and although the data is presumably the same (!), the names associated with them can be very different, and I assume non-numeric Boolean stuff is dependent on the software developer i.e. 0/1 vs. On/Off or whatever. However, the ones I’m using for my EDLA09D3V3 with BackUp Heater (BUH) and ESPAltherma are (first 3/4 most important):
I’m sure that the OEM setup is much the same in general for the Power/Energy Output input, feeds and calculations. I was originally tracking Before BUH R1T, hence the two entries, but I’m only using R2T now.
Above shows the Input processing and resulting feeds. For R2T the steps are
Hope that helps.
I did my own “calibration” today, following the procedure described by @John here:
I measured
I then took the cover off the heat pump and using a two probe thermometer I measured the temperature before and after the backup heater. It’s helpful that they left the pipes uninsulated… 
I powered up the heat pump again and let the pump run to circulate water again. I measured no difference before and after the backup heater - apart from the odd 0.1C wobble in the two measurements. This makes sense to me, the backup heater looks well insulated and I’d given the system almost an hour of pumping around water with no significant heat source.
The temperature probes measured closest to R1T, so I’ve used that as my baseline sensor.
I’m now using the following correction values:
and have applied these as offsets in Home Assistant.
Hi, I recently added ESPAltherma to my MID monitoring setup - with help from this very thread!
I wanted to get the outdoor and indoor temperatures into EmonCMS and also see how accurate Daikin’s own values were. I have the EDLA04.
I run two instances of the My Heatpump app - one for the MID meter (a Sontex Superstatic 749), one for the ESPAltherma. Both are using a SDM230 MID modbus electricity meter to provide electric input data. There are issues about using the Daikin values so since it’s a £35 bit of kit, I don’t think that’s too expensive for those wanting detailed statistics.
So far, they seem very close. My daily COP with the ESP is generally 0.07 lower than with the MID sometimes even less. Total heat output for a day is reported as about 1.5% less with the ESP.
If this trend continues, I would say it’s not worth the outlay for a MID heat meter on a Daikin.
I’m struggling to find out what Class 2 accuracy means in % terms (maybe 2%?).
The extract below from this file suggests it’s only the flow meter element that is certified?
directive defines accuracy of instruments in terms of Class 1, 2 or 3,
with Class 1 being the most accurate. The pair of temperature sensors
does not have an accuracy class, and neither does the calculator, so it is
only the flow sensor which has the associated accuracy. The allowable
Maximum Permissible Error (MPE) for temperature sensors is the same
no matter what the accuracy class of the complete assembly is.```@MBC It’s great that you have done that, and that the results are fairly close. I’m sure lots of us are very grateful for your efforts - I certainly am!
I’m interested in how close the absolute temperatures are in ESP vs MID setups, and the relationship of the ESP temps if you have R1T and R2T (before/after BUH).
Happy to hear the comparison has been useful - especially considering the help you gave me in setting up the ESPAltherma!
I don’t have a BUH so can’t help there.
Obviously the MID meters themselves aren’t 100% accurate but like I say - the differences I’m seeing suggest the Daikin’s own values are very good.
Here’s a chart comparing temperatures. These are returned values - I don’t apply any offsets.
