@iancalderbank

I forgot about your final requirement.

Slowly ramping up the LWT for DHW production is indeed a good idea.

The only suggestion that I would make is start the ramp up with the LWT at the same temperature, or slightly higher, than the DHW cyclinder temperature, otherwise the system could start by cooling the DHW rather than heating.

Posted by: @derek-m

↑

If the IAT is 0.5C below the desired temperature setting, set the required LWT at the value required to run the heat pump at constant minimum output.

LWT that runs the HP at a constant min output, balanced with emitter output, that will be I think about 38-40C for my system. which will obviously force cycling to "no demand" state as the IAT will go over target reasonably quickly. whereas with the lower LWT, the IAT stays almost unmoving , close to target, and we cycle on the HP getting too high an LWT. obviously with the higher LWT the cop whilst running will be worse but does that balance it with having the "off" periods ... an interesting experiment.

that isn't the same as what I proposed though. I proposed keeping the LWT close to the value that matches the heat demand of the house and increasing/decreasing it a small amount to try to lengthen the LWT-went-too-high-for-the-HP based cycles. does that approach seem worth trying to you?

Posted by: @derek-m

↑

The only suggestion that I would make is start the ramp up with the LWT at the same temperature, or slightly higher, than the DHW cyclinder temperature, otherwise the system could start by cooling the DHW rather than heating.

yes of course that makes total sense. will probably automate it to not run unless cylinder is below 25C, then start at an LWT of 30 and work up from there.

Posted by: @iancalderbank

↑

Posted by: @derek-m

↑

If the IAT is 0.5C below the desired temperature setting, set the required LWT at the value required to run the heat pump at constant minimum output.

LWT that runs the HP at a constant min output, balanced with emitter output, that will be I think about 38-40C for my system. which will obviously force cycling to "no demand" state as the IAT will go over target reasonably quickly. whereas with the lower LWT, the IAT stays almost unmoving , close to target, and we cycle on the HP getting too high an LWT. obviously with the higher LWT the cop whilst running will be worse but does that balance it with having the "off" periods ... an interesting experiment.

that isn't the same as what I proposed though. I proposed keeping the LWT close to the value that matches the heat demand of the house and increasing/decreasing it a small amount to try to lengthen the LWT-went-too-high-for-the-HP based cycles. does that approach seem worth trying to you?

 

It is difficult to provide definitive suggestions, since I cannot watch your system operate and see how it responds under different operating conditions, which is the reason why my suggestion was rather vague.

Having said that we have the same thought.

Run the heat pump with the LWT as low as possible, but high enough to stop it cycling. Operate in this way until the IAT has increased to the upper limit, then cause the heat pump to stop operation by lowering the LWT setting.

Allow the IAT to gradually fall over a period of time until it reaches the lower limit, then raise the required LWT to the value calculated above to cause the heat pump to restart.

I suspect that operating the heat pump at a slightly higher LWT may reduce the efficiency slightly, but since this is during milder weather conditions (hence the cycling) the overall affect will be minimal, and may even turn out to be more efficient since the heat pump will not be operating for longer periods of time.

A further thing to consider, and maybe experiment with, is do you also stop the water pump or keep it running?

Hi guys,
I have a Samsung heat pump, split type (part no. AE160JXEDGH/EU and AE160JNYDGH/EU) and in my search to find a solution to integrate it in Home Assistant I found the posts of @william1066 and @tomc in which I saw that they were successful and I decided to order the MIM-B19N modbus module which I connected to the outdoor unit at terminals R1 and R2 because the indoor unit is already connected to F1 and F2. Not having the possibility to run another cable from the house to the outdoor unit, I followed @Steven's approach, the one with ESPHOME so I can connect via wifi. The problem is that in Home Assistant I see all the sensors, only they are unavailable, and in the ESPHOME log there are "Duplicate modbus command found" errors, as if there were 2 modbus modules sending the same commands. I am also attaching the log, maybe one of you can point me in the right direction.

Cosmin

@cosmin can you please explain your cabling and physical setup more clearly?  Are you trying to re-use the F1/F2 cable from the indoor to the outdoor units to carry your own modbus?

@iancalderbank modbus module is connected to my outdoor unit, at R1 & R2 terminals. F1 & F2 are already used for communication between indoor and outdoor units.

Quick question: What is your observed frequency with which data read through modbus changes?

To me it seems that while I obviously can read temperature data every second or so, the same value is reported for 30 seconds and only then the value gets updated and gets reported for another 30 seconds. Do you observe the same behaviour?

Hi all, we have a 9KW Samsung AE090JXYDEH, also in Ireland so maybe a similar set up to the other poster where i have joule thermostats. We have UFH downstairs and radiators in the upper floors, maybe 6 zones.

Can this model be integrated with home assistant?

@cyrusir yes. worth reading some of the previous discussion on this thread but quick summary:

- install modbus MIM-19 module into  samsung outdoor unit

- home assistant PC with uart module , two core cable to the MIM 
- OR ESP8266/32 with uart module, cabled to the MIM. ESP connected to home assistant via ESPHOME.

@iancalderbank thank you, so either way i need a wire from the unit back to home assistant in some way?

@cyrusir I was thinking about this the other day. If you go the ESP route, you could position the ESP board wherever you can run a cable to that has power and WiFi/ethernet e.g. shed or garage