I know this is a topic that has come up before, and as it turns colder so we begin the fight again.
The problem: When OAT falls below 5C, the Samsung controller runs an anti-freeze protection cycle which runs the pumps circulating warm water for 5 mins to prevent the system from freezing when the pumps have been idle for 60mins.
Normally, for most systems, this would never happen as the system should be running constantly when OATs are below 5C (or at least cycling on/off within a 60min window), but because our ASHP is so oversized for our property, we cannot run constantly until temps drop closer to 0C. This means that overnight the pumps will cycle for 5mins every hour which in our case āstealsā hot water from the DHW cylinder, transferring it into the 50L volumiser and 28mm primaries, and we wake up each morning with not so hot water. Note we also have glycol in the system and full home battery backup so itās extremely unlikely the system will freeze.
I know one other user here managed to trick the diverter valve into switching to the circuit rather than the DHW cylinder, but unfortunately that solution was not viable on our Joule pre-plumbed tank due to the way itās wired.
So Iām back to scratching my head and thinking about solutions. Last year I resorted to going out in the night and grabbing hold of the temp sensor in my warm hand to temporarily increase the reading to above 5C once per hour, but thatās not sustainable if I want a full nightās sleep!
Today Iāve had the case off and measured the resistance of the thermistor at different temps (with a pint glass of water and ice cubes). The thermistor is reasonably accurate, so swapping it out isnāt really going to help unless I can find an inaccurate sensor. The sensor is a 10k Ohm at 25C type, and typically reads 20K at 7C and 22K at 5C. The full table of resistance/temperature values are in the attached document.
So Iām considering just replacing the sensor with a fixed 20K resistor and let the heat pump think itās ~7C all year round, but Iām worried what other implications this may have:
Weather Compensation. Obviously the WC curve is based on OAT and would be rendered unless as the ASHP would see a constant 7C. This isnāt really an issue for us as we run at a flat 32C LWT all year round (as the lowest LWT the system can achieve) and manually bump it up a couple degrees on the controller if the weather gets very cold and the house drops below 20C.
Defrost cycles? I donāt know about defrost cycles? Does the controller use OAT at all in determining when to run defrost cycles, and would a constant OAT of 7C affect or prevent defrost cycles from running?
Anti-freeze protection: This is what we are trying to fix - prevent from running.
Anything else Iāve not considered that uses the OAT sensor value?
Iām hoping those with good technical knowledge of the Samsung unit will be able to help bounce around this idea with me to help find a solution.
This suggests that OAT is definitely a factor in defrost cycle initiation.
Personally Iād be wary of doing anything that might override defrost initiation - that L2 line (the usual defrost start condition) extends up indefinitely, and you donāt know when the COND_OUT sensor is seeing -6degC or below (unless you have monitoring equipment on it). So fooling the OAT sensor may save you lots of money most of the time, but wreck your HP if you do need a defrost but somehow prevent it happeningā¦
Iāve been think more about this recently. While Iām probably going to go down the same route as @Michal_S with the shelley relays, I thought about setting FSV #4013 to 7 and leaving the heat on. So that the heatpump comes on below 7 degrees, but, from the installers guide, it looks like the minimum value you can set for 4013 is 14. You can adjust FSVs with the Simple Setting menu, (in FSV). I might try and see if I can set it to 7 using this method later on.
Silly question, but what happens if one were to temporarily set FSV #3011 DHW Tank to Not Use?
If set to Not Use, would the controller logic then not be able to move the diverter valve to the DHW position, as presumably thereās no tank connected??
We have a spell of milder overnight temperatures, so may be a while before I can test, but next time we are due colder overnight temps, Iāll try setting DHW to Not Use (#3011) and see what effect that has.
It turns out I could set #4013 to 10 degrees on mine through the usual FSV menu, but I couldnāt set it any lower than this with the Simple Settings menu.
How this can prevent the pump opening your DHW circuit to AF/AS cycles ? I can set it below 14C with modbus but how this can help me to prevent stealing the heat ?
You can do it and the the heat wonāt be stollen because 3way valve never opens but how do you want to control the DHW ? You will not see the temperature of DHW tank. You would simply need to randomly enable it and check whatās the temperature but usually if you enable DHW and space heating is off it will automatically starts heat the tank. Itās not good workaround IMO. Relays works reliably but there could be issue if shelly will go offline or HA is not working properly. Then the relay wouldnāt switch to DHW heating in case the pump request it and you may end with 55C water in your space heating circuits. If you have radiators not a big issue but with floor heatingā¦. So itās good there is max. DHW running time so it wouldnāt be that big disaster but who knows But I had it like this for a year and no issue. DHW always heated properly. This would be pretty annoying outside heatind session because the mump must run at least 1 anti seize cycle per day. Well itās only 2C removed from the tank per day but anyway! I donāt want the pump to take heat from tankā¦
In winter when the pump is off at night from 00:00 to 6 a.m. and is bellow 5 it is doing 3 - 5 AF cycles almost every hour so in the morning I have 10C less in the tank
Hi @Michal_S, the theory was that it could be used to prevent the AF cycle altogether in the shoulder months by leaving the heating demand on but setting the maximum outdoor temperature to 7. Above 7 degrees is turn the compressor off, and that below 7 degrees the heat pump starts so no need to run the AF cycle. It means you heat the house when you may not want to, but below 7 degrees having a little bit of heating isnāt the end of the world. But it should stop it robbing from DHW though. I imagine you could alter the water outlet temperature in this scenario if you wanted to just run at 30 degrees so youāre not putting in too much unwanted heat into the property.
But this parameter is space heating off temperature. The default value is 14C. This means at the outdoor temperature above 14C the space heating will be turned off.
How this can prevent anti-seize cycles and stealing the heat from DHW ? By setting to 7C you only tell the pump do not start space heating if temperature is above 7C which makes no sense. If there is 8C outside your pump will not heat the space at all.
@Michal_S This is only for use in the shoulder months really, when you donāt want any space heating, but it may dip below 7 degrees at night. In winter you revert back to your normal settings and keep heating on all the time.
Ah, I see but still I guess itās not worth of it. How many times it will drop ? For this it would be better to make a simple automation in HA where you start space heating for a minute once per day to prevent anti seize cycle. Iām using a roomstat and I would need to manipulate target temperature in number because indoor temp is pretty high.
Have you confirmed this would work, or are you just assuming?
Iām not looking to automate a solution (for now), just trying to come up with a more manageable workaround for the relatively small number of days on which this occurs.
I would have to re-enable DHW manually each day. I would do that at the time Iām ready to reheat the DHW, normally in the early afternoon, so it wouldnāt matter to me that a DHW cycle begins immediately. Manually managing that for us is not an issue.
Iām not keen on the relay solution for the simple reason that cutting into wires in the control circuitry on my pre-plumbed Joule cylinder will invalidate my warranty, and Iām still in my warranty period, although I may end up going this route in the long term.
None of the workarounds we have at the moment are great. My current workaround is to simply turn everything off if we donāt want heating overnight and there is a chance the OAT may drop below 5C. This is OK in autumn where we definitely donāt need any heating overnight, but less so where we may need some heat, but would like a setback or off for over 1h. I could use my room thermostat and set a low hysteresis in the hope that the heating wouldnāt be off for more than 60mins at any one time, but thatās not going to help if we want an off period from midnight to 4am or 6am. If setting FSV #3011 DHW Tank to Not Use works, then that would allow me to at least have the heating available as required without losing DHW temperature should the heating remain off for more than 60mins.
Hi, yes. Iāve just tried it. I wrote 0 to 0x4097 address and DHW got deactivated. I canāt select it anymore on wired controller and is displayed in gray. Samsung stupid logic (not a first one) will show you DHW on screen even if itās disabledā¦. Why ? But thatās different topicā¦
What is interesting is that I can still read tank temperature in HA! That means I can fully control my DHW tank if I want to do that so I can program my own heating schedule while I can keep DHW deactivated when DHW is not use which would (hopefully) eliminated opening 3-way valve and stealing heat from the tank. Also seem the complete climate control for DHW is still in use, only the DHW not react because is disabled.
This solution is also not ideal (in case HA stuck or something) but maybe for people who donāt want to deal with relays itās better approach. You also remove one dependencies from the equation ā> shelly relay. Now Iām not sure if I have one or two relays to control it to be honest. I could give this solution a try.
Another idea I have to deactivate the DHW tank immediately when the 3-way valve starts to open without compressor running. This would indicate AF or AS cycle. I hope that as soon as I deactivate DHW tank the pump will switch 3-way valve back to heat spacing circuit and take heat from it. This would be almost perfect solution with as low dependencies as possible. But I can try it only when the temp will be stable below 5C.
Great that you can write the address to deactivate DHW, and still read the temperature probe.
So you have not actually confirmed the 3-way valve will not move to the DHW position if DHW use is disabled?
Weather conditions are such that I cannot physically test yet. I may have to rig something up so I can stick the OAT probe in some ice to simulate temps below 5C to confirm if it works, or not.
I do like your idea of writing some logic to detect the start of an AF or AS cycle, and take action to stop or mitigate. Using that logic, you could essentially create a āDisable AF/ASā function that users could configure as per their preference.
Not confirmed yet. I need to wait for temps below 5C overnight with space heating turned off. I will make an automation which disable DHW for the whole night and letās see if 3-way valve is not moved. But I hope so but canāt prove it yet. When I havenāt had a DHW tank (I bought after first year having HP) then there was no 3-way valve so canāt tell if terminals were still controlled like there is DHW connected but logically I would simply not operate them in case DHW is not connected but with Samsung one never knows so this has to be confirmed.
But logically in case there is no DHW (disabled) the pump need to take heat from space heating circuit so why would it want to switch 3-way valve.
So letās hope for the best and if this works then there is pretty easy to make automation which constantly check compressor and 3-way valve position. If it detect pump change 3-way valve position to DHW and compressor is not running we can clearly assume this is not going to be a regular water heating but AS/AF cycle. I have this logic to operate the relay and it is working perfectly. There can be a small delay of few seconds but this is not enough to steal the heat.
Then hope as soon as I deactivate DHW immediately 3-way valve start to switch back and it takes the heat from space heating circuit.
Not sure if something like this was ever tested at Samsung so guess nobody knows what happens but itās ridiculous we have to do these things.
Yes, I hope this will be the case. But itās not like disabled it only takes the water from space heating circuit and not DHW. Thatās the main goal
Iāve had issue with a similar scenario to allow for solar heat pump control. and when stealing from dhw the 3way indicator does not change
My conclusion is that go either with a relay or scheduled 3011 operation.
Also if yoyāre willing to have an autonomous automation for that, I could propose a pcb to drive the relay locally based on fixe scheduling (maybe with a bypass to generate dhw in a forced way)
For me (at least via modbus) Iām getting information about 3-way valve position - always. I use this for relay because I need to identify itās really an AF/AS cycle or real DHW heating. If the 3-way valve position wouldnāt be communicated I would not be able to do it since I use this automation:
alias: Turn OFF DHW Bypass in case of water heating
description: Turn OFF DHW Bypass in case compressor is running and DHW is required.
triggers:
- minutes: /5
trigger: time_pattern
- entity_id:
- sensor.ehs_compressor_status
- sensor.ehs_3_way_valve_position
- switch.shelly1_34945472d470
trigger: state
conditions: []
actions:
- choose:
- conditions:
- condition: state
entity_id: sensor.ehs_compressor_status
state: "1"
- condition: state
entity_id: sensor.ehs_3_way_valve_position
state: "1"
sequence:
- target:
entity_id: switch.shelly1_34945472d470
action: switch.turn_off
data: {}
default:
- target:
entity_id: switch.shelly1_34945472d470
action: switch.turn_on
data: {}
mode: single
So sensor.ehs_3_way_valve_position always returns itās actual position even in case of AF/AS but in that case the compressor is turned off.
So relay is turned on always but real DHW heating cycle. Then I will turn it off so the 3-way rally open DHW circuit.
@Michal_S I agree with you, I do have the information of the 3way valve during DHW cycling, but not during the daily cycling the DHW when the DHW is enabled but it has not started since 24 hours (thanks to solar DHW heating). I read the value from nasa message 0x409D. But if you have another I can read Iām willing to update that!
So itās good there is max. DHW running time so it wouldnāt be that big disaster but who knows 
