Many heatpump owners disable legionella completely.
A monthly cycle is probably a reasonable compromise if you’re still concerned about it. For more information, see Hot Water Temperature - Scalding and Legionella - HeatGeek
Many heatpump owners disable legionella completely.
A monthly cycle is probably a reasonable compromise if you’re still concerned about it. For more information, see Hot Water Temperature - Scalding and Legionella - HeatGeek
Where is the temperature sensor placed within the tank? How much kWh of heat was put back into the tank afterwards?
Have you considered increasing the reheat threshold to 10° or higher? COP will be higher when starting with a colder tank.
Thanks, I will have a look. Do you really think once per month is enough ? I’m not overaly concerned but I also don’t know what to think about this legionella in general. Never thought about it, never thought on my DHW temperature (electric boiler before). Scalding is not an issue here, I have thermostatic valves almost everywhere an d with e-boiler I had 70C set and never a problem also with small kids. But 70C is not efficient of course for HP. I would run as low as possible without minimizing risks of leg.
Sensor is place right in the middle of the tank. The coils is running from the very top to very bottom with 2.9m2 surface area.
This is what I’ve put into the tank and then antifreeze protection took it out… So around 1.5kW of electricity. Still could be the metering is not precise since this is using samsung’s data which might be not precise and there is some delay. So in reality it can be either better or even worse
Threshold set to 10C is actually a good idea. I was thinking about it but in that case reheat will be triggered only when tak temperature is below 40C in the middle where sensor is placed and that means below at the bottom there is probably 35C which is ideal for legionella
I’ve ordered cheap 3KW immersion heater so place to put into a tank and do disinfection once per week to 70C. Not sure if I can set with this pump intervals to once per two weeks or a month.
Re Legionella Risk.
There are two types of electric showers, one just uses the mains pressure if > around 2bar, so no risk as minimal residence time, ~ 4sec in the 250ml heating can at a minimum flowrate of 3.5LPM. There are also pumped types with a integral pump fed from the CWSC which can often rise to 20C or more in the summer, most people shower at 36C to say 43C, I’ve never read of anyone getting legionella from one of these. The shower itself can’t even be sterelized as the TCO trips the heating elements at 48C to prevent scalding.
Personally, I don’t bother at all; heating my tank from 20° to 45° works fine for my household.
With stratification, I’d expect there to be a cold layer below the hot water, so only a small volume at the 35° temperature, and it won’t stay there for long.
That’s right. Maybe I’m overthinking. I guess I will set it to 48C with reheat when temp drops by 8C. so it starts again at 40C. I probably disable legionella cycle or set it to 55C once a week without immersion heater.
3 posts were split to a new topic: Samsung HT Error Code E912
Hi guys, does anyone knows why heatpup started to heat water when temperature was 44C ??
In field settings I have START temperature set to 8C (STOP 0C), DHW temperature I have set to 48C so I would expect it would start to reheat when temperature drops to 40C. During night I had DHW turned off and it was turned on at 6:00 by automation but why the system start to reheat the water when dT was only 4C ??
@SarahH any idea ? thx.
My Mitsubishi would always start a DHW cycle as soon as DHW was turned back on, regardless of the current temperature. So maybe that’s what the Samsung is doing?
Hi @Michal_S.
I think @Timbones is correct, if you still have your #3011 set to 2.
The Installers Guide says:
I interpret this latter to mean that DHW will start as soon as your programmer asks it to, irrespective of temperature.
Could be but I’ve expected it starts only when the water temperature is below a set offset.
Currently legionella cycle is disabled since I haven’t installed immersion heater yet. I have it at home but I would have to half empty the tank to install it I guess after a heating season I will do it because I need to fix some leaks on cooper piping (well was soldering for the first time so some mistakes have happened :))
But indeed it’s probably because I have #3011 set to 2 - Hysteresis.
So if I understand Sarah, if I want to start exactly when water temp drops below 40C then I have to set #3011 to 1 and START_TEMP (8). Then SET_TEMPERATURE (48) - START_TEMP (8) = 40C. So it does not matter when I enable DHW (I have it OFF during night since I don’t need hot water that time). Is that right ? It’s pretty confusing how the manual is written.
You might be confusing yourself, @Michal_S, calling #3011=2 Hysteresis. Hysteresis is actually set by #3023.
Correct, setting #3011 to 1 should do what I think you want. Then, your DHW won’t start until the tank temp falls to 40degC (in your example) even if the DHW programmer enables heating earlier than this.
Exactly, thanks. I will observe this night and let’s see how it works
Currently I’m working on implementation of some logic to use weather compensation together with load (room) compensation since I’ve found weather compensation itself is not enough. Especially Samsung’s weather law curve is pain in the ass. I wonder who integrated this but it’s dump. Also there are limits for temperatures so it’s unusable for me. 2 point control really…
I don’t know how you control the pump but I’ve found out if there is a cold wind for example heat loses are tad bigger so I need to increase heating curve a bit. Also on sunny days HP tends to overshoot easily so here I would need to decrease flow temp. At the moment I have only simple logic based on dT difference (target indoor temp and actual temperature) but since there are fixed values it’s not good enough. Need to do something more intelligent which monitors if room temperature is steady, increasing or decreasing over a time and modify flow temps accordingly. The only issue is that there is quite big delay after the changes can be registered so it’s quite hard to implement some good logic.
Would Homely solve it?
Probably, but I would have to buy it and I already have modbus installed. Basically I would need to replicate their algorithms to achieve the same but no idea what Homely is doing exactly.
They feed internal temperature, external temperature, flow temperature and weather forecast into a machine learning system that predicts how the internal temperature will change. Then choose the flow temperature that gives the lowest running costs as the required level of comfort.
The machine learning must clearly be trained for the given home as well as with data from other homes. Eg with 30 years programming experience I hope I will not need to program it myself when we get a heatpump!
Yeah, these are exactly the input parameters I already have. Machine learning you say… That seems a bit overengineered to me. Thing is for my home I roughly now what temperatures I need at given outside temperature. But sometime they need some tweak. When I was thinking about it might not be a rocket science. Weather forecast seems to be redundant in this computation.
What I’m thinking is following.
Also for my system running lower than 38C is pretty inefficient because or radiators I have. Often especially in warmer weather I can run with higher flow temps and have better COP.
Of course for you if you don’t have modbus yet is maybe easier to just buy homely and save some hassles
If you know the heat loss coefficient of the property, you can easily calculate how much power is needed for a given indoor temperature. And if you know how much power your emitters will output at dT 50, you can work out what flow temperature they need to be at for the heat currently needed. This is essentially what the weather compensation curve does. I wrote about it here:
Room influence can be added on top of weather compensation, to add a little more heat when the house temperature is below target, or less when it’s over target.
Thanks, this is exactly what I want to achieve. Weather compensation itself is not sufficient because you can set it only linear in Samsung controller. But depending of outside weather - cold strong wind, sun shining, etc. the pure weather compensation can be completely off. So now I’ve implemented some logic which takes into account indoor temperature. If it’s not approaching target temperature I’m adding a offset to water temperature until is not stabilized. I’m just testing if it works as intended.
Definitely I will check your article maybe I can improve some things
BTW, I admire you UK guys how you can have 19C indoor temp I would get frozen lol. Here in Slovakia we are used to 22 - 23C at least
But I have a family in UK and every time we get there we are freezing lol
Room influence can be as simple as adding double the difference between room temperature and target temperature to the weather compensation calculated temperature.
But risk of overshoot if the thermal mass of radators is very high or UFH.
Systems like Homely advoid the need to know the heat loss coefficient of the property and how much power your emitters will output at dT 50, it should also be able to cope when a few rooms (eg bedrooms) have temperature limiting TRV so the emitter output is not fixed. So it is aim at being easyer to setup.