Would be interested in what settings you are using? Watching this thread to see if I can steal shamelessly:)
OK, back at a computer now 
Slowing the flow in my system from 5 to 3 in the FTC5 seems to have made things better. Slowing it to 1 caused it to cycle dramatically because it couldn’t get rid of the heat.
As you can see, this has only been changed whilst it’s warm (early March) so it’s not clear if this slower flow will make things better all year round.
slow-flow is the one at the top above all the others:
The pump has been running a smidge longer which isn’t surprising:
Along with a slower flow I’ve also disabled Force mode on the hot water. Now I flip down to 10 °C to stop it running and flip it higher to get it back on. However, I set it to go up to 50 °C too often and so it worked too hard sometimes. Here’s an example from today where pushing it on to 50 °C (because we had loads of solar) was a bad idea for efficiency - specifically you can see it was reasonable when it popped up from 10 °C to 40 °C (but even that was only 285% efficient):
Anecdotally, it’s not been more efficient running it without Force. The data shows the same thing. I have seen some occasional wonderful runs, but on average it’s pretty much the same as before. I do see it taking a bit longer, and sometimes a lot longer to get warm.
"CanSetForcedHotWater": true,
"EcoHotWater": true,
"ForcedHotWaterMode": false,
Here’s the outcome so far, it’s better than the device doing it’s “heat up when the water drops 15 °C” but not much better. The other confounding factor is the new situation of people taking showers at 22:00 so even if the day was warm the water-heating cycle will be when it’s cold. The x-axis is the average temperature for the day.
Now I’ve dropped the temp to 10 °C to mean “off” it has of course had a dramatic effect on the average target temperature:
In conclusion, I don’t think removing Force has made a dramatic difference for me. However, I’ve got it coded and it’s happy flip-flopping between 10 °C and something else so I’ll leave it doing that and see what happens.
1 Like
I’ve gone for a slightly different approach - I’ve set a schedule so that DHW is prohibited all day except between 1pm and 2pm. As a result, the cylinder is heated up once a day, unless I have excess solar in the morning when the DHW is forced to run earlier (possibly to a higher temperature). Less flexible than fiddling with target temp, but I think it will work well for me.
I figure that efficiency doesn’t matter when there’s excess of free power, and putting it into the heat pump will store more heat than an immersion element would.
I’m mainly interested in running the DHW as slowly as possible, using the minimum amount of power (watts). Hoping that ECO mode will do that, though I’ve also got “Quiet Mode” enabled (an FTC6 feature with little to no documentation) so it might be hard to see a difference.
Here’s a typical cycle from yesterday, before I applied the settings described above:
(power axis lies - 3 kW is actually 2.6 kW)
(cycle stopped prematurely because DHW was set with a maximum duration of 45 minutes)
2 Likes
You lucky thing, we use too much hot water to pull that off, even with the 300 L tank.
I just had a user complaint because the water wasn’t hot enough.
Chasing efficiency can go too far.
I’m going to enable the overnight boost again.
I love being able to run my algorithm and choosing the calculation moment so I can see how it would behave
bash test.sh --calculation-moment=2023-05-01T03:26:10
DEBUG: The power is off but will be more expensive in 60 minutes time and the tank is at 24.0 so aiming for a target of 54.0 °C to warm up water using cheaper electric
DEBUG: Turning on so we can heat the water (Power → true from ManageSpaceHeatingPower.manage)
DEBUG: We’re going to heat the tank to 54.0 °C because it’s at 24.0 °C which is lower than the trigger temperature of 39.0 °C and the flow is at 21.5 °C (SetTankWaterTemperature → 54.0 from ManageSpaceHeatingPower.manage)
Here’s something more tangible.
Effective outdoor temperature (approx): 15.0 °C
12:00 - 12:15
Forced (due to user complaint)
kWh: 0.9 => 2.1
CoP: 237%
Initial: 30 °C
Target: 38 °C
Achieved: 41 °C
12:50 - 13:15
Computer-initiated because the tank was cold
kWh: 1.2 => 4.1
CoP: 328%
Initial: 20 °C
Target: 38 °C
Achieved: 38.5 °C
13:35 - 14:00
Computer-initiated because solar was producing so much
kWh: 1.4 => 3.2
CoP: 221%
Initial: 38.5 °C
Target: 55 °C
Achieved: 50 °C
Conclusion: Heating gently from colder temperatures can give better CoP.
Conclusion: Heating to a lower target can give better CoP.
This is not news.
User comfort is still an important goal, if it wasn’t we wouldn’t need the heat pump.
3 Likes
I’ve had our home Ecodan 8.5kW running now for just over a week (Yey) and have been experimenting with the DHW and came across this thread.
We are on Octopus Flux and will have a solar diverter so my initial plan was to use the FTC6 to only turn on hot water during the night time off peak period. Any solar diversion to the tank would reduce the amount the heat pump would need to do the next night.
Unfortunately certain other members of the family take showers that are too long so some days it would appear the tank isn’t quite enough to see us through the day.
Another thing I discovered was that when the temperature probe is at the bottom of our tank as soon as one shower is had the temp has dropped significantly so if DHW is continuously on it would trigger with every use. I have tried moving the probe to the top of the tank and it only drops significantly in temperature when the tank is nearly depleted.
Therefore my next plan was to keep the temp probe at the top of the tank with DHW always on so the heat pump would kick into action if hot water was nearly depleted as a backup but have a 2am cron job to run a pymelcloud script to trigger Forced Hot Water.
However reading this thread I’m not sure it would be very efficient to do this.
So can someone think of a fairly simple way to achieve what I’m after and still be efficient, that is to have a main hot water run scheduled during cheap electricity at night but the back up of having an extra run if it looks like we are going to run out?
1 Like
I’m fortunate to have a thermal store heated by solar thermal and PV so DHW isn’t an issue for a significant portion of the year, especially now the three children have more or less gone for good, but over the Winter months I heated the AHSP DHW using my Eddi diverter on a scheduled basis to match my Octopus Go tariff rather than have the Ecodan ASHP start up.
Heating DHW isn’t terribly efficient in terms of COP and I figured using cheap rate electricity would be better (although I stand to be corrected!).
As with Nick, I suggest it’s best not to worry too much about this. The gains I’ve made have been marginal and the peace-of-mind that comes with avoiding all the “is there enough hot water” angst means it’s generally best to just get on with.
For example, a couple of days ago we had a power cut so the cronjob didn’t run and the server didn’t come back up so it didn’t even get anacron’d. As a result the water was cold and my beloved wasn’t impressed.
There was also a lot of chaos last night when we wanted three showers in an hour and the system was trying to hold off until it was more efficient. We turned it on with MELCloud and my server turned it back off! Fortunately the second time we turned it on the “oh someone forced this on” code kicked in and left it alone.
Wouldn’t it be better still to run the heatpump during the cheap rate? Even if it only gets a COP of 2, that’s still twice the heat into the tank compared to direct heating. I don’t have solar divert, I just put excess electricity into the heatpump to heat the water until the immersion takes over. (~54°)
There’s the trade-off between having enough hot water for the entire household vs. spending as little as possible to heat that water. You’ll get more complaints about the former.
Your plan above is probably good enough for what you need, and efficiency will probably be fine. Remember: DHW is just 1 tenth of your annual heat demand, so there’s limited value in hyper optimising it.
Heating the water to a higher temperature during the cheap rate, storing more energy in the tank, might allow for longer running showers. Less efficient, but that’s less of a concern while electricity is cheap.
The FTC has a setting for how much of a drop in tank temperature before it will turn on again. You might want to try increasing this to 15° or more. You can also prohibit the hot water during certain hours, e.g. just before cheap period and during peak period.
1 Like
I have realised something about the FTC’s behaviour that might make what I want easy.
I’ve just turned on the DHW from timed off and a hot water cycle started even though the temperature was only a couple of degrees under target (set to come on if drops by 10 degrees).
So… I should be able to leave it on with it monitoring the temperature at the top of the tank (so just before we are about to run out).
But I then also set up the schedule to turn hot water off for a bit before the off peak time and then back on.
I think this should allow us to always run hot water during the night but it would also trigger during the day if we are going to run out.
On days were we don’t use as much or there is solar boost then the additional run shouldn’t happen.
1 Like
I was worried that using the top tank temperature would mean heating of the water would stop too early to heat all of the tank but it appears with our coil base tank the water is agitated enough by thermal currents that this isn’t an issue.
But the stratification in the tank is really good so the bottom temp can be down into the teens when the top of the tank has just dropped a few degrees.
I plan to have our Eddi set up using application example ([Application Examples | myenergi GB)](https://application example 14) connected to the Smart Grid input on the FTC6.
This should let us set an excess solar threshold where under this the immersion will be used as normal but above it the Heat Pump will be triggered to heat the hot water. Need to get this threshold right so it is enough to completely power the heat pump. Were sun is intermittent the heat pump will be allowed to complete a sensible length cycle but this could end up using non solar electricity. However if the threshold is high enough there is a good chance that our battery will be well charge as it is a sunny day and cover any drops in solar rather than using grid. The battery will be set up to take priority over the Eddi for excess solar.
If it doesn’t work out I can always just disable it and use the Eddi and immersion as normal.
1 Like
I was away for a week recently with the ASHP in holiday mode. When it came off that, due to two days of rain etc. the ASHP DHW tank (also heated by the Eddi during PV excess) was below the lower threshold so it was heated back up to 55C by the ASHP.
That took 26 minutes with a COP of 1.24 so yes, more efficient than direct heating.
One of the reasons I use the Eddi is that both it and the EV charger have dynamic load management which I use to avoid blowing the house fuse overnight when everything is trying to operate during the Go period. If the ASHP isn’t already on (despite overnight temperature setback), it could throw another 30+ A into the mix.
2 Likes
Mm, it would appear our temperature probe pockets are too low and too high.
I’m heating our tank during night low rate electricity but on some days this isn’t enough to get us through the day (250l tank!) so I need to sometimes do another run before the evening. This will also depend on amount of solar diversion once we have that installed.
However if we use the bottom pocket I think the hot water would come on with every shower but in the top pocket the temperature doesn’t drop significantly until it’s too late to heat again.
I’m going to stick with it for a bit and continue to adjust the drop that triggers the HP but assuming this doesn’t work what other option is there?
Is there some way I could measure the actual volume of hot water used since the last run? This would seem like the most accurate way to work out if we are going to run out or not.
I’m afraid I haven’t seen anything like that in my adventures so far using the MMSP package that was installed on mine.
How hot are you heating the tank up to? How many kWh are you putting in each night?
Maybe you can use both pockets? Bottom one for heating up the cylinder, top one providing extra info about what’s in the tank. Maybe by analysing both you can decide when to add more heat?
You would need a flow meter on the outlet from (or inlet into) the cylinder to measure exactly how much hot water has been used. Probably best asking how in a new thread.
A cheaper, less invasive, less accurate option might to strap a thermal or acoustic sensor to the pipe to detect when hot water is flowing.
Edit: see this topic for further discussion: Hot water tank temperature monitoring
1 Like
I have been heating to 48 degrees, although I have increased this to 50 while I try to figure our the best thresholds to use.
It might be that eventually the solution is to just have 2 scheduled DHW cycles and lower the temp to 45 degrees so they are as efficient as possible. Probably with something to stop the 2nd one if the solar diverter has heated up the bottom of the tank during the day.
The Ecodan only has one probe that I access the value for through pymelcloud. However it is possible to connect 2 probes to the FTC for larger tanks but I don’t know if it can monitor both or it is just that you can select between it using one or the other from the controller.
Anyone have the 2 probe setup? I think it might be how the larger Ecodan cylinders are set up so anyone with one of those able to comment on how this works?
I think to do what you suggest I would need to add to my external monitoring setup. I currently have a EmonHP and a second Pi that is connected to our solar invertor. What would the simplest way of setting up 2 temperature probes ideally using the hardware I already have?
Another thought I have had about the positioning of a probe to decide when to boost the tank is whether it could be attached to where the connection to the coil comes out. This is near the middle of the tank so might give a more useful reading. As the probe isn’t in a pocket within the tank it wont be a very accurate temperature but it might give a reading I can set a good threshold on.
Over the last few days I’ve been experimenting with attaching the temperature probe to the capped off pipe that comes out of the cylinder for a secondary hot water return.
This is a bit further down the cylinder than the top sensor pocket so has been causing the Ecodan to kick in earlier to heat water when needed so we don’t run out.
A possible down side (although I need more runs to verify) is that as the probe isn’t more ‘in’ the cylinder I think there is a temperature reading lag when heating up. I think this might be meaning the flow temperature is higher and I would assume the actual tank temperature is overshooting a bit.
Maybe I can reduce the temperature to counter that a bit but then that might start effecting the trigger for starting the tank warm up.
Our Electric Meters are hopefully being installed tomorrow so I will be able to see the effect on consumption soon.
1 Like