I was interested in your 24 hour room temp graph which shows how AUTO ADAPT tries to anticipate and over run set temperatures.
Its import to note the Set Room temperature is the Blue Line and the AA actual temperature (aka IAT) is the Black Line. The Green Line is the OAT (outside air temperature or ambient)
I have attached our own copy of last nights 24 hour graph showing 3 incremental rises in target room temperatures (manual adjustment) and how AA behaved rising above the targeted temperature with each manual adjustment then later shutting down to allow the IAT to naturally drop back to the target room temperature. We set back the temperature to 17c at 8.30pm.
Its interesting (sad but true) That the IAT continues to rise during the start of the set back adjustment. Possibly due to the 30 minute monitoring period…..
please ignore the OAT Temp rise in the morning. We have been experimenting with reflected solar gain to adjust the apparent OAT to create a ‘softer’ temperature rise at the start of the day. Every Day we seem to learn something new….The unexpected ones are the best.
The former I think. AA pushes the lower end up and the higher end is constrained by my 43 deg upper limit. If I remove the upper limit, the upper end increases. This makes the variation more but also the average. When I'm back home I'm going to turn the thermal diff off and keep the upper limit (which is the same as my WC curve upper limit. It may start to cycle but let's see.
From my experience the upper limit was ignored when thermo diff was OFF (I'd set it to 42C - it blasted through to 48C in one test). It will be interesting to see how you get on.
Agreed. It didn't do it immediately; in fact for the first couple of days it looked the same as before. But yesterday after the HW cycle it sneaked in a 48.5 deg burst. Next change today will be turning back on but with minimum settings both ways.
-- Attachment is not available --
Thats interesting… I haven’t seen such super high temps. I’ve never thought of containing the upper temperature limit. I have mine set to 53c because the cold end of my WCcurve is set at 53c@-9 outside temp. I’ll have to keep an eye on this 👁
The former I think. AA pushes the lower end up and the higher end is constrained by my 43 deg upper limit. If I remove the upper limit, the upper end increases. This makes the variation more but also the average. When I'm back home I'm going to turn the thermal diff off and keep the upper limit (which is the same as my WC curve upper limit. It may start to cycle but let's see.
From my experience the upper limit was ignored when thermo diff was OFF (I'd set it to 42C - it blasted through to 48C in one test). It will be interesting to see how you get on.
Agreed. It didn't do it immediately; in fact for the first couple of days it looked the same as before. But yesterday after the HW cycle it sneaked in a 48.5 deg burst. Next change today will be turning back on but with minimum settings both ways.
When the LWT increased after the DHW cycle, was this due to the fact that the LWT was higher to produce the hot water, so when the changeover occurred it measured this higher LWT and attributed it to CH rather than DHW?
What is the water flow rate? It would be interesting to see what effect reducing the flow rate would have.
Also what is happening with the IAT and OAT?
Here is a challenge for you Derek:
One of these graphs is recorded during a flow rate of 19lpm and 2 of the graphs are recorded at a flow rate of 14lpm. Which is the two at 14 lpm?
Our ecodan 8.5 has a flow range of 11 lpm to 24lpm the room temp was 20.5c on the last picture and 19c rising to 20c on the other two. Outside temp is 10c.
I have been surprised how adaptable the Ecodan is at maintaining a deltaT of 5degC within these changing flow ranges. Our circulation pump allows flow settings of 14, 16 and a max of 19lpm when in CH heating mode. And a top flow rate of 22lpm in DHW mode.
However I have been planning to operate the HP on the assumption that the faster flow rate will enable the heat pump to reach the higher outputs when the weather is especially cold say at -8. Do you have a view on this? Our HP CAN Seemingly modulate up to around 12kwh at max output.
Meanwhile ive been thinking that operating the pump when it gives a modest 16 lpm flow should give the HP most flexibility in modulating down or up when in moderate winter weather. Do you think this would be the case? I don’t have any evidence of this view just an intuitive idea that it might.
Answers to the quiz: Low flow rate vs high flow rate.
The 11 o’clock graph is operating at 19lpm and the 12 o’clock and 15 o’clock graphs were operating at 14lpm. The 12 o’clock graph has just started to modulate down with a smaller DT even though it is operating at a slower flow rate.
OUR SYSTEM has always had a higher DT when operating at a slower flow rate. I have managed to reduce the flow to 12lpm and here is the graph it has always tended to push for a DT of around 7c.
When operating at 12 lpm this is right at the lower limit of the operating range. So the HP DT appears to be a bit erratic. While the heating cycle rises fast it appears to be trying to drop the DT. Also the cycle appears a little shorter as there is more heat rise in the system. Whereas between 14 and 19lps the system appears to easily regulate the DT to 5c. And when it needs to modulate the mid to fast flow rate seems to comfortably reduce the DT to 3 or 4c when in steady state.
Ecodan does not have variable internal circulation pumps which means they are reliant on a wide range of third party circulation pumps. It’s possible that Ecodan can tolerate a broad range of flow rates and still maintain DT of 5c but I feel they operate to their full output potential if the flow rate can be in the higher part of the range. Unless there is a low loss header when flow rates are perhaps better if they are balanced.
The former I think. AA pushes the lower end up and the higher end is constrained by my 43 deg upper limit. If I remove the upper limit, the upper end increases. This makes the variation more but also the average. When I'm back home I'm going to turn the thermal diff off and keep the upper limit (which is the same as my WC curve upper limit. It may start to cycle but let's see.
From my experience the upper limit was ignored when thermo diff was OFF (I'd set it to 42C - it blasted through to 48C in one test). It will be interesting to see how you get on.
Agreed. It didn't do it immediately; in fact for the first couple of days it looked the same as before. But yesterday after the HW cycle it sneaked in a 48.5 deg burst. Next change today will be turning back on but with minimum settings both ways.
When the LWT increased after the DHW cycle, was this due to the fact that the LWT was higher to produce the hot water, so when the changeover occurred it measured this higher LWT and attributed it to CH rather than DHW?
I don't think so. It takes about an hour or so to heat from mid 30s to 48.5. I would have though any residual peak from the HW would be long gone? Below is a zoom of the day graph. I had a look and since I made the change I've seen 44.5 a couple of times; not as high but still over the set max limit of 43.
Another explanation could be that because the controller has been in DHW mode for a period of time, it may not have been updating the AA algorithm. When it switches back into CH mode, it may see a step jump in some of the parameters, which cause the AA algorithm to over compensate.
AA mode is a type of continuous process control, which performs best when the process is actually continuous. This can be particularly problematic in home heating system where the response times are quite long.
The former I think. AA pushes the lower end up and the higher end is constrained by my 43 deg upper limit. If I remove the upper limit, the upper end increases. This makes the variation more but also the average. When I'm back home I'm going to turn the thermal diff off and keep the upper limit (which is the same as my WC curve upper limit. It may start to cycle but let's see.
From my experience the upper limit was ignored when thermo diff was OFF (I'd set it to 42C - it blasted through to 48C in one test). It will be interesting to see how you get on.
Agreed. It didn't do it immediately; in fact for the first couple of days it looked the same as before. But yesterday after the HW cycle it sneaked in a 48.5 deg burst. Next change today will be turning back on but with minimum settings both ways.
-- Attachment is not available --
I have a theory. When thermo diff adjust mode is ON it opens up a bunch of parameters to help us control/limit it's behaviour. When it's OFF, it's unrestricted. It appears to me that after assessing all the various temps/DT's etc it chooses a flow temp is sees as most efficient, and without the need to cycle. My system cycles a fair bit the milder it gets, so it makes sense that AA mode is pushing up the flow temps to prevent cycling - I certainly observed the highest flow temps at the mildest OAT. There is a bit on your graph that is covering the OAT when it ramped up - it looks(?) to me that it was the mildest part of the day?
In fact, the description against that function in the manual seems(?) to imply just that.
Its interesting (sad but true) That the IAT continues to rise during the start of the set back adjustment. Possibly due to the 30 minute monitoring period…..
I'm assuming the continued IAT rise is the lag effect of the three incremental/cumulative rises in target temp (I guess LWT was at it's highest / rads were warmest just prior to the setback)? In my chart the flow temps were that low the rads held no 'charge' as such.
The former I think. AA pushes the lower end up and the higher end is constrained by my 43 deg upper limit. If I remove the upper limit, the upper end increases. This makes the variation more but also the average. When I'm back home I'm going to turn the thermal diff off and keep the upper limit (which is the same as my WC curve upper limit. It may start to cycle but let's see.
From my experience the upper limit was ignored when thermo diff was OFF (I'd set it to 42C - it blasted through to 48C in one test). It will be interesting to see how you get on.
Agreed. It didn't do it immediately; in fact for the first couple of days it looked the same as before. But yesterday after the HW cycle it sneaked in a 48.5 deg burst. Next change today will be turning back on but with minimum settings both ways.
-- Attachment is not available --
I have a theory. When thermo diff adjust mode is ON it opens up a bunch of parameters to help us control/limit it's behaviour. When it's OFF, it's unrestricted. It appears to me that after assessing all the various temps/DT's etc it chooses a flow temp is sees as most efficient, and without the need to cycle. My system cycles a fair bit the milder it gets, so it makes sense that AA mode is pushing up the flow temps to prevent cycling - I certainly observed the highest flow temps at the mildest OAT. There is a bit on your graph that is covering the OAT when it ramped up - it looks(?) to me that it was the mildest part of the day?
In fact, the description against that function in the manual seems(?) to imply just that.
I agree with your point about off = unrestricted.
When my LWT went up to 48, it was in the process of raising the IAT from 21.5 to 22.5. The set point is 21 so it didn't really need to do this. Also, if you look at the orange energy trace, while it's raising the LWT to 48, it's using well above its minimum. My ASHP's minimum energy consumption is the level just before 3pm. If it really had to raise IAT it could have done it a bit slower, or even better, not at all.
Having said that, these settings do reduce what we normally call ASHP cycling.
I've been watching this tread and trying a few of the settings on my Ecodan 8.6kW.
Currently our FTC6 main controller is in the hallway but we haven't installed the radiator yet so it takes a while for the controller to pick up any effect of its controlling the heat pump.
I had the interval set on 40mins but have upped this to 60mins and am seeing longer runs.
I've also decreased the lower temp to -9 and this allows the flow temp to drop further between runs resulting a longer off period and (I think) an increase in COP as it is easier for the heat produced to go somewhere when run through the, currently undersized, radiators.
Increasing the higher allowed temperature has just resulted in the flow temperature spiking up too much. +5 on top of what seems like an already too high set point temperature choose by AA meant we were hitting 45 degrees with an outdoor temp of around 9C last night. Our design temperature is 45C at -2. I'm going to go back to the +3C min allowed for this value.
When we finally get this radiator in the hall which is both large so improves the radiator output of the whole system but also in the same room as the thermostat I believe I am going to need to decrease the interval again and hopefully AA will decide I need a lower setpoint for the flow temperature.
Its interesting (sad but true) That the IAT continues to rise during the start of the set back adjustment. Possibly due to the 30 minute monitoring period…..
I'm assuming the continued IAT rise is the lag effect of the three incremental/cumulative rises in target temp (I guess LWT was at it's highest / rads were warmest just prior to the setback)? In my chart the flow temps were that low the rads held no 'charge' as such.
I don’t think your first assumption is having an effect on the end of the day. But I can confirm your second point. The radiators were fully heated at about 38c and still being heated when we did the setback.
looking closer at the room temp graph the heat pump had raised the room temp over 2 hours to be 1c above room set temp. The house then took 2 hours to cool back down to set temp. We switched off mid way during a reheat.
So what surprised me (but as you’ve pointed out, perhaps not so surprising) was there was enough energy to continue heating the house.
for me the 2 hour shut down followed by another boost seems a bit too long. Which raises the question is the 30 minute monitoring period too long for our property?
Perhaps I might be better reverting to a 20 minute period. In our case. I believe I’ve identified the cause of the cycling issue and it has gone away by having my WC curve set at 32c at the mild weather end. At that flow temp our output should be above 3kwh which should be above minimum operating level. So it’s possible now that at 10 to 15c outside temps our WCcurve will operate at a min of 32c so rather than shutting down every 20 mins it will hopefully just modulate to a lower output and continue.
Meantime, I’m letting AA continue at these settings.
One has to be careful when interpreting the data obtained, since the increase in measured IAT may not actually be due to the heating system, but by external affects.
Was the increase in IAT from 14:00 to 17:00 due to solar gain heating the outer fabric of the building? This would reduce the rate of heat loss and have the same effect as an actual increase in OAT, but of course the heat pump controller will not know what has happened until the effect becomes apparent inside.
Human activity should also be considered, since a human body itself is deemed to emit 100W of thermal energy. All electrical/electronic equipment will produce some thermal energy when powered.
Having doors open or closed can affect the heat distribution around the home which can affect the temperature measurements.
Possibly one of the biggest factors is time delays. In your chart the OAT was reducing quite rapidly from 13:00 to 18:00 so will have been increasing the LWT, but the actual change in heat loss that occurs inside could be delayed by several hours.
A further thing to remember is the quantity of energy stored within the heating system itself. I think that I once calculated that 1 cubic metre of water contains 3500 times the thermal energy when compared to 1 cubic metre of air.
