That's quite interesting. It makes me wonder if AA has stored a copy of the original WC curve, which it may be adapting over time. What were the settings of your WC curve before you made the changes?
The curve was actually set for another experiment (that was working very well). It's another topic entirely so I won't go into detail here; but I 'inverted' the curve, it was set to 38C at -5 and 42C at +15. The experiment was to have warmer LWT at higher OAT to lessen cycling (now I understand more about DT50 and heat losses etc in my own home).
The system has been in for nearly a year now so I wonder(?) if it has enough historical data available (indeed, will it collect and later utilise any data whilst running in WC or Fixed temp modes?).
That's quite interesting. It makes me wonder if AA has stored a copy of the original WC curve, which it may be adapting over time. What were the settings of your WC curve before you made the changes?
The curve was actually set for another experiment (that was working very well). It's another topic entirely so I won't go into detail here; but I 'inverted' the curve, it was set to 38C at -5 and 42C at +15. The experiment was to have warmer LWT at higher OAT to lessen cycling (now I understand more about DT50 and heat losses etc in my own home).
The system has been in for nearly a year now so I wonder(?) if it has enough historical data available (indeed, will it collect and later utilise any data whilst running in WC or Fixed temp modes?).
That is one of the burning questions, since we don't know precisely how the control algorithms function. I have been trying to apply my knowledge of conventional control system, to try to ascertain what is actually happening inside the controller when operating in AA mode.
Again, I don't know if it only starts 'adapting' once set into AA mode, or whether it has some stored data to which it can refer. Since it would appear that some installer's don't commission systems very well, how useful would such data be even if it was available.
To possibly help reduce cycling, it is necessary to quantify the approximate OAT and LWT at which cycling appears to start, and then raise the warmer end of the WC curve to slightly above this setting. It may be that a setting of 30C @ 15C helps reduce the cycling frequency and extend the running period. Obviously it will be unique for your system and will no doubt vary slightly with weather conditions.
@derek-m What did Toyah once say, 'It's a mystery, I'm still searching for a clue..' 😀 .
My system needs 40C+ flow temps when OAT is in the 12-15C range to reduce cycling. The logic behind trying the inverse curve was to give the system what it wants as such (without excessive overrun) the 38-42 combo was working pretty well.
@derek-m What did Toyah once say, 'It's a mystery, I'm still searching for a clue..' 😀 .
My system needs 40C+ flow temps when OAT is in the 12-15C range to reduce cycling. The logic behind trying the inverse curve was to give the system what it wants as such (without excessive overrun) the 38-42 combo was working pretty well.
Have you tried running with a lower WC curve setting, but using a room thermostat to start and stop the heat pump?
I had a bit of time on my hands this afternoon (and it was raining) so decided to have a quick 'play' with AA mode again. OAT was 12C, and I set my WC curve to 38C at both ends (i.e. it was flat) to see if AA mode made any use of it. Themo diff adjust is back ON again, with a +3 / -5 configuration. I painstakingly noted the compressor frequency at 5 min intervals, and have added them to the timeline at the bottom of the graphs.
It seems that the target temp it has chosen is 32C (so not been influenced by the WC curve by the look of it!?). The first cycle LWT maxed out at around 35C (seems to have hit the +3 nicely) then reduced to 27C (seems to have hit the -5 nicely). The other two cycles are a little less obvious (there appears to be no attempt at the +3). Our DWH comes on at 15:00 so interrupted things, IAT had come up 0.5C in the two hours the heating was running though.
Comfort levels were very good. If I'm correct(?) in that the Ecodan compressor runs in a frequency range between 0-100Hz, the compressor numbers are sitting at the lower end (seems to like 36Hz, one peak at 46Hz when chasing the +3).
That's quite interesting. It makes me wonder if AA has stored a copy of the original WC curve, which it may be adapting over time. What were the settings of your WC curve before you made the changes?
I still feel that AA mode will make reference to variations in OAT, since without it I suspect just a room sensor would fail to control the temperature, without excessive overshooting and undershooting.
Using my Ecodan simulation tool, it shows that the 7C change in OAT from 14C to 7C today, would require the LWT to be increased from 32.39C to 40.19C in a matter of 10 hours or so, which would require a 192% increase in electrical energy.
Or it may have an in-built one?
One thing I notice is that in milder temperatures my LWT doesn't vary much. I've limited my LWT to a maximum of 43 deg and have thermal diff on +-5. It seems to use the max plus adjustment most of the time but never the minus, which makes my LWT between 39 and 42. It would be lower, funnily enough about 5 deg lower, on WC. I think this is because it never really gets to steady state; it gets to set temp plus 1 deg then stops for a couple of hours and repeats, which means it raising the IAT most of the time and not often running steadily. As discussed I'll change the thermal diff setting and see what happens.
This post was modified 1 year ago 2 times by Kev M
Have you tried running with a lower WC curve setting, but using a room thermostat to start and stop the heat pump?
Yes (and everything in-between lol). I was running a curve of 38C at -5 / 35C at +15, this would result in 4-5 cycles per hour during mild outside temps, the heating would only be on for a couple of hours before the room stat turned it off. Any attempt to go lower than 35C would result in 6 or more cycles per hour (not acceptable imo). 42C reduced that cycling to only one or two per hour (again, room stat turning the system off after around couple of hours).
It's a shame that there is no way to extend that 4 minute cycling off period in WC and Fixed modes (at the end of the day it's just having the thermo diff adjust option configurable in the other modes). I think I'll revisit AA mode for a while, that thermo diff adjust lower limit value may just be to handy to ignore.
Have you tried running with a lower WC curve setting, but using a room thermostat to start and stop the heat pump?
Yes (and everything in-between lol). I was running a curve of 38C at -5 / 35C at +15, this would result in 4-5 cycles per hour during mild outside temps, the heating would only be on for a couple of hours before the room stat turned it off. Any attempt to go lower than 35C would result in 6 or more cycles per hour (not acceptable imo). 42C reduced that cycling to only one or two per hour (again, room stat turning the system off after around couple of hours).
It's a shame that there is no way to extend that 4 minute cycling off period in WC and Fixed modes (at the end of the day it's just having the thermo diff adjust option configurable in the other modes). I think I'll revisit AA mode for a while, that thermo diff adjust lower limit value may just be to handy to ignore.
What I find surprising is that you need to run at such high LWT in mild weather conditions. What is your heat loss calculation and heat emitter capacity?
To use a room thermostat, ideally one with a 1C switching deadband, set to 1C above the desired temperature. Then adjust the settings of the heat pump controller so the LWT is slightly above that at which it commences cycling.
The theory is that the heat pump raises the IAT until it is stopped by the room thermostat. The heat pump cannot restart until the room thermostat resets when the iAT falls by 1C. If it functions as I would expect, the heat pump runs for a period and then stops for a period.
What I find surprising is that you need to run at such high LWT in mild weather conditions. What is your heat loss calculation and heat emitter capacity?
We did discuss this on another post recently where I provided this info (you kindly provided a copy of your radiator output calculator spreadsheet). It's a 6kW Ecodan, 5.7Kw heat loss calc, rads total 12000 watts.
What I find surprising is that you need to run at such high LWT in mild weather conditions. What is your heat loss calculation and heat emitter capacity?
We did discuss this on another post recently where I provided this info (you kindly provided a copy of your radiator output calculator spreadsheet). It's a 6kW Ecodan, 5.7Kw heat loss calc, rads total 12000 watts.
Sorry about that, it is one of the pleasures of being old and decrepit. 🙁
What I find surprising is that you need to run at such high LWT in mild weather conditions. What is your heat loss calculation and heat emitter capacity?
We did discuss this on another post recently where I provided this info (you kindly provided a copy of your radiator output calculator spreadsheet). It's a 6kW Ecodan, 5.7Kw heat loss calc, rads total 12000 watts.
I think I was part of this discussion too…
I did some calcs on our system and the radiator capacity at different flow temperatures. And we think a 32 to 35degC flow temp is about how low our system will go before cycling will start.
we have a 21kwh Rad capacity in our installation which translates as follows at various flow temperatures:
So our 8.5 kw heat pump will probably not operate much below 3kw modulation.
I worked these out using the standardised correction table and adapted it to the 5kw deltaT of our heat pump. (Table below)
But it’s not just about the radiator output capability it’s also about the flow capacity and resistance in the pipework so I worked out the heat carrying capacity of all our pipework and came up with this schematic of our Flow Pipework. (Our return pipework is identical)
The importance of 28mm primary pipework taken right to the first junction where 2 X 22 mm pipework takes over and the split of the radiators is approximately symmetrical down both branches to ensure water volumes are more or less balanced.
So we’re hoping all this attention to detail will help avoid excessive cycling. You may have noticed there is some 10mm microbore however only one rad is fed off each branch. This hasn’t caused too much resistance - there is a flow rate of 22 lpm on the DHW circuit and 19 LPM on the heating circuit with the microbore. All the TRVs are now left fully open so that maximum heating capacity is available at all times.
In the knowledge that our capacity is limited to around 32 to 35 we have edited our WCcurve to have a lower limit of 30 LPM at outdoor ambients of 9C and above since any lower will probably invite rapid cycling.
