Posted by: @cathoderay

↑

Posted by: @jamespa

↑

which is uncertain because we don't know the emitter temperature due to the fact that when the hp is off the flow temperature will drop back

Actually, we do know the LWT and RWT when the heat pump is off, see my charts, and so can estimate the radiator temp, say an average of the LWT and RWT.

That would be the case if the water pump keeps running for the whole compressor off period. I believe that some heat pump controllers can be set to run the water pump for several minutes after the compressor has stopped, then run the water pump for several minutes, periodically, to check how much the water temperature has reduced.

I think that some controllers can also be set to not run the water pump after the initial run-on period, so if the LWT and RWT sensors are inside the heat pump unit, their temperature readings are likely to fall much quicker than the water in the heat emitters.

@derek-m - my heat pump appears to keep the flow rate steady, unless the heat pump is completely off. Flowrate (m^hr) added in gold, trace for last 24 hours:

During the setback, flow rate is zero, but when cycling it stays at around 1.4 something, whether compressor is on or off. A zoom in on some recent cycles:

The sensors are definitely in the heap pump, but the water is still circulating, certainly in the primary circuit, but also I am pretty sure in the secondary circuit. The cycling is just the heat pumps way off getting a LWT/RWT average that approximates the Set LWT, and the flow rate doesn't care what part of the cycle the heat pump is in, it just keeps rolling on...I could not stop it if I wished; no one can stop it. Like the Mississippi, it just keeps rolling along. Let it roll. Let it roll on full flood, inexorable, irresistible, benignant, to broader lands and better days...     

PS yes, WSC for those who are trying to place the quote

@cathoderay

So on your particular system, when in setback mode, the whole system appears to be disabled, including the water pump or pumps. When cycling the system is enabled, and only starts and stops the compressor to help control the IAT.

The temperature difference between LWT and RWT when in setback mode is probably due to the location of the sensors. When the compressor is stopped during cycling, LWT and RWT are virtually the same, which is what should be expected if the water pump is still running. It would be interesting to know if the secondary water pump is still operating when the compressor is stopped.

I can foresee at least a couple of problems if employing a lengthy setback during cold weather conditions.

The first would be that the IAT will fall further during lower OAT's, which although this may initially provide greater energy saving, will in turn require more energy to recover the desired temperature, and also take a much longer period of time for this to be achieved.

The second is the actual capacity of the heat pump, since at lower OAT's it would already be running much closer to its maximum output, so that when the additional thermal energy is required due to the first issue, the heat pump would have additional limited capacity to supply this extra thermal energy. A further factor is the reduction in efficiency which occurs when the heat pump is working at its maximum.

This is without any defrosting requirements, which may only make matters worse.

It may be necessary to employ a more controlled setback, of say 2C or 3C below desired. If OAT drops lower than anticipated, IAT may fall down to the room thermostat setting and restart the system. This would therefore have the effect of not only limiting how low IAT falls, but will also reduce the setback period, thus giving the, possibly struggling, heat pump more time to recover to the desired IAT.

Don't get me wrong, I fully applaud your efforts to reduce energy consumption, it is a pity more people don't follow your lead.

Posted by: @derek-m

↑

So on your particular system, when in setback mode, the whole system appears to be disabled, including the water pump or pumps.

That's by intention, achieved by using the room stat timer as an on/off switch, by setting it to 12 degrees for off, 26 degrees for on. I could do the same thing on the wired controller schedule pages, I suppose, but ended up doing it via the room stat. The idea is that off means off.

Posted by: @derek-m

↑

When the compressor is stopped during cycling, LWT and RWT are virtually the same, which is what should be expected if the water pump is still running. It would be interesting to know if the secondary water pump is still operating when the compressor is stopped.

I am not sure about this. If the water pump(s) are still running, warmer water will enter the rads, and get cooled, and the will mean a lower RWT, surely? What I can't explain is how the LWT and RWT approach the same temp during the cycling. Perhaps the clue is in the pattern of cooling seen in the very zoomed in chart: the LWT falls sharply over a few minutes, until it reaches the RWT, at which point they becomes the same, but both are still falling. I will try to determine whether the secondary pump continues to run during the compressor off periods.

Posted by: @derek-m

↑

The first would be that the IAT will fall further during lower OAT's, which although this may initially provide greater energy saving, will in turn require more energy to recover the desired temperature, and also take a much longer period of time for this to be achieved.

This, as ever, is the crux of the matter, how much extra energy during the recovery? And how does that compare energy saved during the setback? And yes it will take longer to recover from a lower IAT, but that is what the auto-adaption is for, to speed up the recovery by boosting the LWT. I still thivk the only way we will get half decent of better answers to these questions is by observing a real world system operating in real world conditions. We just have to wait for colder weather, and then observe what happens.  

Posted by: @derek-m

↑

The second is the actual capacity of the heat pump, since at lower OAT's it would already be running much closer to its maximum output, so that when the additional thermal energy is required due to the first issue, the heat pump would have additional limited capacity to supply this extra thermal energy. A further factor is the reduction in efficiency which occurs when the heat pump is working at its maximum.

Agreed, but I have two cards up my sleeve. The first is to raise only the right hand end of the WCC curve. This will raise the LWT, by varying amounts, at all OATs except the OAT set point for the left hand end of the WCC, currently set at -4. This prevents the WCC pushing the LWT over the top. The second card is the nuclear option, flip the dips and turn my heat pump into a higher capacity model, from 14 to 16kW. Like most people, I prefer to avoid nuclear options. The efficiency penalty caused by higher LWTs is well known as a concept, but not so well understood in detail in the real world. Again, that is why we need real world observations on a real world system. And yes, defrosting will aggravate matters, but only real world data will tell us by how much. I'm expecting the hit to be quite large, but will have to wait and see.

Posted by: @derek-m

↑

This would therefore have the effect of not only limiting how low IAT falls, but will also reduce the setback period, thus giving the, possibly struggling, heat pump more time to recover to the desired IAT.

Agreed (limiting the fall in IAT), but reducing the setback period will also mean more energy consumed per se. It's the balance that matters: setback energy saved vs recovery boost energy cost, which is the greater? I'm not sure the heat pump will struggle given the parameters I am using, which are within its stated normal operating limits. If it does struggle within those limits, then it is not very well designed.

Meanwhile, I think I have made progress on reading the modbus registers:

(a) the DHW on comes back as decimal/hex 6 because in the register both the heating and DHW are on, in binary 110, where the left bit is the DHW and the middle bit the heating, but elsewhere the controller decided only one can be in play at any one time, and given the DHW has priority, it comes on, and the heating gets turned off, when the register has hex 6/binary 110. The diverter valve also makes sure only one of either heating or DHW are on at any one time (mine doesn't have a mid position).

(b) the reply for the WCC setting had me totally baffled for a while. My wired controller manual makes no sense: register address 6 = Curve selection (curve 1-8) ie no curve 9, which is the weather compensation curve. The actual reply is (decimal) 777 - I had to be grateful for small mercies, at least it isn't 666. I finally got the answer from this bit of code:

reg6 = instrument.read_register(6, 0)
print(f'Register6={reg6}')
print(f'Register6 bin={bin(reg6)}')
curve = reg6 & 0xff
print(f'curve number={curve}')

  which outputs the following

MinimalModbus debug mode. Will write to instrument (expecting 7 bytes back): 01 03 00 06 00 01 64 0B (8 bytes)                                                                               
MinimalModbus debug mode. Clearing serial buffers for port /dev/ttyUSB0                                                                                                                      
MinimalModbus debug mode. Sleeping 2.53 ms before sending. Minimum silent period: 4.01 ms, time since read: 1.48 ms.                                                                         
MinimalModbus debug mode. Response from instrument: 01 03 02 03 09 78 B2 (7 bytes), roundtrip time: 0.0 ms. Timeout for reading: 50.0 ms.                                                    
                                                                                                                                                                                             
Register6=777                                                                                                                                                                                
Register6 bin=0b1100001001                                                                                                                                                                   
curve number=9 

The output request, the first line ('Will write to...', but it actually means 'Will read from...') gets one register from device 01 starting at address 06 ie reads register 6.

The output response, the fourth line, has the payload 03 09 and 0309 is hex for decimal 777.

The code then does some bitwise voodoo (curve = reg6 & 0xff) which I believe is a sort of bit masking which can be semi made understandable as follows:

1100001001    777 decimal
  &
0011111111    ff hex
  =
0000001001    9 decimal

What the & does is test each position in the binary representation of the numbers, and returns a 1 only when both numbers have 1 in that position (the first number and the second number have a 1 in that position), giving, as can be seen a binary result of 1001 which is decimal 9, the correct result. 

Another random manual for another Midea heat pump further made things clearer:

I believe this may also be called hi and low bit voodoo, but am not sure. What I am sure about is that Midea want us end users to live in interesting times, and publish misleading manuals to achieve their fiendish ends. It is also why I will not write to the registers unless I am 100% sure that I know what I am writing, and where it will end up.

(c) all of the above (and more) was about trying to find a way to (a) ensure a DHW reheat didn't get prematurely terminated by my auto-adapt script and (b) ensure the heating came on in WCC mode after a DHW reheat, whether it was terminated or ran full cycle. It has since occurred to me that there is a far simpler way of solving this: don't write to the registers when the system is in DHW mode. I am currently tweaking the code to achieve this, a slow process as the DHW only comes on once a day.      

Posted by: @cathoderay

↑

That's by intention, achieved by using the room stat timer as an on/off switch, by setting it to 12 degrees for off, 26 degrees for on. I could do the same thing on the wired controller schedule pages, I suppose, but ended up doing it via the room stat. The idea is that off means off.

I have been pondering whether this is the best way to run my system - us the thermosat just as an "on/off switch", set to an impractically high value like 26C, and adjust the WC curve as low as it will keep the house warm.

However this will mean the heating circulation pump will always be running (except during setback?) rather than only when heat is actually needed. I don't know how much power the pump would use but this seems inefficient. Any thoughts?

@bob77 - with the caveat that I have a heat exchanger, and so a primary and secondary pump - it does look like the secondary pump runs continuously when the heating is on (ie runs all the time when cycling, even when the compressor is off and the LWT/RWT temps are falling), but I think this is the norm with heat pumps, and at least with an on/off setback they get a rest (my secondary pump does stop during the setback, as the heating is off, just as it would be in the summer. With conventional 24/7 continuous running, presumably they run all the time, except perhaps when the system is in DHW rather than heating mode. This suggests the setback might reduce secondary pump power use and also reduce wear, and possibly extend service life.

Posted by: @bob77

↑

Posted by: @cathoderay

↑

That's by intention, achieved by using the room stat timer as an on/off switch, by setting it to 12 degrees for off, 26 degrees for on. I could do the same thing on the wired controller schedule pages, I suppose, but ended up doing it via the room stat. The idea is that off means off.

I have been pondering whether this is the best way to run my system - us the thermosat just as an "on/off switch", set to an impractically high value like 26C, and adjust the WC curve as low as it will keep the house warm.

However this will mean the heating circulation pump will always be running (except during setback?) rather than only when heat is actually needed. I don't know how much power the pump would use but this seems inefficient. Any thoughts?

If your intention is to use your room thermostat as a high temperature limit, then your water pump will need to run when your heat pump is running.

Controllers often have the capability to dictate how the water pump will run when the compressor has stopped producing thermal energy. I think that most controllers run the water pump for several minutes after the compressor has stopped, probably to transfer the remaining thermal energy from the condenser to the heat emitters.

The controller may then run the water pump for several minutes, periodically, to circulate the water to check its temperature.

It may also be possible to instruct the water pump not to run until the compressor restarts.

I would suggest that you check the various options that are specified in the manual.

@derek-m - I think we need to consider whether there is a heat exchanger or not, and if so, which pump we are talking about. In my case there is, and the secondary circuit pump is definitely on/off, and, to answer your previous question, runs continuously when the heating is on (however it is turned on, in my case by having the room stat set very high). The behaviour of the primary circuit pump is less clear. I think on Midea heat pumps the pump starts and there is a short delay before the compressor starts for an initial start up (as for example, in coming out of setback), as I think there is a setting for this, but once it is running, I am not sure whether it runs all the time when cycling, or whether it slows down/stops during the compressor off phase of the cycle. It seems to me that logically it should continue to run, because that is how, as you say, it gets feedback on the LWT/RWT (which strictly speaking become meaningless if the pump is off, nothing is leaving and nothing is returning). On the other hand, the sharp fall in both the LWT and RWT, and their convergence, during the compressor off phase of a cycle, might seems to suggest the circulating pump is off during the compressor off phase.

Having waffled, I now realise the answer is staring us in the face, at least for Midea pumps. The primary pump appears to run continuously, except when the heating is off: see the charts in this post.

Posted by: @derek-m

↑

If your intention is to use your room thermostat as a high temperature limit, then your water pump will need to run when your heat pump is running.

Controllers often have the capability to dictate how the water pump will run when the compressor has stopped producing thermal energy. I think that most controllers run the water pump for several minutes after the compressor has stopped, probably to transfer the remaining thermal energy from the condenser to the heat emitters.

The controller may then run the water pump for several minutes, periodically, to circulate the water to check its temperature.

It may also be possible to instruct the water pump not to run until the compressor restarts.

I would suggest that you check the various options that are specified in the manual.

Forgive my ignorance but do you mean a pump that is part of the heat pump unit, or the external central heating pump? I assumed that the water circulating around the central heating system was pumped by the external pump, whenever the thermostat called for heat, and that this was separate from anything the heat pump itself was doing - in other words if the thermostat said heating was needed then the pump would be going, even if the heat pump itself was off. Is that not the case? I grew up with electric storage heaters, so central heating is a bit of a mystery to me!

@bob77 - I think your understanding is correct, at least for some systems - other heat pumps may/will vary, but on a Midea setup, the secondary pump is fed (on/off) from the heat pump main control board, and according to the Freedom installation guide, it "only runs when there is a heating demand, it is driven by the external run signal from the underfloor heating manifold or room thermostat." This is consistent with what I have observed, when the room stat is calling for heat (and the heat pump is in heating not DHW mode), the secondary pump runs continuously, when the room stat is not calling for heat eg during a setback the secondary pump is off.

Posted by: @cathoderay

↑

@bob77 - I think your understanding is correct, at least for some systems - other heat pumps may/will vary, but on a Midea setup, the secondary pump is fed (on/off) from the heat pump main control board, and according to the Freedom installation guide, it "only runs when there is a heating demand, it is driven by the external run signal from the underfloor heating manifold or room thermostat." This is consistent with what I have observed, when the room stat is calling for heat (and the heat pump is in heating not DHW mode), the secondary pump runs continuously, when the room stat is not calling for heat eg during a setback the secondary pump is off.

Not all heat pumps are midea's, and not all systems have primary and secondary water pumps. Even on systems that do have both water pumps, I cannot see the point of running the secondary pump if the primary pump and heat pump are not operating, though your installer may think this a good idea.

Posted by: @derek-m

↑

Not all heat pumps are midea's, and not all systems have primary and secondary water pumps. Even on systems that do have both water pumps, I cannot see the point of running the secondary pump if the primary pump and heat pump are not operating, though your installer may think this a good idea.

I have never said all heat pumps are Mideas, quite the opposite, nor have I ever said all systems have primary and secondary pumps, again, quite the opposite, so I have no idea what point you are trying to make here.

The design and setup of my Midea system as you well know comes directly from Freedom Heatpumps and their installation manual, and there has been, as again you well know, plenty of discussion over time as to whether their sometimes mandatory sometimes optional heat exchangers/buffers are a good or bad idea, with part of that discussion focusing on the introduction of a second pump into the system, and the penalties that carries.

If you read the line I quoted earlier from the Freedom/Midea installation manual:

it [the secondary pump] "only runs when there is a heating demand, it is driven by the external run signal from the underfloor heating manifold or room thermostat."

It seems to me to say that the secondary pump is controlled by the room (or UFH) stat, when this calls for heat, the pump runs. This comes from Freedom, not my installer.

As it happens, I agree that it seems pointless having the secondary pump running when there is no heat to distribute around the heating circuit, but it seems that is the way Freedom/Midea do it, possibly to the extent that it is hard wired into the circuits. The Midea Engineering Data manual also suggests this is the case:

5.1.1 Space heating

...P_o (6) [the secondary pump] keeps running as long as the unit is ON for space heating...              

with no setting that I can see to alter this state of affairs.

The Engineering Data manual also appears to confirm what I pointed out earlier, but Freedom falied to mention, that the secondary pump is off when the system is in DHW heating mode:

5.1.2 Domestic water heating

...P_o (6) [the secondary pump] stops running as long as the unit is ON for domestic water heating...

There's more Chinglese further down the manual, but the general drift remains as above, the secondary pump runs when the system, in my case room stat, is calling for heat. 

If you don't like this arrangement, you need to take your argument not to me, but to Midea.