My aroTHERM plus is misconfigured

[Bruno]

And mis-installed too?

 

I have an aroTHERM plus installed with a configuration really close to Vaillant's 1st schematic in this document: Heat pump + buffer tank + radiators + DWH

https://www.vaillant.co.uk/downloads/aproducts/renewables-1/arotherm-plus/all-schematics-wiring-notes-1799366.pdf

(pages 3-5)

The difference is that the buffer tank (Vaillant calls it "decoupler") is 100l instead of the 40l in the schematics. Unfortunately that seems to be a rule of thumb here, more or less 10l/kW and there it is...

Does anyone know what can this cause in the equipment's behavior? Is it just a waste of water mass or will it impact it negatively?

 

However... Looking at the configuration and connections, the equipment is configured as if it didn't have any buffer tank (schematics on page 75-77)

Also for this reason the temperature sensor in the buffer tank is installed but not connected anywhere.

So I clearly need to fix the parametrization as well.

 

Toughts? Thanks

 

[J1mbo]

Can you expand on the configuration issue you’re seeing? A larger buffer will generally reduce short cycling and improve overall COP. The VF1 sensor should be connected to the heat pump interface at the appropriate terminal.

[Bruno]

Thank you for your reply.

Well, I have a question about the heat pump's behavior which I will detail in another topic. However, when I was gathering information about that issue I revisited the manual I linked above.

I then noticed that my "basic diagram" is configured as 8, but that is the case when there is no buffer tank, which I do have. Then I noticed the 40l "decoupler", but my buffer tank is actually 100l.

So the heat pump "thinks" it's connected directly to the radiator's circuit, but there is a big water volume in the middle, much bigger than the volume of water in the rads.

Pretty much all other configurations are according to the recommended values (except for min/max flow temperature because the default values are too low for aluminium radiators). I made the following table, some parameters I could not find

 

Meanwhile I already connected the VF1 sensor and changed the basic diagram to 10 and I will follow the consumption/heating pattern during the day, see if there is some difference.

 

[J1mbo]

It will cost you a lot to run the heat pump with a design max of 65*. What is the heat curve configured as? And what was the design flow temperature on the MCS certificate?

 

The min and max are applied to the curve; these settings might not do what you’re expecting. Also, do you have any controllers other than the VRC700?

 

The larger buffer is not a problem at all and actually is required if the system has only low water content radiators. Is it plumbed 4-pipe like the Vaillant diagrams?

 

Hopefully the external temperature sensor has been fitted?

Edited January 8, 2022 by J1mbo

[Bruno]

9 minutes ago, J1mbo said:

It will cost you a lot to run the heat pump with a design max of 65

I know, but it almost never reaches that. Usually it works at 45-55º. 40º is also low and of course, 15º is cold ?

 

The heat curve is configured at 2.0 after I messed around with that for a while. There was no MCS certificate, I'm not from the UK so we don't have that. Unfortunately the installer's knowledge doesn't seem to be bigger than mine, so I'm trying to figure out the best configurations.

 

14 minutes ago, J1mbo said:

The min and max are applied to the curve; these settings might not do what you’re expecting.

Could you please ellaborate? The description in the manual seems to relate to the minimum and maximum flow temperature. And no other controllers, I only have the VRC700 (and the internet gateway, but that doesn't count :-))

 

16 minutes ago, J1mbo said:

Is it plumbed 4-pipe like the Vaillant diagrams?

It is.

And yes, there is an external temperature sensor. Nevertheless it's also mis-positioned as it gets direct sunlight so in sunny days the reading is too high...

[J1mbo]

Ok, so the curve and the outside temperature dictate the flow temperature and, if you change room temperature from thermostat to “temp mod”, the room temperature observed by the VRC700 itself will also modify it.

 

The set point and programmes set in the VRC700 will be overridden by the Internet gateway (assuming you mean Ambisense). The room temperatures observed by Ambisense room temperature sensors is not used in the flow temperature calculation.

 

The min is simply the floor value. If the calculation yields say 30*, the heat pump will aim for 40* anyway as you have it. Likewise the max is a cap.

 

Really the system needs to be designed so a curve in the region of 1.1 is used which will yield 50* at freezing outside when room temperature is set to 20*.

[Bruno]

1 hour ago, J1mbo said:

The min is simply the floor value. If the calculation yields say 30*, the heat pump will aim for 40* anyway as you have it. Likewise the max is a cap.

Yes, that's what I tought it would do. Thanks

 

1 hour ago, J1mbo said:

Ok, so the curve and the outside temperature dictate the flow temperature and, if you change room temperature from thermostat to “temp mod”, the room temperature observed by the VRC700 itself will also modify it.

Actually that's not what the documentation says. The documentation says that termostat will work as "temp mod", except that if the temperature in the room exceeds the setpoint the system will shut down.

That's actually a good thing for me as I'm using the "night" mode to turn the heat pump off (noise in the building as explained here).

 

1 hour ago, J1mbo said:

Really the system needs to be designed so a curve in the region of 1.1 is used which will yield 50* at freezing outside when room temperature is set to 20*

Yeah... That would actually be great, but I don't think my house can go with a curve of 1.1. At least from my experience in the last 2 months of using the heat pump.

 

 

However... I've been monitoring the consumption after I changed the basic diagram from 8 to 10. The consumption for now it's crazy high (just turned on 2h ago) but it's for sure different than it was before. I will keep you informed, but for sure something *is* different just by changing the basic diagram from 8 to 10.

[J1mbo]

In 10 it regulating to the buffer temp, which means the flow temp from the heat pump itself will be slightly higher as there is always a drop across them. In 8 it works to water leaving the ASHP.

 

How long are you running the heat pump each day? 

[Bruno]

23 hours ago, J1mbo said:

In 10 it regulating to the buffer temp, which means the flow temp from the heat pump itself will be slightly higher as there is always a drop across them. In 8 it works to water leaving the ASHP.

The difference I noticed in the consumption is that now the heat pump goes to max power for a long time, until the temperature in the room gets close to the setpoint, then it starts to modulate.

Before, it started with max power for a few minutes, then abruptly dropped the consumption and kept working in a lower power, increasing the power and the flow temperature as the temperature outside decreases. The target temperature was hardly reached unless the day wasn't too cold. 

I have some consumption graphs which I can show when I get to the computer.

The installation is the same (the buffer was there already, I just changed the parameters) 

 

I'm running the heat pump from 13h until 22h30, with 2 DHW cycles in between (30-40min each). 

Edited January 9, 2022 by Bruno

[J1mbo]

that might indicate that it could be worth increasing the speed of your radiator circulation pump.

[Bruno]

I can't really understand why. It's exactly the same setup, just changed a 8 to a 10. Rads heat up nicely, the temperature finally gets to the setpoint, don't know, the behaviour looks better now. There has to be something different inside the controller. 

Anyway today was again a not-very-cold day so the consumption curve is very pleasant. I'll keep following up 

[J1mbo]

If the radiator pump is running at significantly lower volume per hour than the ASHP circuit pump, water “short cuts” through the buffer and hence reaches (ands likely exceeds) the water leaving temperature target very quickly.
 

What set-back temperature is configured?

[Bruno]

The set-back temperature is very low, 12ºC or something like that. As I said, I want it to shut down outside the programmed heating hours ?

 

Regarding the circulation pump you might have a point there. I reduced its speed because of that noise I mentioned, but I can increase it again and compare.

Another thing I can try is to reduce the curve, as far as I could check the buffer temperature is stable enough but the consumption oscillates quite a bit, maybe it's overshooting.

But I need colder days to test it properly, now it's just too easy for the HP ?

[J1mbo]

Shorter running hours obviously results in a lower average property temperature and hence lower total heat loss in any given 24-hour period. That the property isn't being actively heated however doesn't mean it is not losing stored heat to the surroundings (which needs to be 'charged up' again in a future heating cycle). Hence, the cost to run the system doesn't necessarily reduce proportionally as one might expect when containing run-hours because the total thermal demand is serviced in a shorter window, implying higher flow temperatures and hence lower COP.

 

One of the Vaillant documents suggests (sorry I don't have the reference to hand) to avoid swings of more than 3 or 4°C IIRC. I will try and find that.

 

Re pump/water noise. It might be worth rebalancing the system. Start by fully opening all valves - you want the water circulating as easily as possible - and increase the pump speed (at least 2 if it's a three-speed pump). Then balance down (especially smaller radiators and those closest to the pump) until everything is even and hopefully achieving 5°C drop across each radiator. At least one radiator should be fully open both ends, otherwise overall the flow is being unnecessarily restricted. A smaller drop across each radiator increases the panel average surface temperature and hence reduces the ASHP flow temperature by increasing the panel output at lower flow temperature.

[ReedRichards]

3 hours ago, J1mbo said:

That the property isn't being actively heated however doesn't mean it is not losing stored heat to the surroundings (which needs to be 'charged up' again in a future heating cycle). Hence, the cost to run the system doesn't necessarily reduce proportionally as one might expect when containing run-hours because the total thermal demand is serviced in a shorter window, implying higher flow temperatures and hence lower COP.

 

 

Vaillant is one of the few makes of heat pump where this could actually be true.  Using the right control algorithm the heat pump will raise it's water temperature to raise the internal temperature more quickly, although having reached the set temperature it will then reduce the water temperature to maintain the set temperature.  But, as I see it, the majority of heat pumps determine their flow temperature either by the outside temperature or using a pre-set value.  So most heat pumps won't run less efficiently when the thermal demand is greater; they will just run for longer at the same level of efficiency (COP). 

[J1mbo]

7 minutes ago, ReedRichards said:

most heat pumps won't run less efficiently when the thermal demand is greater; they will just run for longer at the same level of efficiency (COP). 

Automatically perhaps. But the settings can be adjusted to lower flow temperatures with longer runtime.

Edited January 10, 2022 by J1mbo

[Dan F]

We finally have our aroTHERM up and running and it has output 226kWh (!) in the last 2- 3 weeks just "booting up" our house.

 

It'a not all fully commissioned yet, and auto-balancing actuactors are still in their boxes so I had some issues over the new year where ASHP seemed to be running non-stop at 100% compressor and wasn't able to increase UFH flow temperature more than 24C.  Installer hasn't been back yet, but I think I fixed it by i) reducing pump speeds (circulation + UFH circuit) to minumum and ii) adjusting each UFH to 1 L/min from fully open.

 

I have heated the slab to 23C and boosted MVHR to help get bedrooms from 17C to 19C and will now try to find a heat curve that will keep the slab at around 22C if possible (suggested by @jack somewhere).  As I think you suggested before @J1mbo, I can then adjust things, if needed, from Loxone by changing the target room temperature which will in turn shift heat curve over and in turn the flow temperature (e.g. to dump cheap energy at night).  We'll see how this goes though, as the heat curve will be based on temperature from one controller rather than temperatures averaged across all rooms.  Unfortunately there is no tested Modbus connectivty yet (even though Vaillant publicized this some time back) and eBus I'm trying to avoid as it's rather low level.  EEBus, which now has bult-in support in Loxone, is useful but rather limited in functionality and you can't input actual temperature or trigger cooling.  Control of cooling, however,  is a seperate exerce which I'll leave to the summer though.

Edited January 10, 2022 by Dan F

[Luke1]

14 hours ago, ReedRichards said:

 

Vaillant is one of the few makes of heat pump where this could actually be true.  Using the right control algorithm the heat pump will raise it's water temperature to raise the internal temperature more quickly, although having reached the set temperature it will then reduce the water temperature to maintain the set temperature.  But, as I see it, the majority of heat pumps determine their flow temperature either by the outside temperature or using a pre-set value.  So most heat pumps won't run less efficiently when the thermal demand is greater; they will just run for longer at the same level of efficiency (COP). 

 

Are you sure this is true?

 

The Nibe system works on degree minutes. In the settings there is an option to add additional heat (immersion) when the degree minutes exceed 700 or another defined number. 

I suspect other systems are similar, hence why so many people complain about running costs when they try to use an ASHP like a conventional boiler.

 

I also note that the Nibe system will try and run the compressor until the degree minutes hit 0. 

If the system is started after a period of blocking then the flow temperature will continue to increase (reducing COP) until either the maximum flow / return temp is reached or it hits 0DM. It will then oscillate at the higher end of the flow temperatures until it reduces the DM to 0.

 

Edited January 11, 2022 by Luke1

[ReedRichards]

35 minutes ago, Luke1 said:

 

Are you sure this is true?

 

 

No, I only really know about the LG heat pump I have myself; I know it is true of that.  This does not have "degree minutes" nor a supplementary (immersion) heater.  I also have a third party controller so the heat pump has no way of knowing how hard it should be working because it doesn't know what temperature I have set the room to be at.  It's a pretty safe bet that anybody else using a third party controller will be in the same boat.  Before buying I looked at Grant heat pumps also and they seemed to be the same as mine.

[J1mbo]

The Vaillant controller also uses degree-minutes (by default compressor starts at -60°, stops at 0°) to regulate flow temperature and really comes in to effect when the ASHP minimum output is greater than the demand, i.e. the set point cannot be maintained and the ASHP over-shoots. In this case, the heat pump cycles around the set point using the degree-minutes metric to average out the flow temperature to the set point over time. This is all done under the control of the heat pump interface, which is a box of electronics that also controls three-port valve, circulation pumps, immersions etc.

 

Separately, the Vaillant system uses another device - VR700, VR700f, or SensoComfort - which is a room-stat and programmer which is cited somewhere in the living space because it includes a room temperature sensor. It computes the desired flow temperature based on outside temperature, inside temperature, set-point, configured curve, and min/max thresholds, and sends the required temperature to the heat pump controller, which deals with running the ASHP itself.

 

In relation to the COPs, run-time and curves: the VR700 (etc) will reduce the flow temperature as the room temperature is exceeded (and vice-versa) by shifting the curve up and down (actually 45° according to the manual). It can also self-tune the curve over time but the curve needs to be approximately right in the first place for it to do so.

 

The original point I was making is that the curve setting (or max/min temps on other controllers) aren't just dependent on the property but also on the number of hours per day that the heating is used and can be adjusted accordingly. Sorry for the confusion.

[Dan F]

7 hours ago, J1mbo said:

The Vaillant controller also uses degree-minutes (by default compressor starts at -60°, stops at 0°) to regulate flow temperature and really comes in to effect when the ASHP minimum output is greater than the demand, i.e. the set point cannot be maintained and the ASHP over-shoots. In this case, the heat pump cycles around the set point using the degree-minutes metric to average out the flow temperature to the set point over time. This is all done under the control of the heat pump interface, which is a box of electronics that also controls three-port valve, circulation pumps, immersions etc.

 

Separately, the Vaillant system uses another device - VR700, VR700f, or SensoComfort - which is a room-stat and programmer which is cited somewhere in the living space because it includes a room temperature sensor. It computes the desired flow temperature based on outside temperature, inside temperature, set-point, configured curve, and min/max thresholds, and sends the required temperature to the heat pump controller, which deals with running the ASHP itself.

 

In relation to the COPs, run-time and curves: the VR700 (etc) will reduce the flow temperature as the room temperature is exceeded (and vice-versa) by shifting the curve up and down (actually 45° according to the manual). It can also self-tune the curve over time but the curve needs to be approximately right in the first place for it to do so.

 

The original point I was making is that the curve setting (or max/min temps on other controllers) aren't just dependent on the property but also on the number of hours per day that the heating is used and can be adjusted accordingly. Sorry for the confusion.

 

Interesting, do you know if there is anything online somewhere that explains the degree-minute concept?

 

Out of inteest what heat curve do you use,  and do you have i) adaptive enabled or disabled ii) room temp modulation avtive or expanded?

[J1mbo]

Degree minutes is the cumulative under or overshoot from target.

 

I visualise it like this. Say the target is 40* and it’s producing 40* and as it happens the controller is reporting -30 degree minutes. Then the flow temperature creeps up and is 41* for a minute. The controller will now report -29 degree minutes.

 

Say an additional zone was activated for heating. A load of cold water enters the heating circuit before the heat pump can respond and the flow temperature reduced to 30*. Now it’s subtracting 10* per minute that this condition holds, whilst increasing compressor output to restore the flow temperature to the target.

 

Backing it off again is probably slow and likely it overshoots, adding to the degree minutes again until either 0 is reached or the flow returns to the set point.
 

If you watch it (not the most interesting of activities) it tends to clock up the changes more quickly, but the above is the gist of it.

[J1mbo]

1 hour ago, Dan F said:

Out of inteest what heat curve do you use,  and do you have i) adaptive enabled or disabled ii) room temp modulation avtive or expanded?

0,65, I think adaptive disabled and room temp mod enabled.

[Bruno]

On 11/01/2022 at 20:36, J1mbo said:

0,65, I think adaptive disabled and room temp mod enabled.

Wow, 0,65? That's really a low, low curve. It gives you under 40ºC flow temperature at 0º outside. Dang ?

 

Actually this temperature curve is what's making me really confuse. As I said, I run the AHSP mostly during the day+1st hours of the evening. Now we are having some colder days but very sunny. So during the day we have some 14-17º. As soon as the sun goes away it drops fast to 7-8º and lower during the evening.

Even if my curve is at 2.0, according to the graph the target flow temperature is under 40º during the day, so the consumption is low but the house doesn't heat (setpoint 21,5 - room temperature eg. 18,5º). Only when the sun goes away and the outside temperature drops, then the flow temperature rises and the house gets comfy. Here is yesterday's consumption graph, it's clear that the consumption rises greatly when the night falls. I would rather having it opposite, because

a) higher outside temperature means higher COP

b) when we return from work we would like to have it warmer already AND the >3,3kW that the HP is consuming sometimes force us to manage other electric devices which we might need to use. When we are not at home, then the AHSP can use whatever energy it requires.

The area between the red lines is DHW.

 

I might consider testing it working 24/7, but if I'm thinking this correctly, it will just shut off during the day (which, for radiators, is even worse).

[Bruno]

An update.

 

We finally got constant low temperature especially at night so I've been messing around with the parameters. And it keeps getting better consumptions, but it still needs further testing.

For the last week or so I've been running it 24/7, limited the max temperature to 55º and minimum to 25º, also disabled "thermostat" and it's now in temperature modulation. It still shuts down for most of the day because of the misplacement of the temperature sensor catching direct sunlight - will fix that shortly, today at luch time we had 12º outside, sensor was measuring 25º ?.

On 08/01/2022 at 14:31, J1mbo said:

Really the system needs to be designed so a curve in the region of 1.1 is used which will yield 50* at freezing outside when room temperature is set to 20*.

And I played with the curve a lot. Last night I was able to get the curve down to 0.9, consuming less than half than the night before ? In nights with 1-3º outside it was consuming average 2kW and it even did several defrost cycles (at least I think that's what I saw in the energy meter), last night it averaged slightly under 1kW . In the morning there was no temperature loss in the room so the low flow temperature is enough. Rads rock ??

 

For now I just need to get the external temperature sensor's placement fixed and set a higher daytime temperature so that it boosts the house a bit more during the day, while we have sunny days.