Without knowing the exact details of your system, how it has been installed and configured, it is difficult to provide a definitive answer.

Possibly the best way to operate your system would be to have the 'Master' heat pump solely for CH, with the 'Slave' looking after DHW and assisting the 'Master' with CH when needed.

Probably the best way to operate such a system would be to have the slave heat DHW during late afternoon, when the master may be having a rest, particularly if you also have solar PV. An alternative, though less efficient, method would be heating DHW on a cheap overnight tariff.

Due to those awful Laws of Physics interfering again, 🙄 there will probably be an operating point at which it is more efficient, and probably cost effective, to run two heat pumps at say 37.5%, rather than one at say 75%. I cannot say for certain, but it is possible that this could occur when frequent defrosting rears its ugly head. 😡 

How such a changeover could be achieved in practice, I suspect would be dependent upon the capabilities of the controller.

@derek-m Thanks for your input. I'd been mulling over the best configuration option for the master and slave and wondered about reversing it to make the slave do DHW. I'll seek a view from the installer on this.

Yep, my own observations (like everyone else here really) indicate that efficiency heads dramatically south as soon as you're into defrost cycle territory, although curiously, it's not always linear. Dry and freezing may well be better than 2°C and wet. My installers advice was to set the WC up such that the highest water temp corresponds to +2°C ambient, as because of the dew point, it's better to do this than the more oft-quoted -2°C set point. 

I've resurrected this thread to pose a couple of specific questions to this august body.... 

I have two 12kW Samsung Gen6 heat pumps operating on a master/slave basis. The master providing heat and DHW, the slave heat only. Both connect to a buffer tank to feed my single zone radiator system. Relevant 402* settings are 4021 Use (BUH1), 4022 Both, 4023 Use, 4024 2°C and 4025 15°C. Effectively, the slave should only be called in at temperatures below 2°C and (I think) when the master is struggling to reach set point after a certain interval. Both pumps have identical WC settings. 

The slave has been called into use several times in the last few weeks. Last week, with OAT circa -1°C and wet, the master was entering defrost cycles every 40 mins. Each time it starts a defrost the backup/slave is signalled to stop. The consequence of this is that the slave, with it's own defrost cycle to deal with as well, ends up cycling on/off a fair amount and (whilst it helps) never allows the LWT to climb anywhere near the target. At -1°C on my settings LWT should have been about 47° but typically it can't get above 42 ish before this cycling causes temperatures to drop, a delay whilst the master builds LWT again after it's defrost and then calls for the slave again...... repeat etc. The system seems to be working hard but this doesn't feel efficient. So, two questions...

1. Is it possible for the slave unit, once called for, to continue to run whilst the master is undergoing a defrost? If so, is that something I can configure in the field settings?

2. A comms link cable was recently added between my master and slave controllers to enable this type of back-up operation, but is it possible to revert to the units operating in tandem (for heating) with this link still in place? Is there simply a field setting change I can make to do this? It'd be great to easily switch between master/slave and tandem running for really cold spells. 

I'd appreciate any help from anyone with two-pump systems or specific Samsung knowledge. I've had a few problems with my system since installation (now resolved) but I'm not confident about making changes and the manual is so hard to fathom it's no help at all!

Thanks

Simon

@dunlorn Given your current observations on your. system I would be very tempted to increase the WC settings on the slave unit 1 or 2 degs higher than the master to make it take more of the workload when called into action.

Thanks @IaAck, I'll definitely give that a go although I still want to have the option of occasional tandem running if I can sort that. 

Posted by: @dunlorn

↑

Thanks @IaAck, I'll definitely give that a go although I still want to have the option of occasional tandem running if I can sort that. 

Could you explain the reason for wanting the heat pumps to run in tandem? I would have thought that if the actual heating demand, is not greater than the combined minimum thermal energy output of the heat pumps, then this would lead to cycling of one or both pumps.

Also it would appear that heat pump efficiency varies with loading as well as LWT and OAT, so having one heat pump running in mid range may be more efficient than two heat pumps running at low output.

Hi @derek-m

Your question identifies the very reason I want to run in tandem, essentially to determine whether (and when) it might be preferable to run like this compared to the effective 'duty-assist' operation.

I live in the Cairngorms and - if this December/January is anything like last year - we'll have many days <0°C and down to -10°C at night. 45% of our annual electricity consumption was in these two months and I'm trying to get in a better position this year. I was wondering whether tandem operation might make sense over this period, effectively having 2 units running mid-range rather than 1 flat out with another cutting in and out.

I've never seen any heat demand calcs but I'm trying to assess it for myself based on observing heat pump generation in stable, straight line conditions on my Hive. Today, at around 2-3°C OAT, the house needs 7kW to keep at 18.5 for example. I'm slowly building a picture. 

@dunlorn Me again.

We too have twin ASHPs. 14kw Ecodans, installed in 2011. I suspect these are now oversized, as we have improved the insulation of our house. These are managed by a step controller which selects either one at a time, or when needed, runs both in parallel, when more efficient to do so or when 1 asap alone cannot meet our heating load. For instance, if we have high demand for hot water, it will run the 2 together. Neither heat pump is master or slave. The controller alternates usage to split the load equally between the 2. 

Each heat pump has its own flow and return pipes and own set of circulation pumps. The 2 flows join just before entering the house.

Most of the time, only 1 asap is on at a time. The 2 work in parallel in extremes.

The controller utilises 2 Mitsubishi FTC 2s as base controllers operating under parameters set by the step system. It was a bespoke set up designed and installed by Ice Energy ( no longer trading ) before Mitsubishi had their own twin ashp controller set up, which does use one as master, one as slave as I understand it.

Our house is open plan and quite large. Parts are solid walled. Annual usage around 24000kWh per annum , inc charging our EV.

One other thing, I have removed the buffer tank, having found this was causing considerable inefficiency just keeping the buffer tank up to temp. The Ecodans are inverter driven so start up without additional power loss and therefore cycling not an issue. Ecodan Service Engineer advised the buffer tank removal.Our system is also large enough to provide enough start up heat to be stored in the flow and return pipework alone, as a sort of mini buffer if you like.

@alanb thanks. Your set-up sounds pretty impressive with the step controller. I just have the 'master/slave' or full-on tandem running possibilities and it's the latter I want to be able to switch to for extremes, even though this will inevitably need to be a manual intervention.

I'm sticking with my buffer tank, for now at least, although I do lose 1-2°C of heat across it.

Regards 

Simon

I have my Ashp due its service and have  2 10kw Aerona and it seems a similar set up to yours (1 pump for heat and 1 pump for water and heat).

The last time the Grant engineer came out he altered something to make it more efficient but can't tell you what he did.

What I don't understand is that when the pumps call for hot water the heating goes off where common sense would tell me one pump would continue the heating and one concentrate on heating  the water.

I'll be asking when my next service is due, and paying more attention.

Hello @bretix

Yes, it's been a learning experience  for me with Samsung control. 

My system did operate in tandem but I knew from experience I didn't need a second heat pump for 90% of the time. For me, a single 12kW unit keeps us cosy down to temperatures around 2°C (and sometimes lower if it's dry and still) so I had the control for the second HP configured so it would come in to support only at temps below 3°C. Samsung's configuration allows for a user defined temp threshold to be used. 

I had it operating in this way all last winter but it wasn't satisfactory. Essentially because Samsung's control didn't just activate with this threshold external temperature - firstly it 'waits' for the 'master' unit to attempt to reach target temp, then when the master HP goes into defrost mode, the 'slave' switches off too, and the same when the master goes into DHW mode. So, at the very time when you really need the support of the second unit, it switches off! There's then a delay whilst the master unit attempts to reach target LWT before the slave restarts! End result is cycling and increased consumption for very little benefit.

I presume, like me, you have a buffer tank where HP flows combine? If so the other issue to be aware of relates to the control and capacity of the secondary side pump. When my slave HP kicked in, it obviously doubled the flow rate into the buffer. However, the secondary pump control is not linked to this AND none of the 3 manual settings it allowed get me anywhere near this doubled inlet flow. So, with the secondary flow to the rads much lower than primary inflow, heat is not transferred well, the buffer tank is effectively 'overwhelmed', RWTs rise and the units cycle. All of the above means the impact of having a second HP is pretty marginal and ends up burning kW for very little benefit to house temperature. 

This year I've had further modifications made and I'm hopeful this will lead to better control and performance. These are...

1. Change control of the second HP so that it gets a direct run signal (from the Hive room thermostat in my case) albeit via a manually operated switch. The heat pump remains energised at all times but it only can receive a run signal when I flick the switch on. A modest, and fairly infrequent, manual intervention that I can operate in response to weather or (more likely where I live) turn on at the start of December and off in February!

2. The second change has been to swap the secondary side pump for a higher capacity unit. Again, this is manually controlled (three settings) but the flow rates better match the range of inflows. So, as I flick the switch to bring HP2 on-line I will also increase the secondary flow rate. This should mean less cycling, less loss across the buffer and more kW into the house when needed.

Good luck on your optimisation journey!