ASHP with UFH advice sought

Well that was easy and I will now have another go!
First the context.
The property was built in 2010 and the insulation level is to modern standard. It has a system boiler supplying a simple ring main that supplies 7 separate circuits, a HW and a small rad circuit and 5 separate U/F heating manifolds located at various remote locations, which control 18 separate circuits. The Polypipe heating manifolds work well, but the control system was never that good when it was installed and so the control system has now been undated. The simple controller has been replaced with 7 Optimum wi-fi controllers, which work well and enable separate control of the H/W & Rad circuits and the 5 U/F heating manifolds with separate temperature control of each of the 18 U/F zones. The property also has 8.5kW of solar fitted, an original 4kW fitted in 2015 and an additional 4.5kW with dc batteries last summer. I also have a Solar iBoost fitted which did work well, but is incompatible with the dc battery system so will be changed to a Powerflow unit before the spring. So very broadly I am now reasonably confident the solar together with batteries to time shift demand, will meet the majority of the property’s hot water and heat demand from April to September. The property is also fitted with a number of air-con units and since October 2 of the units have been very successfully operating for 22 hours a day which has significantly reduced the gas demand.
The property was designed in 2009 and heat calculations were prepared and I have now had 12 years of experience and the heating system is certainly over sized. I have now purchased a 16kW Samsung ASHP (AE160RXYDEG), but even with the air con units that are fitted I have decided to split the heating system and keep the existing gas boiler to feed the H/W and small rad circuit at 60 degrees and use the ASHP to feed the 5 U/F heating zones at 35 degrees. This should be straightforward, but after a considerable review of the ASHP market, it appears there is no clear settled view on what is the ideal or optimum approach.
So the question that I would appreciate any advice or comments is there appears to be three general approaches that various suppliers and installers suggest:-

  1. A 4 port buffer tank to create a hydraulic separation between the ASHP and the heat circuit, whilst this is not a true separation, it deals with the volumetric issue due to potential loss of system capacity when demand reduces and 2 port valves close. In addition it also creates a heat sink that is used during the de-frost cycle. I am reasonably convinced that this approach is probably the best option and I have already purchased a 100 litre buffer vessel.
  2. There is also an opinion that whilst it agrees that a buffer vessel is necessary, the turbalence that is created in the buffer vessel by mixing causes a temperature step change and this will have an impact on the thermal efficiency of the process and lower the overall COP. This view is supported by a paper on the subject from Ulster University.
  3. The third view is a buffer vessel is not needed as modern ASHP’s can modulate the output to better match the heat demand.
    Also and whilst I don’t want to fit at the moment, it is obvious that continuously monitoring the flow temperature in and out of the buffer vessel would be useful and tracking this information and comparing with COP data. Having looked at the various impressive data on the forum, this is certainly something I will look at once the new Samsung unit is operating.
    So in conclusion before I complete the installation any experience, advice and in particular anything associated with a recent installation of latest Gen 6 Samsung unit would be much appreciated.
    And also a thank you to Glyn Hudson, your YouTube tutorial was very useful, I was already aware of the issues associated with glycol and was aware of antifreeze valves, but your experience helped me to make the decision to fit antifreeze valves.

I’ve moved this to it’s own topic as it isn’t specific to

1 Like


I have a Stiebel Eltron WPL 25AS heatpump and associated HSBC200 hot water and buffer tank system. It’s all integrated together, and whilst I’m not an ASHP expert, it’s use of the buffer tank (which is 100 litres) does seem to make the system pretty robust and smooths out temperature shifts as the heatpump kicks in and out.

The Stiebel system has ethernet modbus available, so I can examine this for things like flow rate, flow and return temperature etc, as well as hundreds of other parameters and readings. It enables me to calculate CoP on-the-fly, as well as gather cumulative data, and I find that the instantaneous calculation of CoP using this information is both enlightening and reasonably accurate (at least in terms of variation, if not absolute values).

I’d be happy to share more data on how my system performs if you’re interested.



Hello Andrew,

Thank you for making contact and the offer to share data on your system.

The main question at this point is how a buffer tank affects the performance of the system and in particular the CoP.

So could you please advise how your operational CoP compares with Steibel Eltron’s quoted figures?

There is huge amount of information available on the web about heat pump system design and a lot of examples of poor design where people are only achieving a CoP of 2 – 2.5.

Thank you again for the contact, I look forward to your response and my current aim is to install my ASHP in the next couple of weeks.

Best regards

Robert Kennedy

07721 645208