Buffer tank and secondary pumps. Do I need them?

[jimseng]

Hello.

I don't know what more information is needed to guide me on this but I have been trying to learn about my installation before it takes place and after having watched quite a lot on line (dangerous for someone ignorant as I am) I am wondering if I really need a buffer tank and secondary pumps.

I am half way through a new build of a 158 sqM house which is designed to be very efficient (MVHR and masses of Warmcell insulation)
I have been quoted for a Samsung Gen 7 R290 ASHP, the data sheet says a min flow rate of 7 L/m and max flow rate of 48 L/m and the design flow temp for my system is 35 deg.
There is UFH on both floors with a design flow rate of 7.3 L/s (gf) and 12.1 L/s (ff) and I am not planning on doing any zoning.
The ASHP quote specifies a 30L buffer and secondary pumps so I am assuming that is one pump per manifold. I have seen it stated that buffer tanks are not always necessary and I am wondering why this is, why I have been quoted for one and whether I am fussing about nothing. I'm not looking to shave extra pennies off the installation but I am obsessing over getting the most efficiency out of my system and I am averse to "that's the way way always do it" approach.
There is a possibility that the 30l tank is an expansion vessel and not a buffer tank. If so that would leave the question why do I need secondary pumps.

If it is a buffer tank then I am wondering if I should consider secondary pumps with controllable flow rates rather than set and forget.
Happy Christmas by the way.

[JohnMo]

29 minutes ago, jimseng said:

buffer tanks are not always necessary

Generally never required with sensible design.

 

30 minutes ago, jimseng said:

Samsung Gen 7 R290

What size?

 

30 minutes ago, jimseng said:

I am not planning on doing any zoning

Good.

 

So basically the design should look something like this. Two floors if the same design flow temp no mixing should be needed. If one floor needs a higher flow temp some mixing may be required, this should be fine with an electronic mixer not a mechanical one. But if well insulated upstairs should really need much or any heating. So I would be tempted to flow a single flow temp to both floors and balance as needed with flow rates.

 

So next is system capacity. You will need at least 20L per min kW output of the heat pump. If you don't comply with that add a volumiser on the return line prior to tee coming back from cylinder.

 

Then run open loop from ASHP on weather compensation.

 

Alarm bell - if you are bigger that 6kW, find a different installer.

 

Alarm bell installer talks about a buffer, find a different installer.

 

You want an efficient system you want a heat pump that modulates to at least half your heat loss at -3, then the heat pump ticks away all day at 7 Deg outside and maybe above that. Would suggest you heat loss is is in the 2.5 to 3kW range from what you have said. If so a 4kW would-be good.

[Nickfromwales]

Hi.

 

Something’s not right if your design flow temp is as high as 35°! GF UFH, I assume, is going into a slab? If so, 35° into that will cause the house to overheat. You should be mid-high 20’s and certainly under 30°, if well built, airtight, and you have heat recovery too. 

 

Buffer “delete” may be subject to the manufacturer’s installation instructions, and an MCS reg’d installer may refuse to stray from that. On a recent Heat Geek install I witnessed they put either a 25L or a 50L 2-port volumiser in, on the flow which makes more sense to me, as that was stipulated in the MI’s vs what the low heat demand called for. That was a refurb, prob got to EnerPHit levels, and the flow temps at 25° made the place almost unbearable to work in. 

 

@jimseng  As we don’t know much about your particular project we can’t possibly size a heat pump here yet, but 3.5-4kw will be quite slow to reheat a decent sized UVC; your game plan needs thought which considers your DHW needs / frequency / number of occupants.

 

Heating is a doddle to resolve, DHW needs to be considered in isolation and planned for pragmatically. You can, for eg, use an immersion strategically to boost for times of duress, so many ways to ‘make this happen’.

 

Do you have solar PV in the plan?

[jimseng]

It is a 5kw heat pump. PWM flow control.
According to the UFH designs the pipework length is 920m for the whole house. With 16mm pipe / 12mm ID I make that 104L.

The spec for the heat pump states a  flow temp of 15 - 75 deg.

Is it possible the "buffer tank" is in fact an expansion vessel, or does the fact there are secondary pumps in the quote mean it is in fact a buffer tank?

I'm going to go back to the installer and get more information, this is all useful to me. So the main point of the question is do I need a buffer? The answer seems to be no.

The second is do I need secondary pumps if I don't have a buffer. Are they specified because of the buffer?
The UFH on the ground floor is already installed and the screed is down. (150mm insulation, 75mm screed)

The FF heating is 22m routed boards and 16mm pipes, that won't be going in for some months.

The DHW is possibly for a different conversation although FYI I have opted for a 300l cylinder, despite me being a single person with no kids. It is a house suitable for a family. Yes solar, yes immersion strategy.
 

[SimonD]

2 minutes ago, jimseng said:

Is it possible the "buffer tank" is in fact an expansion vessel, or does the fact there are secondary pumps in the quote mean it is in fact a buffer tank?

 

Yes the addition of the secondary pumps indicates that it is a buffer. I'd be curious to understand what the designer's pressure loss calcs are for the pipework index circuit (the circuit with the most pressure loss)and whether they're worried the heat pump has enough residual head, whether they've specified the buffer because of different flow rates between gf and ff, or whether they haven't thought much about it at all and/or are compelled by some manufacturer requirements. 

 

Although 102l open volume is okay, I would personally look to add in a 50l volumiser on your system at least.

[JohnMo]

14 minutes ago, jimseng said:

Is it possible the "buffer tank" is in fact an expansion vessel, or does the fact there are secondary pumps in the quote mean it is in fact a buffer tank

If they are quoting a buffer it isn't an expansion vessel.

 

A buffer requires secondary pumps, a volumiser doesn't. A buffer provides hydraulic seperation and therefore needs a pump(s) for the heating circuit.

 

Samsung Gen 7 schematic don't compel the use of a buffer, if your are doing an open system. They show a volumiser on the flow to heating circuit.

 

14 minutes ago, jimseng said:

300l cylinder

Cylinder size is basically worked out as follows, 45L per bedroom, plus 40L.

 

Reheat time assuming 300L and a 5kW output will be around 1hr and 20 mins to 50 degs, assuming bulk average temperature is about 30 degs.

 

 

[jimseng]

Quote

I'd be curious to understand what the designer's pressure loss calcs are for the pipework index circuit (the circuit with the most pressure loss)and whether they're worried the heat pump has enough residual head

This is probably my fault but after going through various options I elected to have the UFH designed and installed by a different company to the ASHP supplier. But this is where I am, it's too late to change that now. But I can get the ASHP design right with the right knowlege.

Quote

or whether they haven't thought much about it at all

This is the bit that concerns me. The worst thing I find is those who say "That's the way we always do it". 
So to summarize: I am gleaning from this thread that, given that I want to achieve the most efficient system, I don't need or shouldn't have a buffer or secondary pumps? To me this leaves the flow rate up to the heat pump control system which, from the little I know as of now seems like the best option. I'm going to go back to the installer in the new year and query this and, to be fair,  they are responsive, if a little dismissive of my technical questions. I need to be in a reasonable position to argue the case if I am to change the system design. 

As for the DWH cylinder. I chose a large one and have not had any push-back on it being over-sized for a 3 bedroom house. Yes it is big but to me, and my need for deep, long baths I would be furious if I installed one too small. I would never get over it.

[SimonD]

48 minutes ago, jimseng said:

This is probably my fault but after going through various options I elected to have the UFH designed and installed by a different company to the ASHP supplier. But this is where I am, it's too late to change that now. But I can get the ASHP design right with the right knowlege.

 

No, not your fault at all. The ASHP supplier should be doing at least a basic calculation of the index circuit of the UFH as that will help them to design the ASHP system properly. But I assume from what you say that the people who installed the UFH have installed just the UFH manifolds without mixers and pumps on the manifold?

 

51 minutes ago, jimseng said:

So to summarize: I am gleaning from this thread that, given that I want to achieve the most efficient system, I don't need or shouldn't have a buffer or secondary pumps? To me this leaves the flow rate up to the heat pump control system which, from the little I know as of now seems like the best option.

 

Yes, absolutely. The simpler the better, but for a 3 bed house, there should be no need for any hydraulic separation or additional pumps with a system that runs low flow temps straight into the UFH.

[jimseng]

Quote

But I assume from what you say that the people who installed the UFH have installed just the UFH manifolds without mixers and pumps on the manifold?

Yes. Just the ground floor so far with a manifold and nothing else. 

Quote

basic calculation of the index circuit of the UFH 

Can you explain what that means? I am finding that I have to double check everything the "qualified" professionals tell me because it seems no one cares nowadays. I'm happy to do this myself if I have the means, I really want to get this right before it goes in. 

These images may be of interest, maybe not. 

[SimonD]

4 hours ago, jimseng said:

Can you explain what that means?

 

Of course. The index circuit is the part of the heating circuit that has the greatest pressure loss. So each meter of pipework, each fitting, like elbows, and for each valve the water in the heating system flows through there is a loss of pressure due to resistance. If you start at the outlet of the primary circulating pump of the system, often now inside the heat pump itself, you will have an available head pressure (usually specified by the manufacturer where they often provide graphs for how this changes according to flow rates) and this is what is gradually lost as the water travels through the system. The index circuit is just the total length of pipework that has the greatest resistance. Once this is known, you know whether the pump for the main circuit produces enough pressure to supply the heating circuit at the required maximum flow rate. This is what should primarily tell you if you need hydraulic separation and secondary pumps. 

 

The total resistance of the circuit is done by calculating the total heat load that is carried though each section of pipe from the heat pump to the end of the circuit along with using pipe diameter and resultance resistance of the pipe together with fittings. This should be done with every heat pump design.

 

5 hours ago, jimseng said:

These images may be of interest, maybe not. 

 

Yes, these are useful. There are already some discrepancies here. Earlier you said the design temperature of the system was a flow of 35C but the data re the ufh loops says the design temperature is 45/40. The tables each state the output of each manifold, being 4kW and 4.3K which is obviously more than the 5kW unit. I don't know the Samsung units so don't have the technical specs but you need to know what the heat pump is capable of outputting at your outdoor design temperature, which should be given in your design from the heat pump supplier. The tables suggest, I assume, that each manifold has less than 25kPa pressure drop across the manifold flow and returns whcih equates to about 2.5 meters head, but nothing more specific. However this can be calculated more specifically as the relevant data is there in the tables.

 

What is the estimated heat loss calculated for your house?

[jimseng]

Quote

What is the estimated heat loss calculated for your house?

Assuming the heat loss column is in watts then 3320. I think that tallies with the SAPS data, although that is reams of info and hard to decipher.
I'm very grateful for you taking the time to explain this. 

Quote

I don't know the Samsung units so don't have the technical specs 

I have not managed to find anything detailed for this unit yet, just a single page data sheet, but I am looking, in between the cooking my nieces and nephew expect of me. And if the regurgitated potato the dog left for us on the floor this morning is anything to go by she needs to go to the beach. (She has since re-eaten the potato).

[JohnMo]

56 minutes ago, jimseng said:

cooking my nieces and nephew

That's a strange lunch your cooking up. 

[Marvin]

Most efficient way to run an ASHP is low and slow: low temperature, which will mean longer to heat up. For a simple understanding: the less difference between the out side air temperature and the temperature of the heated water leaving the ASHP the more energy efficient the ASHP is running.

 

We have ours running using weather compensation mode. At this point the outside temp is 4.6C and the ASHP is running at 40.1C. However, we are using all radiators with buffer tank and secondary pump with home temp ranging from 22 to 23 C.

Edited Thursday at 10:46 by Marvin
further thoughts

[jimseng]

1 hour ago, JohnMo said:

That's a strange lunch your cooking up. 

The threat keeps them quiet. I expect the older one is a bit chewy. 

[joth]

48 minutes ago, Marvin said:

Most efficient way to run an ASHP is low and slow:

Not to contradict this, but this is specifically true for energy efficiency.

Some folks talking about efficiency actually want running cost efficiency. With TOU tariffs, in a low energy home, running the heat pump fast and furious for 5 hours of overnight cheap rate electricity can be much lower cost than the low and slow mantra has you believe. So some consideration for higher flow temperature and potentially controls to avoid specific room overshoot can be helpful if chasing low running costs.

 

All that said, to the OP this doesn't really impact the choice of a buffer tank or volumizer, such items are less necessary for short cycling reduction if running the ashp hard, and don't help at all in effective TOU shifting of demand. (An oversized UVC however can be useful if chasing low cost tariffs)

 

 

[JohnMo]

55 minutes ago, joth said:

running the heat pump fast and furious for 5 hours of overnight cheap rate electricity can be much lower cost than the low and slow mantra has you believe

Unless you couple that with a battery. Then low and slow wins.

[jimseng]

Low and slow for me. Minimum amount of electricity consumed for maximum comfort. What comfort actually means will only be established once I move in but "efficient" for me is highest cop. 

[Marvin]

4 hours ago, joth said:

Not to contradict this, but this is specifically true for energy efficiency.

Some folks talking about efficiency actually want running cost efficiency. With TOU tariffs, in a low energy home, running the heat pump fast and furious for 5 hours of overnight cheap rate electricity can be much lower cost than the low and slow mantra has you believe. So some consideration for higher flow temperature and potentially controls to avoid specific room overshoot can be helpful if chasing low running costs.

True @joth

 

I forget we have PV and radiators (not a thick concrete slab for massing) which makes it a different calculation! 

 

I think that, just as with EPC's, there is no one-design fits all and that specific property equals specific design. 

[joth]

4 hours ago, JohnMo said:

Unless you couple that with a battery. Then low and slow wins.

Even with a 10kWh battery we find Fast and Furious ashp program is still a cost saving in winter, because the battery is fully used time shifting the rest of our demand. (2 people working from home, with compute heavy work). For sure a larger battery would change that, but it's not a good capital investment for just a couple months a year. 

YMMV, but designing to work low and slow is surely the sensible starting point 

[JohnMo]

39 minutes ago, joth said:

Fast and Furious ashp program

How do you do your schedule?

[SimonD]

On 25/12/2025 at 08:23, jimseng said:

Assuming the heat loss column is in watts then 3320.

 

If that's the case, then your output from the UFH needs to be around 21W/m2 assuming the 158m2 area is floor area. Very roughly, in screed UFH would output this at a mean water temperature of around 27C or a flow temperature of just under 30C at outdoor design temperature - this would vary depending on actual buildup and floor coverings, like if you decided to have carpets and wood floor etc. Do you know if any of the designs took into consideration the floor coverings you have? And do you know whether you have an in screed ufh throughout the house on both GF and FF?

 

If does look like you have a situation where the design doesn't quite hang together based on the information we have.

 

With your DHW, as @Nickfromwales and @JohnMo have pointed out, your heat up times with the chosen cylnder could be quite long depending on usage and this is something you need to discuss with the designer/installer to get the system sized correctly to work for you. 

 

[jimseng]

Quote

And do you know whether you have an in screed ufh throughout the house on both GF and FF?

Yes, based on my guesses of how the house would end up. Although I don't have the budget for floor coverings. If I can get to a plaster finish that would be a big achievement with the money I have left at the moment.

Quote

If does look like you have a situation where the design doesn't quite hang together based on the information we have.

Can you be  a little bit more specific? I'm going to go back to them in the new year and get more details of what they are planning, now I know a little bit more. The installer originally quoted for doing the UFH install too but it was my decision to go with someone else. I don't know if that was a mistake but it seemed the cheaper route at the time. The ground floor has been screeded and will probably be a tile finish. The first floor UFH is routed boards with probably a wood and rug finish, although it might end up mostly carpet a few years down the line.

[SimonD]

14 minutes ago, jimseng said:

Can you be  a little bit more specific?

 

You have a UFH system that appears to be based upon flow/return temps of 45/40 with a total output of 8.3kW versus your heat loss of 3.3kW. Your heat pump supplier is then suggesting a flow temperature of 35C which would overheat your house with the current UFH design with a 5kW heat pump and you have selected a 300l dhw cylinder. You don't seem to have been informed if this 5kW output of the heat pump is at design outdoor temperatures, or the implications of the large cylinder and heat up times with the current proposed heat pump. And then not to mention the proposal for a buffer and secondary pumps.

 

For the actual UFH output there does need to be a calculation based on proposed floor coverings as they have such a large impact on the system output - one of the two companies should have had this discussion with you.

 

I'm also curious about the figures given in the UFH design tables. For example, your dining room says the ouput is 660w with 150mm spacings and a length of 92m and flow rate of 1.9l/m and your entrance hall/wc has exactly the same output but with almost half the loop length at 57m and exactly the same flow rate. I would expect to see some variation in flow rate and/or output with loops that are so different in length but have the same spacing.

[-rick-]

3 hours ago, jimseng said:

Although I don't have the budget for floor coverings. If I can get to a plaster finish that would be a big achievement with the money I have left at the moment.

Given money is tight I just wanted to check that the quotes you are getting are coming out at zero cost to you? ie, the heat pump, hot water tank and install are all covered by the £7500 BUS grant. Given that you already have the UFH installed the BUS grant should cover the rest.

 

[jimseng]

Quote

'm also curious about the figures given in the UFH design tables. For example, your dining room says the ouput is 660w with 150mm spacings and a length of 92m and flow rate of 1.9l/m and your entrance hall/wc has exactly the same output but with almost half the loop length at 57m and exactly the same flow rate

That is a good point, but perhaps it is due to the pipework layout as the majority of the feeds pass through the hallway? 
The 35c flow temperature was based on what I told them many months ago, but if I set a lower flow temp that is a good thing is it not? But perhaps the implication is a smaller heat pump. 5kw is what was specified by the SAPS calculations with a flow temp of 55. I specified 35 after conversations with various heat pump suppliers. 

 

The quotation for the installation came in at about £2000 including commissioning after taking out the BUS grant. I have signed the contract so I am where I am, although the ASHP supplier is very approachable and only the GF UHF is fixed now.
I don't have to run the UHF at 35, I can set any level and I intend to collect lots of data and learn how to get the most out all my energy demands and sources.