Can you clarify what the problem is that you are trying to solve by balancing, whats the objective/currently 'broken'.  As an aside (or perhaps not) since B > D it seems like a lot of the flow in the tank must be going from A to B not A to C.  Is that the problem you are trying to solve?

@jamespa the problems is two fold -

1/ I have to run the ashp much higher than required to get the right amount of heat in the ufh. ( ie the ashp has to come out at 28° (a) to feed into the ufh at 23° (c). Wasting 5° in the tank.

2/ the returns are much the same and over time as the tank heats up then the return to the ashp get hotter and after so long the ashp shuts down as the two temps meet.

I am looking to match (or as people say) slightly higher, the flow out of the buffer so all temps get better

I.e. a=c and d=b

OK got you.  Thats how a buffer should, ideally, be run.

I know of two basic techniques - 

1. Adjust pump speeds (somehow) so the flow rates in and out are instantaneously the same.  Thats going to be difficult if equipment either side can change its flow rate (eg because it changes its impedance - eg because a valve shuts off).  However if this is achieved it results in a stable thermal gradient through the tank and,  because of 'stratification' you end up basically with hot at the top and cold at the bottom, with the transition being at a fairly constant level in the tank and no or minimal mixing.

2. Adjust pump duty cycle so the average flow rate in and out is the same.  This likewise results in hot at the top and cold at the bottom, but the transition moves up and down the tank as the pump(s) switches on and off, but never reaches either top or bottom.

TBH with three pumps on the emitter side you have a tough gig, and it needs someone much more familiar with practical hydronics than I am to advise what, if any techniques are adopted in practice (Im assuming you don't wish to build your own control circuit).  Was this not considered at all at design time (it should have been) and do you really need three pumps and a buffer tank (I know you said you weren't going to remove it but you may have set yourself a very difficult task if you cant change any of the plumbing).

We have 24 loops as the house is 500m2. Due to the size, we would not get any decent loops if all run from one spot of the house. Hence 3 manifolds and pumps.

Up until this year its worked fine and we haven't bothered looking at how efficient it was. We use 1300kw per year on heating and as it was set to run at 35° we where happy

But getting a smart meter made us look at its usage and found it was set very high. It seems Grant want you to run them at 45 to 50 degrees, but for out house that is way too much heat.

So set about changing it to increase efficiency.  But this brought up the fact the hp was fine, it was the tank (I'll use this name rather than llh/buffer) that is having a problem.

Set 4 temp guage for the pipes and noticed that they where showing a 5° drop across the tank. Now 1° I can accept as the tank would also be used to defrost the ashp. All the various indications are the a should equal c and b should equal d.

The only way to do that is the get the flows on both circuits about the same. So we started to make adjustments on both sides, allow 30 minutes to pass and see if it worked. We have now tried all combinations and though i can get c to be close to a, I can not get b down to d. It stays 3 to 5 degrees higher.

We would love to have the tank have hot at the top and cold down the bottom, but the heat from the ashp does not seem to mix. We get a being 28°, b & c being 26°, and d being about 20°

For the time being we are trying to work with what we have.

Am I correct in saying the ashp can flow is greater than the pumps, or is it the other way around. Once I understand this I'll know which side to make changes ?

Posted by: @uk_pete_2000

↑

We use 1300kw per year on heating and as it was set to run at 35° we where happy

Can you clarify.  Do you mean 1300kWh and is this electricity in or heat out?

Posted by: @uk_pete_2000

↑

. It seems Grant want you to run them at 45 to 50 degrees, but for out house that is way too much heat.

For UFH that's completely ridiculous (as you obviously realise).  Not only does it mean that your heat pump is thermodynamically less efficient, but you will be cycling badly.

Posted by: @uk_pete_2000

↑

Am I correct in saying the ashp can flow is greater than the pumps, or is it the other way around. Once I understand this I'll know which side to make changes ?

As far as I can understand the dynamics of buffer tanks (which, in heat pump systems, are generally bad - buffer tanks that is) - if the ASHP flow is greater than sum of pumps then the thermocline (hot->cold transition) will be pushed down to the bottom so lots of hot water from the heat pump will recirculate thus heating the return and you will (probably) get what you see.

If the ASHP flow is less than sum of pumps then the thermocline (hot->cold transition) will be pushed up to the top and lots of cold water (from the emitters) will recirculate thus cooling the flow.

I think the former is better than the latter, but neither is great.

Im really not an expert in the hydronics of multi-loop UFH systems.  Hopefully someone on here is....

If not then its a case of falling back on first principles.  

1. What is the loss of your house at design OAT and the loss of each arm at design OAT?
2. What flow rate is required from the heat pump at a DT of say 5 to deliver the total heat in (1)
3. What flow temperature is your UFH (or each arm of your UFH) designed for (at design OAT)
4. What flow rate(s) does that result in across the UFH at DT of 5?
5. Sum (4) should equal (2), if it doesn't then you have a problem.

OK so now we have to set everything up to achieve this or thereabouts and then control something so they remain balanced or biased the right way (if there is a right way)

If I were presented with this problem (knowing nothing about the practicality) I would probably set the flow rate of the heat pump to be about 10% above (2) and the FT a degree or so above (3) and then adjust the flows in each individual arm and see what happens.  But it does depend on what other controls you have and what they are capable of.  If you have a couple of sensors in the tank you could set the arms up and then adjust the flow rate from the heat pump to match.  

In summary, this really should have been sorted out at design time or alternatively needs someone (not me) who has practical experience of balancing multi-loop UFH.

Sorry i cant be more helpful;.

That's fine. We use 1300kw of electricity. No idea what heat this produces as installed said we lost nearlly 30,000kw which work out we needed 10kw of heat. Since then heat punk say its 11000kw and epc say we need 8300kw of heat.

Take you choice, as they are all different.

We don't bother with rooms or arms as it's all ufh and there are no valves on the system. It was better to leave it all open as one.

No idea of flow rate on any part, all I have are the specs from each unit.  Most of the time each half has a dt of 5, but as time goes on the ashp goes down until its zero and turns off the hp.

In some ways it was set up at design time. For two years I never cared what temps where used to get the house warm. Over those two years we set the temp to 19° and it would turn off, then come back on when the temp got down to 18°. So the house was always warm but not hot.

All I was concerned about was what it cost, not how it did it. As things changed people get more involved and notice things are wrong.

Also with regards the use of a tank. This is Grants (and other ashp) way of doing things. No tank, no guarantee.

Hopefully some one will have the information 

Posted by: @uk_pete_2000

↑

That's fine. We use 1300kw of electricity. No idea what heat this produces as installed said we lost nearlly 30,000kw which work out we needed 10kw of heat. Since then heat punk say its 11000kw and epc say we need 8300kw of heat.

I think you are confusing kWh (energy) with kW (power = rate of using energy).  Think of the first as distance and the second as speed.  It would help if we can clear this up so we know what figures we are discussing!  Unfortunately its not clear from the figures, nor can I work out for certain which is which.

• You definitely do not use 1300kW of electricity, otherwise you house would be on fire.  But you might just use 1300hWh per year, although if you did its very low so Im doubting that this is the case
• '30,000kW' is even more impossible, but could perhaps be 30,000kWh, representing the total energy loss during the year, which would correspond to a peak heating demand (= heat pump capacity) of about 15kW.  Assuming a COP of 3 this would mean that you would expect to consume about 10,000kWh per year and your max instantaneous electricity consumption when its very cold might be as high as 7-8kW.
• The heat pump/epc figures are also impossible.   Do you mean 11kW/8.3kW, (heat demand = heat pump capacity) or 11,000kWh/8300kWh (total annual energy consumption/demand)?

Posted by: @uk_pete_2000

↑

We don't bother with rooms or arms as it's all ufh and there are no valves on the system. It was better to leave it all open as one.

Your diagram shows three separate pumps on three separate arms, so whilst they may be open they are somewhat independently driven according to your diagram.  Im confused, can you clarify?

Posted by: @uk_pete_2000

↑

All I was concerned about was what it cost, not how it did it. As things changed people get more involved and notice things are wrong.

Also with regards the use of a tank. This is Grants (and other ashp) way of doing things. No tank, no guarantee.

Hopefully some one will have the information 

I hear your pain, but without some further information its going to be difficult to help, which I am sure those on the forum, including me, would like to do.

I sense however that you are frustrated with the questions being asked, so please feel free not to respond to me.  I will not comment further unless you do, and will instead hope that someone else can provide an answer.

Happy to answer any and all question but i have no idea of the various acromns used, or what each one means I just read the meter that's attached to the ashp. So yes it is correct, we only use 1300kwh (as thats what's listed on the bill)

I am only going by what figures the installer / heat geek / heat punk  / epc listed. As far as I know they too say kwh, so it's total energy required to heat the house for a year. None of which really matter as they are way over what we require and use. So we only look at what's used, not what it's given out.

Before you wonder, the house is very well Insulated. The walls have 400mm of warm cell insulation. The floor has 400mm of eps and the roof has 700mm of eps. Triple glazed, with ach of 0.4.

Each arm is a manifolds, with 8 loops on each and one pump. Each loop covers one or two rooms. But the lounge / kitchen has 4 loops in. Piping connects all manifolds togeather, much as the diagram

Running temp sensors on the floor it raises from about 18.7° to 19.7° which will raise the temp of the kitchen to 19.7 over time. Heating is turned off at 16.00hrs and back on about 10.00 hrs

Posted by: @uk_pete_2000

↑

Happy to answer any and all question but i have no idea of the various acromns used, or what each one means I just read the meter that's attached to the ashp. So yes it is correct, we only use 1300kwh (as thats what's listed on the bill)

I am only going by what figures the installer / heat geek / heat punk  / epc listed. As far as I know they too say kwh, so it's total energy required to heat the house for a year. None of which really matter as they are way over what we require and use. So we only look at what's used, not what it's given out.

Before you wonder, the house is very well Insulated. The walls have 400mm of warm cell insulation. The floor has 400mm of eps and the roof has 700mm of eps. Triple glazed, with ach of 0.4.

Each arm is a manifolds, with 8 loops on each and one pump. Each loop covers one or two rooms. But the lounge / kitchen has 4 loops in. Piping connects all manifolds togeather, much as the diagram

Running temp sensors on the floor it raises from about 18.7° to 19.7° which will raise the temp of the kitchen to 19.7 over time. Heating is turned off at 16.00hrs and back on about 10.00 hrs

OK so if you are consuming 1300kWh per year that's about £300 at current electricity prices, less if you have an ashp-optimised tarrif.  So you may feel that its not worth spending much money (or time) to shave perhaps 10% and max 20% off the running cost.  Thus we need to concentrate on cheap, simple, changes. We also need to bear in mind that, at this level, fixed consumption due to pumps and electronics may be contributing materially to the running cost.  Its worth noting that your 13kW ASHP is way over sized for your space heating, which has a peak demand of only ~2kW to the house if operated 24x7 or thereabouts, as ASHPs are designed to do.  A 6kW one would have been easily sufficient and have sufficient oomph for DHW.  However we need to work with what you have!

So, to move forward, lets try to work out whats going on.  You may need to help by thinking about the analysis and spotting anything that seems to be at variance to what you observe, or bringing to bear other things that seem relevant.

Currently it looks like your ashp is pumping more into the tank than the UFH pumps are extracting, because b>d and so water from the top must be reaching the bottom.  However if that is the case, one would expect the hot to cold transition to be well down the tank and so c would be close to a.  But that isn't the case and we have an apparent inconsistency.  It appears that mixing is occurring both at the top and the bottom.

Perhaps the tank isn't stratified because either its not well designed or the pump rates are too high for the design.  If so, and if its consistent with the required energy transfer, turning down pump rates all round may help.  Since your ASHP is well over sized, its very possible that pump rates are much higher than they need to be and you could try turning them down if you have the controls.  What control do you have of pump speeds? 

Before doing that however, lets see if we can do a back of the envelope calculation on heat transfer rate.  Somewhere on your ASHP controller it should (probably) tell you the flow rate.  Can you find this?  

Yes we now know it is oversized, but at the time we didn't so did not question the maths / sizing etc. Only after we have run it and found we did not need one this large are we now looking to get the best from it.

Yes is does only cost £300 a year to run, since the smart meter was installed, we have moved over to Octopus cosy, so it's now half that, as long as I run it at the right time. Hence we noticed that the temp across the tank was not optimal. So correct, no point in spending 100's of pounds replacing this and that to save £50 a year. Spending £50 on a flow restrictor is fine, but to test I can close down various valves for a bit and monitor over 24 hrs. It's just trying to find out which side of the tank to work with.

There are no indications of any flow rate. The grant uses rpm and the lowest is 3000rpm.

the grant has only three setting (see above) and we have been through each one. I agree the best may be the lower one as it suggest a flow rate of 15 to 45 l/m. But then the three pumps togeather have a much higher combined flow rate of 45l/m to poss 150l/m (barring pipe size, which start at 28mm and go down in steps to 15mm pipe loops). So that indicates turning the hp up to max ?

Posted by: @uk_pete_2000

↑

the grant has only three setting (see above) and we have been through each one. I agree the best may be the lower one as it suggest a flow rate of 15 to 45 l/m. But then the three pumps togeather have a much higher combined flow rate of 45l/m to poss 150l/m (barring pipe size, which start at 28mm and go down in steps to 15mm pipe loops). So that indicates turning the hp up to max ?

OK so lets start with some back of the envelope stuff.  Your peak heating load is say 3kW.  So you need 3/(4.2*4) l/s = 0.18l/s=10l/min to carry this (10% more if there is glycol in the system).  You need perhaps twice to get full throttle for the DHW.  I suspect the Grant will complain if turned down this low, and in practice it must be cycling.  This being the case I would start by turning the Grant down as low as it is happy with and then turning the UFH down to match, the objective being to reduce the speed of water entering the tank thus hopefully improve stratification. 

Only if that doesnt work would I try turning everything up, but although you may be able to balance unless you can get stratification to work it wont help.

I may be completely wrong!

James