@derek-m In order to get reliable flow round my pipe work, I would also need bypass valves, new diverter and pump etc - hence the cost. My dilemma is knowing how much the buffer tank is contributing to my low COP. I was quoted a SCOP of 3.68 at LWT of 50 deg C (as per Samsung data sheet). At a LWT of 40 deg C and ambient temperature of 8 deg C, I get nothing like that. 

My assumption is that in steady state conditions if you fix the LWT, delta T and the ambient temperature, the COP should stay the same, whether the house is well insulated and warm or poorly insulated and cold. The ASHP doesn’t care what it is heating. Or have I got this all wrong?

Posted by: @mike-h

↑

@derek-m In order to get reliable flow round my pipe work, I would also need bypass valves, new diverter and pump etc - hence the cost. My dilemma is knowing how much the buffer tank is contributing to my low COP. I was quoted a SCOP of 3.68 at LWT of 50 deg C (as per Samsung data sheet). At a LWT of 40 deg C and ambient temperature of 8 deg C, I get nothing like that. 

My assumption is that in steady state conditions if you fix the LWT, delta T and the ambient temperature, the COP should stay the same, whether the house is well insulated and warm or poorly insulated and cold. The ASHP doesn’t care what it is heating. Or have I got this all wrong?

Hi Mike,

It is the heat emitter end where the problem lies. The amount of heat energy that can be emitted by the heat emitters, is dependent upon the physical emitting area, and the temperature difference between the average water temperature inside the heat emitter, and the indoor air temperature. The average water temperature being approximately half way between the temperature of the water entering the heat emitters and the temperature of the water leaving.

If, to meet the present heat demand, the average water temperature at the heat emitters needs to be 35C, and there is a 5C DeltaT across the heat emitters, then the water entering the entering the heat emitters will need to be approximately 37.5C.

If there is mixing occurring within a buffer tank or LLH, there could possibly be a 5C temperature loss between the LWT from the heat pump, and the water going to the heat emitters. The heat pump would therefore need a LWT of 42.5C to meet the heat demand rather than the 37.5C required by the heat emitters.

The heat pump therefore uses more electrical energy to absorb a smaller quantity of heat energy from the outside air, so is obviously operating in a less efficient manner.

To try to balance the temperatures across a buffer tank or LLH, the water flow rate going in from the heat pump needs to be the same, or slightly higher than the water flow rate coming out and going to the heat emitters.

I don't understand why you need bypass valves if you ensure that there are no zone valves or TRV's that can close and fully shut of the water flow. Your present diverter valve and water pump are adequate at present flow rates, which could be lower if the system is operating more efficiently.

If you were to modify the system as I suggested, if problems occur it would be a simple job to reverse the process.

Initially you could try lowering the secondary pump speed and/or increasing the primary pump speed, to try to balance the flow rates. If you measure the temperature at the inlet flow pipework and the exit flow pipework, this should gives some indication of when the flow rates are reasonably balanced. The objective being to obtain the minimum temperature difference.

@derek-m Thanks Derek. I think we agree that the inefficiency of the buffer tank is in part due to the higher LWT required to provide the same heating energy gained by the house. With a 5 degree increase in LWT the efficiency loss is around 13%. But is it the ONLY reason for making the buffer inefficient? In Brendon's article, there was a 27% efficiency loss between Test 2 and Test 3  (COP 3.4 vs 4.7). I am trying to understand what is the explanation for the other 14% efficiency loss. Would I gain 13% efficiency or 27% efficiency if I removed the buffer?

Posted by: @mike-h

↑

@derek-m Thanks Derek. I think we agree that the inefficiency of the buffer tank is in part due to the higher LWT required to provide the same heating energy gained by the house. With a 5 degree increase in LWT the efficiency loss is around 13%. But is it the ONLY reason for making the buffer inefficient? In Brendon's article, there was a 27% efficiency loss between Test 2 and Test 3  (COP 3.4 vs 4.7). I am trying to understand what is the explanation for the other 14% efficiency loss. Would I gain 13% efficiency or 27% efficiency if I removed the buffer?

I suspect it will very much depend upon the design of your system, the heat loss of your home, the heating capacity of the heat emitters, how many water pumps are installed, and if any of these water pumps are having their speed varied by the heat pump controller.

If the primary water pump is 'man enough' to push the maximum required flow rate around the system, then unless you have a mixture of radiators and UFH, you should not require any secondary water pumps.

The warm water coming from the heat pump should go, via the diverter valve, directly to the heat emitters. The flow through the heat emitters should be balanced, with the heat emitters in the coldest room at maximum allowable flow, and the flow to the other rooms only being reduced to achieve the desired temperature.

The WC curve should be adjusted to achieve the required indoor temperature at the lowest permissible LWT.

I would suggest leaving the buffer tank in place, but change its duty to that of a volume tank, which should help with defrost cycles.

The main problem with having more than one water pump, unless it is absolutely necessary to achieve the required water flow rate, is operating them both in a balanced manner. If the main water pump is adequate, then running a second pump is just wasting electricity, and could actually be disrupting the operation of the primary pump.

I may have read, but cannot remember the details of Brendon's article, but the efficiency savings that can be achieved are dependent upon numerous factors, one of which being how badly the system was operating before the improvements were made.

@mike-h you won't know as every circuit is different. However you will gain something. Doing the exact measurements will probably be more hassle than you want to take lol

Posted by: @derek-m

↑

Posted by: @sunandair

↑

Posted by: @derek-m

↑

Ideally you require the water flowing into the LLH to be the same flow rate, or slightly higher, than the flow rate coming out, which should reduce or eliminate any mixing taking place.

Early test results showing opposite is happening in our system. 

I’ve opened up the primary pump valve until flow rate went up to 15 LPM (was 13) and closed the CH pump then opened it back up to give us a total circulation 13LPM flow. This was a bit tricky since I lost circulation in the CH loop at one point so opened the check valve up a bit more. 
(This must be a fine adjustment since a quarter turn gave over 1LPM change)

Results after half an hour settled operation were not as suggested. In that the Tdrop on the CHFlow thermistor was 4c compared to 1.5c when set with a 13L flow on the primary pump and open valve on CH pump.

I’ve currently reverted to the earlier setting -opened the ch valve half a turn and closed the primary pump valve to give 13 LPM flow. we are now only getting 2c Tdrop. Still not as good as original 1.5c drop. 

(Both pumps are on speed setting one and can only be adjusted manually using the isolation valve otherwise the speed will be 15 LPM)

Posted by: @allyfish

↑

OK to have LLH primary flow slightly higher or equal than secondary, but secondary flow must never be more than primary.

We have been running with a slightly higher flow in the secondary and, as above, it’s been giving us a better result at the moment. Of just a 1.5c drop. It’s a bit puzzling to say the least.

It’s quite hard to finely regulate the flow of the CH pump since if it is over adjusted it simply isolates the flow in the two loops without much of a sound indication. Whereas the primary is directly linked to the flow sensor so is a direct reading.

 

 

I suspect that adjusting valves to vary the flow rate is far from ideal, since the flow rate should really be adjusted by varying pump speed. I am also not certain what the temperature difference is to which you refer, is this across the LLH between the flow in and flow out temperature, or between the flow and return temperatures at each side of the LLH. As simple sketch would be useful.

 

Hi @derek-m The temperature difference of 1.5c is between Primary flow thermistor and Secondary flow thermistor. See sketch for positions and approx distances.

Regarding our pumps - we don’t have FTC speed control with the pumps our installer installed, only manual. There are only 3 push button speed settings. However the slowest speed only produced a flow rate of 15 litres per minute with a DT of 2c. However a further restriction of 2 LPM using the isolation valve resulted in a DT of 5. And a flow rate of 13LPM.

New tests are still showing improved FLOW temperatures across the low loss header with an increased pump speed on the central heating side. 

Process.

Both pumps were first set to the lowest pump speed giving a flow rate at the flow sensor of 15 LPM. Both pumps are identical models.

With the HP set to DHW operation this means that only the primary loop is operating via the hot water cylinder. The central heating loop was obviously inactive.

This allowed the flow rate to be adjusted using the valve of the primary pump. The flow rate was adjusted to 12LPM and readings were taken of flow temperatures and return temperature.

We then switched the central heating on to see what would happen to the DT and flow rate. We found the delta T widened slightly to DT6 and the flow rate increased to 13 litres.

So I am still trying match the flow as you suggested above but have found that when the secondary pump is slower than the primary pump the Flow temperature dropped on the central heating side.

However when it was even, only slightly faster than the primary pump it appears to draw more hot water through the top coupling and there seems to be less mixing. However there could be other reasons for the improved performance. I think it’s  possible that there is more drag on the CH pump since there are obviously more pipe-work, pipe sizes and emitters all of which may be cancelling out the slightly higher pump speed and perhaps the overall flow is closely matched.

I have now achieved a slight reduction in temperature drop across the flow pipes of the low loss header - it now appears to be only 1c differential. 

Can you explain what you meant about the relationship between HP size and total water volume?

Ive been working out pipe and rad volumes and it appears we have a system of about 155 litres 

@sunandair

Where is the DT of 5C, when you close in the valve and where is the valve located?

Some actual temperature values on the diagram would be useful.

Posted by: @derek-m

↑

@sunandair

Where is the DT of 5C, when you close in the valve and where is the valve located?

Some actual temperature values on the diagram would be useful.

not sure of the “where” bit of your question - the DT I mean is the temperature differential between Primary Flow and Primary Return. These temperatures are continually changing so their value was taken at different times with More or less the same differential of 5 degrees.

the temperatures would typically be as follows when the valve adjusted flow was 13LPM

Primary Flow 37c

Primary Return 32c (eg DT5c)

Secondary Flow at same time = 35 or 36c

The valve being adjusted was one of the isolation valves of the primary pump. Note: the valve was only used to fine adjust the flow rate down by 2 LPM. TO 13 LPM

readings were taken from the ftc6 thermistor enquiry screen AND monitored on The MELCloud hourly temperatures. Please see attached pics

Typical Primary Loop / DHW heating cycle

Typical Central Heating Cycle in auto adaptive. 

Sketch detail of thermistor positions relative to pumps. 

@sunandair

The 1.5C - 2C temperature difference between primary and secondary, could just be due to inaccuracies in the readings.

Lowering the flow rate from 15 lpm to 13 lpm would on the face of it appear to be balancing the flows, since the secondary temperature appears to remain the same. It is a pity that there is no secondary return temperature measurement, which could indicate how much mixing, if any, is taking place in the buffer tank. It would be interesting to see if lowering the primary flow rate further, starts to increase the temperature difference between primary and secondary.

What I believe is happening, is that at a flow rate of 15 lpm, some of the primary water is flowing through the buffer tank and mixing with the return water from the radiators. The return water to the heat pump is therefore warmer than it should be, hence the lower DT. The amount of heat energy produced by the heat pump should not be vastly different, since the greater flow rate is balanced by the lower DT.

Are all the rooms reaching the desired temperature or are they too warm? The latter would indicate that the LWT is too high.

Posted by: @sunandair

↑

 

Here are two schematics of what we have.

-- Attachment is not available --

 

-- Attachment is not available --
-- Attachment is not available --

@SUNandAir I'm not adding to the tech input as the others are giving all the help and info that is needed. but, I wanted to say to you, a  "really really excellent" for your diagrams.  good to see someone taking the time to document what they've got properly, and what they think they'd like to change, in a way that can be clearly understood. A first class example for others.

Posted by: @iancalderbank

↑

Posted by: @sunandair

↑

 

Here are two schematics of what we have.

-- Attachment is not available --

 

-- Attachment is not available --
-- Attachment is not available --

@SUNandAir I'm not adding to the tech input as the others are giving all the help and info that is needed. but, I wanted to say to you, a  "really really excellent" for your diagrams.  good to see someone taking the time to document what they've got properly, and what they think they'd like to change, in a way that can be clearly understood. A first class example for others.

 

Thanks for your comments Ian, it’s very much appreciated. Half the battle has been understanding and using the correct terms and  acronyms, and realising that a helper has quite a task understanding what information is being provided and what’s missing.

its great to know that others might benefit if the information is clear. Thanks

Posted by: @derek-m

↑

@sunandair

The 1.5C - 2C temperature difference between primary and secondary, could just be due to inaccuracies in the readings.

Lowering the flow rate from 15 lpm to 13 lpm would on the face of it appear to be balancing the flows, since the secondary temperature appears to remain the same. It is a pity that there is no secondary return temperature measurement, which could indicate how much mixing, if any, is taking place in the buffer tank. It would be interesting to see if lowering the primary flow rate further, starts to increase the temperature difference between primary and secondary.

What I believe is happening, is that at a flow rate of 15 lpm, some of the primary water is flowing through the buffer tank and mixing with the return water from the radiators. The return water to the heat pump is therefore warmer than it should be, hence the lower DT. The amount of heat energy produced by the heat pump should not be vastly different, since the greater flow rate is balanced by the lower DT.

Are all the rooms reaching the desired temperature or are they too warm? The latter would indicate that the LWT is too high.

yes. Quote right, accuracy and 1c increments could lead to a skewed interpretation especially when we are only looking at 1or 2c variations.

ive tried to mitigate some of this by comparing thermistors off the same piece of pipe. Also I’ve tried moving the valve setting up until the flow rate just changes then down until it just changes so I know how broad that 1 LPM Reading is on the FTC screen. 

In answer to your question about the room temperatures we are not noticing any over heating  rooms but I think our HP size is close it’s upper limit in that we had an earlier quote that came up with a larger HP size which we declined. We have 6 rads fully opened and 3 with thermostatic valves set to 4/5 the remaining upstairs rads are set to 2. 

Ive noticed something new on the DHW heating cycle 

• The DT gradually narrows to only 3c towards the end of the DHW heating cycle. Presumably because the water is only circulating TO THE TANK in the primary pipes and being reheated quicker. 
• Another possible FACTOR might be the higher flow temperature towards the end of reheating the tank and the heat exchanger (being 22mm pipe) cannot release the heat into the tank quickly enough. The primary pipework is 28mm.
• the DHW cycle starts with a DT of 5c or 6c then slowly narrows to around 3c. The flow rate remains steady at 12LPM.

Meanwhile, when the HP reverts to the Central Heating and the secondary pump switches on the Delta T is around 6 or 7 and the flow rate increases to 13 LPM.

The Attached show

1. graph show the narrow flow and return DT at the end of the DHW cycle
2. graph showing CH Flow and return
3. temperature of thermistors particularly PRIMARY RETURN thermistor THW2
4. Spare Thermistor from another device reading CENTRAL HEATING return.

Following what you asked about our return temperatures are doing across the LLH tank I set up a thermistor to read the return temp and compare with the FTC reading of the Primary Return.

The picture showing the S3 thermostat shows the return temperature of the central heating before it enters the LLH TANK and the THW2 reading shows the return temperature as it leaves the LLH TANK. THERE APPEARS TO BE some consistency as the CH RETURN  is also 1c lower than the PRIMARY RETURN.

so the mixing appears to be the same... I think that’s what I was hoping for.?.?.