Posted by: @derek-m

↑

Posted by: @iantelescope

↑

@derek-m 

Hi Derek,

From the Water Power Equation:

 

Water_Power ( kW)  =  Specific_Heat_of Water ( kJ/(litres x T)  x  Delta_T  x Flow_Rate (litres/sec)

it follows that for a given Power the multiple of Delta_T and Flow_rate is a constant.

 

On my system , I can set the Flow rate using the Pulse-Width_Modulated input on the Radiator Grundfos Motor.

With Delta_T X Flow_Rate a constant , I can set the Flow_Rate, thus control Delta_T.

 

Is this a fair summary of your argument ?

 

Applying your argument to my current system  I have a Delta_T of 3 C , a flow rate or 10.2 lpm at a load of 2.2 kW.

 

 

 

ian

It actually becomes quite complex, even without taking solar gain, wind chill and rain effect into consideration.

Taking the hypothetical example that I posted earlier, with a calculated heat loss of 12 kW per hour, at -3C outside and 21C inside.

With the outside temperature at a constant 9C, the heat loss would be approximately 6 kW, so to maintain the indoor temperature at 21C would require a supply of 6 kW of thermal energy from the heat emitters. If the supply of thermal energy were increased to 6.5 kW, then the indoor temperature would increase to approximately 22C, and if the supply of energy were reduced to 5.5 kW then the indoor temperature would reduce to approximately 20C. So to keep the indoor temperature at a desired level it is a matter of balancing supply to demand.

A degree of complexity starts to appear when considering the heat emitters, since the quantity of thermal energy that they can provide is not only determined by the average water temperature, but also the actual indoor temperature, the total output capacity (normally specified at a Delta T (DT) of 50C) and also the Specific Heating Capacity (SPH) of the liquid within the system.

In the hypothetical example, to supply 12 kW of thermal energy at an average water temperature of 50C, would require heat emitters with a specified capacity of approximately 24.4 kW. For these heat emitters to provide 6 kW of thermal energy would require an average water temperature of 38C. If the DT across the heat emitters is 5C, then the LWT would need to be 40.5C, with a RWT of 35.5C.

If it were possible to introduce a 20% anti-freeze mixture, then the quantity of thermal energy being supplied to the heat emitters would be reduced from 6 kW to 5.57 kW.

The indoor temperature would still be at 21C, so the building heat loss would still be 6 kW, but the quantity of thermal energy being supplied to the heat emitters would now be 5.57 kW.

The heat emitters still require 6 kW of thermal energy to maintain the indoor temperature at 21C, but are now being supplied with only 5.57 kW, with the flow rate being kept constant. The LWT is being kept constant at 40.5C, so the net effect must be for the RWT to be reduced. This in turn will cause the average water temperature to be reduced, which will therefore reduce the amount of thermal energy being supplied by the heat emitters, with the subsequent effect that the indoor temperature will start to reduce. This process will continue until the thermal energy supplied by the heat emitters balances the heat loss from the property, at a value calculated to be 20.14C.

I have attached an Excel spreadsheet with a table showing the probable effect.

If the flow rate remains the same, then the only way to bring the indoor temperature back to the desired 21C is to increase the average water temperature. This can only be achieved by increasing the LWT.

As you have probably gathered by now the process is not exactly straightforward, since increasing the LWT will probably cause an increase in the RWT, with the pair of them causing the average water temperature to increase. This has the effect of transferring more energy into the property and hence increasing the indoor temperature, which in turn will increase the heat loss from the property. Balance should be achieved when the thermal energy supply and demand are approximately equal.

I have not yet developed formula's to calculate the change in RWT when the indoor temperature is brought back to the desired 21C, which will therefore affect the required LWT.

The effect at the heat pump will be a slight reduction in efficiency due to the higher LWT.

-- Attachment is not available --

 

 

I think what you say is all correct but I also think you may be over complicating matters with relatively small second and third order effects.

The DeltaT and heat loss is driven first and foremost by the emitters not by the heat pump which merely responds to what it sees.  So for example if the emitters lose 6kW at a flow temperature of 45C at an indoor temperature of 21C then they will do this independently of what the heat pump does, how much energy it supplies or (to first order) the flow medium.  If the flow medium is changed to one with a lower heat capacity then the emitters will still, to first order, lose 6kW at a FT of 45C.   The heat pump responds by dumping 6kW into the flow medium to maintain the flow temperature (that is its basic control loop - keep the flow temp at its design value by dumping the right amount of heat into the flow medium).

Now lets look at the second order effect.  Strictly of course the loss from emitter to room is dependent on the difference between the average temperature of the emitter and the room, not the difference between the flow temperature and the room so we need to apply a small correction to the above.  This correction is second order because the difference between the average emitter temp and the flow temperature is small in comparison with the difference between flow temperature and room temperature.  We can calculate the correction it as follows:  

With a FT of 45C and a deltaT of 5, the average is 42.5C and if the room temp is 21 then the average room-emitter temp difference is 21.5C.  If the flow rate is the same with 20% glycol, and given that the flow temp is still 45C, the heat loss will still be very nearly 6kW.  The delta T (flow-return) will be 3886/4200*5 = 5.4C, so the average temperature of the emitter, instead of being 42.5C will be 42.3C and thus the room temperature will drop to 20.8C to maintain the same average emitter to room delta T and a 6kW loss**.  The heat pump will still respond by dumping 6kW into the flow medium to restore the flow temp to 45C.

This is what happens if the heat pump doesn't control flow rate, only flow temperature.  In principle you do therefore therefore need to bump the WC curve up by about 0.2C to maintain the same room temperature, but in practice the the pump speed (if not controlled by the heat pump) should be set up by the installer for the design deltaT, and so he/she will set it a bit higher if there is glycol in the system than if there is water, and the WC curve will be the same either way.  In  the real world, whether there is water or glycol, the pump speed is only set approximately and delta T will not be exactly 5C in either case but the heat loss from the emitter will, thankfully, still be very, very close to 6kW!

Now if the heat pump controls both flow temperature and delta T, by controlling both how much heat it dumps into the flow liquid and the pump speed, the heat pump will do what the installer should do and will cause the pump/flow speed will be slightly faster with glycol than water, deltaT will remain at 5C and the heat loss 6kW.

In summary the 'penalty' for glycol is about a 0.2C increase in FT, but this should be adjusted out either by the heat pump itself or by the installer so can, I think, be ignored.

As mentioned upthread the real penalty which cannot be ignored is any temperature drop across the PHE.  If plumbed correctly this should be, I understand, in the range 1-5C corresponding to a 2%-10% performance penalty.  If the PHE is plumbed the wrong way round you could have as much as 10C drop across it (as you can with a badly plumbed buffer tank) = 20% performance hit.

------    

Footnote

** Actually the room temp will be a tiny bit higher, and the loss a tiny bit lower because the loss to the outside will be a bit less than 6kW at 20.8C, however at this point we are into third order effects which are too small to be worth bothering about!  Another higher order effect, also too small to worry about for this excercise, is that radiator (not UFH) loss is proportional to temp diff^1.3 not temp diff^1.  Yet another higher order effect is that, if the pump speed is not actively controlled by the heat pump, the delta T flow-return will vary with OAT, because the amount of heat that the emitters need to emit varies with OAT.  This actually helps a bit with the WC curve for rads, making the 'perfect' curve a bit more linear than it otherwise would be, and thus meaning that a linear WC curve (which is what many heat pumps have) us a better approximation to 'perfection' than it otherwise would be.

@jamespa 

Actual Room Power Loss Measurement

I agree with you , In Measuring the actual Room Heat Loss I have NOT taken the Heat leakage of the adjacent rooms into consideration.

My Problem is that , no matter what I measure .................my measurement will come up against a perfectly valid objection.

Measuring the Water Temperature from the Temperature of the Copper Pipe is not accurate..........

K types are faster and sometimes more accurate than DS18B20 's semiconductors...........sometimes.

The Truth is that each measurement is inaccurate, or subject to objection.

However,  How many correlations drawn from inaccurate measurements constitute Truth?

Analogy

From the Water Power Equation:

Water_Power ( kW) = Specific_Heat_of Water ( kJ/(litres x T) x Delta_T x Flow_Rate (litres/sec).........................................(1)

The Electrical Equation:

Electrical Power (kW)  = Current (Amps) X Volts(kV)  ...............................................................................................................(2)

Why then not model the Heat Pump with Delta_T being the potential energy , analogous to Voltage , and, Flow_Rate analogous to Electrical current?

Flow Rate Reduction

With the Electrical analogy in mind I have reduced both the Primary ( Heat Pump) and the secondary, Radiator Flow rates.

As predicted by Derek and yourself , the Delta_T increases with the Radiator circuit now approaching 35 C for the first time !

Faulty PWM

My PWM output from the Samsung Control board does not work.

To reduce the Flow rate on the Heat Pump I have used an "Arduino".

No Radiator PWM

My Radiator Grunfos UPM3 motor is fitted with a working PWM motor.

However, No source for a PWM signal is available on the Control Board.

To reduce the Flow rate on the Radiator Water circuit I have used another port on the  "Arduino".

Three Faults?

I have three faults , the position of the Buffer, the faulty Heat Pump PWM signal and the non-existent Radiator PWM Source.

@jamespa 

James, Great script.

Let me read, and print this out.......... to summarise your argument.

Excellent writing , but daunting to the uninitiated!

Leaves my "Electrical Analogy" wanting!

ian

@jamespa 

Given that I have control of the Flow rates of both the Heat Pump and Radiator motors on either side of the Heat Exchanger .......

                          to what variable should I control the Flow rates on the Heat Pump or the Radiator Water pump ?  

Heat_Pump_Flow_Rate =  function ( Delta_T across Heat_Exchanger Inputs ) .............?

and /or

Radiator_Flow_Rate = function ( Delta_T across Heat_Exchanger Outputs ) .............?

Function?

What function should I use , experiment ?

ian

@derek-m 

Many Thanks for the advice about setting  the secondary Radiator Flow  rate first.

Then setting the primary, Heat Pump Flow rate ..........

What other manufacturers use Flow control in this way?

ian

@iantelescope 

I'm not actually sure, and if somebody else knows I'd listen to them.  In the absence of that, for starters I'd control the radiator flow for deltaT 5 and the hp flow ditto.  Don't update it too often, once every minute is probably plenty

Then you need to think about the boundary states eg when the heat pump stops heating because demand is satisfied (ie it's cycling).  Also start up conditions , you don't want it shutting the pumps down to get to 5, so probably you need a min speed.   Notwithstanding this I suspect that just controlling to 5 may still work provided it doesn't floor the logic (a good old analog system wouldn't suffer this problem).

A simple proportional control algorithm with a bit of a dead point around the target temp (to avoid hunting) will likely do the job, or something cruder with steps.  It doesn't have to be particularly accurate (how accurate do you expect installers get it?)

If you can make the two target temps variables you can change independently.

@jamespa

Having rechecked my calculations you are indeed correct, since I failed to make allowance for the change in DT when calculating the thermal energy supplied when anti-freeze is added. Rather than falling from 6 kW down to 5.57 kW, it only reduces to approximately 5.9 kW.

So as you stated, the indoor temperature would fall from 21C down to 20.8C, which would require an increase in LWT from 40.5C to 40.7C to re-balance the system at an indoor temperature of 21C.

Sorry for any confusion that I may have caused, but please don't let my wife know that I have admitted that I was wrong.

Posted by: @derek-m

↑

Sorry for any confusion that I may have caused, but please don't let my wife know that I have admitted that I was wrong.

Rest assured your secret is safe.  One of the joys of this forum is that we all learn from others and benefit from peer review.

That said, my 2 year journey into heat pumps might have been be a lot easier if I hadn't attempted to understand this stuff.  The more I understand, the more I realise how little the (UK installation) industry understands (with notable exceptions), and the more depressed I get for the prospects of heat pumps ever becoming main stream in the UK.  My view is that it's an environmental tragedy created by a deliberately under-resourced civil service, a government that doesn't really care, and MCS which has a government sponsored stranglehold, is almost totally opaque, and is dominated  by the narrow interests of the grant harvesters, not the consumer or the environment. 

My heat pump goes in w/c 16 Oct despite all the hurdles the industry and government creates.

Posted by: @iantelescope

↑

@derek-m 

 

Many Thanks for the advice about setting  the secondary Radiator Flow  rate first.

Then setting the primary, Heat Pump Flow rate ..........

 

What other manufacturers use Flow control in this way?

 

ian

Actually, having given some thought about your system and realised I was wrong and James was right, when it comes to the affect of adding anti-freeze, even to just one side of your system.

If you intend to keep your PHE, then rather than trying to control the DT on both sides of the PHE, it may just be necessary to set the primary and secondary pump speeds for optimum thermal energy transfer from primary to secondary. Then keep the speed of each pump constant whilst allowing variations in LWT, RWT and hence DT, to transfer more or less thermal energy from the heat pump to the heat emitters.

Once the Weather Compensation (WC) curve has been correctly optimised, and your heat emitters have been correctly balanced, it should be possible to keep the indoor temperature fairly constant without the need for other controls.

Posted by: @jamespa

↑

Posted by: @derek-m

↑

Sorry for any confusion that I may have caused, but please don't let my wife know that I have admitted that I was wrong.

Rest assured your secret is safe.  One of the joys of this forum is that we all learn from others and benefit from peer review.

That said, my 2 year journey into heat pumps might have been be a lot easier if I hadn't attempted to understand this stuff.  The more I understand, the more I realise how little the (UK installation) industry understands (with notable exceptions), and the more depressed I get for the prospects of heat pumps ever becoming main stream in the UK.  My view is that it's an environmental tragedy created by a deliberately under-resourced civil service, a government that doesn't really care, and MCS which has a government sponsored stranglehold, is almost totally opaque, and is dominated  by the narrow interests of the grant harvesters, not the consumer or the environment. 

My heat pump goes in w/c 16 Oct despite all the hurdles the industry and government creates.

Having been a moderator on this forum almost since its inception, I must admit that I share many of your sentiments with regard to the lack of knowledge both within the industry and the external decision makers.

Unfortunately, Government grants and the like tend to attract the grant harvesters, which was the case when we had solar PV installed over 10 years ago, which cost approximately twice the price of a similar system now.

Having had a long and varied career as an Instrumentation and Control Systems Engineer, I have managed to reduce our total energy consumption to approximately 3 kWh per day at this time of year, with a 4 kWh daily average electricity consumption throughout the year. I therefore try to use my knowledge and experience to help as many people as possible to minimise their energy usage and hence reduce demand.

I have stated numerous times in the past, if you don't need the energy in the first place, then it does not have to be produced and transported to the point of use. I also believe that there is insufficient emphasis being given to bulk energy storage, where renewable energy, when in abundance, is stored for later use.

@derek-m For my own sake, I have battery storage locally and this allows me a degree of independence; the Tesla Energy Plan fell apart a little while back due (I think) to the Energy Crisis price hikes. I note that Octopus Energy have started a trial scheme (limited to the one battery manufacturer at present) that might provide one solution to ease the high demand - low demand situation as customers might join the scheme to provide a network of local battery ‘banks’. I don't know if any other energy suppliers are doing anything similar yet but it sounds like a good idea worth trying. Octopus aren’t offering this trial for Tesla Powerwalls yet and I know of no way to control the Tesla Battery to export - only allow excess solar production to be exported at present. I believe that the TEP must have had this capability via the app; however, I am guessing that this is how it worked. I have heard of public spirited individuals who would like to see energy storage in the form of battery banks but they feel that such banks are inappropriate in their own area! I still feel smug when my system is exporting though… Regards, Toodles.

@derek-m 

Hi Derek,

I have just been given the date , and the terms under which my Heat Pump will be inspected by the NIC , some 16 months after the "installation" was started.

With endless delays while the NIC and MCS  proceeded to Twice Strike OFF my "installer"  I agree totally with your comments about the industry "Regulators".

NIC,MCS and RECC   .............Charities  ?

if pushed , the " Industry Regulators " will say that "they are merely Charities operating to educate the public about the Environmental crisis".

The NIC,MCS and RECC should NOT be charities but  Regulators, funded by , but independent of Government..

Light Touch Regulation!

"Light Touch Regulation" , imposed on the industry to facilitate competition, has led to wholesale abuse.

"Competition is undoubtedly required , but , the industry Must have an educated Industrial ethic in the manner of the German Environmental industry.

"Only the Germans REALLY KNOW how these Heat pumps work."

This statement , from a visiting French Samsung engineer, vividly  sums up the standard of training in the UK industry.

ian