Yet the best performing heat pumps typically don't use PWM pump control and are open loop...
I have recently been pondering, while trying to help another participant in the forum, why some heat pumps bother to modulate the water pump to maintain a fixed deltaT and others dont. Basically whats the point of the extra control loop, why not just leave the water pump at max chat? I have yet to hear an explanation and if anyone has one I would be interested.
4kW peak of solar PV since 2011; EV and a 1930s house which has been partially renovated to improve its efficiency. 7kW Vaillant heat pump.
No I don’t know what volume, I might look at the sheets again. Is it normal to have it recorded somewhere? It maybe somewhere as the system has had glycol or equivalent and anti leak a long time ago .
Sorry random input , must of thought I was on another thread!!
@jamespa Let me throw my hat into the ring on this one. If one accepts that a delta-T of 5C is required for correct operation of a heat pump, then heat output power in kW is determined by (flow in litres/sec x 5 x 4.2) or (flow x 21). So, if the flow is fixed at “max chat”, we then have a fixed maximum power output at 5C dT. So, if everything is fixed (no PWM flow control), just exactly how does this magic open-loop system adjust automatically to the low heat requirement at 15C OAT and maintain 5C dT?
Private individual. No affiliation with commercial "Heat Geeks" of same coincidental name.
@jamespa Let me throw my hat into the ring on this one. If one accepts that a delta-T of 5C is required for correct operation of a heat pump, then heat output power in kW is determined by (flow in litres/sec x 5 x 4.2) or (flow x 21). So, if the flow is fixed at “max chat”, we then have a fixed maximum power output at 5C dT. So, if everything is fixed (no PWM flow control), just exactly how does this magic open-loop system adjust automatically to the low heat requirement at 15C OAT and maintain 5C dT?
SOFAIK it doesn't have to. My understanding is that there is nothing magic about 5C, its just a sensible design value in most domestic situations so one can work out pump speed, radiator emission and pipe sizing. As a concrete example I was replying to another thread where the owner has a Daikin, and that seems to envisage (for reasons that are obscure) different deltaT for rads vs fancoils vs UFH, but clearly its the same heat pump in all cases so this cant be to do with the heat pump operating efficiently but instead to do with some assumptions they are making about the system (at least presumably)
One could take the argument further and make a case on two grounds for not modulating the pump:
Given that the emission of radiators is a non linear function of (average) temp diff between rad and room, (emission is proportional to temp diff rad-room^1.3) and given that many, not all, heat pumps support only a linear WC curve, there is a mismatch. If you do the math (which I did a couple of years ago when trying to understand how much various WC curves affected consumption) it turns out that a constant flow results in a closer approximation to the 'ideal' WC curve than constant deltaT, arguing for constant flow
Lower deltaT = higher average emitter temperature for any given LWT = higher COP, so set the water pump as high as it will reasonably go and leave it at that value for all circumstances (which is pretty much what my Vaillant does).
I could be wrong but given that there appear to be heat pumps that dont modulate the flow speed and given also that its rather implausible that all manufacturers would have settled on the same deltaT as the 'best' operating point and given that there is no obvious reason why 5C would be 'magic' the most logical conclusion would seem to be that it isnt, its just a convenient design convention which strikes a reasonable balance between COP (better with lower deltaT for the reason set out above) and water velocity in pipework (better with higher deltaT because of the mass flow calcs).
That said, I guess that if deltaT falls too low the feedback loop for LWT might struggle because it hasnt got much of a temperature difference to go on. But at that point you are almost certainly into cycling anyway.
Of course I could be sniffing the wrong substance, I haven't seen anything about this that is definitive. We need a heat pump designer to answer the question!
This post was modified 2 weeks ago 6 times by JamesPa
4kW peak of solar PV since 2011; EV and a 1930s house which has been partially renovated to improve its efficiency. 7kW Vaillant heat pump.
@jamespa You should best address this question to Brendon. I would also be keen for more clarification as it is fairly fundamental. I am not an installer or heat pump designer.
It's only by asking challenging questions that we move forward!
@jamespa You should best address this question to Brendon. I would also be keen for more clarification as it is fairly fundamental. I am not an installer or heat pump designer.
Thanks for the clarification. I am also neither an installer nor a heat pump designer! Just an enthusiastic homeowner with a background in physics and engineering.
It's only by asking challenging questions that we move forward. I will reflect and decide which questions to ask and of whom next.
This post was modified 1 week ago 2 times by JamesPa
4kW peak of solar PV since 2011; EV and a 1930s house which has been partially renovated to improve its efficiency. 7kW Vaillant heat pump.
@jamespa For a fixed pump in question on open loop, dT could be <2C at 15C. Below is another view on the topic with which I concur.
Ok thanks.
Why is 8C ideal (and why is 2C a problem?)
I suspect you know it's not, but to reinforce that, not a single one of the top performing heat pumps (of the couple of hundred with certified heat and electricity meters) on heatpumpmonitor.org have a constant dT of 8C. At warmer OATs they mostly run with a dT between 2 and 4.
@jamespa For a fixed pump in question on open loop, dT could be <2C at 15C. Below is another view on the topic with which I concur.
Ok thanks.
Why is 8C ideal (and why is 2C a problem?)
I suspect you know it's not, but to reinforce that, not a single one of the top performing heat pumps (of the couple of hundred with certified heat and electricity meters) on heatpumpmonitor.org have a constant dT of 8C. At warmer OATs they mostly run with a dT between 2 and 4.
So this is fairly simple and there is no need for anyone to get prescious about their current position. We all learn by being challenged and hopefully we mostly appreciate that none of us knows everything.
In the absence of 'engineering' effects it's unambiguous thermodynamically that lower DT equals higher COP (see Carnot equation).
So @heacol what are the engineering factors that run contrary to the pure thermodynamics and argue for higher DT than the minimum practical in any given situation, or indeed controlled DT which means that most of the time the DT could be reduced by simply operating at the max flow rate reasonable in the particular installation, by not modulating the water pump.
This post was modified 1 week ago 2 times by JamesPa
4kW peak of solar PV since 2011; EV and a 1930s house which has been partially renovated to improve its efficiency. 7kW Vaillant heat pump.
Technically I could perhaps be called a heat pump designer, having designed and built the one gshp system that has heated our home for 3 years. It’s very small (2.5kW heat - we have a lot of insulation) and basic, as are most gshp - just on/off like a fridge, COP of 2-4.5 depending on temperatures. It runs at around a DT of 6C on the internal pump, and around 2C on the external ground pump. The internal pump could run faster - it takes around 10W now, and I can double the flow at 40W or so, halving the DT. The heatpump efficiency is determined by the hot flow temperature (unlike a gas boiler by the return); so we are now trading that 30W extra pump power for less compressor power. I think there would be a subtle overall efficiency improvement by doing this - based on the graph below, and assuming a constant 2.5kW thermal out i calculate 45W electrical compressor power would be saved due to that 3C drop, so overall 15W saved - but then I notice that the pump is then audible, so don’t do it. Somewhere around a DT of 5 is typically close enough to the most efficient system, taking compressor and pump together. This is system specific; but pipe work tends to be bigger for higher power systems, so I imagine that 5C is a good general target.
Nb in the plot below orange is home heating, blue is dhw.
@robl Your analysis is correct, for heating the most efficient Delta T across the condensor is in fact 5 Deg C and most manufacturers will reccomend this, however down to 3 and up to 10, have little effect on perfocmand, but below 3 and above 10, the perfomance of a heatpump will damatiacally reduce. On the evaprator side, on a water/brine source the optimum is between 3 and 4 Deg C. The desired Delta T will be different depending on the design and the reduction is caused by a reduction in liquid in the evaproator or condensor, it is very difficult to absorbe or dissapate hat with a gas, it is verry much more efficient through a liquid. This is also the reason, hydronic vapour compression equipment (air-water or water-water) is 20-30% more efficient than gas distributed systems (air-air)
However, when I design microbore systems, I am happy for the delta T to go as high as 14 at peak, because we are only there forvery short periods of time and the small extra running cost, for this short eriode will never ammortise the cost and desruption of a re-pipe.
When it comes to actual flow temperature, the basic rule of thumb, is a raise in running cost by 2.5 % for every degree you move the flow tmperature above the source temperature (outside air or brine temperature). This sise is compouded, below is a graph to put it in perspective. I am not a mathamatician so it may not be 100% accurate but it is done for representation.
Thinking about installing a heat pump but unsure where to start? Already have one but it’s not performing as expected? Or are you locked in a frustrating dispute with an installer or manufacturer? We’re here to help.
