Question-and-Answer Resource for the Building Energy Modeling Community
Get started with the Help page

Why is Pump Fluid Heat Gain already added to Pump Inlet Node?

EnergyPlus accounts for the heat generated by the pump motor. The heat is added to the water flowing through the pump.

I compared System Node Temperature of Pump Inlet Node and Pump Outlet Node to see how much the water had been heated, and found that they have exactly the same temperature even though Pump Fluid Heat Gain Rate was not zero.

Finally, I found that the temperature of the other Node before the Pump Inlet Node was slightly lower than the temperature of the Pump Inlet Node. The temperature rise was consistent with that calculated from Pump Fluid Heat Gain, the water flow rate and the specific heat of water.

The temperature of Pump Inlet Node is the temperature after Pump Fluid Heat Gain has already been added.

You can see this in any idf files with any pump components. One ExampleFile: 5ZoneWarmest.idf (V23-1-0) with simple PlantLoop as shown below.

This goes against my intuition. In my understanding, Inlet Node of any component should show the condition of the fluid before it is affected by the component itself. Am I wrong? Or are pumps the only exception? Or is this a bug?

edit retag close merge delete

1

@Keigo does this post answer your question? Essentially, EnergyPlus "lumps" the heat capacitance of each side (demand/supply) of the plant loop at the inlet node for that side, as if an imaginary storage tank is located there.

The pump inlet/outlet node temperatures being equal should only happen when the pump is not operating. If the temperatures are equal while there is pump flow, you may have found a bug or it's a very small difference (need more decimal places to confirm).

( 2023-06-13 08:46:09 -0500 )edit

@Aaron Boranian Thank you for your comment! Loop fluid capacitance and pump heat is indeed the answer to my question. "This so-called pump heat is a complicating factor in plant simulation because the pump heat alters the load on primary equipment. A simple method of accounting for pumping heat is needed that doesn’t increase the difficulties of the numerical solution and in EnergyPlus this accomplished by including the pump heat in the loop capacitance model."

( 2023-06-13 12:58:31 -0500 )edit

@Aaron Boranian The pump inlet/outlet node temperatures being equal happens when the pump is operating. They are exactly the same to 12 decimal places. It is not a bug. As Engineering Reference above illustrates, EnergyPlus adds pump heat to the well-stirred tank, which looks like the pump heat is added to the pump inlet node. Then, pump inlet/outlet node temperatures should be the same.

( 2023-06-13 13:15:46 -0500 )edit

I learned something new today. From Eng Ref: "Since EnergyPlus Version 7, this heat is added to the loop capacitance tank(s) rather than at the pump’s outlet and so the outlet temperatue is equal to the inlet temperaure."

( 2023-06-13 13:27:43 -0500 )edit

Sort by » oldest newest most voted

The plant loops iterate many times during a time step. The point where the figure shows 0.05C difference is the PlantLoop interface of the supply and demand side of the loop. The supply and demand side of the loop are simulated independently. When the supply side of the loop is simulation each component is simulated and the result is pushed from the plant supply side outlet to the plant demand side inlet. The demand side of the loop is then simulated with the same "push" of the plant demand side outlet to the plant supply side inlet. The simulation of each side of the plant continues until convergece or max iteration is achieved. If you let the plant iterate more this temperature difference between each side of the plant should get smaller. More complex plants need more iterations. See the ConvergenceLimits object.

ConvergenceLimits,
\memo Specifies limits on HVAC system simulation timesteps and iterations.
\memo This item is an advanced feature that should be used only with caution.
\unique-object
N1 , \field Minimum System Timestep
\units minutes
\type integer
\note 0 sets the minimum to the zone timestep (ref: Timestep)
\note 1 is normal (ratchet down to 1 minute)
\note setting greater than zone timestep (in minutes) will effectively set to zone timestep
\minimum 0
\maximum 60
N2 , \field Maximum HVAC Iterations
\type integer
\default 20
\minimum 1
N3 , \field Minimum Plant Iterations
\note Controls the minimum number of plant system solver iterations within a single HVAC iteration
\note Larger values will increase runtime but might improve solution accuracy for complicated plant systems
\note Complex plants include: several interconnected loops, heat recovery, thermal load following generators, etc.
\type integer
\default 2
\minimum 1
N4 ; \field Maximum Plant Iterations
\note Controls the maximum number of plant system solver iterations within a single HVAC iteration
\note Smaller values might decrease runtime but could decrease solution accuracy for complicated plant systems
\type integer
\default 8
\minimum 2

more

Thank you for your answer. As you say, the temperature difference between the supply side and the demand side may come in part from the difference in convergence of the independent simulation of the supply side and the demand side. However, I think the temperature difference should be largely due to the pump heat. I tried increasing the Design Pump Head [Pa] gradually, eventually to 100 times larger than the original Design Pump Head. The temperature difference between the supply side and the demand side increased in proportion to the pump heat.

( 2023-06-13 13:31:58 -0500 )edit