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

results check on variable primary loop

asked 2018-01-17 17:09:12 -0600

Matt Koch's avatar

updated 2018-01-18 12:14:24 -0600

I am simulating a test case where I impose a pre-calculated load profile upon a variable primary loop with water-cooled chiller. There is only one chiller, one cooling tower, one variable speed chilled water pump and one constant speed condenser water pump. In the OpenStudio HVAC diagram. the chilled water pump is at the chiller inlet and the condenser water pump is at the cooling tower outlet. I am reporting the chiller's evaporator and condenser mass flow rates and inlet and outlet temperatures and the cooling tower's mass flow rate and inlet and outlet temperatures. I am also reporting the chiller's evaporator and condenser as well as the cooling tower's heat transfer rates and the pumps' heat gain rates. The idea is obviously to see if everything matches up for each of the 8,760 h.

Much of it it does. However, I see major discrepancies between the cooling tower heat transfer rate and the sum of the condenser heat transfer rate and condenser water pump heat gain rate. I would have thought the sum of the latter two should have been equal to the former. Furthermore, I see minor discrepancies between the evaporator heat transfer rate and the sum of the load profile heat transfer rate and chilled water pump heat gain rate. Again, I would have thought the sum of the latter two should have been equal to the former.

Moreover, the evaporator inlet temperature is equal to the chilled water pump outlet temperature, as I would have expected. However, the Condenser inlet temperature is slightly different from the condenser water pump outlet temperature, when I would have thought they should be the same. What's more, the condenser water pump outlet temperature is often less than the cooling tower outlet temperature, when I thought it should be greater than that because this pump sits at the outlet of the cooling tower and should add heat. Finally, the condenser outlet temperature never equals the tower inlet temperature.

Also, using the difference between cooling tower inlet and outlet temperature as the range and the difference between cooling tower outlet temperature and wet bulb temperature as the approach, with the air flow rate ratio as provided by the calculation, I would have expected that the CoolTools approach temperature would be equal to this approach temperature at least during some periods, where the cooling tower operates in forced convection mode. However, I find this to be the case at no time.

Certainly I am missing something, such as a perhaps a thermal storage effect. However, I did this both with "Rate" and "Energy" output variables, thinking that the "Energy" output variables might smooth out any rate effects, but I get the same as if I multiplied the "Rate" output variables with 3600.

Any insight would be appreciated.

edit retag flag offensive close merge delete

1 Answer

Sort by ยป oldest newest most voted

answered 2018-01-18 14:24:43 -0600

I believe that you are witnessing the impact of EnergyPlus's methods of accounting for loop fluid capacitance and pump heat.

In short, EnergyPlus puts an imaginary storage tank at the inlet of each plant side -- supply and demand. The volume of each tank is the volume of fluid in that respective side of the plant. The tank temperature is then solved for using an energy balance. This is why you find that supply outlet temperature does not always equal demand inlet temperature, and demand outlet temperature does not always equal supply inlet temperature.

Pump heat is also lumped into the plant loop fluid's capacitance calculations through these imaginary storage tanks.

edit flag offensive delete link more



Thank you for confirming that. I had also stumbled across this same documentation last night, after I had posted my question already. I was in fact trying to set the loop volumes to zero and compare the outcome of that, but the simulation fails when I do that. So I suppose the loop volume has a "soothing" effect on the convergence, but it does produce somewhat non-intuitive results. In fact, I am starting to think that plant loops should not be viewed as traditional loops at all, i.e. those one might be used to from PI&D diagrams.

Matt Koch's avatar Matt Koch  ( 2018-01-18 14:57:57 -0600 )edit

Your Answer

Please start posting anonymously - your entry will be published after you log in or create a new account.

Add Answer


Question Tools



Asked: 2018-01-17 17:09:12 -0600

Seen: 212 times

Last updated: Jan 18 '18