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

high electricity use for chiller

asked 2022-01-19 20:13:07 -0600

mattkoch's avatar

updated 2022-01-21 07:08:26 -0600

I am trying to develop a LEED v4.1 model for a 38,000 ft2 two story school building. The ASHRAE 90.1-2016 Appendix G baseline building for this appears to be a System 6, so one packaged VAV DX cooled, electrically heated system per floor, with PFP terminal units. The proposed building uses an air-cooled chiller and a natural gas boiler. I developed the baseline building first, then changed the two packaged systems out for two air handling units, and the chiller and boiler loops get created automatically. The chiller is water-cooled, so I switched that out for an air-cooled one.

I did this just to get an early indication of any improvement due to switching to chilled water. The packaged system cooling coils have a COP of 3, the chiller has a COP of 5.5. When I run the simulation, the baseline end-uses 1236.60 GJ in cooling electricity, while the proposed system end-uses a whopping 2967.77 GJ This does not seem right, given the COPs used, and destroys any hopes of making LEED points or even complying with Appendix G? Any thoughts would be appreciated.

I am also entertaining passing this project (and its sister project) on to anyone interested in and capable of doing LEED v4.1 energy modeling on short notice (1 month) - I can provide the geometry input at this point. You can contact me at if interested. (OpenStudio Application 1.2.1)

edit retag flag offensive close merge delete

2 Answers

Sort by ยป oldest newest most voted

answered 2022-01-21 07:11:31 -0600

mattkoch's avatar

Looks like this might have been a case of the old EIRFPLR curve for the air-cooled chiller. The OpenStudio default curve is just out to lunch on that. So I used the EIRFPLR as well as CapFT and EIRFT curves from the PNNL ASHRAE 90.1-2016 PRM Reference Manual, and results look more plausible now.

edit flag offensive delete link more


When this happened or I was unable to get detailed performance data, I would often simply use ChillerConstantCOP (not wrapped by OpenStudio as of v3.3.0, so requires an EnergyPlus Measure).

MatthewSteen's avatar MatthewSteen  ( 2022-01-21 13:59:08 -0600 )edit

Thanks Matthew, I appreciate the hint to a constant COP chiller. Sadly, I have spent countless hours developing measures in the past few months. I simply have to get the project out now with what I have. I am just glad that Unmet Hours is replete with experts such as yourself. Otherwise I would not have stumbled across that known EIRFPLR issue. Still, even with that resolved, ASHRAE 90.1-2016 is a very difficult endeavor, especially with its PCI approach and the handicapping with the Building Performance Factor, which happens to be a puny 0.47 in my case. That leaves not much room.

mattkoch's avatar mattkoch  ( 2022-01-21 14:16:21 -0600 )edit

answered 2022-01-23 09:56:45 -0600

Yes the part load curve is likely the culprit. To get that more accurate to your selected chiller, there are some handy curve-fit tools linked at this answer here:

edit flag offensive delete link more

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: 2022-01-19 20:13:07 -0600

Seen: 214 times

Last updated: Jan 23 '22