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

Demand End Use for Fans and Cooling seems low [closed]

asked 2014-11-04 12:45:02 -0500

nfonner's avatar

updated 2020-03-08 11:30:56 -0500

This is a potentially simple problem to solve, but after modeling a typical office building in OpenStudio with typical loads (5p/1000sf, 0.9w/sf lighting, ~ 1w/sf equipment/elevators, 62.1 defaults, etc.) I am seeing drastically low demand in the "Demand End Use Components Summary" for Cooling and Fans.

Cooling is ~ 5 btuh/sf (576.4 sf/ton @ 88tons and 51,000 sf) (sf/ton is pretty high, Fans = see example below Heating is ~ 23 btuh/sf (seems about right)

I know a portion of this is due to the design days in the schedules, sizing vs energy runs, etc. but what could account for the fact that I'm seeing ~30% of the fan demand power compared to that sized in the "Equipment Summary".

Eg. 39.6kW (0.0015 hp/cfm is what I input) from Equipment Summary & 13.8 kW from Demand End Use Components Summary (and LEED summary)

Almost all the defaults are left alone for the template system (Packaged DX Rooftop VAV with reheat) 3.4 COP on DX 2speed Sensible Load sized 55 degF Central Cooling Design Supply Air Temperature 0.0085 humidity ratio (~60 on the psych chart) 5.5 in wc fan pressure rise (represents approximate 90.1 fan power limitation on a VAV) 60% fan efficiency

Thanks for any light shed on this.

edit retag flag offensive reopen merge delete

Closed for the following reason duplicate question by nfonner
close date 2015-08-11 15:59:46.067208


I have noticed that simply switching the "Use Night Cycle" field to "Follow the HVAC Operation Schedule" from "Cycle on Full System if Heating or Cooling Required" sets a peak fan power on a Monday morning in July for each system that more closely matches (~20%) the new fan power calculated in Equipment Sizing. Problem is that most of the year the fan power is still hovering around 10-20% of that peak so the it seems fairly ridiculous in comparison.

The HVAC Operation Schedule shuts the system off during unoccupied hours.

nfonner's avatar nfonner  ( 2014-11-05 21:35:44 -0500 )edit

2 Answers

Sort by ยป oldest newest most voted

answered 2014-11-07 14:24:08 -0500

If your heating and cooling setpoint schedules for the design days include a thermostat setback/setup and your system is being autosized, the resulting capacities and flow rates may be way oversized because the autosizing routine is trying to enable the zone temperature to jump from the setback to the setup temperature in the space of a single simulation timestep. If this system is then run under normal operating conditions, the oversized capacities might allow it to meet the load in a very short space of time, and depending on the size of the window used to determine the peak, you might see funny numbers.

The behavior you are seeing for the night cycling vs. follow HVAC schedule makes sense if the system is oversized. If night cycling is enabled, the zone never gets more than 1-2F above/below setpoint before the system cycles on, at which point the load is pretty low. If the system is scheduled off at night, then when it comes back on on Monday morning, the zone temperature is well above/below setpoint, and the load is much closer to what was experienced during the design conditions.

edit flag offensive delete link more


Thanks for the feedback. I did experiment with the DD cooling schedules vs regular operation. I even updated the other default schedules cooling DDs to reflect actual peak occ./ lighting/ equip. This made the fan power demand (Demand Summary) increase by roughly 11.2 kW. (28.4kW vs 17.2 kW). However, the difference between design (43.7 kW) and regular operation (28.4 kW) is still quite noticeable.

I am used to reviewing system design load profiles in Trane Trace 700 to see how much time the system spends at 0, 10 , 20,... 100% part load based on demand. Does this exist in E+/ OS?


nfonner's avatar nfonner  ( 2014-11-11 13:56:02 -0500 )edit

Sorry, ran out of characters. I realize the night cycling issue presents fairly obvious problems when comparing apples to apples, but I am looking to demonstrate a possible outcome of having the HVAC system run 24/7 based on incorrect modeling assumptions. I have seen this surface a few times reviewing LEED models (not on behalf of GBCI) and wanted to cover this in some upcoming training. I also had the ventilation operating 24/7 in conjunction which is really throwing off my peaks given the air-side economizer cycling, but this is still not quite explaining my comment above.

Thanks again.

nfonner's avatar nfonner  ( 2014-11-11 14:08:02 -0500 )edit

answered 2014-12-01 23:04:44 -0500

nfonner's avatar

updated 2014-12-01 23:05:17 -0500

Just to keep this thread alive, I went back to see what else I could find. Using E+ and the ZoneComponentLoadSummary, I found that approximately 40% + of the Estimated Cooling Peak Load is being countered by a massive Ground Contact Floor load (Sensible-Delayed.) I was using the default ground temperatures in Energy+ (18 degC). Changing using the Building average indoor temp - 2 deg C still resulted in very high negative loads typically not associated with a sizing calculation for cooling. Finally, the slab processor measure (link below) cut the load in half. See examples below.

For me this is still quite high (at ~ 24% negative load,) but short of jumping all in...?

With 18 deg C
For example (Classroom with 1 exterior wall),
The Ground Contact Floor load is (-7,555 W.)
The Grand Total Load is (28,353 W.)

With Slab Processor measure (SurfaceProperty:OtherSideCoefficients) links below
Same classroom:
The Ground Contact Floor Load is still (-5,255 W)
The Grand Total Load is (36,446 W)
After a while of toying with the model to max out the cooling load, the ventilation would not show up any longer on this report, so I just used the same number from previous reporting for comparison.

edit flag offensive delete link more


Question Tools



Asked: 2014-11-04 12:45:02 -0500

Seen: 627 times

Last updated: Dec 01 '14