For what it's worth, the example LS-F report in the help file
documentation also shows only 0's in the roof component. From within
eQuest, go to the Help menu, select DOE-2 Help. Expand the "Volume 4:
Libraries & Reports" and then the "Reports" topics. LS-E and LS-F are
under the loads reports. The text says the monthly space components are
summed across spaces. The definition of the LS-F report excludes plenum
spaces:

This report gives a breakdown of loads on a monthly basis for the entire
building, according to the source of the load. The loads in
unconditioned spaces (ZONE-TYPE = UNCONDITIONED or PLENUM) are not
included; all entries are in millions of Btu/month.

This just explains why they don't show up in the LS-F...I believe that
the mechanism used by eQuest to account for heat gain in the plenums is
to apply that heat gain to the return air before the system, thereby
increasing temperature, humidity, etc. in the RA stream due to the loads
in the plenum spaces connected to a given system. Someone from Hirsch
can jump in if I got that wrong.

I'm curious if anyone else has dug deeper on this than I have, though?

Jeremy R. Poling, PE, LEED AP+BDC

Hey everyone,

Sorry to chime in late... but I have some input to wedge into the
conversation as it's wrapping up:

First: As has already been said, LS-F is inclusive of all 'conditioned'
spaces. I didn't realize the DOE2 entries explicitly (more or less) say
that, and had figured it out for myself via logic/gray hair growth... =)
I'll take this as a reminder that the DOE2 help files are perpetually
more useful than they may seem!

Second: I'm also no developer, but Bruce's conclusions didn't sit well
with me so I did a quick study to confirm... I don't think you can
characterize plenum loads as being *entirely* separated from the
conditioned spaces... Do such loads (really hot/cold plenums) get
directed straight back to the central air handler, or do they also
partially interact with the zones as well?

I know ceilings generated by eQuest wizards by default are not
adiabatic, but do use a construction with defined heat transmittance
properties.

I slapped together a 1-zone, 1-floor dummy model and tried changing the
default ceiling construction (U~0.5) to adiabatic in a parametric run.
Return air path was set to "plenum zones." The plenum zone itself was
changed from "unconditioned" to "plenum:"

Lo and behold: The two runs do indeed have differing
cooling/heating/ventilation consumptions (attached)!

* Investigation of the LS-F reports shows the whole building
deals with identical load components altogether - which is expected.

* Comparing of the LS-E reports reveals the plenum with an
adiabatic ceiling has less heating and less cooling to deal with over
the year. Both runs' conditioned space however deal with identical
loads for each month.

Conclusions!

1. Bruce is NOT to be questioned: Plenums modeled as a return air
path do not transmit their loads to the corresponding spaces below.
They carry their loads directly to the parent system.

2. Return air plenums DO however receive heating/cooling loads
transmitted from the conditioned spaces below via the ceiling surfaces.
It's basically a 1-way street.

3. I'd agree real-world behavior is somewhere in-between:
Return-air plenum ceilings should act as a 2-way street thermally, and a
single plenum can be hot on one side but cold on another at any given
instance. In many cases however, I suspect the assumption of all loads
staying in the plenum is closer to reality than assuming they all dump
into the corresponding space below, from an supply airflow-sizing
perspective.

~Nick

NICK CATON, P.E.