Building Orientation Accuracy?

10 posts / 0 new
Last post

I have noticed that eQUEST does not accurately determine the cooling and
heating loads when rotating the building 180 degrees using the Azimuth
feature under the building properties in the detailed edit mode. I
created and attached a file where there are two classrooms. One
classroom faces north and one faces south.

When I run the simulation I get the following results:

North Facing: Space Cooling: 4.1 MBTU, Space Heating: 21.9 MBTU

South Facing: Space Cooling: 4.3 MBTU, Space Heating: 18.9 MBTU

When I change the orientation 180 degrees, the results are:

North Facing: Space Cooling: 3.9 MBTU, Space Heating: 23.0 MBTU

South Facing: Space Cooling: 3.6 MBTU, Space Heating: 18.1 MBTU

When I rotate the site 180 degrees the north and south facing energy
consumption is different (they should be exactly the same). Also, the
space cooling does not make sense because the building's weather file is
Dayton Ohio and the south facing classroom should use more cooling than
the north facing classroom.

This makes me assume that eQUEST has a problem with building rotation in
the detailed edit mode.

Please check to see if I am assuming this correctly because this will
affect every LEED project.

Otto Schwieterman

Otto Schwieterman's picture
Joined: 2011-09-30
Reputation: 200

Otto,
Your north and south have different inputs. I specifically noted plenums temperature set points and plenum floor weight.
I suggest you clean it up and run it again.
Paul Riemer

Paul Riemer's picture
Offline
Joined: 2011-09-30
Reputation: 0

I fixed the differences and now I get the following results:

North Facing: Space Cooling: 4.1 MBTU, Space Heating: 21.7 MBTU

South Facing: Space Cooling: 4.3 MBTU, Space Heating: 18.7 MBTU

When I change the orientation 180 degrees, the results are:

North Facing: Space Cooling: 3.8 MBTU, Space Heating: 22.6 MBTU

South Facing: Space Cooling: 3.6 MBTU, Space Heating: 17.8 MBTU

These also seem incorrect for the same reason.

Otto Schwieterman's picture
Joined: 2011-09-30
Reputation: 200

Your interior wall types and adjacent spaces are still different from north to south. EQUEST and DOE-2 work well for me, good luck with your modeling.

Paul Riemer's picture
Offline
Joined: 2011-09-30
Reputation: 0

Otto,

I just looked at your revised files - you are still modeling two
seemingly fairly different classrooms. Note your "North" classroom has
adiabatic internal walls/ceiling and your "South" classroom has heat
transferring ones. I haven't scoured for other differences that may
well be in there regarding schedule assignments and such - I don't know
of a fast way to do so.

If you're truly concerned with eQuest's ability to correctly account for
a varying azimuth and want to test it, you could make your task easier
by creating a square classroom with identical walls on each side and
observe identical results with 0,90,180,270 degree azimuths.

NICK CATON, E.I.T.

Nick-Caton's picture
Offline
Joined: 2011-09-30
Reputation: 805

I want to thank everyone for their help.

Originally, I created one north facing classroom and ran the
simulation, then I rotated it 180 degrees. The heating results were
acceptable but the cooling results were odd. The cooling consumption was
greater on the north facing orientation than the south facing
orientation. The weather file is Dayton Ohio so this does not make
sense. I then tried to copy the classroom and rotate one of the
classrooms so I had a north facing and south facing classroom. I did not
realize that eQUEST changed some of the plenum and interior wall values
when I copied the original classroom.

I am still trying to figure out why the north facing classroom requires
more cooling than when I rotate the building 180 degrees and essentially
have a south facing classroom. I would think that because of solar heat
gain the south facing classroom would require more cooling. I have
attached a simplified classroom (I did not copy so there should be no
room for errors).

Otto Schwieterman's picture
Joined: 2011-09-30
Reputation: 200

With that extra background, and glancing at this new baseline, I can
guess where your analysis may have gone awry...

Your assumed behavior, that "energy spent cooling" ought to be higher
with the glass facing South, is certainly logical for a project in the
Northern Hemisphere, as the cooling season solar contributions ought to
be higher facing South (barring any shades, of course). If you'll run a
180-azimuth parametric calc on your "North Baseline" most recently
uploaded, you can observe this is exactly the behavior being modeled
(check out SIM report LS-B to compare solar contributions for each run).

Where I think you're making a leap is assuming a higher "space cool"
consumption figures equal higher cooling loads. If you'll check those
same reports, you'll find you actually do have higher peak cooling loads
when facing south. The lower South-facing "space cool" energy
consumption figures come alongside higher "ventilation" consumption
figures - eQuest appears to be up-sizing your South-facing CFM's and/or
run-time instead of cooling coil/condenser capacities. With some
additional inputs you may be able to lock-in the fan and airflow
capacities, which would force your condensers to do more of the fan's
work (higher coil capacities), resulting in a higher "space cool" total,
if that's what you'd like the reports to show. Ultimately your
South-facing run may simply need to run your fans more often over the
year, so you may not be able to entirely avoid a higher ventilation
total...

Of note, for any orientation, it appears you have a significant number
of under-heated hours. You've may want to review your equipment heating
temperature setpoints and/or space/equipment CFM inputs. You may
already know this, but as an additional reminder: your classroom as
modeled is only going to interact with the exterior conditions on the
single exterior wall, as the other three "internal walls" simply model
space thermal massing and heat transfer to other spaces.

Best of luck,

NICK CATON, E.I.T.

Nick-Caton's picture
Offline
Joined: 2011-09-30
Reputation: 805

Hi Otto,

If I were a betting woman I would place a lot of money on eQUEST functioning
properly vis a vis building orientation. That bug would have been fixed
years ago. The beauty of an 8,760 hour/year interactive program are just
that: 8760 hours of weather data and a program that accounts for all of the
building interactions. The interactions include the shell, the lights, the
people, the coming and going, shutting on and off, etc., etc. You are
getting good feedback from Paul and Nick. Keep the faith and keep fixing
your input. Don't spend too much time thinking the model isn't working
right; spend it instead making the input right.

Regards,
Carol

cmg750's picture
Offline
Joined: 2010-10-05
Reputation: 0

I want to thank everyone for their help. This is what I found out.

As Nick mentioned, I was allowing eQUEST to size my heating and cooling
equipment. I specified the CFM but not the capacity. I did this because
I have different building layouts with different window sizes. I went
back into my file and set the capacity for the heating and cooling
equipment and then rotated the building, I got results as I would
expect. The summer cooling energy consumption was greater for the south
facing classrooms than the north facing classrooms.

When working with LEED, has anyone "locked in" the heating and cooling
capacity before rotating the building? If not, they may run into the
same problem that I had.

Scott, I am using a modified version of eQUEST 3-63b.

Thanks for everyone's help!

Otto Schwieterman's picture
Joined: 2011-09-30
Reputation: 200

Otto,

Regarding LEED, the basic "rules" regarding baseline system heating and
cooling capacities are that they must be auto-sized, then incrementally
increased (in eQuest, using the heating and cooling capacity ratios) as
may be necessary to reign in unmet hours that may occur. Fan sizing is
another can of worms that happens in the middle of that process. My
interpretation has been this procedure applies specifically to the
"normal" (0-degree azimuth) rotation. Where the 4 baseline rotations
exist specifically to "normalize" the variable of building orientation
on utility costs (for LEED), I think it makes sense to allow eQuest to
autosize each new rotation in the same manner as the first.

Your specific case resulted in an inordinately dramatic change in the
ratio of "space cooling" to "ventilation" annual consumption, but that's
because it was just a single classroom with one exterior face. In my
experience, entire buildings with all orientations represented being
rotated vary only slightly in the annual consumption figures. For that
reason, the 'problem' you noticed doesn't really surface as something of
concern. Note that I don't think we're expected to check and meet the
same unmet hours criteria for the rotated parametric runs as we are for
the baseline (zero-degree) calculation - only to run the same, finalized
baseline model at 4 different orientations.

I hope others will join this discussion and share their thoughts. On
the other hand my personal deadlines are looming for this week, so I may
be a bit less active =).

NICK CATON, E.I.T.

Nick-Caton's picture
Offline
Joined: 2011-09-30
Reputation: 805