I was interested to see if anyone would share how they model the
exception to G3.1.1 in ASHRAE 90.1-2007 (or 04). This is the part that
says when you have a baseline building with system 5-8, you're supposed
to model system 3 or 4 for any spaces that "have occupancy or process
loads or schedules that differ significantly from the rest of the
building." It goes on to say that basically the peak thermal loads must
differ by 10 Btu / h-ft2 or more from the average of the other spaces,
or that schedules must differ by more than 40 full load hours per week,
for this exception to apply.

I've had reviewers question certain rooms that I've put on separate
PSZ-AC systems, and so I've sort of defaulted to putting almost all
rooms on the main systems (5-8). However, in looking at some of my
recent models, depending on interior loads and climate, the actual peak
cooling load on a square foot basis varies anywhere from 5-20 btu /
h-ft2 range for offices. So this means there is a big difference
between those models with 5 btu / h-ft2 peak and those with 20. For the
models with 5, almost every space in the building is going to be within
+/- 10 btu/h-ft2 of that average. However, those buildings that average
around 20 will have many rooms that are < 10 btu/h-ft2 peak (ie any
internal space whose light + plug load density is < 3.0 W / sq ft).

To use a simple example as a question: if your particular office
building averaged 20 btuh / h-ft2 peak cooling load, would you model a
separate PSZ-AC system for a core bathroom, whose peak cooling load is
only 3 btu/h-ft2?

There are lots of rooms where I question whether I should do a separate
PSZ-AC/HP system (storage rooms, stairs, mechanical rooms, lobbies,
etc). I can see where these meet the 10 Btu/h-ft2 delta exception.

So does everyone really model a bunch of PSZ-AC systems based
specifically on the average peak thermal loads of the building?

I'm thinking out loud, but if I had an office building that averaged on
the low side (5 Btu/h-ft2) for cooling peaks, it seems pretty easy to
have a perimeter conference room that exceeded 15 Btu / h-ft2 peak.
Seems like an exhaustive process to confirm this delta for every room in
a model to determine whether it needs its own system...

Sorry for the long email. Thanks!

James Hansen, PE, LEED AP