Conservative design cooling load calculations vs baseline sizing

10 posts / 0 new
Last post


When sizing cooling equipment for apartments, the design team on my project
used a worst case scenario approach of maxing out all internal loads. For
instance, lighting would be on during the day, with full occupancy and all
plug equipment running, while the sun is hitting the windows. Adjacent
rooms would be considered without cooling.
For LEED, the sizing for the baseline is based on schedules representing
typical operations for occupancy, lighting, plug, etc. That leads the
baseline to size equipment that has smaller capacity than what considered
for the proposed design.
There's an extra dimension, in the fact the design team sized equipment for
each individual room, reaching their peak load at different times of the
day. So the total cooling capacity is the sum of the worst possible
scenarios. The model currently uses a single zone per apartment, with a not
so worst case scenario.
How do I solve this conundrum? Am I missing something?

Patrick Bivona's picture
Joined: 2012-11-12
Reputation: 0

Hi Patrick,

I just wanted to clarify for you that Appendix G (G3. requires the simulation of sizing runs for equipment selection based on either the peak from the weather file or ASHRAE 99.6% heating and 1% cooling design temperatures. As far as I know this is not something USGBC or GBCI has written a clarifying rule on for LEED.

As far as being frustrated over engineers doing overly conservative load calculations, welcome to life as an energy analyst. The best you can hope for is to use simulation to show them and the building owner that they are over engineering the system and try to get some concession. Ultimately at the end of the day it's the engineer who is stamping the work and taking on the liability that has the final say.


Sent from my iPad

Michael Tillou's picture
Joined: 2011-10-02
Reputation: 0

Hi Mike,

I can't say I'm frustrated yet. Just going through the learning curve. I'm
sure frustration will come in time!

I was musing about the potential imbalance in sizing approach between the
proposed design and the baseline design. If we assume for a second that
engineers also use ASHRAE design days when sizing equipment for the
proposed design, there is still a potentially significant difference
between schedules used for sizing and normal operation schedules. Is the
1.15 sizing factor for baseline cooling enough to cover such imbalance? I
don't know enough to say so yet.


Patrick Bivona's picture
Joined: 2012-11-12
Reputation: 0

To be honest, the only potential "overly conservative" sizing I can see
going on here is if the building is served by a central cooling plant and
that plant has been sized based on the sum of the peak loads and not the
peak coincident block load for the building. If this is in fact how they
sized the system then there is potentially a large cost savings that could
be had by appropriately sizing the central plant.

The "worst case" scenario that you describe is a real scenario that can
happen fairly easily. All it takes is someone to be at home with their
lights on and appliances running on a peak summer day. That doesn't seem
like a very unlikely scenario at the zone level.

As for how the Baseline equipment should be sized, it should be no
different in terms of lighting, plug loads, and solar gains. The only
stipulation on the baseline simulation is that it is oversized by a
prescriptive 25% for heating and 15% for cooling. ASHRAE 90.1 doesn't
strictly define what a "sizing run" is, but it would be considered standard
practice to utilize the peak lighting and plug loads during your sizing
calculations (i.e. don't assume any diversity on the lights and use the
maximum anticipated coincident plug load value).


Robby Oylear's picture
Joined: 2011-09-30
Reputation: 202

(A couple folks responded before I finished... sorry if this appears to parrot any more current responses!)

HI Patrick,

To play devil's advocate, I'll just put forward none of the practices you're lamenting are unusual or even necessarily a bad idea from a design perspective.

That said, it isn't unusual in my experience for proposed model cooling capacities to sum larger than a baseline model counterpart (and this can quickly become an arbitrary figure when you've got many small systems). If however you are looking at many magnitudes difference between your models, that in my experience can be a bright red flag to carefully review your model's assumptions of loads & scheduling against the mechanical designer's - you may be missing critical information.

I would also contend LEED & 90.1 together have very little to say regarding equipment sizing methodology - particularly for determing schedules or which loads are to be considered at 0%, 100%, or anything between for determining capacities. The only related rules that come to mind are the weather items Mike mentions and the basic requirement for the baseline model to match the proposed model generally for all such scheduling inputs.

Your energy modeling software of choice may dictate what degree of control you have in this matter, but I can volunteer that for eQuest/DOE2 a separate set of schedules and weather conditions can be assigned for both heating and cooling design days, independent of normal operation. These inputs are used on the LOADS side of the simulation and pretty much directly effect the results for system capacity autosizing. Coming out of the wizards, you'll find by default eQuest assumes "normal" fractional operation schedules are to be used for heating and cooling design days (HDD/CDD). This at times can be in sharp contrast to the load assumptions used for equipment sizing protocol outside of an annual simulation, and if I were you I'd also consider reviewing such scheduling assumptions with your mechanical designer.

[cid:image001.png at 01CEF197.B330F950]

[cid:image002.png at 01CEF199.83784270]

I'd also reinforce that while energy modeling can inform and provide insight one can't get anywhere else, at the end of the day the burden of such decisions (and the associated liabilities) are in the designer's hands. I try to convey such information in a way that avoids "butting heads," but there are times someone at the table isn't really there to talk, learn, and consider changing their minds. You can try your best to make a case, but ultimately you can't make yourself responsible for what other's paychecks are written for. Done politically, as Mike suggests, you can leverage your expertise to potentially win new design work if that's within your scope of interests, but be careful to avoid burning bridges ;).

Hope that helps a bit!


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


An under-appreciated aspect of residential loads calculations is
"swing." The building and its contents will absorb a lot of energy
if the air temperature is allowed to rise a few degrees during peak
events. Modeling work that I have done in ASHRAE 1199-RP and
elsewhere shows that tolerating an occasional 3 F temp rise above the
set point can reduce the required equipment size by up to 35%
(depending on building mass etc. etc.)

It is excessively conservative for cooling calcs to assume peak
internal gains coincident with peak outdoor conditions. On hot
afternoons, people very rarely cook full turkey dinners with all the
TVs on. If they do, let them warm up a little. Better than paying
for an oversized system and running it at lower part load all the time.

A major (and perhaps the only significant) distinction between
residential and non-residential design philosophy is that in non-res
situations, people are constrained about what they can do to control
their environment. In non-res, the worker must stay at his/her desk
and slave on ... in res, the occupants can close the shades, move out
of the sun, and decide to barbecue when it is hot. Much more
forgiving situation, so it is nuts to double-oversize res cooling
systems for the worst worst conditions. In my humble opinion.

All of which relates to actually designing a good system, as opposed
to doing 90.1 analysis. The two activities probably have minimal
overlap at best.

Chip Barnaby, BEMP

Chip Barnaby's picture
Joined: 2011-10-01
Reputation: 0


What you say makes sense to a point. I would argue that most Owner's would
expect the system to work under the very conditions you describe as
excessively conservative and if they don't the contractor and engineer will
be hearing about it down the line. This is something that should be
discussed on a project by project basis, because when presented properly
the Owner can see considerable savings from such a design as long as they
understand the risk they are taking on by accepting the reduced system size.


Robby Oylear's picture
Joined: 2011-09-30
Reputation: 202

When selecting a residential unit, I would undersize the cooling capacity
and oversize the heating capacity. The energy use with decreasing cooling
capacity is not directly proportional (about 1 kw/ton at 100% capacity to 3
kw/ton at 20% capacity). The energy use of a heating system tends to be
directly proportional to the heating load (up to 90% efficiency). This is
also true for commercial buildings. Residential unit control is usually
ON-OFF. It is more comfortable to let the fan and cooling system run
continuously and let the temperature rise a few degrees under extreme
conditions. 150 years ago there was no air conditioning and mankind

Varkie Thomas's picture
Joined: 2011-09-30
Reputation: 0

This statement is a little troubling to me:

"Adjacent rooms would be considered without cooling."

I would suggest that all conditioned spaces be considered conditioned,
not at ambient temperature, although that is a design decision and must
be made by the design engineer.

Robert Wichert P.Eng. LEED AP BD&C

RobertWichert's picture
Joined: 2011-10-02
Reputation: 201

This is a very interesting conversation.

One that could also be held about sizing domestic hot water systems. ASHRAE
also tends to grossly oversize those, based on what I've seen in existing
buildings on projects where they wanted to replace the system. By
monitoring actual flow and doing the ASHRAE calc, it wasn't rare that
ASHRAE recommended double of what was actually needed.

Anyway, I'm sure you are all aware of this, but I feel I must say it anyway
because it's always good to be reminded of the simple things: you should be
careful when applying the statement "undersize the cooling capacity and
oversize the heating capacity" when selecting PTHPs, especially if you end
up with a grossly oversized cooling side. You should select the system
based on the cooling capacity needed anyway (and add an electric resistance
if need be in certain climate zones), unless you really like humidity and
what comes with it. I've witnessed that mistake before and it's not pretty,
especially in a humid climate. If this weak contribution can prevent
someone from making that mistake then it would have been worth it...

Julien Marrec, EBCP, BPI MFBA

jmarrec's picture
Joined: 2013-01-09
Reputation: 0