[Bldg-sim] Direct Evaporative Coolers & the eQUEST Games

4 posts / 0 new
Last post

Help! I have been modeling a car dealership which uses several
disparate systems in combination to condition spaces.

The baseline HVAC system is Packaged Single Zone with Gas Heat.

The actual design for many of the service bays is:

Direct Evaporative Cooler
Unit Heater(s)
Exhaust Fans

For purposes of modeling the actual design, it seemed appropriate to
model the system as "Evaporative Cool' with "Furnace" selected for heat.
The OA fraction was set to 1.0 (more on this later). Actual design
capacities were input into the model. The unfortunate by-product of
this (as is often the case) was that there were a bunch of unmet hours
(76), mostly cooling.

So after playing around with several settings, the only way to get the
unmet cooling hours down seemed to be to change the system type back to
"Package Single Zone" and set the unitary power to 0 for the
'compressor' which really does not exist. This makes the unmet hours go
away, but seems to create a higher heating load, despite the fact that
the actual design has higher thermal efficiency than the baseline. The
space heating almost doubles from the baseline to the proposed.

My suspicion has been that eQUEST does not realize that the evap.
coolers only run during the summer, and thinks it has to heat 100% OA
year-round. Appendix G says that the minimum outdoor ventilation rates
should be the same (right or wrong) for the proposed and baseline
designs. So...my solution was to set a minimum OA schedule, with the
summer being 1 and the winter at 0. Well...this may be closer to
reality, but now the unmet hours spike to 423!

So...it seems to be the recurring problem of fixing one problem, only to
create another one. Any ideas on how better to simulate this?

Thanks for any tips in advance.


Patrick A. Kearns, P.E.

Kearns, Patrick's picture
Joined: 2011-10-02
Reputation: 0

Two approaches to consider.

1. Ask yourself is whether the evaporative cooler is even capable of
providing the level of cooling you tell it (ie: satisfying your cooling
setpoint)? Typically direct evaporative coolers are not intended to provide
full cooling capability just cooling relief. Address the problem by raising
your cooling setpoint in the spaces being served by the evap coolers. Make
sure to use the same cooling schedule for your baseline model.

2. Simulate the HVAC systems in the proposed design with mechanical cooling
that has the same performance as the baseline but use the direct evap cooler
add-on feature as the first stage of cooling. While your proposed design
technically doesn't have mechanical cooling I believe the intent of Appendix
G would require you to simulate it in order to satisfy your unmet cooling
hours. The add-on evap cooling feature in eQuest is able to deal with the
changes in ventilation air required for direct evap cooling so you can get
rid of the crazy MIN-OA-SCH. The user manuals should give you all the info
you need about the evap cooling keywords you will want to use. I think
there is even a diagram showing how the evap cooling add-on ties into the
HVAC system.

As far as your comment about fixing one problem to create another....welcome
to energy modeling in DOE2.2.

good luck

Michael Tillou, PE, LEED

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

Here might be why:

When you design an evaporative cooler, you wanna use the design wetbulb and
the effectiveness to figure out the the design supply air T and then the
design CFM. Say the location is Salt Lake City, UT. Because the design
wetbulb is 68F, and the resulted supply air T is slightly higher than 68F,
say 70F. Then the designed CFM would be much higher than the value you get
when using 55F as the design supply T. However, eQUEST default supply T is
usually 55F (specified by the keyword: MIN-SUPPLY-T). If you change that to
68 or even higher, your cooling unmet hours are almost gone.

With this change, you will see the CFM surges dramatically, which is
probably because your high internal loads and higher supply T. Therefore,
the previous huge unmet hours is due to the lack of CFM. If by doing this,
eQUEST designed CFM is found to be much higher than the actual design, it is
because you and the mechanical guy use different internal loads (assuming
the mechanical design is sound). You would have to talk to him/her to clear
that up.

Of course, you can imagine how much fan energy VFDs on the supply fan of the
evaporative cooler can save if internal gains are huge. Most of the time,
you can get a lower supply T because the OA wetbulb is lower than 68F.
Therefore, although to erase the unmet hours, your design supply T should
be slightly higher than 68F, ideally you may want to run supply T as low as
possible to save fan energy. So that MIN-SUPPLY-T should be as low as
possible. But I just don't know if there is any other place in eQUEST to set
the design supply T then. If we set the MIN-SUPPLY-T to about 68, supply T
would never go below 68 even if it can. So a lot of fan energy would be

So anybody knows another place to plug in the design supply air T?

*Xiang (Shawn) Liu*

Shawn's picture
Joined: 2011-10-02
Reputation: 0

If you are using Packaged Single Zone system to simulate the evaporative
cooler (without mechanical cooling). Other tricks you have to play to get
the system running with no unmet cooling hours are:

1. Specify the mechanical cooling capacity to a very small number, like 0.01,
so mechanical cooling wont kick off. (As I recall, 0 wont work, because
eQUEST treats cooling capacity of 0 as autosizing, so you would still have
machanical cooling).

2. Find another keyword named EVAP-CL-LIMIT-T in the add-on evaporative
cooling tab, raise it from the default value (usually 90) to over the
maximum OA drybulb of your site. EVAP-CL-LIMIT-T specifies outside air
dry-bulb temperature above which the evaporative cooler will be turned
off. A low EVAP-CL-LIMIT-T will turn off the evaporative cooler when cooling
is still needed, and without mechanical cooling, your unmet cooling hours
will go up because nothing is air conditioning the space.

*Xiang (Shawn) Liu*

Xiang Liu's picture
Joined: 2011-10-02
Reputation: 0