IES VE: Radiant, hollow core ceiling

6 posts / 0 new
Last post

I have a question regarding the capabilities of IES VE and its CFD package
and I was hoping someone might have some experience or insight to help get
me started on the following problem. We have a concept cooling design which
includes a radiant hollow core concrete ceiling through which evaporatively
cooled air is routed. The air which is routed through the radiant ceiling
comes from a central evaporative cooler and is then exhausted at the
exterior of each room after it has passed through the airflow passages in
the ceiling. Just to be clear, this is not a typical radiant system in that
electricity is being used to cool the slab but instead evaporatively cooled
outdoor air is being used in its place. We also want to couple these
effects with natural ventilation which occurs because of the use of operable

We are trying to analyze the effects of these techniques in a cooling season
in Colorado (hence the benefit of the evaporative cooler). We have
conducted rough calculations of the benefits of using such a system using
standard ASHRAE energy transfer techniques and we want to verify our
results. Have you ever heard of this type of system being modeled in IES VE
and/or is it capable of modeling such a system? Is there a way to modify
the "radiant ceiling" option in IES to take into account the saving which
would occur by using evaporatively cooled air in the place of electricity?
If not, is there a way to manually create such a system from the ground up?

Any help would be greatly appreciated. Thank you.

Rebecca Butler

Rebecca Butler's picture
Joined: 2011-10-02
Reputation: 0

Hi Rebecca,

What I would is utilise IESs' good capability in modelling building
physics. The hollow concrete slab could be modelled as a narrow,
concrete "plenum" above the zone to be conditioned via the cooled slab.
This "plenum" zone could then be supplied with the evaporatively cooled
air at estimated rate.

I would not go down the path of chilled beams as they are difficult to
control and esp difficult to monitor in the Vista. Also concrete is not
amongst the material types for chilled ceiling. The heat transfer rate
could get exaggerated with other material types.

Hope this helps.


Minu Agarwal's picture
Joined: 2011-10-02
Reputation: 0

We tried to model radiant systems in IES and it failed miserably (the
creation of a floor plenum, adding energy, trying to see what happened,
a nightmare).

I would suggest at this present date, TRNSYS or ROOM (E+TA), for those
who can wait ASHRAE is sponsoring the development of a radiant system
design and evaluation tool (RP 1383)

Peter Simmonds. Ph.D.

Peter Simmonds's picture
Joined: 2011-10-02
Reputation: 0

Rebecca and others,

Modeling air-cooled (or heated), hollow-core radiant slabs (such as TermoDeck) within IES Virtual Environment is a relatively straightforward matter:

1) Include the hollow-core geometry in your model. This must be segmented to properly capture the changing delta-T over distance as the supply air gains heat while passing through any sections of the floor cavity that are going to be in series rather than parallel configurations.

2) If the hollow core is cooled via natural ventilation, then the volume of the core spaces needs to be coupled to the plenums or zones that include the operable openings. Like windows or other operable openings in the occupied spaces, these openings can be controlled via schedules, interior conditions, climate variables, or formulae including any of these. The coupled MacroFlo dynamic bulk airflow modler within the VE will need to run whenever natural ventialtion is to be acconted for, but not if there were no operable windows and the hollow-coore was to be cooled by mechanically driven supply airflow.

In the case of natural ventialtion, you may also wish to use the MicroFlo CFD tool within the VE to determine the performacne of specific openings, cavities, or occupied spaces under specific conditons. The bulk airflow model and thermal simulation are used to set up initial boundary conditions for the CFD model, and then the results from the CFD run can then inform revision of the operable openings in the dynamic bulk-airflow + thermal modeling. While the CFD work can be valuable in determining performance and refining the model, it is not required to run the dynamic bulk-airflow + thermal modeling of the naturally ventilated spaces, nor is it required for modeling the hollow-core floor.

3) Apply convective heat transfer coefficients to the interior surfaces of the hollow-core material constructions appropriate to the use of this core space as a duct (see ASHRAE Fundamentals for determining appropriate coefficients). You will need to calculate the equivalent air-film resistance, as this is the value you will change, replacing the air-film resistance associated with the default variable coefficient for natural convection in the Constructions dialog. Then open the Derived Parameters dialog from there to confirm that the convective heat transfer coefficient calculated by the VE is what you intended (and thus that you have entered the correct air-film resistance value). Note also that the bottom surface construction of the hollow-core ceiling will be applied as the Ceiling construction for the occupied space below, and therefore the "Outside" surface air-film resistance for this element will be the one facing the interior of the hollow core.

4) If air-cooling or heating of the hollow core is mechanically driven, set up the system in ApacheHVAC with the supply air running through the hollow core geometry as would be done in the actual building. If it is a mixed-mode system, wherein the cavity is alternately cooled by mechanical supply air and natural ventilation, then set up the controls such that these modes hand-off appropriately-i.e., if no overlap is desired, ensure that controllers for both system include common sensed variable and thresholds, and non-overlapping responses at these thresholds.

IES VE is in fact exceptionally well suited to doing this and provides appropriate means of controlling such systems and of accounting for the full range of radiant and convective heat transfer paths involved.

As with any tool, taking the time to first understand how to use the tool properly is important to getting appropriate results.

Timothy Moore
Senior Consultant - Special Projects

Mobile: 415 810 2495
Office: 415 983-0603
timothy.moore at

**Design, Simulate + Innovate with the **

Timothy Moore2's picture
Joined: 2011-10-02
Reputation: 200


Regarding your particular evaporatively cooled hollow-core system in a mixed-mode building:

As you are not planning to naturally ventilate the hollow core itself, you will need to use MacroFlo (the dynamically coupled bulk airflow modeling application within the VE) a bit differently than I had described in my more general email on the topic. Rather than coupling operable opening to the hollow core, you will need to use MacroFlo to model the dynamic performance of the space associated with the operable windows in the building, any controls interlocks you may be planning, and to look at the building performance if these were to overlap (i.e., if there are no control interlocks).

While you may find the MicroFlo CFD module useful for looking at the potential for natural ventilation under specific conditions, it is the MacroFlo dynamic bulk airflow modeler you will need to use to control and assess the role of natural ventilation within the dynamic thermal simulations.

As you are planning to couple the floor cavities to a fan-driven mechanical HVAC system using evaporative cooling, you will need to set up this system in ApacheHVAC. If indirect-evap, you will use spray chamber upstream on the exhaust side of a heat exchanger. If direct evap, just the spray chamber. Obviously you can also use both of these. Indeed, the air-handler can be configured however you wish, including IDDE + DX dehum + Desiccant wheel regenerated by the condenser for the DX, or other unconventional configuration.

The VE is very well suited to modeling mixed-mode systems, evaporative cooling, and the radiant heat transfer that are involved in what you are looking at.

Timothy Moore
Senior Consultant - Special Projects

Mobile: 415 810 2495
Office: 415 983-0603
timothy.moore at

**Design, Simulate + Innovate with the **
Email Disclaimer

Timothy Moore2's picture
Joined: 2011-10-02
Reputation: 200


Thanks for the great summary. Do you know if it's possible to model
Thermodeck in eQuest by tricking the system somehow? I've bumped up the
thermal mass of the floors, but it's obviously not having as big an
impact as correctly modelling the CFD component. I had a thought about
modelling a pre-coil to simulate the damping effects of the slab in
reducing peaks and then manually removing that energy, but I'm not sure
if this is correct. Any thoughts?


Luka Matutinovic, B.A.Sc., LEED(r) AP

Matutinovic, Luka's picture
Joined: 2011-09-30
Reputation: 0