Question-and-Answer Resource for the Building Energy Modeling Community
Get started with the Help page
Ask Your Question

Thermal performance of tents- What is the best way of ground modelling?

asked 2019-03-08 06:51:12 -0500

a.rijs's avatar

updated 2019-05-02 02:37:34 -0500

I use EnergyPlus 9.0.1. I want to model the thermal performance of a tent. The tent doesn’t have any floor insulation. The floor is a kind of plastic foil. Moreover, I don’t have any information about the ground type. My question is- what is the best way to model the ground? I have done research on that and I understood that there are a few methods of ground-coupling modelling: 1. Simple method- assigning ground-contact surfaces to the ‘ground’ outside boundary condition and add the Site:GroundTemperature:BuildingSurface object. 2. Slab preprocessor 3. Site:GroundDomain:Slab object in EnergyPlus 4. Foundation:Kiva object in EnergyPlus.

Which method is the most suitable for me? And actually is there any difference between the slab preprocessor and the Site:GroundDomain:Slab object integrated with EnergyPlus?

Thank you in advance for any help.


I performed a simulation of a tent using Kiva for the extremely cold climate in Canada (weather file CAN_NU_Resolute.719240_CWEC.epw). The tent was heated to the heating setpoint temperature of 15°C. I checked the heat transfer rate through the tent’s floor (1-inch layer of soil). And the results were opposite to my expectations. The results showed heat gains through the floor. I thought that maybe I made a mistake in my model so I did a test with the example model ZoneCoupledKivaSlab.idf. I run the simulation for the same weather file and I obtained heat losses through the floor (according to expectations). After that, I slightly modified the example model by changing only the wall materials’ thickness from the initial values to 0.005 m (Composite 2x6 Wood Stud R19 #1, Composite 2x6 Wood Stud R19 #2, Composite 2x6 Wood Stud R19 #3). I didn’t change the floor/slab construction. I simulated the model again and I obtained completely different results- heat gains from the ground (similarly to my tent model). I attached a chart showing the results. Could you please explain to me this outcome? image descriptionimage description

edit retag flag offensive close merge delete


What wall construction are you changing? The foundation wall, or the above grade walls?

Neal Kruis's avatar Neal Kruis  ( 2019-04-26 10:21:57 -0500 )edit

The only thing that I changed are the materials' thicknesses, please have a look at the picture above (yellow highlights). I changed them for 0.005m.

a.rijs's avatar a.rijs  ( 2019-05-02 02:40:06 -0500 )edit

Here's what I believe is happening: When you make the walls very thin, their interior surface temperature drops very close to the outdoor temperature. The heat flow from the slab is now driven by radiation to the surrounding walls instead of through the (mostly) warmer ground. The way the convection, radiation, and conduction balance at the surface is such that the conduction is now flowing into the space.

Neal Kruis's avatar Neal Kruis  ( 2019-05-09 16:06:01 -0500 )edit

2 Answers

Sort by » oldest newest most voted

answered 2019-03-11 18:14:10 -0500

Use the Site:GroundTemperature object. The slab preprocessor and kiva tools are intended to model a permanent structure that will greatly influence the seasonal heat flux in and out of the ground. A tent is impermanent and the ground temperature will reflect outdoor ground surface conditions. Ground temperatures are typically reported at 0.5, 2, and 4 meters. These are all too deep as you want the first few inches of ground which could exchange heat in the time the tent is there, but absent that data use the 0.5 meter temperatures.

That said, I don't think you can do this analysis without knowing ground type and conditions. There is a huge difference in heat transfer between a rock slab that was in the sun all day, a bed of pine needles, and a muddy field after a rainstorm. Moisture conditions alone will greatly influence the results. I'd suggest digging into ways to model ground surface conditions including moisture, perhaps calculating ground heat flux directly.

edit flag offensive delete link more


Fascinating. I support this answer too. Might I add, or snow? As one whose had some miserable winter camping experiences... the moisture has a huge impact.

pow_skier's avatar pow_skier  ( 2019-03-12 09:34:51 -0500 )edit

Thank you very much for your answer. And how about the Site:GroundDomain:Slab method?

And how is it different from the preprocessor? I know that the preprocessor is 3D and the GroundDomain is 1D. But what are other differences? Because apparently, both methods give different results. See:

a.rijs's avatar a.rijs  ( 2019-03-12 09:55:16 -0500 )edit

Both are 3D. Ground Domain performs [roughly] the same calculations as the preprocessor, but it is integrated into the EnergyPlus timestep and coupled to the EnergyPlus heat balance by applying a boundary condition to EnergyPlus's normal 1D conduction. This method allows you to model internal sources in your constructions, but it is not as fast or flexible as Kiva in modeling other foundation configurations.

My recommendation: Use Kiva unless you have an internal source (e.g., radiant heated slab). Never use the preprocessor or Site:GroundTemperature.

Neal Kruis's avatar Neal Kruis  ( 2019-03-12 10:09:07 -0500 )edit

Thank you for your answers. What do you mean with internal sources in my construction? In my case, the tent is air-conditioned by warm or cold supply air.

a.rijs's avatar a.rijs  ( 2019-03-12 10:32:00 -0500 )edit

Construction:InternalSource is used to model an active heating/cooling system embedded in your surfaces (e.g., chilled panels, in-slab radiant heating).

Neal Kruis's avatar Neal Kruis  ( 2019-03-12 10:34:53 -0500 )edit

answered 2019-03-12 10:00:55 -0500

There is nothing special about a tent. It is an on-grade construction.

I can't think of an application where you'd get better results using Site:GroundTemperature than with Kiva. Kiva will perform a real, two-dimensional heat balance between the floor of the tent and the exterior ground. In contrast, the "correct" Site:GroundTemperature to use for this application will be a complete wild guess. With Kiva you can at least get a somewhat reasonable estimate of ground losses with some informed assumptions about the ground properties and using an actual heat balance.

Model the floor surface as a 1-inch (or so) layer of soil (the plastic foil is likely negligible). Model the "foundation wall" as a similarly thin construction of soil with limited depth.

@mdahlhausen is right about the uncertainty in ground conditions, and they are especially important for uninsulated ground surfaces. They are even more important if the tent is heated.

Do the best you can with the information you have and you'll get a much more reasonable solution than approximating the losses as one-dimensional losses to a mysterious ground temperature.

Full disclosure: I wrote Kiva.

edit flag offensive delete link more


I tried this approach and I have an additional question. It is too long to write it as a comment so please have a look at the edit of my question. Thank you in advance.

a.rijs's avatar a.rijs  ( 2019-04-26 07:24:49 -0500 )edit

Your Answer

Please start posting anonymously - your entry will be published after you log in or create a new account.

Add Answer


Question Tools



Asked: 2019-03-08 06:51:12 -0500

Seen: 1,441 times

Last updated: May 02 '19