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

Why kiva results do not make sense?

asked 2021-03-16 12:40:02 -0500

Roya Rajabi's avatar

updated 2021-03-25 06:45:49 -0500


I used different models for slab (F Factor, Kiva, Ground Domain) But I found the result for Kiva totally different than other models. In the kiva Model, I have more heat loss in Summer than Winter!!! I checked the example file for this method (Kiva), I faced the same issue and the same pattern for the result?

Please find the google drive link that includes the graph and idf file.

Would you please help me to understand the reason?

Regards, Roya

edit retag flag offensive close merge delete


Hello @Roya Rajabi! Can you provide example files and/or plots to illustrate what you are seeing?

Neal Kruis's avatar Neal Kruis  ( 2021-03-17 16:28:56 -0500 )edit

Thanks for your reply.

Please find the google drive link that includes the graph and idf file.

Roya Rajabi's avatar Roya Rajabi  ( 2021-03-18 02:12:33 -0500 )edit

Can you also include the weather file you are using?

Neal Kruis's avatar Neal Kruis  ( 2021-03-18 12:10:16 -0500 )edit

I used Denver weather data I added it to the google drive


Roya Rajabi's avatar Roya Rajabi  ( 2021-03-18 12:15:03 -0500 )edit

2 Answers

Sort by ยป oldest newest most voted

answered 2021-03-18 14:58:03 -0500

updated 2021-03-23 15:51:35 -0500


I worked with a colleague to dig into this a bit, and the explanation is actually a bit different from what I originally theorized:

What is happening here is related to the heat balance at the surface, where convection gain + radiation gain = conduction loss.

The slab surface temperature is actually very close to the air temperature for most of the year and convection is relatively low in magnitude. The largest driver of heat flow through the surface is radiation exchange with the other surfaces in the zone. Because the other surfaces are significantly warmer than the slab in the summer there is considerably more radiative heat transfer to the slab surface. This radiative heat needs to go somewhere, and with such a low convective temperature difference, the path of least resistance is to conduct into the slab.

The conclusion here is that conduction is not a good indicator of the thermal load on a space. If you look at the convective heat gain rate instead, you will see the behavior you were probably expecting to see with conduction.

Initial (incorrect) theory:

Here's one theory about what you are seeing (until I can confirm with a weather file):

The deep ground boundary condition is going to be fairly shallow in your case and will likely be the strongest driver of heat transfer from the slab. You input file specifies "Autoselect" for this boundary condition, which will estimate the depth of the water table based on your location and give it a constant temperature equal to the annual average dry-bulb temperature.

This deep ground boundary temperature is likely:

  • slightly lower than the indoor temperature
  • significantly lower than the outdoor temperature in the summer, and
  • significantly warmer than the outdoor temperature in the winter.

In the winter, the driver of heat loss through the slab is much lower than through the above-grade envelope, but in the summer it could be a much stronger driver.

(I'll update with more details when/if I can get a weather file to test this hypothesis.)

edit flag offensive delete link more


Thanks a lot for your time and your help.

Roya Rajabi's avatar Roya Rajabi  ( 2021-03-24 15:55:46 -0500 )edit

thanks, I found the same for other models.

Roya Rajabi's avatar Roya Rajabi  ( 2021-03-24 23:16:51 -0500 )edit

I haven't seen the other models you're comparing to, but the model you posted has rigid insulation under the entire slab AND exterior vertical insulation going down 1.5 meters. Maybe you're comparing an insulated slab in Kiva with uninsulated slabs in the other models?

Neal Kruis's avatar Neal Kruis  ( 2021-03-25 09:24:06 -0500 )edit

answered 2021-03-18 04:34:50 -0500

Andy Tindale's avatar

Hi Roya, I haven't seen your model but wonder whether the extra summer heat loss could be caused by internal temperatures being higher in the summer. In some cases the summer internal temperatures increase by more than the ground temperatures causing higher ground heat loss.

To test this idea you could try configuring your model to have constant temperatures 24/7 throughout the year (by adding close heating and cooling control) and check how that affects the Kiva ground heat loss.


edit flag offensive delete link more


Hi Thanks for your reply but in the example file the inside temperature is constantly on 20 C for whole the year.

Roya Rajabi's avatar Roya Rajabi  ( 2021-03-18 11:27:04 -0500 )edit

Your Answer

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

Add Answer

Training Workshops


Question Tools



Asked: 2021-03-16 12:40:02 -0500

Seen: 412 times

Last updated: Mar 24 '21