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

BEopt boiler results possibly inaccurate?

asked 2019-07-13 12:47:43 -0500

duerstad's avatar

updated 2019-07-13 14:07:13 -0500

Results for a model I have used for a long time make sense when using a forced air furnace regardless of almost infinite variations with the model. Delivered MMBtu/site MMBtu also show 98% for a 98% AFUE furnace. When using a boiler, however, the model shows the use of far more energy. I understand the boiler is using baseboards with relatively high temperatures, but still . . . . BEopt's choice 10 98% efficient natural gas boiler (Energystar 2019 shows very few boilers over 95%; none of 98%) has a site/delivered ratio of .907 while choice 8 (95% efficient with outdoor air reset) has the same ratio at .869. Output capacity is the same and there are never any unmet hours. The two boilers are delivering 109.6 & 109.7 MMBtu while the furnace heats the house with 58.8 MMBtu delivered. Months ago I asked at how a boiler could have results showing such a disparity in favor of forced air and was assured the software was prejudiced against hydronic. What is going on here or what don't I understand?

edit retag flag offensive close merge delete


Can you provide your BEopt project file via, e.g., Dropbox or Google Drive? I did some quick runs of 98% gas furnace vs 98% gas boiler vs 95% gas boiler with outdoor air reset and all three show similar results.

shorowit's avatar shorowit  ( 2019-07-13 14:29:24 -0500 )edit

I don't do this very often, so if I didn't get it right, please let me know.

duerstad's avatar duerstad  ( 2019-07-13 14:58:09 -0500 )edit

1 Answer

Sort by ยป oldest newest most voted

answered 2019-07-15 10:10:03 -0500

updated 2019-07-15 10:10:49 -0500

There is a bug in the EnergyPlus (v8.8) foundation heat transfer model, which for some reason is only triggered in your boiler runs. At some point in the near future, we hope to release a new version of BEopt that uses the latest EnergyPlus release (where the bug is fixed).

In the meantime, you can workaround the bug by opening BEopt's file "C:\Program Files (x86)\NREL\BEopt_2.8.0\Modeling\" and modifying the line self.use_kiva = True to instead say self.use_kiva = False. Then rerun your file. This will revert to our older foundation heat transfer algorithm that does not hit this bug.

edit flag offensive delete link more


Thanks. It now makes sense. The gas furnace now uses fewer MMBtu to heat the house, but uses more energy for fans/pumps than the hydronic. I'll renew my support for BEopt/OpenStudio at since they would also benefit from checking their own experience against "the numbers." In another question here I was already bumping into the limits of foundation heat transfer and it seems the way to get better information about foundation and insulation at present is LoopCad. Perhaps it was my investigating BEopt's options for slab insulation which set off the problem here. Thanks again!

duerstad's avatar duerstad  ( 2019-07-15 12:03:05 -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: 2019-07-13 12:47:43 -0500

Seen: 317 times

Last updated: Jul 15 '19