Data from BEopt
I just ran my first Optimization run using BEopt. I had many high performance options selected for all the important items (air-tightness, insulation, etc). It took about 5 hours to run all the simulations and in the end, it said the B10 Benchmark was much cheaper vs all my high performance combinations. Am I doing something wrong here? I was looking for the most cost effective solution to find the sweet spot of how well I should insulate, how low I should shoot for in air-tightness, etc. But this simulation says I’d be paying way more per month for incorporating any of those options vs just building to the B10 Benchmark, which frankly, has pretty lame options selected. Any input or advice would be welcomed. I’m a complete newb to “building science”, but really wanted my next build (coming up in a couple months for my own home) to be next-level.
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
Replies
I've been running a lot of BeOpt sims for my upcoming build. That said, I haven't used the optimization mode. I do a lot of modeling and testing work in my day job as a mechanical engineer, and I've learned to be very cautious with complicated models with lots of inputs. I've mainly used BeOpt for making comparisons between specific windows, or mini splits.
It could very well be that under the assumptions used by BeOpt, the B10 standard makes the most financial sense. Your desire for a "next level" build may be a competing requirement with lowest total cost over your repayment period.
B10 benchmark is described on page 14 of this pdf:
https://www.nrel.gov/docs/fy14osti/60988.pdf
You could start from the "Pretty Good House" recommendations for insulation and air sealing, and use Be-Opt to make trade off decisions on how to reach that level.
There are a few articles on the Pretty Good House on GBA, including Pretty Good House 2.0. For a deeper dive that might provide the recommendations Erik is referring to, consider getting the PGH book by Chris Briley, Dan Kolbert, Michael Maines, and Emily Mottram. It's due to come out in late June.
Your return on investment will change a lot depending on your wild guess that you input as the inflation number. I was using 3 or 4 % five years ago seems silly low today.
Your fuel choices and costs will have a big impact on the numbers.
My guess is the cost estimates built into the B10 bench mark are way low given what has happened with pricing in the last two years and not valid today.
Are you entering your own cost per square foot into BEopt from your bids?
Walta
"Are you entering your own cost per square foot into BEopt from your bids? "
No, I'm not. I wanted to compare apples to apples so I'm just using BEopt's internal costs, which I've seen are definitely pretty ridiculously low.
Like any computer program the output is no better than your input. The old saying is garbage in equals garbage out.
Seems to me you are wasting your time unless you input todays real numbers.
I do not see what you can learn by comparing anything to something you can no longer build for the cost in the model.
The truth is the recent code minimums R25 walls + R38 ceiling make for a very good house and be very close to optima in many locations.
Walta
I use BEopt regularly but have not taken the time to figure out the Optimization mode, nor do I trust that their internal cost database is accurate for my market. My approach is to run multiple "cases", changing one variable on each case, then look at how those changes affect total energy use and HVAC design loads, and I use my own rough estimates of what each change would cost. If their built-in costs are not accurate it will greatly skew the results.