GBA Logo horizontal Facebook LinkedIn Email Pinterest Twitter Instagram YouTube Icon Navigation Search Icon Main Search Icon Video Play Icon Plus Icon Minus Icon Picture icon Hamburger Icon Close Icon Sorted

Community and Q&A

BEopt: Too good to be true?

thrifttrust | Posted in General Questions on

Has anyone used BEopt software? Are the results reliable?

I entered my design into BEopt. I had to create many of my own options. For instance, there was no description for my R22 exterior foam. I think I understand the process and how to use it.

My Assumptions:
Detroit, Michigan (CZ5)
1040 sq. ft.
71° heating set point
76° cooling set point
11 ft. walls. (There is no option for my shed roof design so I averaged.)
R40 walls (Nominal R22 external EPS and R21 batts downgraded for framing)
R80 vented roof (24″ cellulose in ≥ 30″ parallel chord trusses.)
R20 8′ basement walls and floor (I think. The category wasn’t clear.)
305 sq. ft. U1.9 windows
2 Fiberglass doors
1 air change per hour at 50 pascals
Hyper-heat mini-split
70% HRV
LED lighting
Electric resistance water heating
No dishwasher
Standard clothes washer
Gas clothes dryer (0 CFM. Existing dryer will be located in detached garage/shop.)
10 KW PV (Meant to cover future electric car.)

I dialed back hot water usage. In retirement, we don’t take frequent showers or wash a lot of clothes.

The results show 4801 KWh/y with 6756 KWh/y PV surplus. This is beyond my expectations. Can I trust the results?

I’ve attached the BEopt output graph.

Douglas Higden

GBA Prime

Join the leading community of building science experts

Become a GBA Prime member and get instant access to the latest developments in green building, research, and reports from the field.


  1. thrifttrust | | #1

    Oh, single story and high solar gain glass.

  2. GBA Editor
    Martin Holladay | | #2

    I'm not going to double-check your software run.

    Here's what I know:

    1. You've designed a small house with a very good envelope.

    2. A 10-kW PV array is definitely on the big side.

    3. It's impossible for BEopt to predict electrical usage, which depends on plug loads and occupant behavior.

    4. The occupants of the Montague house in Massachusetts used only 1,959 kWh per year in their 1,152 sf house. The house had a 5-kW PV system, and the system produced 4,892 kWh the first year -- two and half times as much electricity as the occupants used. For more info, see Occupant Behavior Makes a Difference.

    5. To make a small house net-zero, you need anything from 3 kW to 10 kW of PV. How much PV you need depends on your behavior.

  3. NormanWB | | #3

    If you do oversize the PV array and plan to connect to the grid via net metering, make sure the return on investment makes sense. In my case, Duke Energy buys any surplus at roughly half of the retail rate, which can put dent in your PV return calcs.

  4. thrifttrust | | #4

    Thanks for your advice.

    I'm confidant I got my inputs right, I'm just a bit taken aback by the good news. Especially the low cooling load. We've always had low electric bills but we traditionally heat, cook, bathe, dry our clothes using gas and cool with window AC. A combustion free home is a novel concept for me and it's a bit scary, but I'll trust the BEopt gods :) Building the home will consume most of our assets but we should come out of it with no mortgage or energy bills.

    Surrounded by the Great Lakes those arctic air masses coming from Wisconsin, Minnesota and Canada are warmed as they pass over water before reaching lower Michigan. During the recent cold wave our temperatures rarely dipped to 0° F. In the process, evaporation from the lakes kicks up a lot of cloud cover, which helps keep temps up, but it also means Michigan has abysmal winter solar potential. A recent GBA blog gave Michigan in a marginally positive grade for passive solar, but it was based on 15¢/KWh electricity. I pay 10¢, so for me, south windows lose more than they gain. Besides powering a car, I think the PV system should be oversize and mono-crystalline to best capture energy from cloudy-bright skies. That said, I'll reduce the system to the 7KW range. This will also allow me to move it from the house roof to that of the garage/shop.

    Do to lack of sun and cheap energy, Michigan has been slow to adopt solar. Utilities offer net metering. ( Maybe they feel guilty generating half their power from coal. I know I do.) Anyway, I don't trust them to continue the program forever, so the oversize array is a hedge. The crew and I will install the hardware and the wiring should be a modest up-charge to the overall electric installation. I'll still get the 30% credit and panel prices don't seem to have gone up yet. I see this as a once-in-a-lifetime opportunity.

  5. GBA Editor
    Martin Holladay | | #5

    It sounds like you're doing your homework. Good luck.

  6. GBA Editor
    Martin Holladay | | #6

    Many homeowners don't know what inputs to use for plug loads -- and even if they have an idea, actual usage depends on behavior. That's why I wrote that BEopt can't predict electricity use for plug loads. I wasn't disparaging the accuracy of the program.

    Thanks for sharing your experience, and for providing context to Douglas's numbers.

  7. Expert Member
    Michael Maines | | #7

    Douglas, I use BeOpt for most of my projects. It's not perfect but a lot better than going on nothing, and a lot easier (and cheaper) than some of the other options.

    To get hot water numbers that low with an electric resistance water heater you must have really dialed back the input for usage--did you use their input for 50% of average usage, or even less?

    I assume you meant to write that you have U-0.19 windows?

    Your heating load works out to 3.6 kBtu/sf/yr, which is 75% of the Passivhaus standard. That's very impressive, and possible, though your lack of confidence on the basement inputs and my own experience make me wonder if there is a mistake, especially since you say your insolation is not very good. (Though that can be a good thing for cooling load.) You must have fairly moderate heating degree days.

    Martin, if you have not used BeOpt in a while you should give it a try. You can adjust inputs for plug loads, hot water usage and most of the other major energy users. The building envelope inputs are a bit limited but overall I tell clients that the model is probably within 10-20% of accurate, maybe more or maybe less, largely depending on their behavior, and that its best feature (for me) is the ability to quickly and easily compare the impact of adjusting various assemblies. Of course the program's main claim to fame is the automated "path to net zero," which requires a level of trust in their cost assumptions that I don't have. (Or custom inputs for costs, which I also don't have enough of to be useful.)

  8. thrifttrust | | #8

    Thanks Martin, and Michael for the vote of confidence In BEopt.

    Yes indeed, U.19 windows. Single glazed plain glass windows would do better than U1.9. This is based on the only quote I've sought thus far. It is for Jeld-Wen premium vinyl triple glazed. I got the quote as a baseline, but I'm surprised by the low U numbers. I didn't know they even offered triples. I'm concerned that I can't find their air infiltration rates, but as someone at GBA pointed out, To get Energy Star rating they must meet a "pretty good" standard. How does 30 USD/sq. ft. compare to other brands?

    I consider the R20 external basement wall insulation I entered to be good. My problem is that BEopt seems to have no provision for specifying my R20 sub-slab insulated floor. There is also no provision for specifying basement fenestration. I just added the basement windows to the main floor wall totals. I realize that sub-surface load calculations are difficult. They change with depth and season. I suppose above grade portion of the basement walls should be treated like those of the main floor.

    The miscellaneous and water heating loads are based on our current usage. 2 waste conscious occupants, no dishwasher, mostly cold laundry, front loader, ≈ 2 loads/week.

    Can the BEopt results be used to size the mini-splits? Can anyonw tell me how big they should be? The plan is for one wall unit near the center of the main floor and a slightly smaller one in the basement.

  9. Expert Member
    Michael Maines | | #9

    Douglas, even the PHPP struggles a bit with accurately modeling basements--it's complicated for all of the reasons you mention, and more. You're correct, BeOpt does not have an option for sub-slab basement insulation, but for as accurate as their basement inputs are, it averages out. Better to err on the side of less insulation. If it's a walk-out basement, I model it as a basement and as above-grade and estimate a weighted average of the two. If you have a lot of basement windows you might try this approach, or you could reduce your insulation input to compensate. BeOpt, like most high performance home builders, seems to encourage slab on grade construction.

    On the output screen, at lower left, choose "Graph type--HVAC capacities." That tells you the heating and/or cooling load, depending on the situation. Then choose "loads not met" and/or run an hourly simulation (from the Tools menu) to see how much of the time the system is undersized, and what a common low load is for your situation, to gauge what the minimum capacity should be. Then talk to a good HVAC engineer and compare notes--there are a lot of fine points that I'm still trying to learn, after studying this for years and running dozens of BeOpt models. Unfortunately most HVAC pros are not used to building envelopes as robust as yours so you may have to do some searching, to keep them from insisting on a head in every room.

    Most window manufacturers don't list air infiltration rates, and there is no direct input for it in BeOpt. When I want to compare double hung windows vs. casements or tilt/turns, I just adjust the overall infiltration rate--say 1.0 for casements vs. 1.2 for double hungs. It's a guess, but a somewhat educated one. The difference in a new window isn't enough to greatly affect energy use, though it adds up over time. $30/sf is a good price.

Log in or create an account to post an answer.


Recent Questions and Replies

  • |
  • |
  • |
  • |