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

Total energy use from manual J

blamus20 | Posted in Mechanicals on

I have a manual J report and its a bit over my head. I can see where it shows peak equipment loads to size a/c and furnace, but where can i find the total energy use to heat/cool the house per year? I want this info in kwh so I can size my solar roof to shoot for net zero.


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. Expert Member
    NICK KEENAN | | #1

    The NREL manual has the information you need:

    Find the page for your location, and it will tell you climate and solar availability information for your location.

    The Manual J calculates peak loads for a given interior and exterior temperature. The assumption is that load is directly proportional to the difference between interior and exterior temperature. If you take the BTU/hr at peak and multiply by 24 you get BTU/day. If you divide that by the degree difference between inside and outside on the design day, you get BTU/degree-day. You have to calculate it separately for heating and cooling.

    In the NREL manual it has heating and cooling degree-days per year for your climate. Multiply the BTU/degree-day by the degree-days to BTU/year.

    OK, now you have to get kwh/year from BTU/year. If you're going solar electric, most likely you're using a heat pump. The manufacturer will publish efficiency numbers with the BTU/kWh ratio. Heat pumps are less efficient the greater the temperature difference, so if you can get a performance curve the most accurate way is to calculate kWh for each month, using the average temperature for that month. If that's not available just use the average efficiency.

    The NREL book also has solar radiation, in BTU/sf. Your panel manufacturer will publish an efficiency number, multiply the available radiation for the year by the efficiency (assuming your panel isn't shaded or tilted). Convert BTU to kWh (1 BTU=0.000293071 kWh) to get kWh/year per square foot per year. From that calculate how many square feet you need.

    This is assuming you have net metering. If you don't have net metering it gets more complicated.

  2. blamus20 | | #2

    oh man, this is more involved than I expected. I gave it a try following your instructions but then got stuck pretty quickly.

    So my "Equipment Load" under Heating Summary in my manual J is 55727 Btuh. So I multiply by 24, gets me 1337448 Btu/day.

    Im in Boulder CO so I use Denver's numbers. I assuming I would use the Avg monthly temps to calculate the temp diff for each month.

    Then I divide 1337448 by the temp diff for every month to get 12 different Btu/degree-day. Then I multiply the 12 Btu/degree-day by the listed Heating Degree Days. Which then gets me a Btu/year for each month of the year? Surely thats not right.....

    1. Expert Member
      NICK KEENAN | | #4

      In your Manual J it should have an interior and exterior design temperature for heating and cooling. What are those?

      1. blamus20 | | #5

        Under Winter design conditions: design TD 63F

        So 55727Btuh x 24h = 1337448 Btu/day
        1337448/63F (or C?) = 21229 = Btu/Year

        I've attached my Manual J and the heatpump specs.
        Which is the efficiency number I should use?

        I actually just need a very rough number in kwh to "prove" to xcel energy to allow me to install a larger PV system than 10kh. They have a "rule of thumb table" where for a 3.5k-4k sqft house the max is 13kw, which they obtain from national averages for houses that size and then +20%. But my understanding is that is way too small for an all electric house because the national average house uses gas for space heating and water heater. I have a 3ksqft, 3-12 shed roof facing perfectly south in sunny colorado at 6000ft, and I'm getting the tesla solar roof so adding more capacity is not an option in the future. Right now they designed a 10kw system which mean only 30% of the roof tiles are solar active. I am trying to get a rough idea if I should shoot for a 20kw or even 30kw (getting really pricey though) system, but before they are willing to design those systems I have to get approval from xcel energy.

        And that's before the possibility of an electric car in the future. (xcel needs purchase receipts for every piece of equipment that adds to their kW numbers)

        Maybe someone here has done a large all electric house in CO and can give me some ideas on how many kw they consume per year?

        Thank you for all your help.

        1. Expert Member
          NICK KEENAN | | #6

          1337448 Btu/day @ 63F delta means 21,229 BTU/degree-day.

          Multiply that by your annual degree-days of heating to get annual heating BTU.

          1. Expert Member
            NICK KEENAN | | #7

            There are three efficiency numbers that can be provided.
            Coefficient of Performance (COP) is typically given for heating, it's a ratio of of kilowatt-hours of heat out to kilowatt-hours of electricity in. To use this you need to convert BTU to kilowatt-hours.

            Energy Efficiency Rating (EER) is typically given for cooling. It's the ratio of BTU of cooling to kilowatt-hours of electricity. It is given at a specific temperature.

            Seasonal Energy Efficiency Rating (SEER) is average EER over a typical cooling season, a weighted average of the temperature over the whole season.

          2. blamus20 | | #8

            Ann heating degree-days is 6016 (65F) in the NREL doc.

            21229BTU/degree-day * 6016 = 127715669 BTU/year

            can I use HSPF instead of COP? Thats what I can find for now.

            127715/11.4 (HSPF) = 11203kWh

            does this sound about right?

        2. Expert Member
          NICK KEENAN | | #13

          Yeah, HSPF is the heating equivalent of SEER, it's an attempt to average performance over the entire heating season. The units are BTU/kWh so the way you divided it is correct.

  3. CramerSilkworth | | #3

    Manual J is a fairly blunt instrument, even just for load calcs, as is the degree-day process already suggested. What you really need is an energy model, but even that is probably at best +/-10% accurate.

  4. charlie_sullivan | | #9

    Note that everyone is talking about your HVAC load, but you'll have other electric use in the house, domestic hot water, cooking, laundry, lighting, electronics, and other miscellaneous plug loads. Maybe an electric vehicle to charge soon if not already.

    1. blamus20 | | #10

      yes exactly. xcel gives a spreadsheet that allow you to input additional equipment and have pre determined electric usage for all of them. (see attached) that includes everything you stated. E.g. they say water heater 3000kwh. But there is no mention of electric space heating. So I plan to input that in "other" at the end. I wonder why thats missing.

      1. charlie_sullivan | | #11

        That spreadsheet sounds like more of a crude estimate than the calculation you are doing for space heating. There's a lot there where you could do better than their estimates. Better would be to have some measurements from how you actually live.

        1. blamus20 | | #12

          exactly, this is just a very crude estimate. But this is a new build so theres no measurements. I just need to know how much PV i need to account for the space heating. I'm not expecting to be accurate. If I go too big, then I am future proofed for things like EVs. If I go too small, then I waste the opportunity to have a bigger system in the roof, but of course then it'd cost me less to install. Im not sure if i can afford a 30kw system ($80k roof?) But regardless, to go bigger than 10kw I have to prove that I have such and such equipment that will consume x-amount of electricity. Would 10kw be reasonable to claim to account for the heat pump space heating?

          1. Expert Member
            NICK KEENAN | | #14

            Solar panels are rated in kiloWatts, energy usage is in kiloWatt hours. The final piece of the puzzle is how many hours of sunshine a year you get, which tells you how many kW you need for that many kWh. Your utility may have a standard number they use which would be best.

          2. Expert Member
            NICK KEENAN | | #15

            My experience is that since solar panels are subsidized, you want to get the largest array you can get and still get the subsidies.

            Throw your cooling load in there too, even if you never intend to run AC.

  5. blamus20 | | #16

    the tesla proposal for a 10.17kw system says will produce 13080kwh on my roof. Maybe its possible to extrapolate what a 20kw or 30kw system will produce?

    I'm at the stage of accepting the quote so yes, I plan on adding as much as I can and ask for quotes on larger systems to see how much they are, if I can be net zero, if I can afford it, and what the subsidies look like as I go larger. Right now for this 10.17kw system they quote me 37k, with 6.6k tax credit, so net cost about $30.5k A quick click around using their estimator tool shows going from non solar tiles to solar active is a price increase of about $2/w, which is cheaper than conventional panels, which makes sense because I'm paying for the glass tile anyway even if its not solar active. My limited understanding is that they only subsidies the solar active tiles, so the more of them i have, the more subsidy, and more importantly, the more of my roof is subsidized, since the solar system is the roof. So if thats true, then you are right, I really should go as big as I can afford........

  6. Expert Member
    NICK KEENAN | | #17

    If a 10.17 kW system produces 13,080 kWh per year, that means it gets the equivalent of 1,286 hours of full sunshine per year. That sounds about right. That number will scale proportionately with any size array.

  7. blamus20 | | #18

    I have a related question about solar costs and ROIs

    Is it generally true with today's PV prices ($2-3/W) that spending money on getting more PV is better than spending that money on upgrading windows from U0.3 to U0.2 or 0.1 (triple glazed), purely from a ROI perspective. Not considering comfort or even air sealing differences etc.

    I have a lot of glazing, but using Andersen 100 which is a high value choice, and I got a quote from Alpen with their ZR-6 (U0.16) that is double the andersen cost. Now as I'm considering putting an extra $20-40k into my solar roof to get 10-20kw I wonder if its worth upgrading the insulation envelope instead. My gut feeling is that its not.

    I do understand that these windows are not an apples to apples comparison because andersen 100 is a fibrex composite while the alpen ZR6 is fiberglass. But no one makes a composite triple glazed for comparison. I'm not allowed to have vinyl frames in my area for wild fire reasons. I think they melt too fast or something. So I don't think i'm allowed uPVC either. And uPVC also have much thicker frames from what I've seen.

    1. Expert Member
      NICK KEENAN | | #19

      There aren't a lot of generalities in this biz.

      If you have a Manual J where they show their work, you can put in the values for different windows and see how that changes the peak heating and cooling loads. Then repeat the math that we just did and see how your annual energy usage changes. Then take the same dollar amount and see how much solar it buys, multiply that by the annual hours of sunshine to get the annual production. Then you will be able to answer the question, do you get more bank for your buck increasing production or reducing consumption?

      The numbers you want to change for your windows are the U-factor, which affects both heating and cooling, and the SHGC (Solar heat gain coefficient) which affects your cooling load.

      1. blamus20 | | #21

        I just dug out my old Beopt model and ran it with the windows upgrade (u0.3 to u0.18). Given that this is an old/outdated model but its still the same building shape, all glazing and floor sqft within around 20% (a guess) of what actually got framed. the difference is only 600kwh/yr. Is that even possible!? That's nothing!

        Another interesting find - It thinks my heating will only take 5000kwh/yr which is about half of what my manual J calcs above showed. This I do believe can be the case because its a passive solar design where 50% of the glazing faces directly south, so I anticipate a decent amount of solar gains which the manual J does not account for per heating season. I have almost 700sqft of glazing facing south and 350sqft pretty equally distributed on the other aspects.

        Regardless, installing a large PV roof seems like a good idea if just for future proofing for hot tubs and EVs just seems like the best investment if I already have R30 walls and R60 roof.

  8. blamus20 | | #20

    Yeah....I only have the paper report for the manual J. I guess if I commit to spending a day on Beopt I would get my answer. I did play with that years ago and I seem to remember upgrading windows to triple pane did surprisingly little, compared to things like air sealing etc. For all the cases I have found online where people did this analysis (not many) the conclusion seem to say that the payback period for triple pane upgrade is about 30+ years while PV is around 10. This nudges me to invest more in PV instead. And I can only see that becoming more and more the case as windows are only getting more expensive while the cost of PV is dropping over time.....

  9. sribe | | #22

    Hi Bernard,

    I am a bit south of you (10-20 miles) and a bit higher (8,900'), but I can tell you from experience that I don't trust Manual J numbers for mountain houses. I am in an old house (1974) with electric boiler feeding hydronic baseboards, and have been working on some upgrades, so I know my actual heating loads very well. The Manual J I paid for was rather high for some rooms, and 50% on the low side for the ones with window walls and overhangs. Granted, your house is a much simpler footprint, so maybe it's easier to calculate.

    I got 15kW of solar a couple of years ago, and the production you've been quoted for your 10kW seems reasonable--it's proportionally a bit better than mine, but I have some partial shading. The high-altitude UV intensity really helps us.

    Anyway, yes, the payback on PV these days is better than the payback on a lot of envelope upgrades once you get to a decent point. I'm looking at window upgrades mostly because with my current windows I can't actually get enough heat through a radiant floor to meet design requirements here (-11F).

Log in or create an account to post an answer.


Recent Questions and Replies

  • |
  • |
  • |
  • |