More Musings of an Energy Nerd
Oversized furnaces, heat pumps, and air conditioners are the norm in most American homes. That’s a problem, since oversized equipment often provides lower levels of comfort and costs more than right-sized equipment.
Designing and installing a forced-air heating and cooling system is a multi-step process that should include heat-loss calculations, heat-gain calculations, equipment sizing, and duct system design. In a typical residential installation, however, some of these steps are performed sloppily or entirely omitted.
The purpose of a heat-loss calculation is to determine the design heating load, which is used to select the heating equipment needed during worst-case conditions: that is, during a very cold day in winter with no solar gain through the windows. Heat-gain calculations are used to determine the design cooling load, which is used to size the home’s air conditioner. These heat-gain calculations also assume worst-case conditions: a very hot day in summer with maximum solar gain through the windows.
What is Manual J?
According to most U.S. building codes, HVAC contractors are required to perform the system design steps listed above using calculation worksheets developed by the Air Conditioning Contractors of America (ACCA). Three of the most important of these worksheets are called Manual J, Manual S, and Manual D. For many years, these worksheets have been the standard methods for calculating residential heat loss, heat gain, equipment size, and duct sizes.
In the old days, heat-loss calculations were done with paper-and-pencil worksheets. For more information on old-school heat-loss calculations, see How to Perform a Heat-Loss Calculation — Part 1 and How to Perform a Heat-Loss Calculation — Part 2 .
Manual J, Manual S, and Manual D calculations are no longer performed with a pencil and paper; contractors now use software programs to complete these worksheets. Hank Rutkowski, the mechanical engineer who wrote (and continues to…
Weekly Newsletter
Get building science and energy efficiency advice, plus special offers, in your inbox.
This article is only available to GBA Prime Members
Sign up for a free trial and get instant access to this article as well as GBA’s complete library of premium articles and construction details.
Start Free TrialAlready a member? Log in
6 Comments
I purchase WrightSoft for $500 which is reasonable. Doing an initial pass on a house design before turning it over to a third party designer my opinion is that the home designer is responsible for the heat loss calculation, so the budget has enough money for comfort. The third party is to verify the heat loss and size equipment. This is basic theory taught in most engineering classes - with the math simplified.
Do we know how accurate Manual J is in practice? I seem to recall reading here that even an earnestly done calc is typically off by plus or minus 20%. Further do we know to what extent calibrating heat calcs to utility billing history is an improvement? I have heard it said that manual J ignores occupant behavior. For example by assuming that all zones are heated and cooled equally.
It would be interesting to see if there are some studies on the topic. I'm a bit of a Manual J novice, but I'm extremely experienced in PV system generation modeling, and I'm sure there are some of the same dynamics. When we talk about accuracy, the areas I'm thinking of are:
- How accurate are the design temperatures for the location being modeled? This can be quite localized, and of course it is changing over time. 1km of distance can be material for solar generation differences, but our design temp data is more regional.
- How closely does the behavior of the actual building construction match the behavior of the construction types and materials in the model? IE, without considering whether the modeler picked the right box, how closely does a particular wall assembly in the software match the behavior of that same wall assembly in the real world?
- How closely did the modeler select the correct dimensions, material types, volumes, air leakage numbers, etc.? IE, how well did the modeler do in representing the real world in the model?
- How closely does occupant use and behavior match the model expectations?
My gut is that if the design temps are accurate for the location, and the modeler does a very good job with the material selections, and that the building design matches the construction types provided, then my guess is that it is pretty good, probably better than +/- 20%, maybe more like +/-5%. But that is a lot of assumption to make. My guess is that the biggest source of inaccuracy is just the modeler not matching the model to the actual building, since that is the hardest part of the process.
Of course, "all models are wrong, and some models are useful" so it doesn't actually need to be perfect in order to help you make better decisions, it just has to be fairly close, and accurately reflect the directionality of changes - should I put my money into better windows or more air sealing?
Please make this article available outside the paywall. We need for the info in this article to be widely available to those millions of decision-makers in owner-occupied homes across the country.
Dear Martin: Thanks for another great article! You covered another key point with Manual J reports: bogus / fudged numbers simply dropped into a 'stock' report by an unscrupulous contractor. As Birchwood Bill noted above, I strongly encourage homeowners to buy and use ACCA certified Manual J software - either online, e.g. CoolCalc, or downloadable software, e.g. Wrightsoft. For the list go to: https://www.acca.org/standards/approved-software
Creating your own Manual J analysis is important for four reasons: (1) Accuracy - you are responsible for collecting the data, and you can examine your results, (2) You will have a Manual J report in hand when you talk to HVAC contractors. If they scoff and tout their "rule of thumb" expertise, thank them for their time and show them the door. (3) This gives you an opportunity to evaluate the opportunities for insulation and air-sealing improvements - and the actual cost of not doing so. (4) You may learn a lot about your home, because you have to measure every window, check whether they are double / triple pane, argon filled and coated to block or admit sunlight, understand the insulation and air sealing in your walls, attic and basement, check the state of your ducts (do they run through an unconditioned attic?) etc etc etc. Manual J can be an exercise in energy efficiency - and cost reduction.
Here is a complement for Manual J.: Pearl Certification for your home when you sell it - or when you buy one. Their assessor ran me through the wringer with the house we sold. It had lots of insulation, air sealing and window upgrades - but he found a few things I missed. The 'level' of their certification = real money. See: https://pearlcertification.com/homeowner-product-page
Reminds me of when the sales rep for a large regional installer informed me, at the conclusion of calculating a Manual J, that we needed a two ton system for our <500 sqft carriage house in Washington, DC. I didn’t know anything at the time; but, somehow I knew the smaller option (1.5 ton-two zone) would be adequate. His manager made me sign a waiver… which I did. That was five years ago and it’s been fine.
Log in or become a member to post a comment.
Sign up Log in