How Well Does an ACCA Manual J Load Calculation Predict my Own Home’s Performance?
It’s been hot here in southwest Virginia. Yesterday we reached 100 deg F for the first time in several years, so this week is a great example of “design” conditions. At these temperatures, my ancient air conditioner starts to struggle to maintain the thermostat setpoint. At 5 pm, with 100 deg F ambient temperatures, the temperature was 80 degrees at the thermostat, which was five degrees above setpoint.
I was curious how closely an ACCA Manual J load calculation would predict these conditions. The Manual J is a procedure for estimating the amount of heating and cooling a building requires, based on inputs such as the insulation values, windows, building orientation, and so on. Note that Manual J is used only for residential buildings, since larger commercial building behave fundamentally differently from homes. Whereas the heating and cooling loads in homes are driven by the building envelope, loads in commercial buildings are typically impacted more by internal loads, such as lighting, as well as ventilation loads.
Our home is a brick “four-square” house built in the 1920’s, and is approximately 2,000 sq. ft. The foundation is mostly basement (uninsulated) and partial crawlspace (R-13). The attic insulation is decent but not great (R-30), along with the walls in the rear addition (R-15). We’ve also done some air sealing and installed rigid foam insulation on the rim joists. The existing brick walls have not been modified over the years and probably have little to no insulation. The A/C unit is an old R-22 split unit, nominally rated for 2-1/2 tons of capacity. I’ve been waiting for it to pass on to the next world so I can install a more efficient unit, but every season it just keeps on going . . .
Before doing a load calculation, I would expect that the 2.5-ton unit would be on the small side for this size and age of home. It’s important to note that we don’t usually size cooling equipment to maintain setpoint on record hot days; instead, we typically size the equipment for ~99% of the hours in the year. If we size only based on the absolute worst scenario, then the equipment will be significantly oversized for the rest of the year. Drawbacks of oversizing include poor humidity control, equipment wear and tear from short-cycling, higher energy usage, and higher first costs.
Anyway, I went ahead and put together a Manual J (8th edition) using a commercially available software package, Right-Suite Universal. First, I ran the scenario using normal design conditions[i]. Then, I ran the calculation assuming 100 F outside, and adjusted the indoor design temperature until my total cooling load was ~30,000 btu (2.5 tons, the nominal size of my existing equipment[ii]). Last, I compared the predicted results to the actual results.
Results
First, the equipment sizing per the Manual J resulted in a ~3-ton unit, slightly larger than what’s installed.
Next, using 100-degree design conditions in the software, combined with 2-1/2 tons of cooling load, corresponded with an indoor temperature of 85 deg F, five degrees hotter than the actual indoor temperature during a 100-degree day. It’s important to note that the Manual J loads are steady-state loads, whereas in my test, we were only at 100 degrees for a few hours. Especially with a brick house, the thermal inertia of the mass of the house can help reduce the peak. If the outside temperature stayed at 100 degrees all day and night, the indoor temperature would probably creep up more in line with the Manual J results. In the real world, thankfully, we are only at peak temperatures for brief periods of time.
Conclusion
I have found Manual J to be fairly conservative in estimating cooling loads both in my own house and in other residential projects I’ve completed, which is in line with studies that show Manual J to slightly oversize equipment on average. 2-1/2 tons is a bit too small for our house for most people (though it’s fine for us since we don’t need much cooling). If I were sizing this equipment for someone else’s home, I would specify a 3-ton unit sized per Manual J, ideally with a dual-stage or variable-speed compressor and fan so that the equipment can better accommodate part-load conditions for the majority of the year.
As long as the inputs are correct in the model, there is no need to upsize equipment beyond the Manual J results for an additional safety factor. Another thing to watch out for is adding extra internal loads, etc., “just to be safe”. No need to exaggerate inputs.
A breakdown of the Manual J loads.
[i] Note that humidity is a large factor, too, but I have only shown the dry-bulb conditions for simplification here.
[ii] 2.5 tons is the equipment’s nominal capacity, based on one unique set of conditions. My equipment’s actual capacity at actual conditions is probably a bit lower than this number, but is close enough for the purposes of this little experiment.
[iii] This is the capacity of the equipment that is maintaining these conditions over a short period of time, and not the actual load. The actual load could be higher, meaning that if it stayed 100 deg F continuously, the house could continue to warm up. For practical purposes, though, we know that the 2.5-ton unit can maintain the indoor temperature at 80 degrees when it is 100 deg F outside for a typical peak day.
[iv] Per ACCA Manual S procedure.
