INTRO (0:00 to 0:30)
Voiceover, cold open on a thermostat reading 75 with a weather app beside it reading 112: Every install decision you will ever make starts with one number: how much heat is coming into this house on the design hour. Not roughly. Not per square foot. A calculated number, in BTU per hour, with a sensible piece and a latent piece. The calculation is ACCA Manual J, the equipment match is Manual S, and a master tech owns both. Four and a half minutes, the whole skeleton.
ON-SCREEN: M37: Load Calculation and Equipment Selection
ON-SCREEN: Manual J = the load. Manual S = the machine. Manual D = the ducts (M38)
MAIN (0:30 to 4:15)
[0:30-1:15] The load is physics you already know
Voiceover over M37-manual-j-inputs-map.svg, highlighting each arrow as it is named: F3 taught you heat flows hot to cold through every resistance in its path. A house at 75 in air at 112 is taking heat through six doors at once: solar gain through the glass, scaled by orientation and SHGC. Conduction through walls and ceilings, set by R-values. Infiltration leaking hot outside air through the envelope. Internal gains from people and appliances. And the duct penalty: ducts in a hot attic leak 20 to 30 percent of your air and soak up heat the whole trip. Add it up at design conditions, the 1 percent values, never record extremes, and split it into sensible and latent. That is Manual J. No rules of thumb survive contact with it.
ON-SCREEN: highlight each input on the map as named
ON-SCREEN: Design point this course: 112 F outdoor, 75 F indoor. 1 percent design, not records
[1:15-2:00] Three houses, one square footage, three loads
Voiceover over M37-envelope-heat-gain-phoenix.svg, the three stacked bars: Here is why square feet per ton is dead. Three homes, 2,000 square feet each, same design day. Eighties block house, original single-pane glass, leaky envelope, ducts in a hot attic: about 52,000 BTU per hour. Nineties frame house, partially upgraded but ducts still in the attic: about 38,000. A 2020 spray-foam build with the attic sealed inside the envelope: about 26,000. The 500-square-feet-per-ton rule gives all three the same 4 tons: too small, sloppy, and nearly double, in that order. The envelope sets the load. The floor area just holds it up.
ON-SCREEN: 52,000 / 38,200 / 26,000 BTU per hour. Same 2,000 sq ft
ON-SCREEN: One rule of thumb, wrong three different ways
[2:00-2:45] What oversizing actually costs
Voiceover over M37-oversizing-failure-chain.svg, following the chain left to right: Oversize is not a safety margin, it is a failure chain. The big unit hits setpoint in minutes and slams off. Short cycles mean comfort swings, because rooms far from the thermostat never get a long runtime. They mean weak moisture removal, because a coil only dehumidifies after it has run wet for several minutes. They mean multiplied starts, and starts are what kill capacitors, contactors, and compressors; capacitors alone are a fifth of all service calls. And the big unit demands more airflow through the same old ducts, which is the C12 high-static chain with a new nameplate. Undersize has one cost: it loses a few degrees on the few hottest afternoons. That asymmetry is why the sizing window leans where it does.
ON-SCREEN: Short cycle: swings, weak latent, start wear, duct static
ON-SCREEN: Recall: capacitors = 21 percent of service calls
[2:45-3:45] Manual S: real capacity, not nameplate
Voiceover over M37-manual-s-selection-window.svg, then a screen capture of an expanded performance table: A nominal ton is 12,000 BTU per hour at the AHRI lab point: 95 degrees outside, humid 80/67 air entering the coil. Nobody's design day lives there. At 112 to 115, capacity falls roughly a percent per degree, and dry desert air entering at 62 to 63 wet bulb shifts the coil's work from latent to sensible. So Manual S reads the manufacturer's expanded table at the real ambient and the real wet bulb, interpolating between rows. Worked example: a nominal 3.5 ton labeled 42,000 delivers about 36,700 at 112. Against a 38,200 load, that is 96 percent: inside the single-stage window of 90 to 115 percent of total load. Staged equipment gets up to about 120, variable capacity up to about 130, sensible must cover sensible, latent must cover latent, and when two machines both fit, take the smaller one.
ON-SCREEN: AHRI point: 95 F, 80/67 entering. Your design day: 112 F, about 63 wb
ON-SCREEN: Nominal 42,000 delivers about 36,700 at 112 F
ON-SCREEN: Window: single-stage 90 to 115 percent. Smallest that covers wins
[3:45-4:15] The flow and the cross-examination
Voiceover over M37-load-calc-software-flow.svg: Every load calc tool walks the same flow, whatever its logo: design conditions, construction assemblies, room-by-room geometry, infiltration, internal gains, ducts, outputs. Your job is honest inputs and then cross-examination: BTU per square foot inside the vintage band, latent fraction believable for the climate, component breakdown matching the house you walked, and the old equipment's runtime history used as a witness. A calc that survives that is a number you can defend. One that ships unexamined is just confident garbage.
ON-SCREEN: Inputs honest. Outputs cross-examined. Then Manual S
OUTRO (4:15 to 4:30)
Voiceover, closing card: Block load picks the box. Room-by-room sizes the air. Watch the v2 video to see a real house walked input by input, the calc reviewed, and the machine selected off the table at 112. Then M38 takes the room-by-room sheet and designs the ducts.
ON-SCREEN: Next: v2 field demo. Then M38: Manual D duct design