What it is

A load calculation gives you a number — say, 34,000 BTU/h of cooling load. Manual S is the next step: turning that number into a specific piece of equipment that actually delivers it under the conditions at the job. The trap most people fall into is thinking "34,000 BTU/h, so I need a 3-ton" and grabbing any 3-ton unit. That's not how it works, because a unit's real capacity changes with the conditions it's running in, and the nameplate "3 tons" is just a rating at one specific test point. Manual S is the discipline of matching delivered capacity to the load.

How it works

Equipment is rated at standard test conditions. Out in the field, the actual conditions — outdoor temperature, indoor wet-bulb, and airflow — are almost never the test conditions, and capacity moves with them:

  • Hotter outdoors → less cooling capacity (the condenser rejects heat against a smaller temperature difference and the compressor works harder).
  • Indoor wet-bulb changes the split between sensible and latent capacity. Drier indoor air shifts capacity toward sensible; more humid air pulls more toward latent.
  • Airflow (CFM) shifts capacity too — more airflow leans sensible, less leans latent.

So a "3-ton" unit might deliver more or less than 36,000 BTU/h depending on the day, and — critically — its sensible and latent portions shift around. Manual S uses the manufacturer's expanded performance data (the detailed tables/curves that show capacity at many conditions, not just the single nameplate rating) to find the unit whose delivered sensible and latent capacity at your design conditions best matches the sensible and latent loads from the Manual J.

In the field

The selection logic:

  1. Start with the load calc outputs: total cooling load, broken into sensible and latent (and the heating load separately). You're matching to these, not to a round tonnage.
  1. Pull the candidate equipment's expanded performance data at your design outdoor temp, design indoor wet-bulb, and intended airflow. The nameplate number alone isn't enough — you need capacity at your conditions.
  1. Check that delivered sensible capacity covers the sensible load and delivered latent capacity covers the latent load at those conditions. A unit can have enough total capacity but the wrong split — too little latent for a humid climate, for instance.
  1. Aim to match, not to wildly exceed. Standard practice keeps cooling capacity within a modest band of the load (generally not grossly oversized) so you avoid the short-cycling/humidity problems of oversizing. Slightly under on a rare design hour is usually better than chronically oversized.
  1. Match the indoor and outdoor units (the AHRI-matched combination). Capacity ratings are for matched coil/condenser/air-handler combinations. A mismatched coil changes the real performance and may not be rated at all.
  1. Reconcile heating and cooling. Heat pumps especially: the unit that's right for the cooling load may be over or under for the heating load. You balance the two, often with supplemental heat covering the gap rather than oversizing the compressor for winter.

Normal values & targets

  • Nameplate tonnage is a rating point, not a guarantee. A 3-ton (36,000 BTU/h nominal) unit delivers more or less than that depending on conditions — read the expanded data.
  • Cooling selection band (typical practice): keep total cooling capacity reasonably close to the load — modest, not grossly oversized. Excess capacity causes the humidity/short-cycling problems covered in the oversizing article.
  • Sensible and latent both have to be covered at design conditions. Total BTU/h matching isn't enough if the split is wrong for the climate.
  • Capacity drops as outdoor temp rises — the unit makes its rated capacity at the test point, less on a hotter design day. Select for the design day, not the test sheet.

Always use the actual manufacturer expanded performance data for the specific matched system; these are concepts and typical practices, not fixed selection rules.

Common faults & what they mean

  • Selecting on nameplate tonnage alone: ignores that real capacity shifts with conditions and that the sensible/latent split may not fit the climate. Total tons can be right while the unit still underperforms on humidity.
  • Mismatched coil and condenser: voids the rated capacity; real performance is unknown and often poor. Use AHRI-matched combinations.
  • Oversizing because "the load was 34k so I rounded the 3-ton up to a 3.5-ton": reintroduces every short-cycling and humidity problem; match the load instead.
  • Sizing a heat pump's compressor up to cover the heating load: oversizes it for cooling and hurts summer humidity — use supplemental heat for the winter gap instead.
  • Skipping the latent check in a humid climate: unit has enough total capacity but leaves the house damp because its latent capacity at design wet-bulb is too low.

Tech tips & gotchas

  • Manual J and Manual S are a pair. The load calc without proper equipment selection is half a job; you can do a perfect load calc and still ruin it by grabbing the wrong box.
  • Expanded performance data is the whole point. If you're only looking at the nameplate, you're not doing Manual S — you're guessing with extra steps.
  • The right answer is often "match the load," not "go bigger." Headroom is built into design conditions already; piling on more capacity is how comfort dies.
  • Airflow is part of the selection. The same equipment delivers a different sensible/latent split at different CFM — set airflow to suit the climate (lower for more dehumidification in humid areas, within manufacturer limits).
  • Document the selection against the load and the conditions. It defends the install and helps the next tech understand why this equipment is here.

Safety / code notes

  • Use AHRI-certified matched system ratings for capacity claims; unmatched combinations may not meet rated or code-required efficiency.
  • Many jurisdictions require documented load and equipment-selection calculations for permitted installs — confirm local code.
  • This article teaches the concept and method of equipment selection in plain terms. It does not reproduce ACCA Manual S procedures or any manufacturer's performance tables — use the actual standard and the equipment's published data for a real selection.