Duct Design and Sizing Reference
Why this reference exists
Most field service techs work on existing ductwork that was sized + installed by someone else. When it fails, retrofits become guesswork: "feels like the supply is small, let's go up a size." This costs hours of labor + leaves customers with rooms that are still hot/cold. Proper duct design (per ACCA Manual D) is the framework that makes retrofits + new ducts reliable.
How airflow works
Air at a given pressure moves through ducts at a velocity. Pressure drops along the run. The HVAC equipment has a static pressure budget (typically 0.5 in WC total external to equipment). If the duct system requires more pressure than that, airflow drops below design + the system underperforms.
Components of duct pressure loss:
- Friction in straight runs
- Fittings (elbows, transitions, takeoffs)
- Filter resistance
- Grilles + registers
- Coils + heat exchanger
Design budgets ~0.05-0.08 in WC per 100 ft equivalent length for straight ducts.
Sizing methods
Equal Friction (most common): designer picks a target pressure drop per 100 ft (typically 0.08-0.10 in WC), then sizes each branch to match that target. Result: each run is the same friction-per-foot.
Velocity Sizing: limit air velocity by location:
- Main trunk: 800-1,200 FPM
- Branch ducts: 600-900 FPM
- Returns: 600-700 FPM
- Final outlet velocities: 400-600 FPM for comfort
Static Regain (commercial): sophisticated method matching velocity changes at each branch. Rare residential.
For residential: Equal Friction at 0.08 in WC per 100 ft is the working default.
Duct material
Sheet metal (galvanized steel):
- Most efficient airflow (low friction)
- Most durable
- Requires sealing at joints
- Site-fabrication possible
- Industry standard for trunks + main branches
Flex duct (insulated):
- Easy install
- Higher friction (longer effective length than equivalent rigid)
- Easy to crush + reduce airflow
- Typical run-out from trunk to register
- Spec: UL 181 rated, R-6 to R-8 insulation
- Max suggested length: 6 ft per run-out
Rigid fiberglass duct board:
- Used in some installations, especially in unconditioned spaces
- Lower noise transmission
- Less durable
- Falling out of favor in residential
Smaller-diameter ducts (high-velocity / mini-duct):
- 2-3" diameter ducts at higher velocity
- Unico, SpacePak; older homes without space for traditional ducts
- Niche
CFM per room
ACCA Manual J + D process:
- Calculate heat gain + heat loss per room (Manual J)
- Determine CFM needed per room (delta-T of system + room load)
- Size duct branches to deliver that CFM
Quick approximation:
| Room type | CFM/ton |
|---|---|
| Bedroom | 50-100 |
| Living room | 100-200 |
| Kitchen | 100-150 |
| Bathroom | 30-50 |
| Master suite | 150-250 |
Total CFM ≈ 350-450 per ton of system capacity (heat pumps 400-500 per ton).
Trunk + branch sizing
For a 4-ton system at 400 CFM/ton = 1600 CFM total:
Main supply trunk: 16" round OR 12x16" rectangular sized for ~1200 FPM = 1600 CFM.
Branch reduces stepwise as it feeds rooms. Run-outs to single registers are typically 6" or 8" round.
Use a friction chart (ASHRAE handbook, ACCA Manual D, or duct calculator app) to size each section.
Returns
Returns are often undersized. Industry rule: return CFM should equal supply CFM. Returns sized for 600-700 FPM (lower velocity than supply for quiet operation).
A 4-ton system needs ~1600 CFM of return air. Single 20x25 return grille = ~700 CFM. Most homes need MULTIPLE return grilles.
Bedrooms with doors closed: need either a return grille IN the bedroom OR transfer grilles in walls/doors OR jump ducts in ceiling. Without these, closed bedroom door = pressurization + reduced airflow.
Sealing
Duct leakage is the dominant inefficiency in most residential systems. Studies show 20-30% air loss to attic/crawl is common.
Sealing methods:
- Mastic + fiber mesh tape (UL 181 listed): durable, premium
- Foil tape (UL 181-listed): acceptable for short joints, less durable than mastic
- Aeroseal (proprietary aerosol seal sprayed into ducts): premium retrofit; 90%+ leakage reduction
- Duct boot sealing: foam + mastic at register + return grille boots
DO NOT use:
- Cloth-backed "duct tape" (the irony - fails fast)
- Acoustical caulk (not rated for duct)
Insulation
In unconditioned attics + crawl spaces, ducts need insulation:
- R-8 minimum per IECC
- Vapor-barrier on exterior
- Spiral-tape seam joints
Without insulation: condensation drips in summer + huge heat gain/loss.
Acceptance criteria (duct design)
Per Manual D:
- Total external static pressure within equipment range (typically 0.5 in WC max)
- CFM per room within ±10% of design
- Velocity within range (no whistling, no rumbling)
- Leakage less than 5% measured by Duct Blaster
Common design pitfalls
- Undersized returns: starves system, raises static pressure, reduces airflow
- Long flex runs: 12 ft of flex equivalent to 6 ft of rigid; design assumes shorter
- Crushed flex behind insulation: reduces effective area
- No room-by-room CFM calc: hot/cold rooms
- Wrong supply register: high-velocity register in bedroom = drafts + noise
- Bedroom door closed without return path: pressure imbalance
- Unsealed ducts in attic: 30% energy loss
- Trunk too small: high velocity, noisy, equipment static pressure exceeded
Diagnostic tools
For service trade techs:
- Static pressure manometer (Magnehelic, Dwyer): measures total external SP at supply + return
- Anemometer or flow hood (TSI, Alnor): measures CFM at registers
- Duct Blaster (Retrotec, Energy Conservatory): measures duct leakage
- Smoke pencil or smoke pen: visualizes airflow path + leaks
A 5-minute static pressure test on a complaint call reveals 60-70% of "system not enough" issues are duct, not equipment.
Retrofit decisions
Common scenarios:
"Bedroom is hot":
- Measure CFM at bedroom register
- Compare to design CFM for that load
- If short: undersized branch, crushed flex, closed damper, return path
- Fix at source, not by upsizing equipment
"System struggles in summer":
- Measure system static pressure
- If > 0.5 in WC: ducts or filter too restrictive
- Solution: bigger filter housing, redesigned duct, sealed leaks
- NOT a bigger AC
"Loud system":
- Measure register velocity
- If > 600 FPM: oversized airflow OR undersized duct
- Solution: balancing dampers, redesigned duct
- Sometimes a larger filter housing (lower restriction)
Customer talking points
When duct issues are root cause:
References
- ACCA Manual D Residential Duct Systems
- ACCA Manual J Load Calculation
- SMACNA HVAC Duct Construction Standards
- ASHRAE Handbook (Fundamentals, Systems + Equipment)
- IECC R403.3 + 403.4 (duct sealing + insulation)
- Manuall internal: Annual HVAC System Maintenance, MERV Ratings and Air Filtration Reference