HVAC
Duct design and friction rate field guide with ACCA Manual D
Run Manual J to S to D in order, set the friction rate from available static and total effective length, size every trunk and branch to it, and verify the airflow in the field.
Direct answer
Manual D is ACCA's residential duct design procedure: it sizes every duct so the blower delivers each room its design airflow within the equipment's available static pressure. The friction rate, available static pressure times 100 divided by total effective length, in inches of water per 100 ft, is the budget you size every run against.
Key takeaways
- Friction rate equals available static pressure times 100 divided by total effective length, in inches of water per 100 ft.
- Run the ACCA chain in fixed order: Manual J for load, Manual S for equipment, then Manual D for the duct.
- A workable friction rate usually lands between about 0.06 and 0.18 in. wg per 100 ft; do not default to a flat 0.10.
- Total effective length is the longest supply run plus the longest return run, with every fitting counted as equivalent length.
- Common residential velocities: supply trunks 700 to 900 ft/min, branches around 600, returns 600 or lower near living space.
Manual D, and the friction rate as the budget you spend
Manual D is ACCA's procedure for designing a residential and light-commercial duct system so the blower moves the right amount of air to every room without running out of static pressure on the way. It is not a chart you look a size up on. It is a method that ends in a single number, the friction rate, and then sizes every trunk, branch, and run to that number.
The core idea is a budget. The equipment has a fixed amount of static pressure it can spend pushing air through the duct, the filter, the coil, and the grilles. Manual D works out how much of that is left for the duct alone, spreads it over the longest air path in the house, and turns the result into inches of water per 100 ft. That per-100-ft figure is the friction rate, and you size every duct so it never spends more than its share.
Get the friction rate right and a 6 in branch and a 100 ft trunk both come back inside the same static budget, which is the whole point. Get it wrong, or skip it and eyeball the sizes off habit, and you build a system that either starves the far rooms or roars at the registers. The friction rate is the discipline that keeps the far room and the blower table on the same side of the math.
Why can't you size a duct without the load and the equipment first?
Because the duct exists to carry an airflow you do not know yet, delivered by a blower you have not picked yet. The sequence is fixed: Manual J for the load, Manual S for the equipment, then Manual D for the duct. Run them out of order and you are sizing pipe for a flow rate you are guessing at.
Manual J is the room-by-room heating and cooling load. It comes first because the load sets the airflow each room needs, which is the CFM you will eventually size the branch to deliver. Manual S selects the equipment to match that load, and Manual S is the ANSI-recognized procedure for residential equipment selection. The reason Manual S matters to the duct is the blower data that rides with the selected unit: the rated external static and the airflow it makes at each speed tap. That rating is the static budget Manual D divides up.
Manual T sits in the chain too, covering air distribution, the register and grille selection that puts the air into the room without dumping it on someone. The sequence is sometimes written J, S, T, then D. The order is the law of the work. You cannot pick a duct size against a static budget the equipment has not been chosen to provide, against an airflow the load has not been calculated to require. A duct sized before the load is a duct sized to nothing.
Available static pressure: what the duct actually gets
Available static pressure is the static the duct runs get after the components take their cut. The equipment's blower is rated for a maximum external static, and the filter, the coil, the supply registers, the return grilles, the balancing dampers, and any accessory like a humidifier or a UV rack all spend part of it before the duct sees a foot of it. Subtract the component allowances from the rated external static and what is left is the available static pressure, the ASP that belongs to the duct.
This is the step amateurs skip, and it is the most expensive one to skip. Design the duct to the full rated external static and you have spent the filter's and the coil's share twice. The installed system then runs well past its rating, the blower slides down its curve, and the airflow falls off where you cannot see it without a manometer. The duct static pressure guide covers measuring that installed static and reading it against the blower table; here the same budget runs the other direction, from the rating down to what the duct can have.
Use realistic allowances, not optimistic ones. The filter is the one that climbs as it loads, so the clean number is the floor. The coil under a wet load runs higher than the dry catalog figure. When the equipment manufacturer publishes a component pressure drop, use it; when you have to estimate, estimate generously, because a tight ASP is what keeps the duct from being designed to a friction rate the blower cannot actually deliver.
| Static budget item | Example allowance (in. wg) | Notes |
|---|---|---|
| Rated external static (from blower table) | 0.50 | The whole budget; read the actual unit, not a rule of thumb |
| Filter | 0.10 | Clean is the floor; rises as it loads |
| Cooling coil | 0.15 | Wet coil under load runs higher than the dry figure |
| Supply registers and return grilles | 0.06 | Free area and face velocity drive it |
| Balancing dampers and accessories | 0.05 | Humidifier, UV, zone dampers if present |
| Available static pressure for the duct (ASP) | 0.14 | What remains for supply plus return runs |
What is total effective length?
Total effective length is the longest air path the blower has to push against, measured in feet, with every fitting counted as the straight duct it acts like. It is the longest supply run plus the longest return run, the two added together to form the critical path. You do not add up all the duct in the house. You find the one supply run and the one return run that are hardest on the blower and size the system around that worst case.
The trap is that fittings, not straight pipe, usually dominate the number. A run might be 40 ft of straight trunk and branch, then carry a supply plenum takeoff, two elbows, a branch takeoff, and a register boot. Each of those fittings is counted as an equivalent length, the feet of straight duct that would cost the same static, and the Manual D tables assign a value to every fitting type and size. A single hard 90 can carry tens of feet of equivalent length. Stack four or five fittings on a run and the fittings can outweigh all the straight duct in it.
This is why a short, fitting-heavy run can be harder on the blower than a long, clean one. Measure the straight footage off the plan, then add the equivalent length of every fitting on the supply side of the worst run and every fitting on the return side of its worst run. The two longest effective lengths added together is the TEL the friction rate divides into. Leave the fittings out and the friction rate comes back too generous, every duct gets undersized, and the system runs short of air it looked fine to deliver on paper.
- TEL
- Total effective length, the longest supply run plus the longest return run, with each fitting counted as its equivalent length of straight duct
- Equivalent length
- The feet of straight duct that would cost the same static pressure as a given fitting, read from the Manual D tables
- Critical path
- The worst-case air path the system is sized around, the longest supply plus the longest return effective length
How do you calculate the friction rate?
Multiply the available static pressure by 100 and divide by the total effective length. The result is the design friction rate in inches of water per 100 ft, and it is the pressure drop per 100 ft you allow every duct to spend. With an ASP of 0.14 in. wg and a TEL of 175 ft, the friction rate is 0.14 times 100 divided by 175, which is 0.08 in. wg per 100 ft.
The 100 is in the formula because the friction chart and the ductulator are built around pressure drop per 100 ft. The friction rate is not a target you pick off a shelf. It falls out of the equipment you selected and the duct path the house forced on you, which is why the same nominal tonnage can need different duct sizes in two different houses.
A workable friction rate usually lands between about 0.06 and 0.18 in. wg per 100 ft, the band ACCA describes as the wedge. Come in under 0.06 and the ducts get large and expensive for no airflow gain. Run over about 0.18 and the velocities climb into noise and the runs start starving. The old habit of designing everything to a flat 0.10 is a guess, not a calculation, and ACCA has been blunt that the friction rate is not a constant. Run the number for the system in front of you.
FR = (ASP × 100) / TELASP = ESPrated − component losses- ASP
- Available static pressure for the duct, in inches of water column, after the components are subtracted from the rated external static
- FR
- Friction rate, the design pressure drop per 100 ft of duct, in inches of water column per 100 ft
- ESP (rated)
- The maximum external static pressure the equipment's blower is rated to push against, from the blower table
Reading the friction chart and the ductulator
Once you have the friction rate, the friction chart and the ductulator turn a CFM into a duct size. You enter with two numbers, the airflow the run carries and the design friction rate, and the chart hands you the round duct diameter that delivers that flow at that pressure drop, along with the velocity it runs at. A ductulator is the same chart wrapped on a slide wheel, and it is still the fastest tool on a tailgate.
The chart works in round duct, so rectangular trunks need the equivalent diameter, the round size that carries the same flow at the same friction. The common conversion is the Huebscher relation, and it is built into every ductulator and duct calculator, so in practice you read the round size, then look up the rectangular dimensions that match it on the equivalent-diameter table. Round is the more efficient shape for a given flow because it has less surface for the air to drag on, so a round duct and its rectangular equivalent are never the same area.
Read both numbers the chart gives you, not just the size. Velocity comes off the chart with the diameter, and it is the second check that keeps a run inside the noise limits. A size that satisfies the friction rate but lands at 1000 ft/min in a bedroom branch is the wrong size even though the friction math passed, which is why the velocity column matters as much as the diameter.
De = 1.30 × (a × b)0.625 / (a + b)0.25- Ductulator
- A slide-wheel or app version of the friction chart that converts CFM and friction rate into a duct size and velocity
- Equivalent diameter
- The round duct size that carries the same airflow at the same friction rate as a given rectangular duct
What duct velocity is too high?
Velocity is too high when the air makes noise the occupant hears or when it starts costing static the friction rate did not budget for. Common residential practice keeps supply trunks roughly in the 700 to 900 ft/min range, branch runs around 600, and returns at 600 or lower, with the lower numbers near the living space where ears are. These are common design ceilings, not hard code limits, and the equipment's sound data and the project's noise criteria control the call.
The reason returns run slower than supplies is location. Return grilles sit in hallways and living rooms, so a return moving air fast turns into a whistle in the quietest part of the house. Supplies can run a little faster because the trunk is usually in a chase or an attic, but push a branch boot past its limit and the register roars no matter how well the room is balanced.
There is a floor as well as a ceiling. Air moving too slowly in a supply branch loses its throw and dumps cold air at the boot instead of carrying it into the room, and on heating it can stratify. The friction-rate method tends to handle this on its own, since equal friction naturally drops velocity as the trunk reduces and the flow splits off, but check the velocity column on the short, high-flow runs near the air handler, which is where the speed runs highest and the noise complaint starts.
| Run | Common residential velocity (ft/min) | Why |
|---|---|---|
| Supply trunk | 700 to 900 | In a chase or attic; tolerates more speed |
| Supply branch | about 600 | Feeds the boot; noise shows up at the register |
| Return trunk | 700 to 900 | Away from occupied space |
| Return grille and branch | 600 or lower | Sits in the living space; whistles when fast |
Equal friction, static regain, and constant velocity
Three methods size a duct system, and they are not interchangeable. Equal friction is the residential and light-commercial method, the one Manual D is built on. Static regain is the larger-commercial method. Constant velocity belongs to industrial exhaust. Pick the wrong one for the job and you are doing more math than the system needs or less than it deserves.
Equal friction holds the same friction rate across every run, which is exactly the friction-rate budget this guide describes. Its quiet advantage is that velocity falls on its own as the trunk reduces and air peels off to the branches, so the far end naturally runs slower and quieter. It is simple, it suits constant-volume systems, and it is why a hand ductulator and a friction rate get a house designed correctly without a computer.
Static regain is the move on big, high-velocity commercial systems with long runs. It sizes each successive section so the static recovered as the air slows down offsets the friction loss in the next section, holding a near-constant static at every branch and terminal. It evens out the pressure available at the diffusers on a long distribution main, but it is more work and it is overkill on a house. Constant velocity, holding a fixed speed through the whole system, is for exhaust that has to keep particulate or fumes entrained so they do not drop out and settle in the duct. For residential and most light-commercial supply air, equal friction is the method, and the other two are tools you reach for when the building tells you to.
Sizing the trunk, the branches, and the room CFM
Start from the room airflows the load handed you. Manual J gives each room its heating and cooling load, and that converts to a design CFM per room, the number the branch to that room has to deliver. Add the rooms on a trunk and you have the CFM that trunk carries before its first takeoff. Size from the air handler outward, the trunk for the full flow, then each section for the flow that is left after the branches ahead of it have peeled off.
The trunk reduces as it sheds air. A supply trunk that leaves the air handler at 1200 CFM does not stay one size the whole way; after it drops 400 CFM to the first few rooms, the next section carries 800, and at the design friction rate that is a smaller duct. Reducing the trunk holds the velocity and the friction rate roughly constant down its length, which is the equal-friction method doing its job. A trunk that never reduces wastes metal and lets the near branches hog the air.
Branches and takeoffs each get sized for their own CFM at the same friction rate. The takeoff type matters here, because a takeoff is a fitting with its own equivalent length, and a high-loss takeoff eats into the budget you sized the branch against. Size the branch off the chart for its room CFM, pick a takeoff that does not throw away static, and run the branch to the boot at the register. The discipline is the same at every scale: the flow sets the size, the friction rate sets the standard, and every section answers to both.
Why is my return undersized?
The return is half the system and it is the half that gets value-engineered down to one undersized grille. The supply gets all the design attention, room by room, while the return becomes whatever filter-back grille fit the wall. Then the blower has to pull the whole airflow back through that one choke point, the return static climbs, and the system runs short on air it was supposedly designed to deliver.
The physics is simple and it is unforgiving. Whatever the supply pushes out, the return has to pull back, so the return path needs comparable area to the supply, not a fraction of it. An undersized return shows up as a high negative static on the return side of the manometer, and it cannot be fixed from the supply side no matter how many supply branches you open. The duct static pressure guide walks through splitting the reading to localize a high return static; in design, the move is to size the return path to the airflow up front so it never becomes the bottleneck.
Design enough return area, and design the path, not just the grille. A single central return needs a duct sized for the full system CFM at the friction rate, the same as the supply trunk. Multiple returns or transfer paths spread the pull and run quieter. Where doors close off rooms with supplies but no return, the air needs a way back, a transfer grille, a jumper duct, or an undercut door, or the room pressurizes and the supply backs down. A closed bedroom door with a supply and no return path is the most common return failure in a finished house, and it is a design oversight, not an install error.
Registers, grilles, and throw
The duct delivers the air to the boot, and the register or diffuser puts it into the room. Selecting them is Manual T territory, air distribution, and the two numbers that matter are the airflow the outlet has to pass and the throw it has to make. Throw is how far the air carries into the room before it slows down, and it is what mixes the supply air with the room air instead of dropping it cold on the floor under the register.
Size the supply register for its branch CFM at a face velocity that throws the air where the room needs it without making noise. A register too small for its flow spikes the face velocity, whistles, and throws too hard; a register too large dumps the air short and leaves a dead spot. Manufacturer selection tables give the throw at a given flow for each register, and that is the data you select against, not the nominal neck size.
Return grilles get sized for a low face velocity, because they sit in the occupied space and noise is the constraint. A return grille run too fast is the whistle in the hallway. Size the grille's free area, not its overall dimension, for the flow at a face velocity that stays quiet. The grille that looks big enough by its outside size is often half free area behind the bars, which is exactly how a return that measured fine on the wall measures undersized on the manometer.
Fittings and the cost of a hard 90
Fittings are where a duct design quietly bleeds static, because each one carries an equivalent length that the straight-duct footage hides. A hard, mitered 90 costs far more equivalent length than a long-radius elbow of the same size. A high-loss plenum takeoff costs more than a smooth one. A register boot turning the air 90 degrees into the room is its own equivalent length on every branch. Stack them and the fittings can outweigh the straight duct on the run.
Design the bad fittings out before you size up to cover them. A long-radius elbow instead of a hard 90, a smooth saddle takeoff instead of a punched hole, a boot with a gentler turn, each one buys back equivalent length and lets the run hold its size. The cheapest way to widen the static budget is rarely a bigger duct. It is a better fitting that costs less of the budget in the first place.
Count every fitting in the TEL, then design to make the worst run shorter in effective length, not just larger in diameter. The run with the air handler takeoff, two elbows, a branch tee, and a boot is carrying its fittings as the bulk of its effective length, and swapping two of those for low-loss versions can move the friction rate enough to keep the whole system inside the budget. Fittings ignored in the TEL is the single most common reason a design that passed on paper starves the far room in the field.
Sheet metal, flex, and the flex penalty
Duct material changes the loss, and flex is where good designs die in the field. Properly stretched and supported, research shows flex duct runs close to sheet metal on pressure drop. The problem is that it almost never stays properly stretched. The spiral wire interior is rougher than smooth metal to begin with, and the moment flex is left slack, compressed, or sagging between supports, the loss climbs fast.
The numbers are brutal and worth carrying. Studies have found that flex left at just 4 percent slack moved on the order of a third less air than metal duct. At 15 percent compression the pressure drop runs several times the stretched value, and at 30 percent compression, the kind you get with flex draped over joists, it can climb to roughly ten times. This is why a common move is to design flex to a lower friction rate than metal, on the order of 0.05 in. wg per 100 ft against the 0.10 you might use for sheet metal, or to run the full Manual D with flex's real roughness. You are buying back the loss the installer's slack will add.
Lined duct adds its own friction. Internal acoustic liner roughens the surface and shrinks the clear area, so a lined trunk carries less air than its sheet-metal dimension suggests and has to be sized for the lined inside, not the outside. Use flex for the last short connection to a boot where it belongs, keep it stretched and supported, and run the trunks and the long branches in metal where the loss stays where you designed it. Flex everywhere, installed loose, is a duct system designed twice and delivered once.
Duct leakage and the sealed-duct assumption
Every Manual D number assumes a sealed duct. The design moves the CFM you calculated from the air handler to the register, and it has no allowance for air that escapes through unsealed joints into an attic or a crawlspace along the way. Leak the duct and the far room gets less than the design said, no matter how clean the friction-rate math was.
Leakage is the loss nobody measures because the air is gone before it reaches a grille to read. A return leaking in an attic pulls hot, dusty air into the system; a supply leaking in a crawlspace dumps conditioned air you paid to move into a space nobody occupies. SMACNA owns the duct construction and the sealing classes that control this, the gauges, the joint and seam reinforcement, and the leakage classification the duct is built and sealed to. The design assumes the duct is sealed to a class; the install has to actually hit it.
Seal to the joints, not the tape brand. Mastic on every transverse and longitudinal seam, the takeoffs, and the boot-to-drywall connection is what makes the sealed-duct assumption true. The duct static pressure guide treats leakage as the static problem that hides because the air escapes before a register reads it; in design, the rule is to draw and specify a sealed duct, and to remember that a leaky duct turns a correct design into an undersized one in the field.
Sizing for high-CFM and data center airflow
High-CFM work changes the geometry but not the budget. A data center or a large commercial space moves far more air than a house, and the design often runs through a plenum, an underfloor raised floor or an overhead supply, rather than a tree of branch ducts. The number that drives the air is still a static budget spread over a path, and the velocity still has to stay inside its limits or the system makes noise and burns fan power.
Underfloor distribution turns the raised floor into the supply plenum, and the pressure under the floor, not a duct friction rate, drives air up through the perforated tiles to the equipment inlets. That plenum static is small, and the design job is keeping it even so the far racks get air, the same discipline as a residential trunk that reduces so the far branch is not starved. Overhead ducted supply to a containment aisle is closer to conventional duct design at a larger scale, sized for the high CFM at a velocity that stays quiet and efficient.
The thermal target the airflow has to deliver comes from ASHRAE TC 9.9 on the data center side, and the cooling material covers the envelope the equipment wants. For the duct or plenum designer, the carryover from residential is exact: find the budget, spread it over the path, hold the velocity, and stop the leaks. A wide-open near tile starves the far rack the same way a wide-open near branch starves the far room, and the fix is balancing the path, not adding fan.
How do you verify a duct design in the field?
You verify a duct design by measuring the airflow it actually delivers against the airflow it was designed to deliver, and the static it actually runs against the budget you designed to. The design is a prediction. The manometer and the flow hood tell you whether the prediction came true, and on a real install it often does not on the first reading.
Measure the external static first and read it against the equipment's blower table, because that frames everything. If the installed static came in higher than the available static you designed to, the duct is fighting more than the budget allowed, and the airflow is already low before you check a single register. The duct static pressure guide covers measuring TESP and splitting supply from return to find which side carries the excess. Then plot the fan airflow off the table and check the room airflows at the outlets against the design CFM.
When the install does not match the design, find the gap before you blame the equipment. The usual culprits are flex left slack, fittings worse than the ones in the TEL, a return that came in undersized, or duct leakage the design assumed away. A full test and balance is how the airflow gets set to design and documented, and the air balancing report guide covers the proportional method and the report the engineer signs. Design predicts the airflow; balancing proves it and sets it, and the two belong in the same job folder.
Field example: friction rate for a 3-ton system
Take a 3-ton air handler rated for 0.50 in. wg external static, moving about 1200 CFM. Subtract the components: 0.10 for the filter, 0.15 for the coil, 0.06 for the registers and grilles, and 0.05 for the dampers and accessories. That leaves an available static pressure of 0.14 in. wg for the duct.
Now the path. The longest supply run measures 75 ft of straight duct plus fittings, a plenum takeoff, two elbows, a branch takeoff, and a boot, that add about 60 ft of equivalent length, for 135 ft. The worst return run is 25 ft of duct plus a grille and an elbow worth about 15 ft, for 40 ft. The total effective length is 135 plus 40, which is 175 ft.
The friction rate is 0.14 times 100 divided by 175, which is 0.08 in. wg per 100 ft, inside the wedge. Now size each run at 0.08. The trunk at 1200 CFM and 0.08 comes off the chart around a 14 in round or its rectangular equivalent, running near 800 ft/min. A 150 CFM bedroom branch at 0.08 lands around a 7 in round at roughly 560 ft/min, quiet and inside the branch limit. Notice the friction rate, 0.08, not the default 0.10, fell out of this house's equipment and path, and a flat 0.10 would have undersized every run.
| Run | CFM | Friction rate (in. wg/100 ft) | Size (round) | Velocity (ft/min) | TEL (ft) |
|---|---|---|---|---|---|
| Supply trunk (first section) | 1200 | 0.08 | 16 in | about 860 | 135 (worst supply path) |
| Supply trunk (after 400 CFM drops off) | 800 | 0.08 | 14 in | about 750 | part of critical path |
| Bedroom branch | 150 | 0.08 | 7 in | about 560 | on the 135 ft path |
| Return trunk | 1200 | 0.08 | 16 in | about 860 | 40 (worst return path) |
| System critical path TEL | 175 |
Common mistakes
- Eyeballing duct sizes off habit with no friction rate behind them.
- Designing to the full rated external static without subtracting the filter, coil, registers, and dampers first.
- Leaving fittings out of the total effective length, so the friction rate comes back too generous.
- Undersizing the return to a single grille while the supply gets all the design attention.
- Running flex everywhere and counting on metal-duct loss when the slack flex runs several times higher.
- Defaulting every job to a flat 0.10 friction rate instead of calculating it for the system.
- Sizing for friction rate and never checking the velocity column, so a branch passes the math and roars at the register.
- Forgetting the return path for closed-door rooms, so a transfer grille or jumper was never designed in.
Field checklist
Want this checklist to run itself on every job — with photo proof and a signed record crews can hand the customer? That's FieldOS.
What to document
Come the complaint that the far room runs short, the design record is what explains why each run is the size it is and shows the friction rate behind it held up. The record is what tells the next person why a run is the size it is, and it is what a commissioning agent checks the install against.
Capture the available static pressure and how you got it, the total effective length with the fittings counted, and the friction rate that fell out. Then, for every run, record the CFM, the friction rate it was sized to, the size, the velocity off the chart, and its place on the critical path. Note the material, metal or flex, because the friction rate assumption depends on it, and note the return path for closed-door rooms. The table below is the core run-by-run record.
| Run | CFM | Friction rate | Size | Velocity | TEL |
|---|---|---|---|---|---|
| Supply trunk | Design flow at the air handler | Design FR (in. wg/100 ft) | Round or rectangular | ft/min off the chart | Worst supply path |
| Each branch | Room design CFM from Manual J | Same design FR | Round size | ft/min, checked vs limit | On the critical path |
| Return trunk and grille | Full system CFM | Same design FR | Sized to full flow | ft/min, kept low | Worst return path |
| System | Total CFM | ASP x 100 / TEL | Supply plus return TEL |
Standards and references
The residential duct design chain belongs to ACCA. Manual J calculates the load, Manual S selects the equipment to the load, Manual T covers air distribution and register and grille selection, and Manual D designs the duct using the available-static and friction-rate method this guide describes. Manual S and Manual D are ANSI-recognized procedures. Run them in order, J then S then D, with T feeding the outlet selection.
SMACNA owns duct construction: the sheet metal gauges, the joint and seam reinforcement, the pressure classifications, and the sealing classes that control the leakage a Manual D design assumes away. The duct you draw to a friction rate has to be built and sealed to a SMACNA class to actually deliver it. AMCA rates the fan performance behind the blower table you design against. ASHRAE carries the commercial side, with handbook duct design guidance, Standard 62.1 for ventilation, 90.1 for energy, and TC 9.9 for the data center thermal envelope.
Above all of it sit the equipment manufacturer's published blower and component data, which give you the rated external static and the component pressure drops the whole budget starts from, and the project specification, which can be stricter than any rule of thumb. Cite the body that owns the point, use the manufacturer's data for the numbers that drive the budget, and confirm the current edition of each manual and standard, because they revise on their own cycles.
Units, terms, and conversions
Duct design carries a few names and units, so the same quantity reads differently across a Manual D worksheet, a manufacturer sheet, and a metric drawing.
Static pressure and friction rate are in inches of water column, written in. wg, in. w.c., or in. H2O, all the same thing, where 1 in. wg is about 249 pascals. Friction rate is that per 100 ft of duct. Airflow is CFM, cubic feet per minute, in the field, and liters per second or cubic meters per hour in metric sources, where 1 CFM is about 0.472 liters per second. Velocity is feet per minute, often written FPM, or meters per second in metric. Duct size is a diameter in inches for round and two dimensions in inches for rectangular, with the equivalent diameter tying the two together.
- Manual D
- ACCA's residential and light-commercial duct design procedure, built on available static pressure and the friction rate
- ASP
- Available static pressure, the rated external static left for the duct after the components are subtracted, in in. wg
- TEL
- Total effective length, the longest supply plus longest return run with fittings counted as equivalent length, in feet
- Friction rate
- The design pressure drop per 100 ft of duct, ASP times 100 divided by TEL, in in. wg per 100 ft
- Equal friction
- The residential sizing method that holds the same friction rate across every run, dropping velocity as the trunk reduces
- Ductulator
- A slide-wheel or app friction chart that converts CFM and friction rate into a duct size and velocity
- Equivalent length
- The feet of straight duct a fitting costs in static, read from the Manual D fitting tables
FAQ
What is a friction rate in duct design?
A friction rate is the pressure drop per 100 ft of duct that a Manual D design allows every run to spend, in inches of water column per 100 ft. It equals the available static pressure times 100 divided by the total effective length, and a workable value usually lands between about 0.06 and 0.18.
How do you size a duct with Manual D?
Calculate the friction rate from available static pressure and total effective length, then enter a friction chart or ductulator with each run's CFM and that friction rate to read the duct size and velocity. Size the trunk for full flow, reduce it as air drops off, and check the velocity stays in range.
What is available static pressure?
Available static pressure is the static left for the duct after the components take their share. Start from the equipment's rated external static off the blower table, subtract the filter, coil, registers, grilles, and accessory losses, and what remains is the available static pressure the trunks and branches are designed against, in inches of water column.
What is total effective length in Manual D?
Total effective length is the longest supply run plus the longest return run, with every fitting counted as its equivalent length of straight duct. It is the critical path the system is sized around, not the sum of all the duct, and fittings often outweigh the straight footage on a run.
Why is my return undersized?
Returns get value-engineered down to one grille while the supply gets the design attention, but the return has to pull back everything the supply pushes out. Size the return path for the full system CFM at the friction rate, the same as the supply trunk, and give closed-door rooms a transfer path back.
What friction rate should I use for flex duct?
Design flex to a lower friction rate than metal, often around 0.05 in. wg per 100 ft against 0.10 for sheet metal, because flex left slack or compressed runs several times the stretched pressure drop. Keep flex stretched and supported and run trunks in metal, or do a full Manual D with flex's real roughness.
Equal friction vs static regain: which method do I use?
Use equal friction for residential and light-commercial duct, the method Manual D is built on, because it is simple and drops velocity as the trunk reduces. Use static regain on large, high-velocity commercial systems with long runs, where it holds a near-constant static at every branch. Constant velocity is for industrial exhaust.
How much velocity is too much in a residential duct?
Common practice keeps supply trunks roughly 700 to 900 ft/min, branches around 600, and returns at 600 or lower near the living space. These are common design ceilings, not code limits, and the equipment sound data and project noise criteria control. Past them, registers and return grilles start to whistle.
Why does my duct design pass on paper but starve the far room?
Usually fittings left out of the total effective length, flex installed slack, an undersized return, or duct leakage the design assumed away. Count every fitting in the TEL, design flex to a lower friction rate, size the return to full flow, and seal the duct to a SMACNA class, then verify airflow against design after install.
Do I run Manual J or Manual D first?
Manual J first, always. The load calculation sets each room's airflow and the whole-house total, Manual S selects equipment to that load and gives you the blower data, and only then does Manual D size the duct. You cannot size duct against a static budget and an airflow the load and equipment have not established yet.
People also ask
Codes cited in this guide
This guide is written and reviewed against the published standards below. Always confirm the current adopted edition with the authority having jurisdiction.