How the two differ in kind
Two ducts serving the same load can be sized by two different rules and come out different sizes. Neither is wrong; they hold different quantities constant, and understanding which one tells you what the resulting system will do.
The VELOCITY method fixes the air speed in each section — often a table of recommended velocities by duct type and by the noise criterion of the space served — and sizes the duct so that the design flow moves at that speed. Pressure loss per metre then falls out of the geometry and varies from section to section.
The EQUAL FRICTION method fixes the pressure loss per unit length instead, commonly expressed in pascals per metre or inches of water per hundred feet, and sizes every section to that same friction rate. Velocity then falls out, and it falls out in a useful way: as air is delivered into branches, the remaining flow is smaller, the duct at the same friction rate is smaller, and the velocity DROPS along the run. The result is a naturally tapering trunk with velocity highest near the fan and lowest at the far end — which is also where the quiet rooms usually are.
That behaviour is why equal friction is the default for ordinary low-velocity systems. It is simple, it produces a sensible taper, and on a broadly symmetrical layout the branches end up with roughly comparable resistance, so the system is close to balanced before anybody touches a damper. What neither method does is balance an ASYMMETRICAL system. Air takes the path of least resistance, so a short branch near the fan receives far more than its share while the longest branch is starved — and no amount of correct sizing by either rule changes that, which is what balancing dampers and, on larger systems, static regain sizing exist to address.
The factors that actually differ
Show
| Velocity method | Equal friction method | |
|---|---|---|
| What is held constant | Air velocity in each section. | Pressure loss per unit length, the same in every section. |
| What varies | Friction rate, section by section — so total pressure loss is not predictable from the layout alone. | Velocity, which tapers down along the run as air is delivered to branches. |
| Noise control | Direct, since velocity is the variable most closely tied to regenerated noise and you are setting it. | Indirect but often favourable, because velocity is naturally highest near the fan and lowest at the quiet end. |
| Ease of use | Straightforward but section-by-section, with a judgement about the right velocity for each. | Simplest of all — one friction rate, then read every size off a chart or calculator. |
| Balance on a symmetrical layout | Not inherently balanced; branches of equal length happen to match, others do not. | Reasonably good, because equal length of equal-friction duct means equal resistance. |
| Balance on an asymmetrical layout | Poor, and it needs dampers. | Also poor, and for the same reason. Neither method fixes a short branch stealing air from a long one. |
| Where it is used | Sections where velocity must be controlled for a specific reason — near a sensitive room, in exhaust carrying particulate that must stay entrained. | The default for ordinary low-velocity commercial and residential systems. |
| Energy | Depends entirely on the velocities chosen; low velocities mean big ducts and a low-energy system. | Depends on the friction rate chosen; a lower rate means bigger ducts, a bigger first cost and lower fan energy for the system's life. |
| Fittings | Counted separately, as equivalent lengths or loss coefficients, in both methods. | The same — and fittings are frequently the majority of a real system's loss, which is why sizing the straight duct alone is not a design. |
| What comes after either | A pressure loss calculation along the index run, then fan selection, then balancing. | Identical. Sizing is the first step, not the design. |
Which one, and when
Choose velocity method when…
- A specific section must not exceed a velocity — a duct crossing a quiet room, a terminal connection.
- The system carries something that must stay entrained: dust, chips, fumes in an industrial exhaust.
- The duct is a fixed size already and you are checking what it can carry.
- A noise criterion is driving the design and velocity is the variable it responds to.
Choose equal friction method when…
- An ordinary low-velocity supply or extract system — this is the normal default.
- The layout is reasonably symmetrical, where the method's self-balancing tendency is worth something.
- Speed of design matters and a single friction rate across the system is adequate.
- As the starting point before checking velocities in the sections where noise matters, which is the usual practice.
Now run your own numbers
This page holds no prices on purpose — a national average is wrong for almost every real project. Quantify both options with your dimensions and your local quotes.
Frequently asked questions
- Why does equal friction make velocity taper along the run?
- Because at a fixed friction rate, a smaller duct carries less air at a lower speed. As the trunk delivers air into branch after branch, the flow remaining downstream falls; sizing each new section to the same pressure loss per metre therefore gives a smaller duct, and the smaller duct carrying the reduced flow does so more slowly. The result is a trunk that steps down in size and in velocity as it runs away from the fan. That is a genuinely useful accident: velocity, and therefore regenerated noise, is highest near the plant where nobody is sitting and lowest at the end of the run where the quiet rooms usually are. Sizing by velocity instead would hold the speed up all the way to the last branch, which is why velocity sizing is normally applied to particular sections rather than to a whole system.
- Which friction rate should I use?
- One chosen deliberately, because it is a first-cost against running-cost decision rather than a standard value. A low friction rate produces large ducts: more sheet metal, more insulation, more space in the ceiling void, a higher first cost — and a lower pressure loss, so a smaller fan, less energy every hour the system runs, and less noise. A high rate does the opposite. Typical values in general use sit in a band that has more to do with historical convention and available ceiling space than with any optimum, and where energy performance matters the rate should be pushed down rather than taken from a table. The check afterwards is velocity: a friction rate that produces acceptable sizes on the trunk can produce unacceptably high velocities in small branches, which is where the noise complaints come from.
- Does either method balance the system?
- Neither does, on any layout that is not symmetrical, and expecting otherwise is the commonest disappointment with equal friction. Air divides between branches according to their resistance, so a short branch close to the fan presents far less resistance than a long one at the end, and it takes correspondingly more air — whatever rule was used to size the ducts. Equal friction is close to balanced only when the branches are of similar length and similar fitting count, because then their resistances are genuinely similar. Real systems are not, which is why balancing dampers are fitted at branches and why commissioning means measuring flows and adjusting until they match the design. Sizing is what makes balancing possible with a reasonable damper setting rather than a nearly closed one; it does not replace it.
- What is static regain and when is it used?
- A third method, used mainly on larger and higher-velocity systems, which sizes each section so that the increase in static pressure from the reduction in velocity at a branch take-off offsets the friction loss in the next section — so the static pressure at every branch is approximately the same. The consequence is what makes it attractive: branches see comparable pressure whatever their position, so a long system is much closer to balanced without heavy damper throttling, and the fan does not have to be sized for the worst branch while everything else is strangled back. The cost is that it produces larger ducts toward the end of the run rather than smaller ones, which uses more material and more ceiling space, and the calculation is more involved. On a short residential system it is not worth the complexity.
- How much do fittings matter?
- Usually more than the straight duct, which is why a sizing calculation alone is not a design. A bend, a tee, a transition, a damper or a terminal connection each impose a loss that is frequently equivalent to many metres of straight duct, and a compact system with a lot of fittings and short runs is dominated by them. They are accounted for either as an equivalent length — the length of straight duct that would cause the same loss, added to the run before the friction calculation — or as a loss coefficient applied to the velocity pressure. Either way, the quality of the fitting matters enormously: a long-radius bend with turning vanes costs a fraction of what a square mitred elbow costs, and a well-formed tee costs far less than a bullhead connection.
- What velocities are acceptable?
- It depends on what the duct is passing through and what the space needs, which is why published tables are given by application and by noise criterion rather than as a single figure. The pattern is consistent: main plant-room ducts tolerate the highest velocities, trunk ducts in ceiling voids less, branches less again, and the final connection to a terminal least of all — because that is closest to the occupant and the terminal itself is a noise generator. Exhaust systems carrying particulate invert one of these constraints: they have a MINIMUM velocity below which the material drops out and settles in the duct, so they are sized for transport rather than for quiet. The practical approach for ordinary systems is to size by equal friction and then check velocities against the table, resizing the sections that exceed it.
- Round, rectangular or flat oval?
- Round wherever it fits, on every measure that matters, and rectangular where the space does not allow it. A round duct has the least surface area for a given cross-section, so less friction, less material and less heat transfer; it is stiffer, so it needs less reinforcement and leaks less at joints; and it is quieter, because a flat panel is a better radiator than a curved one. Rectangular duct exists because ceiling voids are shallow and a round duct of the required area often will not fit, and it is sized by the equivalent diameter that gives the same pressure loss — which is not the same as the equal-area diameter. The penalty grows with the aspect ratio, so a very flat rectangular duct performs considerably worse than its area suggests, which is what flat oval exists to improve on.
- What comes after sizing?
- The pressure loss calculation along the index run, which is the path with the highest total resistance from fan to terminal — that figure, plus the terminal, the filters, the coils and any attenuator, is what the fan has to deliver. Sizing produces duct dimensions; it does not produce a fan selection, and selecting a fan from the friction rate alone underestimates the system badly because it omits every fitting and component. After that comes the noise calculation per frequency band, then fan selection against its curve at the design point, then installation, then commissioning and balancing — measuring the flows actually achieved and adjusting them. A system that was sized correctly and never balanced delivers the design flow to some rooms and not others, and nobody finds out which until somebody complains.