How the two differ in kind
Air moving along a duct loses pressure to friction, and friction depends on the surface it is moving along. A rigid duct's interior is smooth sheet metal, and its published resistance is what it delivers for its whole life. A flexible duct's interior is a spiral wire with a liner stretched over it, and its published resistance is what it delivers WHEN PULLED TAUT.
That condition is the whole subject. Flex installed slack — sagging between supports, or a two-metre run cut from three metres and left compressed — has an interior that is no longer a smooth bore but a series of ridges, and the measured resistance rises steeply. At around thirty per cent compression the loss can be about four times the rated figure, and at that point the run is doing more to the system than any design decision made about it.
The trap is that nothing reveals it. The duct is the right diameter, the run is the right length, the calculation returns a sensible number, and the diffuser at the end delivers a fraction of its design airflow. The symptom reads as an undersized fan or a bad design, and it is neither: it is an installation variable that swamps every design variable on the same run.
The factors that actually differ
Show
| Rigid sheet metal duct | Flexible duct | |
|---|---|---|
| Resistance as installed | As published, for its life. Smooth bore, and it does not change. | As published only if taut. Compressed, several times higher, invisibly. |
| Speed of installation | Slow. Fittings are fabricated, joints made and sealed, supports engineered. | Fast, and forgiving of a route that is not quite where the drawing put it. |
| Fittings and direction changes | Every bend is a fabricated fitting with a known loss coefficient — and a long-radius bend is dramatically better than a tight one. | Bends are made by bending it, which is the convenience and also the problem: a tight radius in flex is a large and unquantified loss. |
| Leakage | At joints, controlled by sealing to a leakage class and verifiable by a pressure test. | At the collar connections, where the inner liner and the outer jacket must be banded SEPARATELY. Banding only the jacket is a common and invisible leak. |
| Noise | Transmits fan noise along its length and can drum. Lined sections or a silencer are the remedy. | Absorptive — the liner and insulation attenuate, which is a genuine reason to use a short flex run just before a diffuser. |
| Support | Hung on a designed spacing; it holds its own shape between supports. | Needs frequent support with a wide strap that does not pinch. Under-supported flex sags, and a sag is a compression. |
| Cleaning and life | Cleanable, inspectable, and lasts as long as the building. | The corrugations trap dust and the liner can tear. Generally replaced rather than cleaned, and it has a service life measured in years rather than decades. |
| Insulation | Wrapped separately, which is a second operation and a separate quantity. | Comes insulated inside its own jacket, which is part of why it is quick. |
Which one, and when
Choose rigid sheet metal duct when…
- The run is long, or the system is a trunk carrying most of the air.
- The building is commercial, where leakage class, cleanability and life expectancy are specified rather than assumed.
- The duct is exposed or accessible, where appearance and durability matter.
- Static pressure is tight and there is no margin to lose to an installation variable.
Choose flexible duct when…
- The run is short — the last stretch to a diffuser, where its absorptive properties are a bonus.
- The route is awkward and rigid fittings would mean fabricating one-offs.
- There is genuine misalignment to absorb between a rigid branch and a fitting position.
- Speed and cost dominate on a domestic installation, and somebody will actually pull it taut.
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
- How much does compression really cost?
- More than almost any design decision on the same run. Published flex duct resistance assumes the duct is fully extended, and measured data shows loss rising steeply with compression: at around thirty per cent compression the resistance can be about four times the rated value, and the effect is worse in smaller diameters. The practical consequence is that a run cut generously and pushed into place — which is the natural way to install it — can throttle a branch that the calculation said was fine. The fix costs nothing: cut flex to length rather than to convenience, pull it taut between supports, and take the slack out before banding the second end. It is the highest-return five minutes on a ductwork installation.
- Can I use flex for the whole system?
- On a small domestic system it is done constantly and it is where the complaints come from. The problems compound: every run has the compression risk, every bend is unquantified, the total system resistance ends up far above the design, and the fan rides up its curve delivering less air at a higher pressure. On anything with a trunk carrying most of the air, the trunk should be rigid — it is the section where resistance costs the most and where a compression penalty is applied to every branch downstream of it. The standard arrangement exists for a reason: rigid trunk and branches, short taut flex runouts of a metre or two to the diffusers.
- How is flex supposed to be connected?
- As two separate connections, and this is the detail most often half-done. The inner liner is pulled over the collar, past the bead, and banded tight — that is the air seal and it must be mechanically fastened, not taped alone. The insulation and outer jacket are then pulled over and banded separately, which is the vapour seal. Banding only the jacket leaves the liner loose inside it, so conditioned air leaks into the insulation and out of the joint, and nothing about the installation looks wrong. Mastic on the collar under the liner is normal practice. The whole connection takes a couple of minutes and is invisible once the ceiling is up.
- Which is quieter?
- Flex, substantially, and it is a real design tool rather than a consolation. Its liner and insulation absorb sound, so a short flex run between a rigid branch and a diffuser attenuates fan noise and airborne crosstalk between rooms served by the same branch. Rigid duct does the opposite: it conducts fan noise along its length and thin sheet can drum with pressure pulsations. Where a rigid system needs the same benefit it gets an acoustic lining or a silencer, and a silencer trades pressure drop and eventually self-noise for the attenuation — which is the same bargain the flex makes, made deliberately and specified.
- Can flexible duct be cleaned?
- Poorly, and that is usually the reason it is replaced instead. The corrugated interior traps dust in the valleys between the wire turns, and mechanical cleaning risks tearing the liner — at which point the duct leaks into its own insulation and has to come out anyway. Rigid duct is cleanable with brushes and vacuum, inspectable with a camera, and it does not have a wear-out mode. Where indoor air quality is a specified requirement — a healthcare or laboratory building — that difference generally rules flex out of anything but the final connection, and sometimes out of that too.
- How far apart should flex be supported?
- Closer than instinct suggests, with wide straps rather than wire or tie-wraps, and with the duct pulled taut between them. Manufacturer guidance typically puts supports well under two metres apart and limits the permitted sag between them to a small fraction of the diameter — because a sag is a compression, and a compression is the resistance penalty this whole page is about. The strap has to be wide enough not to pinch the duct at the support, since a pinch is a local restriction at every hanger. And flex should never be used to bridge a gap it has to hang across: if the route needs a span, the span is rigid duct with the flex connecting at the end of it.