A fan meets its curve on the test bench — then falls short on site. The system curve is why, and how to spec around it.
Reviewed by Jitamitra application engineering
Share on LinkedInEvery underperforming fan generates the same phone call: “your fan is undersized.” Sometimes it is. Far more often, the fan is running exactly on its published curve — and the system wrapped around it is no longer the system anyone designed. The fan never chose its operating point. The system chose it.
A fan has one curve: pressure it can produce against the flow it moves, at a given speed and gas density. A duct system has another: the resistance it demands, which rises with the square of the flow through it. Double the flow and the system asks for roughly four times the pressure to push it. The fan can only ever run at the single point where its own curve crosses that system curve. Move the system curve and the crossing point moves with it — the flow and static you actually get are decided at that intersection, not on the datasheet.
This is why a fan system curve matters as much as the fan. The bench test proves the fan makes its pressure and flow at the test point. What it cannot prove is where your ductwork will make it operate. If the resistance the fan sees in service differs from the resistance assumed at selection, the fan lands somewhere else on its curve — usually low on flow, and it looks “undersized” when it is nothing of the kind. (The square-law behaviour behind all of this is worked through in the fan laws and the cube law.)
Design static is a calculation made before anything is built. Field static is what the finished, populated, aging installation actually presents. They diverge for ordinary, physical reasons:
A fan sized to a clean filter and a pristine duct is sized to a system state that exists for about a week. Treat the filter’s dirty-condition pressure drop and the expected duct build-up as part of the required static — because in service, that is exactly what they are. This matters most on high-resistance duties: a boiler ID/FD fan pulling through economiser, bag filter and a long flue path can lose a large fraction of its margin to fouling alone, and a process blower feeding a cartridge collector behaves the same way.
When a fan underperforms, one afternoon of measurement settles whether it is the fan or the system:
The first-hour version of this check, done at handover before anything is blamed, is laid out in the commissioning checklist for the first hour.
The fix is upstream, at enquiry. Give your fan supplier the real duct layout — not the idealised schematic — so the system-effect allowance is added before selection, not discovered after. State the inlet condition honestly. Put the dirty-filter and fouled-duct resistance into the required static, and name the operating range you actually need to hold, not a single clean-day point. A duty defined this way lets the supplier place the fan-curve/system-curve intersection where you want it across the whole service life — the discipline behind specifying the duty point.
Do that, and control becomes a choice rather than a rescue. If the operating point still needs trimming in service, damper, IGV or VFD control moves it deliberately — instead of a dirty filter moving it for you.
Talk to us about your fan’s real operating point →
Jitamitra Electro Engineering · Fan-engineering notes, written for the engineer.
Sources & basis. This note is drawn from Jitamitra's own fan-engineering practice. No customer, vendor or commercial details appear here, and every figure stays within the range we actually build and serve — no invented numbers.
Usually not. The fan never chose its operating point — the system chose it. A fan can only run at the single point where its own curve crosses the system curve. If the resistance in service differs from the resistance assumed at selection, the fan lands somewhere else on its curve, usually low on flow, and looks undersized when it is nothing of the kind.
Four ordinary physical reasons. Filters and screens foul, so resistance climbs week by week. Ducts narrow as deposits build in dust and fume service. As-built rarely equals as-drawn — an extra elbow to clear a beam, a balancing damper left half-shut since commissioning. And an elbow landing hard against the inlet distorts the air before it enters, a penalty no fan curve shows and no bench test reproduces.
The system, not the fan. A fan sized to a clean filter and a pristine duct is sized to a system state that exists for about a week — every day after that moves the operating point back. A boiler ID/FD fan pulling through economiser, bag filter and a long flue path can lose a large fraction of its margin to fouling alone.
One afternoon of measurement settles it. Verify speed with a tachometer — belts slip and VFD settings drift. Measure the pressure drop in sections, intake to exhaust, and find which section has grown beyond design. Then plot the measured point against the fan curve at the measured speed and density. On the curve means the fan is fine and the system moved; off it, we can talk about the fan.
The real duct layout, not the idealised schematic, so the system-effect allowance is added before selection rather than discovered after. State the inlet condition honestly. Put the dirty-filter and fouled-duct resistance into the required static. And name the operating range you actually need to hold, not a single clean-day point. A duty defined that way lets us place the intersection where you want it across the service life.
No. This note is drawn from Jitamitra's own fan-engineering practice, not from any one job. No customer, vendor or commercial details appear here, and every figure stays within the range we actually build and serve — no invented numbers. What it gives you is the method, which you then run against your own measured system.
"Add the system resistance" is easy to say. Here is a real run added up, so you can see which term actually decides the fan.
Take our earlier hood at 1,287 m³/h through a 160 mm duct at 17.5 m/s: 20 m of straight, two long-radius bends, a filter, discharge to atmosphere. Velocity pressure at that speed is 18.7 mmWC.
The filter is 87 per cent of it. Every metre of duct and both bends together are under 15 per cent. That is the point of totalling a run rather than arguing about it: the fan is decided by the filter's dirty condition, and a selection made on clean filters is a fan that works for a fortnight.
Two fans at 2 × 50% is not a promise of 50% when one trips. System resistance rises with the square of flow, so the surviving fan does not sit at half duty — it rides back up its own curve to a higher pressure and a lower resistance point, and typically delivers 60–70 per cent of the two-fan flow, not 50.
That is usually good news and occasionally the opposite: the survivor is doing more work than the paper says, so check its motor is rated for the single-fan operating point, not just the shared one. A 2 × 50% set whose motors are sized only for the paired duty will trip the survivor on the day you need it.
Flow, static, gas temperature, application — or attach a spec, GA drawing or a multi-fan schedule. Engineer to engineer.
ISO 9001:2015 quality system · performance-tested to IS 4894 / ISO 5801 / AMCA 210 method · witnessed FAT on request.
*For our standard range. ATEX and special projects need 5 to 7 working days.
This site sets no cookies of its own. Google Analytics loads only if you accept. It sets two cookies, so we can see which fan topics people actually read. Decline and no analytics loads — the site works exactly the same. What we collect →
