A clogged fan filter announces itself early: the pressure across the media rises and the fan load changes, weeks or months before the enclosure temperature moves. You can catch it with a differential pressure switch set around 0.05-0.1 in WC, a fan-current or airflow check, or a simple run-hour based replacement schedule. Set the threshold with the fan running and the filter clean, and the alarm arrives before the heat does.
Why temperature alone misses it
Enclosure temperature is the sum of three things: the heat the equipment makes, the temperature outside the box, and how much air moves through it. A blocked filter only changes the third term. On a cool day or at light machine load, a fully loaded filter can hide completely. Then a warm week or a heavier production schedule pushes the same box over its limit.
That is why a high-temperature alarm is not filter protection. It is a report that the protection has already failed. Filter monitoring measures the restriction itself, so it is independent of weather and load. Note that a vented, filtered enclosure trades rating for airflow; if you are weighing that trade, see putting a fan on a NEMA 4X enclosure: what it costs you and enclosure vent or sealed box: which stays dry outdoors.
Four ways to detect a loading filter
Differential pressure switch
Measure the pressure drop across the filter: one sensing port outside the enclosure, one between the fan and the filter. As media loads, the drop rises. Industrial switches exist for exactly this range; furnace and air-handler makers use the same technique for the same problem. The worry that the signal is too small is partly fair: the drop across a clean enclosure filter is small, so spec a switch with an adjustable setpoint down to roughly 0.05 in WC, and confirm the range with a manometer before you rely on it. It gives a dry contact your PLC or alarm relay can use, and it needs no calibration afterwards. This is the first choice for a critical box.
Fan current or power
A fan pushing against a blocked intake moves less air, and most axial fans draw less current as a result, not more. So a plain overcurrent trip usually never fires. What works is a monitoring relay that alarms when current falls outside a window, or true power for AC fans. The catch: the signature depends on the specific fan and filter, so you must record the clean and blocked values yourself and set the window between them. It also catches a seized or broken fan, which the pressure switch does not. Reasonable as a second signal on an existing cabinet; weak as the only one.
Airflow flap or vane
A light flap in the airstream, held open by airflow and detected by a proximity switch, gives a direct flow/no-flow signal. It is simple and cheap, but it only distinguishes gross blockage, not gradual loading, and the flap is one more moving part in a dusty airstream. Better suited to proving that a fan runs than to warning that a filter is half full.
Run-hour timer and scheduled replacement
If the fan runs on a predictable schedule, filter life is roughly proportional to run hours. A run-hour meter, or the fan run-time counter already in your PLC, turns the filter into a maintenance item: replace at a set number of hours, chosen from how fast filters actually load at that site. It costs almost nothing, needs no sensors, and reliably beats waiting for a temperature trip. What it cannot do is react to an unusually dirty spell. For many non-critical boxes, this is all the monitoring you need.
Which method for which box
| Method | Cost | Alarm type | Catches slow loading | Catches fan failure | Best for |
|---|---|---|---|---|---|
| Differential pressure switch | Low | Dry contact, settable | Yes | No | Critical enclosures, first choice |
| Fan current / power window | Low | Contact or analog | Yes, after proof test | Yes | Boxes that already have a fan circuit to monitor |
| Airflow flap | Lowest | Flow / no flow | Only when nearly blocked | Yes | Proving the fan runs |
| Run-hour replacement | None | PM schedule | Yes, on average | No | Predictable duty, non-critical boxes |
What you need
- A differential pressure switch with an adjustable setpoint down to about 0.05 in WC, or a manometer plus a switch
- Two lengths of sensing tubing: one terminating outside the enclosure, one terminating between fan and filter
- A manometer or low-range gauge for the initial readings
- A spare set of clean filters for the proof test
- An input to land the contact on: a PLC digital input, an alarm relay, or a monitoring relay
If the cabinet has a fan because cooling, not filtration, was the real problem, check the sizing first with the enclosure heat calculation guide and the comparison in enclosure cooling: fan vs air conditioner vs heat exchanger.
Steps to set up filter monitoring
- Measure the clean baseline. Install a new filter, run the fan at normal speed, and read the pressure drop across it with the manometer. Write the number down; it is your reference for every check afterwards.
- Block the intake partially, for example with a sheet of cardboard over half the filter, and read the drop again. The difference between clean and blocked tells you whether a switch in the 0.05-0.1 in WC range will separate the two states. If the blocked reading is barely above clean, the filter area is oversized for the airflow and you should pick the current method instead.
- Mount the pressure switch with one port sensing outside air and the other sensing the plenum between fan and filter. Keep the tubing short, sloped so condensation drains out, and away from direct hose spray.
- Set the trip point between the two readings, closer to the clean value. A common working rule is to alarm when the drop reaches roughly twice the clean-filter baseline, provided that lands above about 0.05 in WC. The goal is warning at 50-70 percent loading, not at blockage.
- Land the contact on your PLC or alarm relay as a normally-open alarm, and prove it: with the fan off, no alarm; with the fan running and the filter clean, no alarm; with your test blockage in place, alarm.
- If you chose current monitoring instead, record clean-filter current, then blocked current, then set an undercurrent or window alarm between them with the real filter in place. Repeat the check at the next filter change, because fan bearings age too.
- Record the baseline and trip values on the cabinet drawing or in the maintenance system, with the date, so the next technician can tell drift in the filter from drift in the sensor.
Checks when done
- Alarm clears when you fit a fresh filter and returns when you block the intake
- No nuisance alarm during fan start-up or at the lowest expected ambient
- Tubing is secure, sloped and clear of washdown spray
- The alarm appears in the same system as the temperature alarm, with a different message
- The clean baseline is written where the maintenance team will find it
Mistakes that cost you the warning
- Setting the trip point near the blocked reading instead of near the clean one; the alarm then fires with the box already hot
- Mounting both sensing ports inside the plenum; the switch then reads fan static, not filter loading
- Trusting an overcurrent relay on an axial fan, which draws less air and less current as the filter loads
- Forgetting that the signal only exists while the fan runs; on duty-cycled fans, gate the alarm with a fan-run signal or delay it a few seconds after start
- Letting condensation collect in a horizontal sensing tube until it blocks the port and reads as a permanent clog
- Servicing the filter without re-checking the baseline, so a wrong switch setting hides a real loading trend
How to stop it coming back
Turn the alarm into a schedule. Log the pressure reading at every filter change; after two or three changes you will know the loading rate for that site, and you can set run-hours or calendar interval so the alarm becomes a backstop rather than the first notice. Keep filters stocked for the site's real interval, not a generic one.
If loading is consistently fast, the fix may be upstream: pre-separating dust before the filter, doubling filter area to cut the clean pressure drop, or closing the box and cooling it without outside air. A sealed enclosure with a heat exchanger or air conditioner needs no filter monitoring at all, and the layout advice in component layout inside a sealed enclosure to manage heat applies once the air stays inside. Airflow itself is covered further in cooling louvers and enclosure ratings: what vents really cost; the same duct and filter pressure questions come up in how compressor room filters affect cooling airflow.
When to re-specify the enclosure
If filters clog in weeks, if the alarm trips every washdown cycle, or if the enclosure lives in salt air or chemical mist where outside air was the wrong answer anyway, stop managing the filter and remove it. We build sealed 304 and 316L stainless enclosures to your size and cutouts, rated NEMA 4X / IP66 by the manufacturer; not UL listed, delivered in 10 business days from drawing approval. Compare vented against sealed with the custom enclosure options, or shortlist by environment in the enclosure selector. For a sealed box, the monitoring you need is temperature and humidity, not filter pressure; start from the enclosure heat calculation guide.
Enclosures for this job
Catalog sizes that suit the job this article answers, picked from its subject; open one for its full specifications.
| Enclosure | Material | Outside H x W x D / display | Rating |
|---|---|---|---|
| 16 × 12 × 8 in | 304 stainless | 16 x 12 x 8 in | Rated NEMA 4X / IP66 by the manufacturer; not UL listed |
| 24 × 20 × 8 in | 304 stainless | 24 x 20 x 8 in | Rated NEMA 4X / IP66 by the manufacturer; not UL listed |
| 30 × 24 × 10 in | 304 stainless | 30 x 24 x 10 in | Rated NEMA 4X / IP66 by the manufacturer; not UL listed |
Questions
Is the pressure drop across an enclosure filter really measurable?
Yes, but it is small. A clean filter media on a typical enclosure fan drops well under 0.1 in WC at normal airflow, and the alarm point is usually set around 0.05-0.1 in WC. Switches with that sensitivity exist because furnace and air-handler makers use the same range for the same job. Measure the actual clean-filter drop with a manometer before you set anything.
Can I just monitor enclosure temperature instead?
Temperature tells you the problem has already arrived. A cool week or a lightly loaded drive can mask a fully blocked filter, and the heat problem appears suddenly when ambient rises or load increases. Filter monitoring gives you warning independent of weather and machine loading, so you service the filter on your schedule instead of after a trip.
Does fan current monitoring work for AC and DC fans?
It works, but it needs proof-testing. An axial fan on a clogged intake moves less air and its motor load usually falls, not rises, so a simple overcurrent relay set high will never trip. Many monitoring relays can alarm on undercurrent or on current outside a window. Record the clean-filter current and the blocked current with the real filter in place, then set the window between them.
How often should enclosure fan filters be changed?
Set the interval from the site, not from a table: dusty shops and paper mills may need monthly service, a clean indoor room may need it twice a year. A run-hour timer or filter alarm replaces guessing. Whatever you choose, log the clean-filter pressure or current at each service so you can see how fast the filter loads at that location.
