Filter-bank differential-pressure records before coil freeze-up review

Direct answer

Before a coil freeze-up review, record the equipment ID, area served, filter-bank ID, filter stage, filter count, size, type, MERV when known, orientation, seals, bypass gaps, visible loading, wet or damaged filters, upstream and downstream pressure points, gauge or sensor identity, zero/calibration status where required, fan command, fan speed or VFD status, airflow setpoint or operating mode, damper/economizer position if relevant, coil status, filter-bank differential pressure, BAS trend or alarm history, clean-filter or changeout basis, replacement action, after reading, photos, open exceptions, and release or hold decision.

The record belongs before filters are thrown away, alarms are cleared, panels are closed, ice melts, or the crew writes that a dirty filter caused the freeze-up. Filter DP is useful evidence. It can show restriction across a filter bank and it can support a maintenance decision. It does not, by itself, prove the airflow through the coil, the refrigerant condition, the control sequence, the economizer condition, the fan performance, or the sole cause of freezing.

Use this field note as documentation guidance only. Manufacturer instructions, project specifications, TAB requirements, commissioning procedures, BAS standards, code requirements, refrigerant rules, qualified technician judgment, and site safety procedures control the actual diagnosis, service, repair, reset, and release.

Do not turn a dirty-filter alarm into a cause statement

A dirty-filter switch, BAS alarm, or high differential-pressure reading is a starting point. It is not the whole freeze-up review. The same alarm may mean loaded filters, a changed airflow command, wrong alarm threshold, tubing issue, sensor drift, closed damper, different operating mode, or a filter bank that was never compared to the correct final-resistance basis.

The record should separate four things: what was observed, what was measured, how the reading was interpreted, and what action was released. That separation matters when a frozen coil affects downtime, warranty, tenant complaints, water damage, or a maintenance dispute.

A useful review lets the next person answer basic questions without guessing. Were the filters actually installed? Did the taps span the filter bank or another restriction? Was the fan at design airflow, minimum speed, night setback, smoke mode, purge mode, or manual override? Did the filter DP exceed a manufacturer or project limit, or only a generic alarm value?

Start with the exact filter bank

Identify the unit first: AHU, RTU, MAU, FCU, fan coil, CRAC, heat pump, or other equipment tag. Record the area served, system side, filter-bank label, prefilter or final-filter stage, rack arrangement, access side, and whether the bank is part of a return, mixed-air, outside-air, or recirculation path.

Then record the filters. Capture size, depth, quantity, type, model or part number where visible, MERV when known, airflow arrow orientation, gasket or header condition, clips or retainers, filter frame condition, missing filters, collapsed media, loaded corners, wet filters, damaged filters, and visible bypass. A high reading is easier to interpret when the bank is complete and sealed. A low reading can also matter if a filter is missing, bypassing, or ruptured.

Service history belongs in the packet. Record last replacement date if known, work order number, spare filter source, whether all stages were changed, and whether the removed filters were saved or photographed. If the crew replaces filters before documenting the condition, the review loses one of its clearest pieces of evidence.

Measurement points are part of the reading

Filter-bank DP should show the pressure difference across the filter bank, not a mystery number from somewhere in the cabinet. Record the upstream and downstream locations, probe or tap type, tubing path, high and low port connections, and whether the points span the intended filter stage.

Record the instrument too. A wall gauge, handheld manometer, BAS transducer, pressure switch, or temporary test setup should be named. Note zero status, range, units, calibration or inspection date where the procedure requires it, tubing condition, condensation in tubing, plugged tips, loose fittings, and whether the reading was steady or fluctuating.

Panel and access-door status can change readings. State whether doors were installed, gaskets sealed, access panels latched, belts guarded, and normal operating covers in place. Do not use this checklist as permission to drill test holes, bypass safeties, reach into moving equipment, or open energized or manufacturer-restricted compartments.

Record the operating mode with the number

Differential pressure changes with airflow. A filter-bank reading without fan and airflow context can be misleading, especially on VFD fans, ECM blowers, VAV systems, staged equipment, economizer cycles, smoke-control modes, and BAS overrides.

Record fan command, fan speed, VFD hertz or percent when available, airflow setpoint, occupied or unoccupied mode, heating or cooling call, compressor or chilled-water status, coil valve position if relevant, outside-air and return-air damper position, relief or exhaust status, and whether the system was in normal automatic operation or manual test.

After a filter change, repeat the reading in the same mode where possible. A before reading at full airflow and an after reading at minimum speed may look like a dramatic improvement even when the filter was only part of the story.

Compare to the right changeout basis

A new filter has an initial resistance. A loaded filter has a higher resistance. A filter manufacturer or project standard may publish or specify a rated final resistance, final pressure drop, or changeout limit at a stated airflow. Those values are not universal for every rack, fan, building, filter depth, face velocity, or operating mode.

The packet should name the basis used for the decision. That may be filter manufacturer literature, equipment manufacturer limits, the design or TAB report, commissioning record, owner maintenance standard, BAS alarm schedule, infection-control requirement, cleanroom requirement, or project specification.

Do not write "filters exceeded limit" unless the limit is named. Write the actual DP reading, the operating condition, the changeout basis, and whether the reading was above, below, or not comparable to that basis. Calendar replacement can still be part of maintenance, but a freeze-up review needs more than "filters were due."

Keep filter DP separate from other pressure readings

Filter-bank DP is a component pressure drop. Total external static pressure, return static pressure, supply static pressure, coil pressure drop, duct drop, damper drop, cabinet drop, and filter drop answer different questions.

If the crew is also taking static pressure readings, list them separately. A loaded filter can raise return-side restriction, but a dirty coil, undersized return, closed damper, blocked grille, collapsed flex, failed actuator, wrong blower setting, or measurement error can create similar field symptoms.

This is the reason the article is a record packet. It does not ask the reader to ignore filter DP. It asks the reader not to turn one component reading into a complete airflow diagnosis.

Coil freeze-up review needs more than filter DP

Restricted airflow can contribute to a frozen evaporator coil. Dirty filters, blocked returns, closed dampers, dirty coils, blower problems, low load, controls problems, economizer issues, and refrigerant problems can produce similar or overlapping symptoms. The filter-bank record should preserve evidence, not close the case too early.

Capture freeze-up evidence before it disappears. Photograph frost or ice location, iced suction line, wet cabinet, condensate overflow, drain-pan condition, coil face condition, filter condition, access panel condition, and any BAS alarm screen or trend used in the review. Record whether the system was shut down, thawed, left fan-only, locked out, or held for qualified service under site procedure.

Keep open items visible. If refrigerant checks, coil cleaning, freeze-stat testing, economizer sequence review, fan inspection, belt check, airflow measurement, controls trend review, or TAB review remain open, write that status. A clean after-filter DP reading does not prove the coil is safe to run if another freeze-up cause remains unresolved.

Minimum filter-bank DP packet

Use the service ticket, PM form, TAB form, commissioning checklist, BAS alarm report, or owner maintenance system first. Add this packet where those forms do not connect filter condition, differential pressure, fan mode, coil evidence, and freeze-up release clearly enough.

Before coil freeze-up review checklist

Run this check before filters are discarded, BAS alarms are cleared, or the service note assigns cause.

• Confirm the exact equipment tag, area served, filter-bank ID, filter stage, and rack arrangement.
• Record filter size, depth, type, model, MERV when known, quantity, airflow arrow direction, seals, clips, gaps, and visible bypass.
• Photograph loaded, wet, collapsed, missing, damaged, or poorly sealed filters before replacement.
• Identify upstream and downstream pressure points and verify they span the intended filter bank.
• Record gauge, manometer, transducer, or pressure switch identity, zero status, units, tubing condition, and calibration/inspection status where required.
• Record fan command, fan speed or VFD status, airflow setpoint, operating mode, damper/economizer status, and heating or cooling call.
• Record filter-bank DP, BAS trend or alarm history, time stamps, and whether the reading was steady.
• Compare the reading to the named manufacturer, design, TAB, owner, BAS, or project changeout basis.
• Keep filter DP separate from total external static pressure, coil pressure drop, return static, and supply static readings.
• Photograph coil freeze evidence, coil face condition, drain/pan signs, and relevant BAS alarms before evidence is lost.
• Record replaced filters, after readings in the same operating mode where possible, and any remaining airflow, refrigerant, controls, economizer, fan, or coil review.
• Write the final status: released to operate, released with monitoring, partial release, held, locked out, or recheck required.

Weak and strong service notes

Weak note: coil froze because filters were dirty, changed filters, unit back on.

That note does not show the filter bank, installed condition, measurement points, DP reading, fan mode, changeout basis, coil evidence, BAS trend, after reading, or other possible freeze-up causes. It also turns a maintenance action into a cause statement.

Stronger note: AHU-3 serving west office reported low airflow and frozen evaporator coil at 07:40. Unit held under site procedure before filter replacement. Filter bank FB-3A documented as final-filter stage, twelve 24 by 24 by 2 pleated filters, MERV 13 per filter label. Photos show heavy loading at lower row, two loose retainers, no missing filters, and no visible bypass gap at access door. DP taps verified upstream and downstream of FB-3A; handheld manometer zeroed before reading. System recorded in occupied cooling mode with supply fan at 100 percent command and outside-air damper at normal occupied position. Filter-bank DP, BAS dirty-filter alarm trend, and freeze-stat alarm time attached. Manufacturer filter data and owner PM standard listed as the changeout basis. Filters replaced with same model and orientation; retainers corrected. After reading taken in same fan command and logged. Coil ice photos, drain-pan water, and wet insulation photo attached. Released only for thaw and supervised restart. Refrigerant check, coil-face cleaning review, economizer trend review, and freeze-stat functional review remain open before final cause note.

The stronger note works because it says what the filter evidence proves and what it does not prove. It supports a maintenance action without pretending the whole freeze-up review is complete.

Common mistakes

The first mistake is using one universal dirty-filter pressure value. Filter changeout depends on filter type, airflow, system design, manufacturer data, owner standards, and project requirements.

The second mistake is missing fan mode. A VFD fan at minimum speed and the same fan at design speed can show very different DP across the same filter bank.

The third mistake is treating MERV as a pressure-drop verdict. MERV helps describe particle-capture performance. It does not prove whether the installed filter size, depth, rack, and system airflow are acceptable.

The fourth mistake is replacing filters without before photos or after readings. The removed filter condition and the same-mode after reading are often the evidence needed later.

The fifth mistake is confusing component drops. Filter DP, coil DP, total external static pressure, return static, and supply static should not be blended into one unexplained number.

The sixth mistake is clearing ice, alarms, and trends before the review is documented. Some freeze-up evidence disappears quickly once the system is off or thawing.

Questions that come up

Is high filter DP enough to say dirty filters froze the coil? No. It can support a restriction finding and a filter replacement decision, but coil freeze-up review also needs airflow, coil, fan, refrigerant, controls, damper, economizer, load, and safety context.

What if the filters are new and DP is still high? Record the filter model, MERV, size, depth, face area, rack condition, fan mode, airflow evidence, and measurement points. A new high reading may point to wrong filter selection, undersized filter area, high airflow, blocked downstream component, measurement error, or a system that needs design/TAB review.

Should filters be changed by calendar or by DP? Many maintenance programs use both time and condition. For a freeze-up review, write which basis controlled the decision and preserve the actual DP and operating mode.

Does MERV 13 cause frozen coils? Not as a blanket statement. A higher-efficiency filter can add resistance in some systems, but pressure drop depends on filter construction, depth, face area, loading, airflow, rack fit, and system capability.

Can a BAS trend replace a field reading? Sometimes it may be accepted by the owner or commissioning plan, but the record should still identify the sensor, points, trend range, fan mode, alarm threshold, and whether field verification was performed.

Compliance and safety limits

This field note is not HVAC design, TAB certification, commissioning approval, refrigerant diagnosis, charging instruction, control-sequence approval, freeze-stat testing procedure, BAS programming, electrical instruction, confined-space instruction, ladder or roof-safety plan, or manufacturer service instruction. The manufacturer, engineer, AHJ, TAB provider, commissioning authority, BAS provider, owner standard, qualified technician, and site safety plan control the work.

Do not use this checklist to bypass lockout, electrical safety, moving fan parts, access-panel rules, refrigerant handling rules, hot or cold surface protection, ladder or roof access controls, confined-space procedures, control safeties, condensate overflow protection, mold or water-damage procedures, or manufacturer restrictions on operation after freezing. The packet preserves the evidence. It does not authorize unsafe operation or an unsupported cause statement.

Sources checked

• EPA, What is a MERV rating?Used for MERV as a filter performance comparison rather than a complete airflow or pressure-drop answer.
• ANSI/ASHRAE Standard 52.2-2017Used for resistance, initial resistance, final resistance, rated airflow, and rated final resistance context.
• Building America Solution Center, High-MERV FiltersUsed for pressure drop, initial and final pressure drop, airflow, blower response, and system sizing context.
• ASHRAE, Filtration and Disinfection FAQUsed for filter upgrade and airflow-resistance context around higher-efficiency filtration.
• National Comfort Institute via ENERGY STAR, Measure and Interpret Static PressuresUsed for static-pressure measurement, component pressure-drop, manometer, and test-point context.
• AAF International, Side Access HousingsUsed for filter-bank housing and gauge context around static pressure, filter resistance, and pressure drop.
• AIRAH, Filter Skills WorkshopUsed for monitoring filter resistance, design final resistance, and filter service context.
• Carrier, Will Frozen AC Fix Itself? A Guide To Frozen AC CoilsUsed for coil-freeze context around restricted airflow, dirty filters, blocked vents, and refrigerant issues.

Related guides