Quick checks before adjustments
This guide investigates pressure below the approved range in a room intended to be positive to its reference space, or pressure that repeatedly fluctuates. Confirm the required direction first. Negative-pressure containment rooms need their own approved response; increasing their pressure may defeat containment.
- Identify the exact boundary: room name, reference room, required direction, operating mode and approved limits.
- Record the alarm before resetting it: time, pressure, duration, door state and affected work. Apply the site's deviation response if an action limit or containment boundary is affected.
- Look for an obvious event: an open door or pass box, a fan alarm, an operating-mode change or work just completed.
- Check the trend: distinguish a brief door event from slow recovery, a sustained low reading and repeated oscillation.
- Confirm the measurement: a local display and BMS may share one sensor. Agreement between them is not an independent check.
Operators can record displayed information and visible conditions. Instrument connections, fan access, balancing, control changes and airflow visualization belong to trained, authorized personnel using the site procedure.
For aseptic pharmaceutical areas, pressure deviations require investigation and recovery must be confirmed. Apply the facility's quality process to affected operations. [2]
Understand what the reading means
Differential pressure is a comparison between two locations. For a reading defined as Proom − Preference, a positive result means the room is at higher pressure than that reference. Mark both locations on the drawing. Do not combine readings taken against different references as if they were directly comparable.
The size and resistance of transfer and leakage paths determine how much pressure the offset can sustain. An open door changes those paths. Evaluate door-open behaviour and closed-door recovery separately against the approved design.
Use the room's approved pressure range, alarm delay and recovery criteria. Pharmaceutical guidance defines pressure, temperature and other limits according to the facility and process; its example values must retain that scope. [1]
Choose a starting point
These questions suggest the next investigation step. They do not calculate the probability of a cause or declare the room fit for use.
Start by confirming the reading
Use Step A below to establish a reliable room-to-reference measurement, then compare it with the approved limits.
Open Step ASelections stay in this browser tab and reset on reload. No readings or answers are sent by this guide.
Detailed investigation
Record the finding from each check. A plausible cause becomes a supported finding only when the measurement or inspection demonstrates it.
A. Verify the pressure measurement Instrumentation technician
- Collect
- Sensor and reference-instrument IDs, ranges, calibration status, room and reference locations, units, time and door state.
- Check
- Compare a suitable calibrated micromanometer at the same approved pressure points, under the same conditions. Inspect accessible tubing for reversal, disconnection, kinks, blockage or moisture. Check the displayed units and BMS range scaling.
- Interpret
- If independent readings disagree beyond the allowed measurement uncertainty, investigate the measurement chain before rebalancing. If they agree and the pressure is outside limits, continue to Step B or C.
- Correct
- Repair the demonstrated tubing, configuration or instrument fault through the approved procedure. Zero adjustment requires the manufacturer's specified pressure condition; never zero away a real room differential.
- Verify
- Repeat the comparison, record as-found and as-left results, and test the alarm through the approved method.
Manufacturer example: Dwyer's 616KX manual describes zero-pressure adjustment, selectable output signals and slower response with longer tubing. Follow the manual for the actual installed model. [3]
B. Separate door events from boundary leakage Operator + maintenance
- Collect
- Pressure trend with door and pass-box events; time from normal door closure to return within the approved band.
- Check
- Observe whether doors close and latch normally. Inspect seals, sweeps, hatches and recent service penetrations. Compare the current boundary with drawings and previous records.
- Interpret
- A brief drop during normal opening and acceptable recovery is a different event from a room that stays low after closure. Slow recovery may involve airflow or controls as well as leakage.
- Correct
- Repair a confirmed failed closer, seal or unplanned opening. Keep designed transfer grilles, door undercuts and pressure-relief paths as specified; they are not automatically defects.
- Verify
- Repeat the approved door sequence and record pressure, recovery and adjacent-room behaviour. Do not wedge doors or defeat interlocks to run an informal test.
WHO discusses door design, gaps, interlocks and their influence on room pressure relationships. [1]
C. Measure the room air balance Air balancing specialist
- Collect
- The latest approved balance report, supply and return terminal schedule, process-exhaust schedule and current operating mode.
- Check
- Measure supply, return and exhaust volumes with appropriate calibrated instruments and the approved test method. Record each terminal and process state. Sum flows in one unit, such as m³/h, and compare both totals and individual branches with the design.
- Interpret
- A reduced mechanical offset suggests a changed balance. Normal totals with poor room pressure direct attention to leakage, transfer paths or the reference space. Measurement uncertainty matters when the offset is small relative to total flow.
- Correct
- Resolve the confirmed cause, then rebalance to the approved requirements. A new process extract may need additional make-up air and a design review. Never reduce required extraction simply to improve a pressure display.
- Verify
- Recheck pressure and airflow in the affected room and connected rooms, including the approved normal process modes.
Air balancing involves measured flows and adjustments. In flow-tracking systems, supply and exhaust must remain coordinated as demand changes. These principles do not supply a universal cleanroom airflow offset. [4] [5]
D. Find what changed in the air system HVAC technician
- Collect
- Fan command and feedback, VFD alarms, terminal flows, filter pressure-drop history, damper records and the maintenance timeline.
- Check
- Review fan and controller information at the approved interface. During authorized isolated access, examine the components indicated by that evidence, such as a slipped drive, fouled coil, incorrectly installed filter or shifted damper.
- Interpret
- Compare filter pressure drop at comparable airflow. A changed flow also changes resistance readings. One weak terminal directs the investigation toward its branch; a zone-wide reduction calls for checks of common plant and controls.
- Correct
- Restore the confirmed component fault to the manufacturer's and site's requirements. Verify the installed filter type, orientation and seating following filter work.
- Verify
- Record restored airflow and pressure. Where the work affects the installed HEPA boundary or qualified conditions, carry out the required integrity or requalification checks.
Pressure drop or room pressure alone does not prove HEPA integrity. FDA's aseptic guidance treats installed leakage and airflow performance as distinct checks. [2]
E. Investigate repeated oscillation Controls specialist
- Collect
- Time-aligned trends for pressure, fan speed, damper command and feedback, supply and exhaust flows, door events and operating-mode changes.
- Check
- Establish whether the movement follows a door event, exhaust start, schedule change or repeating controller action. Use a recording interval that can capture the event. Compare commanded movement with actual feedback.
- Interpret
- A repeating pattern is a clue, not proof of incorrect tuning. A sticking actuator, delayed sensor response, changing reference or interacting control loops can produce similar symptoms.
- Correct
- Have the specialist verify the control sequence, sensor response and actuator operation before changing tuning. Preserve existing settings and document one controlled change at a time.
- Verify
- Demonstrate stable operation and acceptable response to the agreed disturbances, including the approved exhaust and occupancy modes.
Greenheck's laboratory application guide illustrates interacting flow-control loops and the importance of response time. Its laboratory design values are not cleanroom acceptance criteria. [5]
F. Check neighbouring rooms and the reference Facilities engineer
- Collect
- A marked room-pressure diagram, names of each reference space and simultaneous readings across the affected zone.
- Check
- Look for a changed corridor pressure, common AHU condition, central exhaust demand, schedule override or recent work in a connected room. Confirm the drawing's intended air paths.
- Interpret
- A smaller room-to-corridor differential can result from the corridor becoming more positive. A local room adjustment alone may not address the initiating fault.
- Correct
- Resolve the demonstrated zone or reference problem and coordinate the cascade. Fire and smoke-control modes remain subject to their own approved life-safety sequence.
- Verify
- Record the complete set of relevant room-to-room relationships under the approved operating conditions.
Common misdiagnoses
| Assumption | Better evidence |
|---|---|
| The fan is running, so airflow is correct. | Measure delivery. Run status confirms operation, not the required airflow at every terminal. |
| High filter pressure drop proves this is the cause. | Compare flow, filter type and previous readings before attributing the room fault to loading. |
| Increasing the pressure setpoint will solve it. | Confirm available airflow, balance, leakage and controller response first. |
| Every opening must be sealed. | Distinguish unintended leakage from a designed transfer or relief path. |
| The gauge recovered, so the investigation is complete. | Confirm sustained operation, required airflow and neighbouring-room performance; complete the site's release process. |
| More completed checklist items mean a more certain cause. | Record what each check found. Completion counts and preset scores do not demonstrate causation. |
Worked example: a new process extract
HYPOTHETICAL TRAINING EXAMPLE
In this invented example, a room has an approved closed-door range of +10 to +20 Pa relative to its corridor. It normally reads +14 Pa. After a new process extract starts, two independent instruments agree on about +3 Pa. Doors are closed and no new boundary defect is found.
| Condition | Supply | Return | Exhaust | Net offset |
|---|---|---|---|---|
| Before the change | 2,400 | 2,100 | 150 | +150 |
| Extract operating | 2,400 | 2,100 | 300 | 0 |
| After approved rebalancing | 2,550 | 2,100 | 300 | +150 |
The readings support an additional exhaust demand without coordinated make-up. The engineer checks system capacity, reviews the change and approves revised balancing while retaining the required extraction.
After the work, the example room measures +13 Pa, passes its defined recovery checks and maintains adjacent-room requirements. That measured result closes the technical loop. The restored +150 m³/h offset alone does not predict or prove +13 Pa.
These numbers were created to explain the investigation. They are not records from a 360S project, recommended setpoints or an automatic airflow-sizing rule.
Verify recovery and prevent recurrence
- Repeat the original failed check under comparable conditions and record before/after values.
- Observe performance over the site-defined period, including representative normal door and process events. Record the observation duration.
- Confirm the connected system: room pressure, required airflows, adjacent pressure relationships and any affected temperature or humidity requirements.
- Restore monitoring: verify alarms and record the removal of temporary overrides through the approved procedure.
- Complete the close-out: confirmed cause, repair, residual issues, relevant retesting and the responsible person's acceptance. An engineering repair is separate from any quality authorization to resume a sensitive process.
For prevention, add a specific action tied to the cause: door-closer inspection, tubing-route protection, filter trend review, updated balance records, or coordination of new exhaust equipment with the HVAC design. Assign an owner and due date.
WHO's pharmaceutical HVAC guidance includes documented maintenance, current system records and qualification appropriate to changes. [1]
Sources and application scope
This is an original 360S educational investigation sequence, developed from the existing pressure-cascade guide and the references below. It is not a diagnostic probability model or a compliance certificate. Use the current facility requirements and equipment-specific instructions.
- WHO TRS 1019, Annex 2 (2019). HVAC for non-sterile pharmaceutical products, interpretation of guidelines. Relevant sections: premises, monitoring, pressure relationships, qualification and maintenance. Pharmaceutical scope.
- FDA: Sterile Drug Products Produced by Aseptic Processing (2004). Sections IV.C and IV.D cover clean-area separation and filtration. Aseptic pharmaceutical scope; not a universal requirement for electronics or other cleanrooms.
- DwyerOmega: Series 616KX User's Guide, revision 3 (2025). Manufacturer-specific installation, output selection and zero adjustment. Use the actual device manual for field work.
- Greenheck: Benefits of Air Balancing and Commissioning for Make-Up Air Units. General HVAC balancing and commissioning principles.
- Greenheck: Variable Volume Laboratory Systems, an Overview. Flow tracking and control-loop coordination. Laboratory examples require application-specific engineering.
References consulted 2 October 2026. Site-specific procedures and any applicable regulatory requirements take precedence for operating decisions.