Work Instruction: Smoke Study (Airflow Visualization) Execution and Documentation

This is a ready-to-use, task-level work instruction for performing and documenting a cleanroom airflow visualization (smoke) study, as distinct from the particle counting and HEPA integrity tests covered elsewhere in the qualification protocol. It sits under the parent protocol <<FILL: qualification protocol ID, e.g. the cleanroom classification and qualification protocol>> and should be used alongside that protocol’s test records. Replace every <<FILL: ...>> placeholder. A filled specimen follows. This is general guidance to adapt and verify, not regulatory advice.

Field	Entry
Work instruction number	`<<FILL: WI-ID, e.g. WI-FAC-031>>`
Parent protocol	`<<FILL: qualification protocol ID>>`
Version / effective date	`<<FILL>>`
Applies to	`<<FILL: Facilities/HVAC SME, Validation/C&Q, Aseptic operations>>`

When to use this work instruction

Use it whenever Grade A airflow needs to be visually demonstrated: initial qualification, periodic requalification (risk-triggered or scheduled), after any change to equipment layout, RABS or isolator configuration, filling line tooling, or HVAC that could plausibly alter the sweep pattern over the critical zone, and whenever an in-operation excursion investigation needs to interpret an event against the documented airflow pattern for that zone.

Tools and materials

Smoke/fog generator producing a visible, neutral-buoyancy aerosol (commonly a glycol or DI-water fog, or a controlled fume), rated safe for the cleanroom classification in use.
Two or more video cameras or a single camera repositioned for multiple defined angles, with a means to timestamp or voice-annotate the recording.
The room’s classification diagram showing the critical zone, HEPA locations, and return/exhaust points.
The scripted intervention list for this room (who, what action, at what point in the sequence) agreed with aseptic operations in advance.
Personal protective equipment and gowning appropriate to the room grade; confirm the fog agent is compatible with the room’s product and does not itself present a contamination or safety hazard.

Task A: prepare and script the study

Step	Action	Acceptance for this step
A1	Confirm the room is in its qualified HVAC state (airflow balanced, no open maintenance work) before scheduling.	Room status confirmed with Facilities/HVAC; no outstanding work that would invalidate the test.
A2	Write the intervention script: every routine and non-routine intervention the process allows (line setup, weight checks, stopper or component additions, sample pulls, jam clearing, glove or gown changes), each with who performs it and where.	Script is specific enough that the same interventions shown in the smoke video are the ones used in real production and in the in-operation particle count.
A3	Confirm camera positions cover the full critical zone from multiple angles, including the angle that would reveal smoke curling back over an open product path.	At least `<<FILL: number, e.g. 2>>` camera angles agreed and positioned; no critical sub-zone left uncovered.

Task B: film the at-rest pattern

Step	Action	Acceptance for this step
B1	With the room at rest (equipment running, no personnel present), introduce smoke at the defined challenge points and record continuously.	Unidirectional sweep is visible across the full critical zone with no personnel disturbance.
B2	Record the sweep direction, any eddies, dead spots, or in-rushes observed, with timestamp references into the video.	Observations are written contemporaneously, not reconstructed from memory afterward.

Task C: film the dynamic (in-operation) pattern with interventions

Step	Action	Acceptance for this step
C1	With the room in operation and the defined crew present, perform each scripted intervention in sequence while filming continuously from the agreed angles.	Every intervention on the script in Task A is actually performed and filmed; a skipped intervention is documented and re-scheduled, not waived silently.
C2	For each intervention, observe and note whether the air sweep continues to move away from the product, or whether the intervention draws air from a less clean area toward the exposed product.	Each intervention has a recorded observation, not a blanket “acceptable” covering the whole session.
C3	Where an observation is ambiguous (for example, a brief, small disturbance that self-corrects quickly), replay the footage and have a second qualified reviewer confirm the read before finalizing.	No ambiguous observation is resolved by a single reviewer’s unchecked judgment.

Task D: assemble the report

Step	Action	Acceptance for this step
D1	Compile the annotated video (or representative stills with timestamps) with the written observations from Tasks B and C into the smoke study report.	Report references the specific video timestamps that support each conclusion; the video is retained, not just the summary.
D2	State the overall acceptance decision against the criteria in section “Acceptance criteria” below, with any finding routed per Task E.	Decision is traceable to the specific observations, not a standalone conclusion unsupported by the footage.
D3	Route the report for review and approval through the parent protocol.	Reviewer explicitly confirms the video supports the stated conclusion before approving.

Task E: what to do if you observe a failure

Stop and document the specific intervention, camera angle, and timestamp showing the adverse pattern (turbulence, in-draw, dead spot, or stagnation over the product path).
Do not characterize it as “acceptable” or “minor” without a second qualified reviewer’s independent confirmation from the footage.
Raise a deviation per <<FILL: deviation management SOP-ID>>, assess whether the finding affects the room’s qualified or routine-use status, and determine whether the study must be repeated after a corrective action (equipment relocation, airflow adjustment, intervention technique change) before the room resumes the affected use.

Acceptance criteria

There is no particle number here; the criterion is observational, judged against the questions a reviewer must answer for both the at-rest and dynamic film:

Air sweeps unidirectionally over the critical zone and moves away from the product in both states.
No upward currents, eddies, or stagnant zones appear over the open product path.
Every scripted intervention was filmed, and each one shows air continuing to move away from the product, or the finding was dispositioned per Task E.
The written report’s conclusion is fully supported by the retained video; a reviewer re-watching the footage reaches the same conclusion.

Filled specimen

The following shows the report summary for an example Grade A filling line study, so you can see the level of detail expected. The company, line, and specifics are illustrative; replace them with your own.

Field	Entry
Room / line	Filling suite FL-02, Grade A zone within Grade B background
Study type and date	Dynamic, scripted interventions, 11 September 2026
Interventions filmed	7 of 7 scripted (line start, stopper bowl refill, in-process weight check, sample pull, jam clear, glove change, line stop/restart)
Finding	Stopper bowl refill at minute 14 showed a brief 2-second smoke curl toward the open bowl from camera angle 2; second reviewer confirmed from footage; self-corrected within 2 seconds with no product exposed at that moment
Disposition	Logged as an observation, not a failure, because no product was exposed during the 2-second event; a technique change (slower approach angle) recommended and added to the operator training; study not required to repeat since the underlying airflow pattern itself was not adverse
Overall result	Pass, with the technique recommendation tracked as a CAPA-lite action
Reviewer / approver	A. Mensah (Validation), signed; R. Gomez (QA), signed, 12 September 2026

Note what makes this defensible: the brief disturbance was found, filmed, independently confirmed, and dispositioned against whether product was actually exposed at that moment, rather than smoothed over as “acceptable” with no supporting detail.

Common mistakes this work instruction prevents

Filming only the at-rest state and calling the study complete, with no dynamic or intervention footage.
A script that lists generic interventions rather than the specific ones this room’s process actually performs.
Camera angles that miss the critical sub-zone where the real risk sits.
A report that concludes “acceptable” while the retained footage, if anyone watched it, shows visible turbulence over open product.
Treating a borderline finding as self-evidently minor without a second reviewer or without assessing whether product was exposed at that moment.
Never repeating the study after a layout, tooling, RABS/isolator configuration, or HVAC change that could plausibly alter the pattern.

How to adapt

Set your work instruction number and link it to the parent qualification or requalification protocol.
Replace the generic intervention list with your room’s actual scripted interventions, agreed with aseptic operations.
Set your camera count and angle requirements based on the room’s geometry and critical zone layout.
Point the Task E escalation to your real deviation management procedure.
If your site uses rapid smoke-pattern analysis tools instead of pure visual review, reference their validation record here.

Header