Software Helps Tidy Up Clean Rooms

Concern over cross-contamination between different cleanliness classes within a clean-room garment-changing area led an in-house engineering team at the Belgium facilities of Alcon Laboratories, a pharmaceutical company based in Ft. Worth, Texas, to carry out a design study simulating air movement, heat transfer and distribution of contaminants through the rooms.

Staff

Share

Concern over cross-contamination between different cleanliness classes within a clean-room garment-changing area led an in-house engineering team at the Belgium facilities of Alcon Laboratories, a pharmaceutical company based in Ft. Worth, Texas, to carry out a design study simulating air movement, heat transfer and distribution of contaminants through the rooms.

Using Windows-based software, the engineers were able to model and test unconventional design concepts to see how they met clean-room regulatory requirements and consequently make cost-effective design changes quickly. Without this simulation, the installation would have gone ahead with the original design and might have encountered potentially serious problems when it came to commissioning and validation of the room.

Clean-room cartography

The layout of the area affected comprised an unclassified area, where technicians changed from their laboratory clothes into their clean-room garments. Then, depending on what they were working on, the technicians passed through areas C or D—with area C offering a higher level of cleanliness.

The original ventilation system design utilized 12 standard ceiling-mounted HEPA filters supplying clean air, which was then removed via low-level extractors in the unclassified areas. The results of the modeling software showed the deficiencies of this strategy, with contaminated air flowing from the dirtier to the cleaner areas of the room.

The in-house engineers took the data from this first study and used it as the basis for a design study, simulating a series of scenarios and comparing their performance with the baseline computer model.

Through this parametric design study, which was aimed at assessing the sensitivity to change of the ventilation scheme, a final, improved design evolved.

In the first instance, a floor-to-ceiling partition wall was installed between the unclassified area and the Class C area, providing a physical barrier to airflow between these zones.

High-capacity HEPA filters, combined with a coarse-grained cloth, were installed above the doors entering the Class C production areas which created a unidirectional airflow. The HEPA filters across the remainder of the room were replaced with high-capacity HEPA filters connected to two ceiling-mounted air socks, creating a radial pattern of airflow dispersal.

Predictive airflow modeling indicated the success of this installation, with the air flowing from the cleanest areas in the room back toward the low-level extracts in the unclassified area. The engineers then validated these predictions by performing a series of actual smoke tests prior to final commissioning of the premises.

Michel Wellens, the project manager at Alcon, was pleased about the decision to use a modeling software, as it saved both time and money.

"The investment made in this design study has more than paid for itself," he explained. "The money and time saving made in avoiding design changes in the final stages of the project means that we have finished with a robust, reliable design that was proved to work as intended at the conceptual design stage of the project."

For more information on Flovent software by Flomerics, circle 102 on the Reader Service Card on page 73 of this issue.