CleanroomTestingandCertification_inhouseGHPtraining

Thermal Anemometer• Single point• Turbulent air flow velocity varies thru time• Time constant –Moving average results in more stable reading 251

Iso-axial Measurements• Probe must be oriented toward airflow• Applied to both tube array and thermal anemometer – Must be within 22.5 of flow axis 22.5 angle 252

Thermal Probe Orientation Error 253

Instrument Calibration• Instruments calibrated in ideal conditions –Laminar to near laminar airflow• Cleanrooms are turbulent flow 254

Laminar to Turbulent Flow 255

Work Height Flow Vectors• Iso-axial readings only under filters / screens• Airflow is not vertical at work height readings – Flat surfaces • Vial accumulation table 256

Room Design Affects Flow Vectors• Flow angle deviates toward air returns – Velocity increases approaching returns• Curtains or barriers are necessary to maintain vertical airflow at work height 257

Is Velocity Uniformity Important?• 3D Anemometer study– Interface mid line between filters– Each flow vector determined from 1,800readings– 3 axia (x, y, z) at 20 Hz for 30 seconds• Mock filling line – 600 mm x 4800 mm– Curtains to 760 mm above the floor– Filters butted end to end• Minimum “ceiling grid width” 258

Let’s talk “CLEAN”• ISO 14644 cleanliness Classes 1 through 9 – Pharmaceutical critical areas are ISO 5 – Microelectronics are ISO 2 or ISO 3• How do they do it? – ULPA filters rather than HEPA filters • No gasket or gel seal leaks – Air returns through grated / raised floor – Gore-Tex gowning serves as a filter media – Filtered, negative pressure head gear – Extensive use of isolators 259

Microelectronics Air Velocity• Never at 0.45 m/s or 0.50 m/s!!– Turbulence under objects in airflow path• Ceiling grid• Tall equipment– Energy cost prohibitive for 20,000 filterfacility• Typical design velocities of 0.30 m/sto 0.35 m/s– Ceiling heights are now over 7 m 260

Conclusion• Velocity measurements prone to error– Low velocity pressure– Turbulent air flow– Iso-axial sampling errors• Poor measurement repeatability• Statistical outliers in a normal distribution• Too much emphasis is placed on velocity• Is regulatory “guidance” driving bestpractices? 261

ParticleCounters Theory and Operation 262

ObjectivesAt end of session, you will be able to:• Explain how a particle counter operates• Explain and quantify the operating limits of particle counter• List uses for a particle counter in cleanrooms and HEPA filter testing 263

Laser Particle Counter 264

Particle Counter Response 265

Cumulative vs. Differential DisplaySize (um) Cumulati Differenti ve al 0.3 10 10 0.2 25 15 0.1 42 17 266

Particle CounterBackground Noise 267

Variance in ParticleCounter Calibration 268

FED-STD-209D Figure 1: Classification Graph 269

Counting Error as a Result ofSizing Error for Aerosol Size 270

Optical Coincidence 271

Electronic Saturation 272

Beware of Published Upper Limits• JIS Standard B-9921-1999 is a particle counter design standard• ISO 21501-4 calibration standard taken from B-9921• Maximum concentrations are determined by the pulse width of a particle – Assume 1 millisecond for a 0.3 um particle – It is assumed one can then count 1,000 particles per second – True if particles are exactly 1 millisecond apart in sample – Possion distribution have random spacing – Therefore, maximum concentrations are grossly overstated – Typically 100,000 counts per minute 273

ISO 21501-4 : 2007 274 Particle Counters• A very progressive calibration standard – Size calibration – Verification of size setting – Counting efficiency – Size resolution – False count rate – Maximum particle concentration (calculated) – Sample flow rate – Response rate• Counters made prior to 2007 may not meet 21501-4

ParticleCountingStandards 275

Particle Counts• Introduction – Brief History• A Standard• Statistics – Assumptions – Number of Locations – Sample Volumes – Calculation• Sequential Sampling 276

Particle Counts• A Standard – A way to talk–ISO 14644-1 and 2 (1999)• Scope–Exclusively in terms of concentration ofairborne particles–Only particles based on lower limit sizesranging from 0.1 to 5 um.–Does not address the physical,chemical, radiological or viable nature ofairborne particles 277

Particle Counts• ISO 14644-1 and 2 (2015) –Terms and definitions • 3.1.1 cleanroom –Room within which the number concentration of airborne particles is controlled and classified, and which is designed, constructed and operated in a manner to control the introduction, generation, and retention of particles inside the room 278

Particle Counts• ISO 14644-1 and 2 (2015)– Terms and definitions• 3.1.2 clean zone– Defined space within which the numberconcentration of airborne particles is controlled andclassified, and which is constructed and operated ina manner to control the introduction, generation,and retention of contaminants inside the space– Note 3: A defined space within a cleanroom or mightbe achieved by a separative device. Such a devicecan be located inside or outside a cleanroom. cleanzone(s) can be 279

Particle Counts• ISO 14644-1 and 2 (2015)–Terms and definitions• 3.1.4 classification–Method of assessing level of cleanlinessagainst a specification for a cleanroomor clean zone–Note 1: Levels should be expressed interms of an ISO Class, which representsthe maximum allowable concentrationsof particles in a unit volume of air. 280

Particle Counts• ISO 14644-1 and 2 – Classification formula 281

Particle Counts ISO Maximum concentration limits (particles/m3 of air) for particles equal to and FED STDclassification larger than the considered sizes shown (concentration limits are calculated 209Enumber (N) in accordance with equation (1) in 3.2) equivalent 0.1 µm 0.2 µm 0.3 µm 0.5 µm 1.0 µm 5.0 µmISO 1 0.351676 Class 1ISO 2 3.516757ISO 3 35.16757ISO 4 351.6757 Class 10ISO 5 100,000 23,651.44 10,176.25 3,516.757 831.76 29.25 Class 100ISO 6 35,167.57ISO 7 351,675.7 ClassISO 8 3,516,757 1,000ISO 9 Class 35,167,572 10,000 Class 100,000 Room Air 282

Particle Counts – Table 1 283

Why Special? Looking at Annex 1Volume 4, EU Guidelines to Good Manufacturing Practice, MedicinalProducts for Human and Veterinary Use – Annex 1, Manufacture ofSterile Medicinal Products (corrected version) - 2008Grade Maximum permitted number of particles/m3 equal to or greater than the tabulated size 0.5μm A 3,520 At rest 0.5μm In operation B 3,520 5.0μm 5.0μm 20 ISO Class 4.8 3,520 20 See note to left ISO Class 7-ish (5 except 5.0 um was 29) 29 352,000 2,900C 352,000 2,900 ISO Class 7-ish 3,520,000 29,000 ISO Class 8-ishD 3,520,000 29,000 ISO Class 8-ish not defined not defined 284

Other Places – Annex 1• For classification purposes in Grade A zones, a minimum sample volume of 1 m3 should be taken per sample location. For Grade A the airborne particle classification is ISO 4.8 dictated by the limit or particles  5.0 m. For Grade B (at rest) the airborne particle classification is ISO 5 for both considered particle sizes.• It takes 36 minutes (35.3) to sample 1 m3 with a 1 cfm counter• Minimum sample volume for 29 particles  5.0 m is 0.69 m3 with a 1 cfm counter would be 25 minutes (24.4) Grade Maximum permitted number of particles/m3 equal to or greater than the tabulated size At rest In operation 0.5μm 5.0μm 0.5μm 5.0μmA 3,520 20 ISO Class 3,520 20 See note 4.8 (5 except to leftB 3,520 29 5.0 um was 352,000 2,900 ISO 29) Class 7- ish 285

Particle Counts – Table 1C.7 Adaptation of the macroparticle descriptor toaccommodate consideration of  5 m particle sizefor ISO Class 5 CleanroomIn order to express an airborne concentration of 29particles/m3 in the particle size range  5 mbased on the use of an LSAPC, the designationwould be “ISO M (29;  5 m ); LSAPC” and for 20particles/m3 the designation would be “ISO M (20; 5 m ); LSAPC” (see Table 1, Note f). 286

Particle Counts• Statistics –Assumptions • There are more smaller particles than larger. 287

Particle Counts• Assumptions – More smaller particles – For ISO Class 5 • 100,000 → 0.1 m and larger • 23,700 → 0.2 m and larger • 10,200 → 0.3 m and larger • 3,520 → 0.5 m and larger • 832 → 1.0 m and larger • 29 → 5.0 m and larger 288

Particle Counts• Assumptions – More smaller particles– For ISO Class 5 76,300 between 0.1 and 0.2 m 13,500 between 0.1 and 0.2 m • 100,000 → 0.1 m and larger 6,680 between 0.1 and 0.2 m • 23,700 → 0.2 m and larger 2,688 between 0.1 and 0.2 m • 10,200 → 0.3 m and larger 803 between 0.1 and 0.2 m • 3,520 → 0.5 m and larger • 832 → 1.0 m and larger 289 • 29 → 5.0 m and larger

Particle Counts• Assumptions – More smaller particles – But this is an assumption without a precise relationship • ISO 14644-1 Section 4.3 (2015) states “Particle number concentrations for different threshold sizes in Table 1 do not reflect actual particle size and number distribution in the air and serve as criteria for classification only.” – It is possible an area might meet the requirements of the Air Cleanliness Class at one particle size but not at another. 290

Particle Counts• Statistics –Assumptions –Number of Locations • Statistically valid sample 291

Particle Counts• Number of Locations – Derive the minimum number of sample location, NL from Table A.1. – > 1000 m2 : N = 27( A/1000)NOTE 1 If the considered area falls betweentwo values in the table, the greater of the twoshould be selected.NOTE 2 In the case of unidirectional airflow, thearea may be considered as the cross section ofthe moving air perpendicular to the direction ofthe airflow. In all other cases the area may beconsidered as the horizontal plan area of thecleanroom or clean zone. 292

Particle Counts• Positioning the sampling location (A.4.2)• In order to position the sampling locationsa) Use the minimum number of locations NL from Table A.1b) Divide cleanroom or clean zone into NL sections of equal area,c) Select a representative sampling location within the sectiond) Position PC prove in the place of work activity or another specified point. • Additional sampling locations may be selected for locations considered critical. • Additional sections and associated sampling location may be included to facilitate subdivision into equal sections. • For non-unidirectional airflow cleanrooms or clean zones, locations may not be representative if they are located directly beneath non-diffused supply air sources. 293

Particle Counts• Statistics –Assumptions –Number of Locations –Sample volumes • Statistically valid sample 294

Particle Counts 295

Particle Counts ISO Maximum Sample Volumes in Liter (raw) FED STDclassification 209Enumber (N) 0.1 µm 0.2 µm 0.3 µm 0.5 µm 1.0 µm 5.0 µm equivalentISO 1 2,000 833 2,000ISO 2 200 84.4 196 Class 1 8.44 19.6 Class 10ISO 3 20 571 56.8ISO 4 2 241ISO 5 0.2 0.844 1.96 5.68 24.04 690 Class 100ISO 6 0.02 0.0844 0.196 0.568 2.404ISO 7 0.0568 0.2404 68.259 ClassISO 8 0.00568 0.02404 6.8259 1,000ISO 9 0.000568 0.002404 0.68259 Class 10,000 Class 100,000 0.068259 Room Air 296

Particle Counts ISO Maximum Sample Volumes in Liter (raw) FED STDclassification 209Enumber (N) 0.1 µm 0.2 µm 0.3 µm 0.5 µm 1.0 µm 5.0 µm 2,000 equivalent ISO 1 200 20 833 2,000 571 Class 1 ISO 2 84.4 196 ISO 3ISO 4 2 8.44 19.6 56.8 241 Class 10ISO 5 2 2 2 5.68 24.04 690 Class 100ISO 6 2 2 2 2 2.404 68.259 ClassISO 7 1,000ISO 8 2 2 6.8259 Class 10,000 2 2 2 Class 100,000ISO 9 2 2 2 Room AirA.4 The volume sampled at each location shall be at least 2 liters, with a minimum sampling time ateach location of 1 minute. 297

Particle Counts ISO Maximum Sample Times assuming 1 cfm (28.3 lpm) LSAPC FED STDclassification 209Enumber (N) 0.1 µm 0.2 µm 0.3 µm 0.5 µm 1.0 µm 5.0 µm 71 equivalent ISO 1 8 30 71 Class 1 ISO 2 1 84.4 196 21 ISO 3ISO 4 1 1 1 56.8 9 Class 10ISO 5 1 1 1 1 1 25 Class 100ISO 6 1 1 1 1 1 3 ClassISO 7 1,000ISO 8 1 1 1 Class 10,000 1 1 1 Class 100,000ISO 9 1 1 1 Room AirB.4.2.2 The volume sampled at each location shall be at least 2 liters, with a minimum sampling time ateach location of 1 minute. 298

Particle Counts• Statistics –Assumptions –Number of Locations –Sample volumes –Calculation 299

Particle Counts• Calculation – Average the individual samples at each location – The cleanroom or clean zone meets the air cleanliness classification if the average at each location does not exceed the concentration limits – The dirtiest sample location determines the ISO Class 300