Air Techniques International

Most current standards and recommended practices require a minimum of 10 micrograms per liter (ug/l) as an upstream aerosol challenge. While higher concentrations can be used, 10-ug/l is the minimum that may be used and still have a valid leakage certification test. All of ATI's aerosol photometers were designed to operate at upstream challenge concentrations between 10 and 100 ug/l.

2) Will using aerosol to test filters cause excessive loading (clogging) and shorten the filters usable life?

Poly-dispersed aerosol is used to challenge the "integrity" or leakage of the plenums, framework, connections, etc., and the filter media itself. Therefore people ask, "How much aerosol is needed?"

The example below uses 50 micrograms per liter as an upstream challenge.
Example: Let's use a filter bank that is 10 feet x 10 feet and utilizes (25) 24" x 24" x 12" HEPA filters. The normal 24" x 24" x 12" HEPA has an average of 65 pleats. The pleat size is 22 1/2" x 10 1/2". If we take this information and multiply it, we come up with the average HEPA filter containing approximately 213 square feet of media. If you multiply this by 25 (for the 25 HEPA filters in the bank) we come up with a total of 5,325 square feet of media absorb. We will also calculate the amount of liquid we will aerosolize and let this 5,325 square feet of media absorb. To figure out how much liquid we will introduce to these filters, we have to calculate how long it will take to scan the filter bank. If you use a photometer with a rectangular isokinetic probe, it will take 40 minutes. This is calculated by figuring the scanning rate of 10 feet per minute. Since the filter bank is 10 feet wide, one scan per minute across the face of the filters would be allowed. Since the probe dimension is 3 inches and the bank is 120 inches high, 3 into 120 equals 40 strokes. The test time will be 40 minutes (each stroke taking 1 minute, it requires 40 strokes to cover the entire area).

The next calculation is the amount of liquid that will be converted into an aerosol and spread over the 5,325 square feet of media. This equals 0.00283 grams per cfm (calculate 100 micrograms per liter x 28.3 for one cubic foot per minute). Since the system is a 25,000 cfm system, we multiply 25,000 x .00283 grams per cfm which totals 70.75 grams of liquid we will aerosolize to obtain a challenge of 100 micrograms per liter in this 25,000 cfm filtration system. Because we will run this test for 40 minutes, we will multiply 70.75 x 40 minutes and come up with a total of 2,830 grams of liquid that we will aerosolize to challenge the integrity of the 25,000 cfm filtration system with 100 micrograms per liter. The 2,830 grams is equal to 99.8 ounces of oil spread over 5,325 square feet of media (equivalent to a 12 inch wide roll of filter media just under a mile long). How can we clog this filter system? If we decide to only use 36 ounces to test the filter system, we could then challenge it with 36.1 micrograms per liter (equal to three-12 ounce containers).

3) Where should upstream challenge aerosol be introduced into a filter system that is being tested?

Adequate aerosol mixing upstream can usually be obtained by introduction at least ten duct diameters upstream of the filters, or by introducing it upstream of baffles or turning vanes in the duct. When neither of these methods is practical, a Stairmand disk located four to six duct diameters upstream will provide satisfactory mixing. A Stairmand disk is a plate with the same geometric shape as the duct section that blocks the central half of the duct area. Air flowing past the disk creates vortices on the leeward side that compel turbulent and thorough mixing of the introduced aerosol and the dilution air stream. The required diameter of the stairmand disk may be calculated using the following information.

The Stairmand disk diameter equals the pipe (or duct) diameter divided by the square root of two (1.414213).

4) What type of PAO does the FDA require for filter leakage certification tests?

Several differing opinions on the correct PAO to use have arisen recently. The FDA, in their original CGMP (Current Good Manufacturing Processes) release was specific in the type that was approved as a replacement for DOP (DEHP).

"The Chemical Abstracts Service (CAS) number which identifies this product also remained as 68649-12-7. Other reported alternatives used in the industry include DOS (Di-(2-ethylhexyl) sebacate) and Ondina Oil. However, no manufacturer has yet submitted all the necessary data to evaluate these alternatives. As such, Emery 3004 PAO with the CAS number 68649-12-7 still remains an acceptable replacement for DOP."

ATI's position is that PAO with a CAS # 68649-12-7 is acceptable by FDA definition. The full text of the referenced CGMP document is available here.

5) How much compressed air does a Type III-A Laskin nozzle aerosol generator need?

Each Laskin nozzle consumes approximately 2.64 cfm (75 Liters) of air at 20 psi (1.4 bars) and total air consumption is proportional to the number of nozzles in use. Therefore the maximum compressed air requirement for a three nozzle generator is 7.92 cfm at 20 psi and a six nozzle generator would need 15.84 cfm at 20 psi.

The applied pressure of 20 psi should be measured at the nozzle inlet and must remain constant to allow calculation of the aerosol generator output. Air compressor performance specifications that will show output volume (cfm) versus pressure (psi) are readily available from most vendors and will make selection of an appropriate compressor less difficult.

6) What is the particle size distribution of a Laskin nozzle generator using PAO?

7) What is the particle size distribution of a Laskin nozzle generator using DOP?

8) What is the PAO particle size distribution of a Thermal generator (TDA-5A/5B)?

TDA-5B Thermal Condensation Type Aerosol Generator using PAO (Poly-alpha Olefin) operating at standard set up parameters of 408° C (765° F) with a 50 psig inert gas supply pressure.

9) What is the DOP/DEHS particle size distribution of a Thermal generator (TDA-5A/5B)?

TDA-5B Thermal Condensation Type Aerosol Generator using DOP (Dioctyl Phythalate) operating at standard set up parameters of 408° C (765° F) with a 50 psig inert gas supply pressure.

	Number	Surface	Mass	Volume
	Particle Size	Particle Size	Particle Size	Particle Size
median (nm)	287	354	384	384
mean (nm)	296	361	392	392
geo. mean (nm)	275	346	377	377
mode (nm)	300	385	400	400
geo. st. dev.	1.54	1.35	1.33	1.33

*The TDA-5B aerosol distribution listed above is characteristic of the operating conditions and settings present at the time of testing. Particle size distributions generated during field usage will change depending upon the ambient temperature, humidity and equipment settings in use.

10) What is the PAO particle size distribution of a Thermal generator (ATI-5C)?

ATI-5C Thermal Condensation Type Aerosol Generator using PAO (Poly-alpha Olefin) operating at standard set up parameters of 408° C (765° F) with a 50 psig inert gas supply pressure.

	Number	Surface	Mass	Volume
	Particle Size	Particle Size	Particle Size	Particle Size
median (nm)	399	500	548	548
mean (nm)	412	513	561	561
geo. mean (nm)	383	490	537	537
mode (nm)	414	514	573	573
geo. st. dev.	1.51	1.37	1.34	1.34

*The ATI-5C aerosol distribution listed above is characteristic of the operating conditions and settings present at the time of testing. Particle size distributions generated during field usage will change depending upon the ambient temperature, humidity and equipment settings in use.

11) What is the recommended shelf life of PAO (CAS # 68649-12-7)?

The manufacturer's stated shelf life for PAO (CAS# 68649-12-7) is three (3) years in the original, sealed and unopened container. The shelf life of an open container, that is tightly sealed between uses, is one (1) year.

ATI does not recommend the use of the TDA-5B or any other thermal condensation based aerosol generator for use in the certification of Biological Safety Cabinets. IEST RP034 specifically cautions against this practice and NSF/ANSI 49-2002, Annex F specifically calls out for the use of a Laskin nozzle style generator or equivalent during certification. A TDA-5B generator is not considered a Laskin nozzle equivalent aerosol generator.

13) What happens if the photometer sample flow drops below its normal operating range of 28.3 lpm +/- 2.8?

Reduction of the photometer sample flow to a level below the lower tolerance limit, while using the internal reference settings for the 100% level, results in an increased displayed leakage value. This higher value is the result of an increase in residence time of the aerosol passing through the light scattering chamber's area of observation. The longer residence time of the aerosol results in a more conservative evaluation of the filter under test. If an actual upstream aerosol sample was being used to establish a 100% point, the unit’s aerosol response is self correcting for the decreased flow rate and standard accuracy conditions would be maintained.

The straylight value is an indication of the optical conditions of the Light Scattering Chamber (LSC) at a given point in time and as such, do not have a tolerance range established. Increases in the straylight usually occur through ordinary usage over the annual calibration cycle. Sudden and/or catastrophic increases in the straylight value typically are the result of operator error or component failure and render the unit inoperable. Some examples of catastrophic failure are introduction of liquid to the scattering chamber, formation of condensation on the optics, large accumulations of solid particulate and breakage of discrete components within the LSC. Photometers which remain operable, even with elevated straylight values, are still capable of accurately measuring filter leakage values.

A reduction of the sample flow to a level below the lower tolerance limit, while using the internal reference settings for 100% levels, results in an increased displayed leakage value. This higher value is the result of an increase in residence time of the aerosol passing through the light scattering chamber's (LSC) area of observation. The longer residence time of the aerosol results in a more conservative evaluation of the filter under test.

An increase in sample flow, however uncommon, results in the opposite effect. A shorter residence time decreases the displayed leakage value and results in a liberal assessment of the filter performance and the potential for passing a "bad" filter.

These effects are the reason for the ±10% limits paced upon the sample flow requirements in aerosol photometer testing protocols.

If an actual upstream aerosol sample was being used to establish a 100% point, the units aerosol response is self correcting for the change in flow rate and standard unit leakage accuracy conditions would be maintained.

The response time of any aerosol photometer is determined by several factors.
Sample flow rate, sample tube volume and the refresh rate of the photometer display all affect the time necessary to achieve a valid response. In most cases the refresh rate is the most limiting factor.

A typical photometer will completely exchange the sample air within its flow path approximately 4 times per second assuming the following:

1) A 1 CFM continuous sample rate (28.3 lpm)
2) 12 feet of 3/8 inch sample tubing (0.25 inch inside diameter)
3) Display refresh rate at its minimum (most frequent) value

28.3 lpm converts to 472 ml per second. A 12 foot length of sample tubing has an internal volume of 116 ml.
This means that the entire volume of the sample tubing is replaced 4.1 times per second.

Because the display only updates approximately once per second, this becomes longest interval event that must occur for a valid value to be displayed.

17) Can a mono-dispersed "Hot" aerosol penetrometer use PAO-4 (Emery 3004

) in place of DOP?

Testing was performed in 1992 by the U.S. Army to determine a DOP replacement candidate material for use in "Hot" mono-dispersed aerosol filter penetrometers. The results of that study concluded that PAO-4 (Emery 3004

) was an acceptable replacement for DOP in not only mono-dispersed "Hot" testing, but in poly-dispersed "Cold" testing applications as well. The full text of that report is available at CRDEC-TR-333.

18) Is there a "Portable" compressor available, capable of supplying the air volume required for 2+ Laskin nozzles?

Unfortunately we have been unable to identify a specific portable compressor having the capacity to source 18 CFM at 20 psi. In all known instances a compressor of that capacity is wheeled, and incorporates a reservoir tank, and therefore is not readily portable.

Most common applications where the TDA-4B is used require only a portion of the full output capacity. Typically 1 to 2 nozzles. For these applications we have had great success recommending a Thomas compressor that is available through McMaster-Carr. The specific model is McMaster-Carr part number 8280k999, Thomas part number 1820CO. This is a 1-1/2 HP, oil-less compressor that operates on 115V/60 HZ and has a 50 psi maximum output pressure. The manufacturer, GD-Thomas, has rated the output at 6.0 CFM while maintaining 40 PSI.
In those applications that require full-output an in-house compressed air source is typically utilized.
Please contact ATI Customer Service if you need any further assistance.

The liquid volume capacity of the TDA-4B Aerosol Generator is 250 cubic inches. This means that slightly more than one gallon (4 liters) is necessary to fill the TDA-4B to the centerliine of the sight gauge.

A leakage rate of 0.010% would be reported as 0.0075% due to the 25% reduction in response.

A high reference response condition is where the photometer reports a concentration value greater than physically present. This has the effect of over-reporting leakage levels and potentially failing filters that having passing characteristics. Better than “Worst case scenario.

Conditions of example:

Relavent Notes:
The reference tolerance on all ATI photometers to date is ± 10%.

When utilized for filter leakage certification, every effort should be taken to use an actual sample of the “Upstream” aerosol concentration. The Reference feature should be used for obtaining a working 100% set point only as a last resort.

Displayed % Leakage values are limited to 135% of the chosen 100% set point regardless of whether an aerosol sample or Internal Reference was used. A photometer having a value of 65 chosen as the reference factor has a reading ceiling of up to 135% of the 65 chosen as the reference factor, or 87.8-ug/l. The 135% display limit is caused by range limitations of the amplifier.

1) A 1 CFM continuous sample rate (28.3 lpm) 2) 12 feet of 3/8 inch sample tubing (0.25 inch inside diameter) 3) Display refresh rate at its minimum (most frequent) value

1) A 1 CFM continuous sample rate (28.3 lpm)
2) 12 feet of 3/8 inch sample tubing (0.25 inch inside diameter)
3) Display refresh rate at its minimum (most frequent) value