Impact of Changes to ISO Standards on Contamination Control Programs by Leonard Bensch, Pall Corporation ABSTRACT Fluid contamination is generally considered the primary cause of hydraulle component wear and reduction in system reliability and performance. Because of this, most companies have some type of contamination control program in place, which includes the application of filtration and fluid condition monitoring by particle counting. The Impending adoption of four new of revised ISO fluid power standards is going to have a major impact on such contamination control programs. The primary changes are resulting from the obsolescence of AC Fine Test Dust (ACFTD), a test contaminant that has been utlized for several decades for primary calibration of automatic partic'e Counters and for fiter testing. Although there are replacement dusts, none of them have an identical particle size istribution to ACFTD, therefore, all associated test resulls are somewhat different. This paper presents typical changes found with the new dusts including the impact on automatic particle counter calibration, resultant particle sizes and counts, and fiter performance - both efliciency of particle capture and dirt capacity. I should be pointed ‘out that, although laboratory resulls and performance interpretations will change, this is an artifact of the testing only, and fitter performance and actual contamination levels in the feld wil romain the same as before, INTRODUCTION AND HISTORICAL BACKGROUND Prior to detailing the Impact of the changes that will occur, this section presents background information and the previous status of the various parameters and test procedures. PARTICLE COUNTING - ~ The first method of counting particles in a hydraulic fluid sample was most likely through the use of a microscope, This technique, which is stil in use today, involves filering a sample of the fluid through a fine membrane fier then examining it under a microscope. Particles ate sized, usually based on their longest chord dimension, and counted. The results are generally roported as number of particles per unit volume of fluid in a specific particle size range. Liquid optical automatic particle counters came into being for hyciraulic fluid analysis in the late 1960s and their use has gradually replaced microscopic techniques in most industries. ‘Such automatic particle counters, termed APCs, {generally function through the principle of light obscuration, “A light beam, most commonly from a laser diode Source, is transmitted through the fluid passing through the APC sensor. Particles in the fluid sample absorb or ‘scatter light in proportion to their size, The transmitted or scattered light is measured by a photodector and the resuling signal strongth is used to determine the particle size. In the late 1960s a project was initiated to develop a calibration procedure with the goal to ensure that particle ‘counts obtained with an APC agreed as closely as possible with counts obtained by optical microscopy. The calibration procedure, which ultimately became an international standard, ISO 4402 [1], is sill in use today by most fluid power laboratories around the world. It is based on microscopic sizing and counting which was performed on ‘AC Fine Test Dust using the longest chord dimension. This calibration procedure gave users confidence in the use ‘of APCs and they are currently ullized in laboratories and the fold to measure the particle size distribution of a fluid sample, either ondine or collected from the operating system in a sample botle. AIR CLEANER FINE TEST DUST — The most commonly used test dustin the fluid power industry was Air Cleaner Fine Test Dust, also called AC Fine Test Dust or ACFTD which was originally sold by the AC Spatk Plug Division (later the AC Rochester Division) of General Motors Corporation. It was manufactured by collecting dust, primary silica, from a certain area in Arizona then ball miling and classifying it into @ consistent particle size distribution including particles sizes from roughly 0-100 um, Because of the consistent particle size distribution of ACFTD and its regular particle shape, believed to be representative of contaminants found in typical hydraulic systems, it was chosen in 1969-1970 for the development of the previously mentioned calibration procedure for automatic, liquid borne, optical particle counters. ‘The particle size distribution for AC Fine Test Dust defined in ISO 4402 and shown graphically in Figure 1 is used to set the electronic threshold lavels that define the particle sizes measured in a particle counter. In addition to ABC calibration, AC Fine Test Dust as well as other similer dusts like AC Coarse Test Dust were used for many ‘other purposes in the contamination control field. As an example, ACFTD was used for fiter performance testing to measure both the efficiency and dit capacity of hydraulic and lube oil filers, It was also used either as recelved orafter separating into specific particle size intervals for testing the contaminant sensitivity of hydraulic components. In 1992, AC Rochester discontinued manufacturing these AC Duss, and SAE and ISO Technical Committee TC 22 started efforts to find replacement dusts, 10,000 _— ST 1.000 2 100 = 10 4 1 — Particle size, ym Figure 1. AG Fine Test Dust Size Distribution Based on Longest Chord Dimension FLUID CLEANLINESS CLASSIFICATION STANDARDS - The ISO Cleanliness Code, ISO 4406, 1987 [2] is the most widely used International standard for representing the contamination level of industrial fluid power systems. ‘The ISO 4406 classification is a two number code, 0.g., 14/12, based on the number of particles greater than 5 yim and 15 jim respectively. it was expanded to three numbers in ISO DIS 4406, 1996 [3] by the addition of a code number representing the number of particles greater than 2 ym, This standard was approved by international ballot but was subsequently withdrawn prior to publication because of the changes which were imminent to particle size dofinitions due to @ new APC calfbration procedure being drafted, ‘The NAS 1638 cleanliness standard [4] was originally developed for aerospace components but is stil widely used for industrial and aerospace fluid power applications. NAS 1638 is comprised of fluid cleanliness classes, each class defined in terms of maximum allowed paricie counts for designated particle size ranges, See Table 1 MULTEPASS FILTER TEST METHOD - In the lote 1960s and early 1970s the hydraullc fiter Multipass test method was developed to measure filter performance, primarily efficiency of particle capture and dirt capacity. This IMultt-pass test later became an International Standard, ISO 4572 (5] and is sill widely used. Because of the characteristics of ACFTD, which made it valuable for APC calibration, it was also ideal and selected for use in the fier test, ‘The test protocol requires adding ACFTD as a slumy into the reservoir upstream of the filer. Particle ‘counts are taken, both upstream and downstream of the fiter, throughout the Mull-pass test. ‘These particle counts are used to calculate the efficiency, expressed as a filration or Bela ratio (upstream count divided by downstream ‘count at a given particle size), and are obviously dependent on the APC calibration. The dirt capacity for the fiter tested is also reported as the amount of ACFTD injected to cause the fiter to reach its terminal differential pressure. The particle size distribution of the fest dust has a significant impact on both the filer efficiency and dirt capacity measurements, The effects of test dust on fiter efficiency are compounded if one also factors in the ‘changes due to particle counter calibration and the corresponding basic changos in definition of particle size.Table 1._NAS 1638 Cleanliness Classos Tianna Conshatey TS Pari (Gaia! 1007), Sica, PETE Ta as] 9550 SOTO STO O73 7 o 250) [~ a | zi 2} F009 q 3 2,006] a) mee 2a BL 8000] ® aan 7} 309 I} 809 S| 709 4060] To. — Te ToL 9 3 0 a} —75 TTS Tz] 20] — 5H 2 ra. sao 8 761.04 REPLACEMENT TEST DUSTS “The work by SAE and ISO on finding dusts to replace the AC Dusts has culminated in the development of a new 180 standard: ISO 12103-1, 1997 [6]. This standard defines and designates four new test dusts as listed in Table 2 Table 2. New and Replacement ISO Test Dusts 180 Designation | Common Name Oiher Names TSO T2103-AT | 150 Ulirafine Test Dust PIV 0-10 pin Test Dust (ISO UFTD) TSO T2103 - AT [180 Fine Test Dust PIF Fine Test Dust (ISO FTD) SAE Fine Test Dust TSO 12703-AS [SO Medium Test Dust PTI 5-80 ym Test Dust (ISO MTD) ‘SAE 5-80 sm Test Dust TSO T2108 AF | ISO Coarse Test Dust PTI" Coarse Test Dust (Iso CTD) | SAE Coarse Test Dust PIT Powder Technology, Inc. ‘The new test dusts are currently being manufactured by Powder Technology Incorporated (PTI) from the same silica based material used by AC Rochester so that the chemical characteristics are similar to the AC Dusts. As ‘opposed to the AC Rochester method, PTI processes the Arizona dust with a jet mill and classifies it into woll- Controlled particle size distributions. ISO MTD (ISO 12103-A3) and ISO FTD (ISO 12103-A2) have the closest particle size distribution to AC Fine Test Dust. ‘Although the replacement ISO contaminants are processed from the same silica based material that the AC Test Dusts were processed from, they differ in their particle size distributions. Figure 2 depicts plots ofthe particle size vs, number distribution of the dusts in terms of number of particles in one microgram of dust greater than a given particle size, Note that the number of particles in 1 yg of dust is also equal to the number of particles per mL in a1 g/L suspension of the dust. The partice size distributions reported in Figure 2 are based on measurements with ‘an automatic particle counter calibrated with ACFTD in accordance wrth ISO 4402.10,000 4,000 10 Particles )yg> size Lege. avaea pi Particte size, pm Figure 2, Particle Size Distributions of AG and ISO Test Dusts (Calibration per ISO 4402) Note again that neither ISO MTD nor ISO FTD has a particle size number distribution that is equivalent to that of ACFTD, Both 1d ISO ETD exhibit higher particle counts than ACETD for sizes above 4 about 10 um, Because of these aifferent distributions, all laboratory results when using the new dusts are expected {o vary somewhat from the original results with ACFTD. REVISED APC CALIBRATION PROCEDURE For some time, the industry had become aware thal, when more sophisticated instruments such as scanning electron microscopes are used for analysis of ACFTD, a substantial increase in the numbors of particles are reported compared to the distribution based on optical microscopes, given in [SO 4402. This was especially true for particle sizes below about 10 ym. Therefore a project was started at the National Fluid Power Association to dovelop a new APC calibration method based on a contaminant whose particle size distribution could be traceable to the US National Institute of Standards and Technology (NIST). The result was a now method, ANSV/(NFPA) 72.9.6R1 (1990) [7] that used mono-sized latex particles suspended in oll with sizes traceable to NiST. Usage of this method was strongly discouraged, since, shorty after its introduction, it was found thal poor agreement was ‘obtained between different types of APCs calibrated with latex particles. "APCs made by different manufacturers and APCs ullizing diferent light sources (such as laser dlode or white light) or diferent measurement principles (light scattering or light extinction) yielded different particle count results when analyzing ACFTD or similar samples. ‘This Is due te differences in the optical properties of latex and silica, It was concluded that the APC calibration contaminant should be optically similar to the contaminants typically found in fiter testing and field samples. The 72,9.6R1 procedure has subsequently been withdrawn by NFPA, Based on pravious experience, |SO 12103-A3 Test Dust (ISO MTD) was selected as the best candidate due to the fact that it contains less sub-micron particles, which can cause saturation of an automatic particle counter, and is more easily dispersed. A project was undertaken by NIST in 1993 to cortify the particle size distribution of suspensions of |SO MTD as a reference material to be used for APC calibration. This effort has resulted in the NIST Standard Reference Material SRM 2806 [8] consisting of a 2.8 mg/L. suspension of ISO MTD in MIL-H-5606 hydraulic uid. The cesuls of their analysis show a significant aference in the particle size distbution of ISO MTD. a5 mgnsured wh elgavon microscope NIST) compared to revo Fels wih af APC calbraled wit ACEO Sr ISS IUD Gee Figs 3 The APC data reported Tor TSO MTD in Figure ® ore The avoragos Yom an inlernational round robin test program sponsored by ISO TC131/SC8NVG9,“+ counte by APC (ISO 4402) “@-Elcctrom Microscope (NIST) Figure 3. Particle Counts for [SO Medium Test Dust - ISO 4402 vs, NIST Note from Figure 3 that at sizes below about 10 jum, significantly more particles were observed by NIST than with the ACFTD calibration. The higher particle counts are a result of the enhanced sensitivity of scanning electron microscopy when compared to optical microscopy performed in the 1960s, AL sizes larger than 10 um, fewer particles were observed by NIST. This is primarily because NIST reported the projacled area equivalent diameter Of particles, which is smaller than tho fongost chord dimension used to generate the published ACFTO size distribution. 180 Technical Committee TC134/SC6 Is updating the current APC calibration procedure, ISO 4402, using the Standard Reference Material SRM 2808, The revised calibration procedure, was approved at international ballot, land is now being circulated for a final belot but the ISO standard number has changed from 4402 to 11171 (FDIS. 14171) [9], This calibration procedure includes many other enhancements to ensure better resolution, accuracy, repeatebilly and reproducbilily; howovor, the offect of the now ISO MTD dust and NIST counts will have the largest impact, Also, a new designation, um(c) is recommended for reporting particle sizes using the new ISO 11171 procedure, with the "(c)" referring {o ‘certified’ sizes traceable to NIST, so that the resulls are easily distinguished ‘rom those obtained with ISO 4402, In addition to the new primary calibration procedure, !SO 11171, 1SO is developing @ new procedure, ISO 11943 [10}, for calibration and verification of onine pariicle counting and dilution systems. Although. primarily for automatic ondine particle counting equipment for Mult-pass fiter test stands, ISO 11943 is also applied to on-ine particle counters used for contamination analysis of operating systems inthe fed, REDEFINITION OF PARTICLE SIZES Based on the size dstibutions of ISO_MTD in Figure 3, for particle sizes below about 10_um, the particle size Seteuned Br ST farce one tou (nasal i eter Tan Ue corosponding paride size per So ‘Toone dfeoneeinceases vi decoasiia parle sz—Ae an evargta he aries coun for {ym (80 4402 Ealbration) of about 2000 paricieshig covesponcs (oe paficie ize of 42 um(@) (NIST size dstibuion, Thus, Stor ated To a te na eT or pte Soe fe NIST aloe tipo be chor In Theo Sefiniion 180 4402), og. 2 pmvis 46 amie) ang um 1s 6 yrmfe) Above about 10 pm, the opposite relalionsi Sets ir hat how parle sizes wil bo lower than The old denon. aq. 15-um fs-about 136 umfe) The Telationship between parlcie sizes dofined by the ISO 4402 and NIST size distributions for ISO MID is shown in gure 4, below, andin Table 3, Note that these changes wil nol impact parlicie count ical Microscope, as these counting methods have not changed, Note that because of possible diferences between automatic parlicle counters and the accuracy of their original APC calibration, these relationships may vary slightly and must be determined for each APC to be used. It should also be noted that even though a particle size or count may change because of the new calibration, the actual Contamination level ina system will not be influenced and will remain the samme, 5New size, ume) 1 0 100 (Old size (um) - ACFTD (150 4402) Figure 4, Estimated Changes to Partice Sizes with New Calibration (NIST) Table 3. Comparing Particle Sizes Obtained by Difforent Calibration Methods [AR AGETD Size (180 | Comesponds toa NIST 4402), ym of. | Size (ISO 11171), am(e) fi 70) az 2.0 46 27, 50) 30) BA a3 60) 30) 6a 70, TH. 70.0 os 12.0 115 15.0 136. 755 44.0 20.0 475 25.0 212 30.0, 24.9 20.0) Str 50.0 38.2 00.0 70.0" * based on latex calibration por 180 FDIS 11177 IMPACT OF NEW CALIBRATION ON PARTICLE COUNTS: Particle counts obtained with an APC calibrated with the new ISO 11171 procedure will difer from the corresponding particle counts obtained with the APC calibrated per {SO 4402 at any particular size, Users of particle count data must be made aware of the APC calibration method and how to interpret results when using the ew method as compared to ISO 4402, As a first approximation, historical automatic particle count dala may be ‘converted from ISO 4402 sizes to the new NIST sizes using Table 3, ‘The values in Table 4 illustrate the effect of the (wo calibration procedures on apparent parlile counts. As shown in the table, iF one changes from ISO 4402 to the new NIST calibration without making adjustments to the sizes being monitored, significant differences arise in particle concentrations. For particle sizes smaller than about 10 bm, apparent increases in particle concentration will be reported which could prompt unnecessary action. The magnitude of the difference increases with decreasing patticle size. For sizes larger than 10 ym, the reverse 6‘occurs and apparent decreases in concentration result from the new calibration. Failure to recognize this variation as the result of a change in calibration, rather than a change in contamination level, may lead to misinterpretation of particle count data and inappropriate action. Table 4. Typical Effect of New Calibration on Particle Counts Parle Size | Old Gouni (ACFTD) Now Count (UST) |_um or mic) _| {particles / mL) (particles / mL) 2 4470) 24,900 5 1870 3,400 18 179, 105 25 40 14 CHANGES TO CONTAMINATION CODES CHANGES IN ISO 4406 - The ISO Commitles TC131/SC8 is now issuing for ballot @ modified coding method based on the new calibration procedure. For APC counts, the revised procedure, ISO FDIS 4406.2 [11], uses three code numbers, corresponding to the concentrations of particles larger than 4 pm(c), 6 pm(c) and 14 um) with the ew calibration method. The new 6 ymm(c) and 14 mic) sizes correspond closely to ISO 4402 sizes of approximately 5 jim and 18 jim (see Table 3). These sizes were chosen so that no significant shift in code number ‘occurs due to changes in the APC calibration method. For optical microscopy measurements per SAE ARP598 [12] oF ISO 4407 [13], the calibration is unchanged and the two digits will remain the same as before at § ym and 48 lum. Thus the second two digits of the actual code will be similar regardless of the calibration or measurement technique used. In some instances, companies had proviously adopted for their own use a 3-digit form of the ISO 4406 code similar to ISO DIS 4406:1994 with the third digit representing 2 um. For these companies the change to three digits used in the new ISO 4406:1909 willbe relatively easy because the second 2 digits will In most cases remain the same as before (only the particle sizes are adjusted), and no changes are necessary to specifications. Because the first digit for the new code wil represent 4 jum(c) rather than 4.6 um(c) which more closely corresponds to the old 2 4m, fan increase in the parlcle count and ISO code for this digit will be noticed. Typically this increase will be in the range of 0 to 2 ISO digits. As 2 rule of thumb, specifications should be increased in the first digit by about 1 code level with the second two digits remaining unchanged. For example, a specification of 17/15/12 based on 2, 5, and 48 4m (ACFTD calibration) should be changed to ISO 18/16/12 based on 4, 6, and 14 um(c) (NIST calibration) Figure 5 reflects the Impact on the ISO 4406 code for automatic particle counts conducted on two typical fuid samples, Note that for the new “NIST” calibration per ISO 11171, the cleanliness classes are determined by particle counts > 4, 6, and 14 um(c)._ In each case, the second two digits of the code are nearly identical to the previous 180 4406 2-digit code. Indeed, differences only result when observed particle concentrations correspond to transitions between two code numbers. As a result of the change in size classes with the revised code, the ISO code designation for a particular sampie will nat change appreciably, regardless of whether the new or old calibration method Is used. This permits equipment users and others who specify cleanliness levels in terms of an 180 code to maintain their historical databases and thelr existing cleanliness specifications without change, 400,000 —— ad 1s0.coses 10,000 5 SE set & IN. RE Signo © ooo P| fe g Some ae 2 see fe D asene 4 100 a 2 SR iayaoe 3 hae Figure 5 Effect of Calibration on Particle Counts and ISO Contamination Codes 7CHANGES IN NAS 1638 ~ In order to implement the new NIST calibration with minimal impact on NAS cleaniiness classes, it has been proposed to revise the reference patlcle size ranges in NAS 1638 based on the particle size relationship in Table 3, Therefore the actual particle sizes measured would be similar regardless of the method of calibration. As an examplo, 5 ym por |SO 4402 calioration is replaced with 6 umm(c) per NIST calibration. Because ‘optical microscopic particle counting procedures are not being changed, no changes are required in the cleanliness standards provided the size ranges counted correspond to the original size ranges set by the standards. In addition to the change in particle size, it has also been proposed that NAS 1638 be changed 0 uilize cumulative particle counts, the sare as in SAE AS4059 [14] (also the samo as ISO 11218 [15} Table 5 exemplifies the changes proposed in the NAS 1638 and AS4059 cleanliness classes for APC analysis. It ‘should be noted that, because the frst size range has been extended to 4 jim or 4 um(c), new contamination limits ‘ae included. One additional change, pariiculary for SAE AS4050 is the addition of the >100 pm size or 70 ym(c) por ISO 11171. This is similar to the previous NAS 1638, Table 5. Proposed Modification of NAS 1638 and SAE AS4059 Cleanliness Classes Sie, ROFTD] Tiaimam Contamination Us Catalin, (partes! 100 ml) ronteal [SF 3B] > 15 | =z] >50 [> 100] Regoscope | um | wm | ym | um | am [Sans] St] st] oa] oa] oe 70] sr 2800 | wie) | ume | ume) | wm | ve) | wove CA | Of 300) 82} 780] 304 sa] 70a} 360) 805} F030] 3] 7 3.120] 1,220] 277] sf} 7 af 8285] 2850] a2] 79] 77]. 3] 2 af 72.500} 4,60] — aes] — Te | i Bf 75.000) 8,790] 1,790] 308 a 3 350,000) 79,500] 3.460] 7g "e 7700, 500] 38,600] — 6.990] 7.229 x ‘a{ 200,000) — 77 500] 73.500} 2.485 ai ‘af 400,000) ~7156;000] 27-760) 4.900] 720] 128 Fol 800.000] —317,000] 55,400] 8800] 258 Fi] 7,600,000] 628,000) 717,000] 18,600] 3 TZ] 3.200, 000) 3260, 000} 222,506] 38,200 Toa IMPACT OF NEW MULTI-PASS PROCEDURE ON FILTER PERFORMANCE, REVISIONS TO THE MULTI-PASS PROCEDURE ~ The Mult-pass test procedure, ISO 4572, is currently ‘undergoing substantial changes which, because of the quantily of changes, necessitated a new ISO number. The procedure is currently undergoing ballot as ISO FDIS 16889 [16]. The primary changes in the Mult-pass test from the original version wil be: + Scope applies to fier elements that exhibit an average fitration ratio greater than or equal to 75 for particle sizes lees than or equal to 25 ym. This extends the procedure indefinitely to sizes below 10 jim, which was the ‘minimum of the original procedure. + Anew test dust, ISO MTD (|SO 12103-A1), Is specified as opposed io AC Fine Test Dust. ‘+ The contaminant injection rate (base upstream gravimetric level) is allowed to be one of three conditions below; ‘a change from the single level of 10 mg/L alowed in 1SO 4572, a. Condition 1: 3mgil. b. — Gondition 2; 10 mgit. . — Gondition 3:15 mg/l.Better test equipment validation procedures are incorporated to help ensure test accuracy and repestabily Tighter tolerances are maintained on test parameters. Ondine APC is required; no botle samples are allowed for particle counting purposes. ‘Automatic particle counls are taken at a maximum of one-minute intervals throughout the test Ten average (up and downstream) particle counts are reported based on time with esch point representing 10 % of the total tes ime. + Ten average fitralion ratios for each particle size measured are reported based on the 10 average counts above. + Beta ratios are to be designeted using the symbol “()" to signify they were measured using the 1SO 16989 procedure with NIST traceable APC calibration. A Beta ratio of 200 at 5 ume) would be designated a8 Bs. + Overall average upstream and downstream particle counts are reported, based on the average of the 10 averages above, + Overall average firation ratios are reported based on the overall average counts ometer values corresponding to average filtration ratlos of 2, 10, 78, 100, 200 and 1000 are calculated and reported Filtration ratios (upto values of 100,000 ve. micrometer size are poted Fiation ratios vs. dlta-p are plotted Filvation ratios vs. time are plotted Both retained capacity and total mass injctod are calculated and reported. Flu conductivity is measured and antistatic additive is recommended to maintain conductivity. System volume is allowed toe flexible inthe range of 1/4-12 of the flow per minute valu. Implementation of these changes should result in a more repeatable and reproducible test method. IMPACT OF NEW TEST METHOD ON FILTER EFFICIENCY MEASUREMENTS — Filler efficiencies, reported as filration, Beta, of B ratios are significantly affected due to the changes being implemented in the Mull-pass test method. This is primarily due to the new test dust, use of online automalic counters calibrated to the new NIST traceable procedure, and new methods for calculating average Beta ratios. The impact of the changes will vary for different filers, depending on the influence of the test dust and the dagree of change in the particle size of interest due to NIST calibration (see Table 3). In general, Beta ratios tend to be higher for fers challenged with ISO MTD. than with ACFTD, but the magnitude of shit is unpredictable. Implementation of the new APC calibration procedure wil generally tend to make finer fiters appear coarser and coarse fillers spear finer. It should be pointed out that any changes in performance measurements are merely an artifact of the fest contaminant and Calibration method and are not a reflection of any expected changes in actual fter performance. In fact, unless the fiter design itso is changed, the performance in an operating system wil not change. Tests have been conducted on Pall Industrial Hydraulics standard filer media using the new Mult-pass test method, |SO FDIS 16889, described above, Average Beta ratios obtained and comparison lo results from the previous ISO 4572 method are shown in Figures 6-10. Table 6 shows a summary of the new Pall fiter ratings, um(c), at Beta =2, 10, 75, 100, 200, 1000, and 10000. 10,000 arr) 400 —- 10 1 ° 1 2 3 4 Partlete size, um or pine) Figura 6. Filtration Ratlos for KZ MediaFiguro 7. Fitration Ratios 104 4,000 100 0 ‘uration ratio, Beta 4 Figure 8, Fitration Ratios for KN Media 410,000 1,000 i £100 g E z 1 a 4 68 @ 7 io, nim oF me) for KP Media SE tt so . 2 . 8 we Parts sie, wan oF (6) Figure 9, Filtration Ratios for KS Media 10Filtration rate, Reta ee es 1 we x as Particle size, ni oF m(e) Figure 10. Filtration Ratios for KT Media ‘Tablo 6. New Filter Ratings for Pall Media using ISO FDIS 16889, Pall Mocks um(c) rating for Beta value 2 10. 75 100 200, 4,000_[_ 10,000) 1 = 2 2 2 2.0) 25. 32, KP, 2 2 34 33) 38 30) 67, KN, 24 34 50) 52, 57 7.0) 88, KS, 32 55 | 83 Br ‘a7, 72.0 NIM KT 72 tio 15.8 165 182 22.0 NIM IMPACT OF NEW TEST METHOD ON FILTER DIRT CAPACITY MEASUREMENTS ~ The retained dirt capacity {or a fter element will be affected by the replacement of ACFTD with ISO MTD in the Mult-pass test ISO 16889, since the dusts differ in terms of particle size distribution and morphology. This is shown in Table 7 using results from ISO document TC131/SC8/WG9 N100. In nearly all cases, the resuits for a particular type of filter represent ‘an average of two or three tests. Based on these data, there appears to be no specific trend in retained dirt ‘capacity, Capacity may be somewhat higher or lower with ISO MTD, depending on the specific itor boing tostod. Because each lype of fiter performs differently with ISO MTD, no general factor can be given to convert ACFTD capacttios to ISO MTD capacitios and an actual Mult-pass test must be conducted. Table 7. Impact of ISO MTD Dust on Retained Dirt Capacity, Filler | Measured Rolained Dit Gapaciy, g Beta |AGFID | ISOMTD | Relalive Capacly {ACFTO) (ISO MTDIACFTD) T_T p16 78 a7 089 2 185 209. 118 3 310 394 127 4 [p.=33_ | 322 367 iit S| Bu=t3_| 300 258 0186" 6 | Boats | 865 04 0.93) 7 | Bp= 86 | 547, 07, 0.98. S| Bo=25 | 520 520 100 Subsequent to the data reported at ISO, the author has reviewed comparative results from a number of industrial ‘hydraulic fter elements including those made by Pall as well as other manufacturers. A typical increase in dirt ‘capacity is found when using ISO MTD of about 10% to 40% over values measured with ACFTD. When ISO MTD dir capacity data are desired, because each fiter element performs differently with the new dust, itis necessary to conduct actual tests on every fiter grade.It should be pointed out that a dirt capacity increase or decrease with ISO MTD does not imply that the fiter will exhibit a longer or shorter actual service life. In fact, there will not be a change in field service life as a resull of the laboratory test dust change. Dirt capacity values are too often used to try to predict relative field service life [17]. In reality, field service life is dependent upon a great many factors, and the dirt capacity measured in a laboratory has ile to do with field service io. CONCLUSIONS Changes fo fuid power filtration and contamination standards that use AC Fine Test Dust are necessitated bacause the dust is no longer being manufactured. Four new or revised ISO standards are affected by this change - ISO 11171, [S018889, ISO 11943 and ISO 4406, The new APC calibration method, ISO 11171, utiizes NIST certified calibration samples, instead of an obsolete tost dust. The revised Mull-pass test, ISO 16889, uses this calibration and replaces ACFTD with ISO MTD. ISO 11943 for on-line APC calibration is based on the revised calibration ‘method and validates the entire on-line particle counting system, including a dilution system when used, 1SO 4406, the revised solid contamination level code, uses the new calbration method end adds a third size, 4 yr, to the code for APC results. Sizes for the revised code wore chosen to allow users of the code to retain their historical data base and leave cleanliness level specifications based on the code unchanged. The adoption and implementation of these now and revised international standards will cause the following ‘changes to occur: Particle sizes when using an APC calibrated with NIST certified ISO MTD will change; Particle counts when using an APC calibrated with NIST certified ISO MTD will change; ‘+ Cleanliness codes wil change unless APC reported particle sizes are adjusted in accordance with the revised 180 4406:1909 or the proposed revised NAS 1638. Reported filter Bela ratios will change when using the SO 16889 Mult-pass test method; + Reported fter dirt capacities will change when using the ISO 16889 Mult-pass test method, ‘The projected changes in each of these parameters have beon discussed in this paper. Users should always keep in mind that although laboratory performance values wil be changing, this is an artifact of the testing only, and fier performance and actual contamination levels in the fleld wil romain the same as before. REFERENCES 4. 180 4402, "Hydraulic fuld power - Calibration of automatic-count instruments for particles suspended in liquids - Method using classified AC Fine Test Dust contaminant’, 1991, 2, 1S0 4406, "Hydraulic Muid power - Fluids - Method for coding level of contamination by solid particles", 1987, 3, DIS 4406, "Hydraulic uid power - Fluids - Code for defining the level of contamination of solid pattcies, Revision of frst edition (I80 4406; 1987)", 1994. 4. NAS 1638, “Cleanliness Requirements of Parts Used in Hydraulics Systems", 1992. 5, 1S0 4572, "Hydraulic fuld power -Fiters - Mult-pass method for evaluating fitration performance", 1981. 6. 180 12103, "Road vehicles - Test dust for fitor evaluation - Part 1: Arizona test dust’, 1997. 7. ANSI(NFPA) 72.9.6 R1, “Hydraulic fluid power - Calibration method for liquid automatic particle counters using latex spheres", 1990. 8, National Insitute of Standards and Technology - Standard Reference Material 2606 - Medium Test Dust (MTD) in Hydraulic Fluid, 1997. 9, ISO FDIS 1171, "Hydraulic uid power ~ Calibration of liquid automatic particle counters", 1999. 40. 1S0 FDIS 11943, * Hydraulic fuld power -on- calibration and validation’, 1999, 2 liquid automatic particle-counting systems - Methods of 211, ISO FDIS 4406.2, "Hydraulic fluid power - Fluids - Code for defining the level of contamination of solid patticles”, 1999, 12, SAE ARP508B, “The Determination of Particulate Contamination of Liquids by the Particle Count Method", 1986, 13. [SO 4407, "Hydraulic uid power - Fluid contamination - Determination of particulate contamination by the counting method using a microscope", 1991 14. SAE AS40598, “Aerospace - Cleanliness Classification for Hydraulic Fluids", 1992. 18, ISO 11218, “Aerospace ~ Cleanliness classification for hydraulic fuids", 1993. 416, ISO FDIS 16889, “Hydraulic Muid power -Fiters - Multipass method for evaluating fitrtion performance of filter element 47. Bensch, L. E, ‘Dit capacity: The overrated fter rating factor’, Machine Design, June 23, 1983. B