Classification, Control and Regulation of Operating Fluid and Oil Purity

One of the main reasons for failures of modern equipment is contamination of operating fluids.  Failures are the result of the widespread use of contaminated operating fluids in manufacturing, and operation of oil filled equipment.

Up to 90% of hydraulic drive failures occur to some extent due to contamination of operating fluids by particulate matter.  The situation is further complicated by the variety, nature, quantity, and quality and the impact of the contaminant on the parts and components of the machinery.

A strong recent tendency is the growing power and speed loads on components and assemblies, as well as the requirements of reliability and length of service life.  Hence, the increased sensitivity of equipment to contaminants.  At the same time, it is necessary to minimize all costs related to design, construction and operation.

The optimal solution to purity requires a custom approach to individual machine systems.

Maintaining reliability of various equipment systems requires adequate diagnostic skills of service personnel. Each worker must not only possess the required knowledge, but also increase that knowledge by practical experience and validate the required qualifications.  This, in many cases, can be problematic.

Classification of oil and operating fluid quality is very important. To detect contaminants, they should be duly classified.  Systematization allows for classification, i.e. division of contaminants into classes according to their characteristics.

For the latter, both quantitative and qualitative indications may be used because only their combination can give a full and objective picture of the problem.  It should be noted that today’s methods of research uses quantitative indications first and qualitative indications second since the latter are more varied and complicated and require time for analysis and dissemination.

As of today, there are several purity classifications described in the corresponding standards and regulations: GOST 17216, NAS-1638, SAE, company «Cincinnati».

Table 1

Classification of Operating Fluid Purity

Liquid purity class Number of contaminant particles in 100±0,5 cm3 of liquid at maximum particle size in microns Max contaminant weight, %
From 0.5

to 1

From 1 to2 From.2 to


From. 5 to10 From. 10 to25 From. 25 to50 From. 50 to100 From. 100 to200 Fibers
00 800 400 32 8 4 1 None AA AA Not regulated
0 1600 800 63 16 8 2 None
1 Not regulated 1600 125 32 16 3 None
2 Not regulated 250 63 32 4 1
3 Not regulat-ed 125 63 8 2
4 250 125 12 3
5 500 250 25 4 1
6 1000 500 50 6 2 1 0,000032
7 2000 1000 100 12 4 2 0,000064
8 4000 2000 200 25 6 3 0,000125
9 8000 4000 400 50 12 4 0,00025
10 16000 8000 800 100 25 5 0,0005
11 31500 16000 1600 200 50 10 0,001
12 63000 31500 3150 400 100 20 0,002
13 63000 6300 800 200 40 0,004
14 125000 12500 1600 400 80 0,008
15 25000 3150 800 160 0,016
16 50000 6300 1600 315 0,032
17 12500 3150 630 0,064


  1. None — the condition of the liquid when no particles of a given size could be detected in the sample, or the number of particles detected in multiple samples is less than the number of samples.
  2. AA — Absolute absence of contaminant particles.
  3. The dependency of the liquids purity on the weight of the contaminant with consideration of the number of contaminant particles in the liquid is a reference value. Weight is given for particles with average density 4х103 kg/m3 and liquid density 1×103 kg/m3.

The type of the classified indication is defined by the subject. Since the subjects are dispersed systems (aerosols, suspensions) or their dispersed phases (sediment, centrifugate, micropowders, dust etc), the main classification indication is the size consist of the contaminant. It allows to express purity class through the amount of contaminant in accepted size groups in a unit of volume of the studied liquid.

The sequence of classes in such classifications is defined by incremented parameter principle.

Table 2

Industrial liquid purity classification (NAS-1638)

Purity class Number of contaminant particles in 100 ml of liquid at maximum particle size in micron
5-15 15-25 25-50 50-100 >100
00 125 22 4 1 0
0 250 44 8 2 0
1 500 89 16 3 1
2 1000 178 32 6 1
3 2000 356 63 11 2
4 4000 712 126 22 4
5 8000 1425 253 45 8
6 16000 2850 506 90 16
7 32000 5700 1012 180 32
8 64000 11400 2025 360 64
9 128000 22800 4050 720 128
10 256000 45600 8100 1440 256
11 512000 91200 16200 2880 512
12 1024000 182400 32400 5760 1024

The current classifications of contamination thererfore, consist of consecutive dispersion sequences based on geometric progression with a module of two (2). The same principle is used for building the scale of particle size or fractions.

Since the consecutive parameter values are a geometric progression, the principle of universality for development of classification scales is justified.  Similar cascades with doubling or graduation of limiting values also exist in evolutionary systems.  It should also be remembered that terms of the geometric progression with module 2 correspond to the numbers of R10 row, which is used as the base in machine building. The nominal operating parameters of hydraulic drives (volume, pressure, mean consumption, rotation speed etc) are corresponded to a certain number from the R10 row.

The uniformity of the contamination is defined by its quantity, while the quality, in turn, shows non-uniformity.  The latter quality is a structure of serious influence on abrasive wear of machines and the procedure of industrial fluid purification.

Parallel existence of several classifications of liquid contamination is due to the complexity and variety of quantitative indications. They are however, common in accepting contamination classes according to granulometric composition.  This means that contaminants are classified by quality and may vary only in terms of size-consist (quantitatively).  E.g, GOST 17316 «Industrial purity; Classification of fluid purity» suggests to classify contaminants based only on their quantity, which only allows for consideration of size and consists of single granulometric composition for all classes.

This composition is quite coarse compared to similar contaminant characteristics used in existing classifications.

Appearance of a coarse qualitative parameter as the standard at the time was due to the need to prevent inconsistency of production and products in meeting the new standard requirements.

Table 3 demonstrates the differences in the current classifications of contaminants. The analysis shows significant variations in contamination limits from developers of each classification. This is due to the physical properties, conditions and the dynamic of change of the contaminants for each specific type of machinery.

Table 3

Classification Granulometric characteristic (% content) of contaminant classes in particle size groups, micron Reduction ratio Note
5-10 10-25 25-50 50-100 100-200
Gost 17216 64,2 32,1 3,2 0,4 0,1 2 For all 19 classes
«Cincinnati» 72,7 23,7 3,2 0,34 0,03 3 For all 12 classes
SAE, ACTM, AJA 72-78 23,7-24 2,7-3,7 0,45-0,55 0,03-0,05 3-4 For all 7 classes
NAS-1638 5-15 15-25 25-50 50-100 100-200 5,7 For all 14 classes
82,3 14,5 2,63 0,49 0,008
Termal Control 5-10 10-25 15-25 25-50 50-100 100-200
59 26 10,6 3,7 0,5 0,2 1,6 1st classes
90,5 7,4 1,6 0,4 0,06 0,04 10 2nd classes
92,0 6,8 0,9 0,2 0,03 0,01 12 3rd-4th classes
93,7 5,6 0,6 0,16 0,01 0,005 15 5th-9th classes
ISO-4406 5-15 Above 15
87,0 13 (11)* 8 For all classes
93,6 6,4 (5,8) 16
96,8 3,2 (3,0) 32
98,4 1,6 (1,5) 64


  1. The number in parenthesis shows the amount of 15-25 micron.
  2. ISO–4406 is a reference for classification.

The data shows that a complete picture of fluid purity requires knowledge of the size consistency as well.

The data also indicates that in the general case presence of particles larger than 25 microns among the contaminants is below 4%.  The percentage therefore, of particle sizes 5-10 and 10-15, or 5-15 and 15-25 can be quite high. It is important to know that 10-25 micron sized particles are considered the most dangerous for modern machines.

The granulometric characteristics in practice is shown qualitatively by a reduction ratio equal to the amount of particles between 5 and 10 microns or between 5 and 15 microns to the amount of particles between 10 and 25 microns or between 15 to 25 microns.

It is now possible to classify contamination in accordance to the reduction ratio Ki, also given in table 3.

The lower the ratio, the less uniform and low dispersive the contaminant is and vice versa.

The GOST 17216 standard classifies dispersions with a reduction ratio equal to two (2), which is the lowest indication.  It is one for all classes, which makes the existing classification rather coarse.  This became the reason for the difficulties encountered in the estimation of contamination according to this classification.  In some cases, classification was impossible.

The differences in granulometric composition of contaminants are caused by the nature of the original composition on the one hand, and by mechanical crushing, saturation with products of wear, filtration or separation of operating fluid on the other hand.

Therefore, the current classifications consider different evolutionary stages of contamination, characterized by several levels of quality (Termal Control, ISO–4406).

1st limit (Ki=2). Contaminants are the result of soil dust entering the fluid, which is the first and regular source of contamination. This level can be used to characterize liquids in storage, in transportation, filled into machines without prior purification or after purification with coarse filters.  Hence the rule that fresh operating fluid must always be treated as contaminated and should only enter the system through filters or power purifiers.

2nd level (Ki=4). This level is common for liquids during filling and operation in drives with power purifiers. For this case the classifications NAS, SAE and “Cincinnati” were developed.

3rd level (Ki=8). This is common for liquids operated in hydraulic systems without filters or with coarse filters. In this case, a rather intensive dispersion of dust particles occurs.

A special physical model of contaminant granulation was adopted as a basis for the 3rd level. It is based on the assumption that in a system without a separator or a filter, ideal progressive dispersion occurs, i.e. all particles are crushed mechanically in all size groups, without shards. The probability of destruction of each particle at any stage of granulation is considered constant and not dependent on size or the presence of other particles. Influence of other factors on the presence of particles is excluded.

In most existing classifications, the rows of size groups or classes are arranged in geometric progression with module 2.  For the purposes of control, particles are included in the group in the order of the largest size.  Since any particle is in reality a three-dimensional body, it can easily transfer to a lower group when granulated into 8 parts (23).

4th level (Ki=16). This level is determined mostly by a temporary change in the selective ability of the filters when the liquid is operated in hydraulic systems equipped with fine filters (25 microns and less).  It is obvious that classification by dispersion for each level and conditions of use is required.

Classification of fluids according to GOST 17216 complies with the conditions of their use in accordance with the 1st level.  At the same time, other levels, shown in Figure 4, correspond to GOST 28028 classification.

Table 4

Conditions of use (limit level) Granulometric composition of the contaminant, % Granulation factor, Ki
Particle size, micron
5–10 10-25 25-50 50-100 100-200
No less than No more than
1 64,2 32,1 3,2 0,4 0,10 2
2 78,1 19,5 2,15 0,25 0,05 4
3 87,5 10,93 1,37 0,17 0,02 8
4 93,3 5,85 0,75 0,11 0,01 16

Table 5 contains examples of possible granulation factors at certain conditions. Using special devices, the dispersion of contaminants in the fluid is determined. The corresponding purity class is selected according to the method given in GOST 20028, which implies determining the amount of solid particles with the size of 10 to 25 microns.

Table 5

Oil storage and use conditions Number of oil samples Quality and quantity of oil contaminant indicators (ranges)
Purity class Ki
1. Oils stored and filled into hydraulic drives of tractors at assembly line 174 4-11 0,8-22
2. Hydraulic drive of combine harvester during running test 62 8-12 3-18
3.Hydrostatic transmission of a combine harvester during running test 62 6-9 2-8
4. Hydraulic drive of casting, forging and pressing machinery of a vehicle factoryduring purification by:
– filter 427 9-16 6,5-60
– centrifuge 6-19 8-398
5. Hydraulic drive of processing centers of machine building factory 84 5-11 2,8-27

Careful analysis of granulometric composition of particulate matter in samples from table 5 shows that the above levels of granulation factor can be used in practice.  Tracking of dynamics of change of this parameter along with the control of purity class change, produces a more objective estimation of contaminating and purifying processes in fluids.  This significantly improves and increases the efficiency of control of industrial purity of operating fluids.

With the data from tables 4 and 5, oil contamination index (РТМ2-Н90-2-76) and contaminant weight (GOST 17216-2001) can be controlled.  It is easy to notice that the correlation between purity classes in terms of dispersion, granulation index and contaminant weight, given in GOST 17216, cannot be confirmed.

This is easy to see. For instance, every purity class according to contamination index (from 10 to 15) correlates to 4 classes of purity for particle size group class, having sizes from 10 to 25 micron depending on granulation index. Each purity class is correlated with up to 7 classes (from 7 to 13) of the 10-25 micron size group.

The results, therefore, show that controlling contamination of operating fluid by weight and contamination index is almost impossible in practice.

In most cases, the inconsistency of purity classes, obtained by using various methods and approaches, is due to the differences in measurement and reflection of dispersion and granulometric consistency of contaminants. Therefore, purity classes from different classifications must not be correlated.

Since NAS-1638 purity classes are very widely used in the world.  One of the examples is the assumption of class 11 in GOST 17216 correlating to class 7 in NAS-1638. There is a catch however.  Though the size groups of 5-10 and 5-15 microns in both classes are similar, the amount of contaminants of the size groups 10-25 and 25-50 microns in liquid with purity class 11 can be 3 or 1.5 times higher than in liquid with purity class 7.  The impact of such particles on machines will not be equal. It is possible to misrepresent a higher oil contamination in terms of quality and quantity by a lower oil contamination when indicating class 7 according to NAS-1638 instead of class 11 according to GOST 17216.

The conclusion is that control and classification of contaminants’ quality is no less important than quantity. Also, the importance of quality control and classification along with quantity is indirectly implied by methods developed in ISO 4406-87 and ISO 4406-99.

If contaminants are only classified by quantity, and quality is constant, codification of contamination should be done with one classification number. Accordingly, classification, which uses both quality and quantity indications, can be coded by at least two classification numbers. ISO 4406 correlates each particle number range with code number (table 6).

Table 6

Number of particles in 1 cm3 Classification number Number of particles in 1 cm3 Classification number
from To (including) from To (including)
2500000 Abve 28 80 160 14
1300000 2600000 28 40 80 13
640000 1300000 27 20 40 12
320000 640000 26 10 20 11
160000 320000 25 5 10 10
80000 160000 24 2,5 5 9
40000 80000 23 1,3 2,5 8
20000 40000 22 0,64 1,3 7
10000 20000 21 0,32 0,64 6
5000 10000 20 0,16 0,32 5
2500 5000 19 0,08 0,16 4
1300 2500 18 0,04 0,08 3
640 1300 17 0,02 0,04 2
320 640 16 0,01 0,02 1
160 320 15 0,00 0,01 до 1

A contamination code is selected by the first classification number according to the general volume of particles in the contaminant with the size above 5 microns.  Then the second classification number is selected based on the amount of particles in the contaminant with the size above 15 microns.  Then the two numbers are written in one line separated by a slash.

Example. A classification number is 18/13.  This means that a 1 ml volume of the liquid containts from 1300 to 2500 particles with the size above 5 micron and 40 to 80 particles with the size higher than 15 microns.

If a granulation ratio is used, the code of contamination may be recorded in a more informative way such as 15-5.2.  This means the following: 12 is the purity class in terms of the amount of particles with sizes 10-25 microns, according to GOST 17216, and 5.2 is the amount of granulometric content according to the granulation index Ki.

In the mid 1990s it became apparent that the principles of contamination classification used in GOST 17216-21 are not fully adequate to the growing requirements of control, regulation and ensuring purity of operating fluids.

Improvement of the regulations on industrial purity is also important. A norm is a planned technological and economic parameter containing optimal mean indications such as consumption rate and flow rate.  These accepted qualitative and quantitative norms express allowable parameter values.  Norms are an economic category indicating the actual level of technology development. Since the technology keeps developing, the norms must be changed from time to time to stimulate increasing quality of the existing equipment.

Often industrial purity norms and industrial purity requirements are mistakenly thought to be identical. This however, this is not always true.  Sometimes, the requirements for industrial purity of components are different from those of the complete system.  The latter must generally comply with the established norms. Requirements for industrial purity are more dynamic than the corresponding norms.

There are several recommendations today regarding industrial purity of fluids in hydraulic drives.  For instance, the Pall company recommends that clients never use hydraulic systems with contamination worse than level 16/13 according to ISO-4406.

Some of the recommendations by HYDAC for filter selection for general purpose industrial hydraulic equipment are given in table 7.

Table 7

Hydraulic drive Purity class Recommended absolute, micron
NAS-1638 ISO4406
Gear pump 10 19/16 20
Safety valves
Piston pumps 9 18/15 10
Vane pumps
Pressure valves
Servo-valves 7 16/13 3

A qualitative characteristic of allowable contamination levels is common.  For example, to maintain purity at the level of 11-12 class according to GOST 17216-71, nominal 10 micron filters must be used; this will ensure operation of hydraulic equipment without significant reduction of performance. There is data in the literature showing that if nominal filtration fineness is improved from 25 to 5 microns, it is quite possible to reduce intensity of efficiency drop of aviation pumps and hydraulic motors by 7-8 times, and to increase their effective service life. At the same time, perpetual machines are not always desirable. Since in each specific hydraulic drive a certain optimal level of purity should be maintained according to economic necessity.  An abstract statement such as “the better purification, the better reliability” should be replaced by a more practical one, such as “purity class and filtration should be as good as is necessary in the specific situation.”

The British Hydraulic Power Association developed a method of selecting filter equipment.  It is based on the so-called weight factor which considers seven technical and economic parameters. These include sensitivity of equipment to contamination, as well as the projected service life of the equipment.

Currently, the experience of operating hydraulic equipment is sufficient to include several factors for optimization and requirements of oil purity.  This can best be illustrated by effective pump service life.  Since optimization is not the purpose, but only a tool to extend pump service life, it is necessary to estimate the potential service life of the equipment.  Such an estimate can be made using certain assumptions.  Thus, the wear of friction parts is directly proportional to the load.  This statement is the basis of estimating the service life of many machines and mechanisms.  Of course this is more important for hydraulic drives of machines, agricultural and road construction machinery.  The above systems are mass produced and must be compliant with the strict requirements of unification and standartization of components and parts. The types of positive displacement hydraulic equipment are physical type-sizes. The main parameter of these machines (volume) in the type-size sequence, is presented as geometric progression with module 2. They are also grouped by rotation speed, type of oil used, operating temperature, filtration fineness, and industrial purity.

Since such hydraulic machines are tribosystems, many of their operating parameters are identical. The differences are mainly in relative velocity (friction) and proportionate volume. Therefore the estimated service life of hydraulic machines of one type size group should be calculated based on the similarity principle and other such parameters as volume, speed ratio, power ratio or load factor.  The specific parameter configuration depends on which parameter of the type size group is assumed constant.

The source data for estimation of service life expectations of hydraulic pumps G12-2M and G12-3M are given in the reference book “Machine Hydraulic Drives” by Sveshnikov.  The method of calculation is given in table 8.

Table 8

Parameters Weight/dimensions
1 2 3
Type-size (model) of the pump
G12-3 G12-2
m , kg 8,2 20 30
q , cm3 8 12,5 16 25 32 40 63 80 125 160 224
D = q1/3 2 2,3 2,5 2,9 3,2 3,4 4,0 4,3 5,0 5,4 6,0
nnom , min-1 960
Tnom , hour, minimumfilter 40 micron 7000 3000 2000
Cn=nхD Cn»Dх103
CT = Tnomх Cn 23,8х106 15х106 12х106
Tpr = СT /Cni(х103) 12 10 9,5 8,3 7,5 7 3,8 3,5 3 2,2 2
Tnom , hour, minimum filter 25 micron 10000 4000 2500
CT = Tnomх CN 34х106 20х106 15х106
Tpr = СT /Cni(х103) 17 15 13,8 12 10,5 10 5,0 4,7 4,0 2,8 2,5

The result of the calculation shows that each of the pump type size groups has its own service life expectency.  All other parameters being equal. For example, the G12-31АМ pump of the І size-weight group has an expected service life of approximately 17,000 hours.  In turn, basic pump G13-33M has an expected service life of only 10,000 hours.  It is the latter value that is taken as a norm. This means that requirements of the purity of the operating fluid of the first pump can be significantly lowered.

Currently available data shows the full importance of filtration fineness for equipment service life expectency.  The graphical representation of estimated life cycle is shown in Figure 1.

Fig. 1. Influence of nominal filtration fineness of operating fluid on pump lifetime depending on nominal power

The graph shows that if the filter with nominal filtration of 25 microns is replaced with a 10 micron filter, the expected service life of the G12-33M pump can be extended by 10,000 hours from 10,000 to 20,000 hours.

Presentation of available data in the form of graphs allows to set norms and requirements more objectively for each pump load mode.  Another facto,r which should be considered when optimizing norms and requirements to liquid purity, is the volume.

In case of a decrease or an increase of contaminant content in the constant volume of fluid, the concentration also changes.  Not only the total amount of contaminants, causing wear of the pump, but also the amount of that part of the contaminant that can enter the contact area of a friction pair at any moment in time.  In the opposite case, with the change of the volume of operating fluid with a constant concentration of contaminants, there will only be a change in the total amount of contaminating particles causing pump wear, since the amount of the contaminant able to enter the contact area remains the same.

Fig. 2 (a, b) shows examples of influence of contaminants and the volume of the liquid with constant level of contamination on the reduction of the relative volumetric efficiency of gear pumps.

Fig. 2. Influence of contamination and load on gear pump wear

The graphs show that increasing concentration of the contaminant with a constant load leads to reduction of contact tension caused by abrasive particles. The curve is parabolic.

Fig 2.c shows the dependency of the changing pump supply factor on load mode. The test conditions were as follows: nominal pressure 10 MPa, operating liquid contains 0.08% of M20 contaminant. The graph has a complex nature, close to exponential. The extreme point is located at the point of kz=0,75.

Pump loading was cycling. Cycle duration was six seconds. In such conditions, the extreme point for the supply factor is 1.5 seconds of pump operation. Further reduction of cycle causes the contaminants in the contact area to remain the same, reducing the wear of the friction surfaces.

Increasing operating pressure (load on the pump’s parts) causes proportional increase of parts wear, but after passing a certain critical point, the intensity of wear starts to slow down (Fig. 2(g)).

For the purposes of this experiment the critical operating pressure when injecting the artificial contaminant is 3 MPa.

The results shown prove that the current recommendations to use absolute filters for absolute purity are interesting from the technical point of view, but dubious from the economic standpoint.

The necessity of using any method is largely determined by the amount of money the customer is willing to pay for the achieved result.

See more video about transformer oil purification

обновлено: July 1, 2016 автором: dannik