Diesel Engine Fuel Filters

It is a common belief that the engine fuel filter, installed by the engine manufacturer, will provide the required level of contamination protection in order to achieve reliable operation. This is a fair and reasonable assumption because the intended function of a diesel engine fuel filter is to “achieve a fuel cleanliness that will enable the fuel injection components to function reliably and within their designed operating parameters throughout the life of the component.” However, there are several key aspects regarding the application of engine fuel filters that call into serious doubt their ability to achieve this result or the more stringent cleanliness level being sought by HPCR fuel injector manufacturers.Most major engine manufacturers have determined that fuel must meet or exceed a cleanliness level of ISO18/16/13. The Worldwide Fuel Charter (WWFC) also supports this. It is therefore fair and reasonable to assume that the engine fuel filter must have improved in order to achieve the fuel injector OEM required cleanliness levels of ISO12/9/6.

A typical passive diesel fuel filtration system installed on a HPCR injected engine uses both a primary filter and secondary filter to achieve the desired cleanliness level. The primary filter is typically installed on the suction side of the low-pressure fuel pump, with the secondary filter on the pressure side. Again, manufacturers will modify these designs depending on requirements of the engine and their research into what they believe works best.

The primary filter is generally required to remove larger particles that can damage the low-pressure pump and also separate undissolved water from the fuel. The secondary filter is generally designed to remove the smaller particles, around 4μm, which are known to damage the downstream engine components such as the high-pressure fuel pump and the fuel injectors. Some new filter designs use a single-stage system that incorporates the water removal capability of the primary filter with the high efficiency particle removal of the secondary filter.

Two-stage systems sometimes use a coalescing or silicon treated filter for water removal and a surface filter for particle removal. The two-stage system typically utilizes media derived from cellulose or a cellulose/glass composite. The secondary filter can be a finer cellulose composite media, or in the most advanced systems now being used on HPCR injection systems, a fully synthetic glass fiber filter media.

Single-stage systems typically use a single filter with a multilayer composite structure. The current state-of-the-art filter designs now use a melt blown glass media formed on or laminated to, a nanofiber substrate support and typically bonded with a Phenolic or Epoxy resin.

Synthetic wetlaid microglass filter media

Synthetic wetlaid microglass filter media

When referring to diesel fuel cleanliness levels and the cleanliness code ISO4406:1999, the most critical part of the code is the first digit (i.e. 12). This is the digit that represents the concentration of 4μm, and larger, particles in the fuel, which are the most damaging because of their relatively close size to clearances in the HPCR injector and HP fuel pump.

For a diesel engine fuel filter to remove contamination from the fuel system it must be very efficient at and below 4μm. The efficiency of a filter is typically reported as a Beta Ratio and it is the recognized industry standard for reporting filter efficiencies. The Beta Ratio is defined as the difference between the number of particles larger than a given size upstream of a filter to the number of particles of the same size downstream of the filter. More simply defined as:

beta-ratioFilter Beta Ratios are tested for using the ISO16889 Multipass test standard.  The test system is represented below.


ISO16889 Multipass Test Circuit

Interestingly, some filtration suppliers will choose not to report the filter efficiency as a Beta Ratio. They instead prefer to report the efficiency as a percentage (%) efficiency. The efficiency of the filter can be calculated directly from the Beta Ratio:

efficiency-formulaFor example, the % efficiency of a filter with a Beta Ratio of 10 at 4μm is calculated as follows:

efficiency-sampleSince 90% seems like a relatively high percentage, the casual observer may be drawn to the conclusion that the filter is highly effective. In the filtration world, this is not the case.  In terms of the quantity of damaging particles passing through the filter, the differences in efficiency percentage are significant and have enormous implications in terms of wear on an engine’s fuel system. The table below identifies the two reporting methods and their corresponding data. The example assumes a total of 100,000 x 4μm particles challenging the filter.

Filter Efficiencies

Traditionally, diesel fuel filters utilized in EUI injection systems have had efficiency requirements of Beta 75 at 10μm (β10≥75) or 98.7% efficient. The latest HPCR injection systems now require Beta 1000 at 4μm (β4≥1000), or 99.95% efficiency in order to obtain a far cleaner fuel downstream, protecting vital components.

The Multipass ISO16889 test method was primarily designed to test the efficiency of filters that were intended for recirculating systems, such as those found in oil based hydraulic and lubrication systems. As such, the ISO16889 test system recirculates the entire volume of fluid, passing it through the filter multiple times. Diesel fuel filters installed on engines also use this test method to determine their efficiency. The problem with using the ISO16889 test procedure for inline engine filters is that an engine fuel system does not recirculate the entire volume of fuel. A filter within a diesel engine fuel system is considered as a “single pass” filter application. The filter only has a single chance to capture contamination before the fuel moves on to the injectors and engine.  This is a much different application than the ISO16889 test, whereby the filter has multiple chances to capture contamination until its terminal differential pressure is reached.

Reporting a diesel engine fuel filter efficiency, using the Multipass ISO16889 standard, is somewhat deceptive as to its actual performance in the application, because the test does not accurately model how the engine fuel filter is used in a real world environment. The reported Beta Ratio, or filter efficiency, can be misleading to a prospective client regarding the actual performance they will see in a diesel engine application. The efficiency of a filter in a single pass application, such as an engine fuel system, is vastly lower than a traditional Multipass application as modeled in the ISO16889 test. Many engine filtration suppliers and manufacturers, however, actively market their products to the contrary by suggesting that the engine fuel filters have a high Beta Ratio and will thus achieve a low cleanliness level, or perform at the ISO16889 standard as well as achieving a low cleanliness level. Little or no information is provided to the end user on what the performance of the product will be when installed on the actual engine. What must be understood is that a diesel engine fuel filter, installed into an actual diesel engine fuel system, does not perform even close to the efficiency levels at which are advertised.

There are numerous other test procedures that are sometimes used to test fuel filters under different conditions and with different contaminants; they all, however, cause additional confusion for the end user and do not accurately simulate a real world installation. One such test is the ISO19438:2003 test. This test was specifically designed to evaluate the performance of internal combustion engine fuel filters with a constant flow rate of between 50 liters/hour and 800 liters/hour. This test is again a Multipass filtration test, with a constant flow rate, something that an engine fuel filter will typically not see in the real world. Other test methods known to exist are as follows:  While the reporting of Beta Ratios or % efficiency remains widely accepted as the standard method for reporting filter efficiencies, even the most reputable manufacturers of engine fuel filters may not always employ it as a means of advising their customers to their filter performance. In some cases it is almost impossible to find the information in marketing materials or technical documents.

There are many manufacturers that prefer to use slogans or descriptions rather than hard data to advertise filter efficiencies and would rather rely on the strength of branding as a reputable means of advising customers of the filter performance. Slogans such as “Ultra High Efficiency” or “Advanced” are often used to indicate that the filter has an extreme level of performance above and beyond a standard product. However, it is only when the test data of the filter is observed that the slogan becomes appropriate to use. One particular filter used in this example, having been advertised as “Ultra High Efficient,” actually has an efficiency of 98.7% above 4μm (β4>75) according to the ISO16889 Multipass test. Clearly this is well below the required level of filtration efficiency for a HPCR engine fuel system and well below what a customer would expect from an Ultra High Efficiency filter.  Testing of engine fuel filters under actual engine operation conditions can show a dramatic difference to the efficiencies that are reported from laboratory testing such as the ISO16889 Multipass test.

With engine based fuel contamination control, it is important to understand a few issues which are often forgotten or overlooked. The first issue is that of the fuel tank and the cleanliness level of the fuel within it. Once fuel enters the fuel tank, tests have shown that the cleanliness level can rise by 1-2 ISO codes, due to the contamination left in the tank or generated during operation. Tests have shown that cleanliness levels prior to the Primary Fuel Filter can be on the order of ISO22/19/16 when ISO18/16/13 cleanliness fuel is supplied from a dirty tank.  Fuel flow through an engine filter is not consistent, however, most test methods do not simulate this during filter element tests. In a real world application, as engine speed increases, so too does the velocity of fuel flow through the filter. As such, engine fuel filters are continually challenged by flow surges throughout their life cycle, and because filtration efficiency is directly affected by the flow of fluid through the media (flux rate), flow surges can dramatically affect the ability of the fuel filter to remove contaminants or retain previously captured particles. This issue becomes even more challenging with particles below 4μm. The figure below represents the effect of flow surges on a typical diesel engine fuel filter installed on a mining haul truck. The rapid rise and high quantity of 4μm particles (shown in blue) is a direct result of increased vehicle acceleration and fuel flow through the filter. During these periods, the level of contamination downstream of the filter was 10 times dirtier than that under steady low flow conditions and at its peak reached a cleanliness level of approximately ISO26/13/11. A key point to note here is the enormous gap between the 4μm code and the 6μm code. This gap clearly shows that particles around the 4μm size range are dramatically affected by engine speed and fuel flow. This is a primary concern because these particle sizes cause the most damage within the fuel system. Such massive changes in downstream fuel cleanliness levels are alarming.

Effect of Flow Surge on Diesel Engine Fuel Filter

Effect of Flow Surge on Diesel Engine Fuel Filter

Existing fuel filter test standards do not yet consider flow surges, however, the proposed ISO DIS 23369 Multipass method of evaluating filtration performance of hydraulic filters under cyclic flow conditions may provide a more accurate means of measurement. It is, however, not yet widely used.  Vibration is an additional cause for concern to the efficiency of engine fuel filters.  As fuel filters are typically mounted directly to the engine, they experience extreme levels of high frequency vibration under highly variable conditions. These include the actual engine vibration due to engine operation and the movement of the vehicle over rough terrain. Vibration is known to directly affect the contamination retention capabilities of the filter media, yet little has been done to simulate these issues with suitable testing procedures or standards.

vibrationAccording to Cummins, the figure above illustrates the effect of vibration on contaminant removal in a laboratory environment. The test monitored upstream and downstream contamination levels of a low Beta Rated filter while subjecting it to random vibration at known, constant levels of acceleration. The results indicated that the vibration adversely affects particle removal efficiency, with the Beta Ratio of the filter decreasing with higher levels of vibration. Interestingly, the analysis also shows that the steady-state Beta Ratio (at a given acceleration) decreases with increasing magnitude of the acceleration. This test clearly identifies that contaminant removal data obtained using existing filter test standards, in the absence of vibration, is an overestimate of the removal efficiency under real world conditions. The illustration above also shows two dips in the data where the Beta Ratio actually drops below a Beta value of 1. This is clear evidence of a filter unloading more contamination than being captured when certain conditions are met.

Further evidence supporting the detrimental effects of flow surge (acceleration) and vibration can be observed from tests conducted on a Caterpillar 793C haul truck at a large Australian Iron Ore operation. This test utilized an online laser particle counter to monitor cleanliness levels from the engine fuel system, and also before the and after the dispensing fuel filter on the bowser. Photos are referenced below with the following table illustrating the results following testing.  The data in the table above identifies that contamination levels in the 4μm code rose from a code of 13 when filling the truck with fuel to a code of 16 in the engine fuel system at 1500RPM and after the engine fuel filter. This represents an increase of 3 ISO Cleanliness Codes or 8 times dirtier fuel.










To put this into perspective, for an engine fuel filter to achieve a cleanliness level of ISO12/9/6 when being presented with ISO18/16/13 fuel, the filter must remove contamination that is a minimum of 64 times dirtier (possibly 128x) than required by the injector OEM, do it in a single pass of the filter, under high cyclic flow surges and subjected to high frequency vibration. All of these issues calls into serious doubts about the ability for most diesel engine fuel filters to provide cleanliness levels at or better than ISO12/9/6 when vehicles are fueled with an ISO18/16/13 cleanliness fuel as recommended by most OEM’s.  The distribution of particle sizes within diesel fuel is vastly different from other hydrocarbons seen in the industry. Lubrication oils, engine oils, and hydraulic oils typically have a wide distribution of particle sizes within a given sample. A typical sample of diesel fuel, however, comprises of approximately 80-97% of particles at or below 4μm in size. With such a high percentage of particles in the fuel, at the same size that causes the greatest damage in a HPCR injection system, filter companies have responded by reducing the filtration rating on engine fuel filters to below 4μm. This transition to a lower micron rated filter has resulted in additional pressure on the life of the filter element. Correspondingly, most fuel filters have not increased in physical size to combat the additional load.

In many industries, engine fuel filters are typically changed on a time based planned maintenance (PM) strategy measured in hours worked. The OEM of each engine, or asset, typically advises the PM event timing, along with additional information being provided by industry specialists. Most engine manufacturers in most locations around the world typically schedule engine fuel filter changes at 500 hours. As filter micron ratings have reduced to target contaminants below 4μm, we are now starting to see some engine manufacturers advising their customers to once again reduce the fuel filter PM intervals from their current 500 hour intervals to 250 hours on new equipment fitted with HPCR injection systems. This new requirement is starting to place additional pressure on existing maintenance strategies as additional PM intervals must also be refined or considered to ensure engines or assets are not being taken out of service just for a fuel filter change.

Reducing the PM interval of an engine fuel filter from 500 to 250 hours is a direct response by filter manufacturers to ensure that the engine fuel filter will not completely block during its time in service due to reduced micron rating. It is, however, a false economy and not required should the asset fuel tank be delivered with a fuel cleanliness that does not overload the engine fuel filter in the first place.

Engine fuel filters do not employ a bypass valve, which would enable fuel to bypass the filter in the event that it becomes blocked with contaminants. Rather, the engine management system constantly monitors engine fuel filter differential pressure and will regulate the fuel flow to the engine based on the condition of the filter. If the engine management system indicates that a fuel filter has a high differential pressure, the fuel flow is automatically reduced causing a loss of power, and possible machine speed.

The situation again circles back to the specification of fuel cleanliness levels being delivered to the machine at refueling. While the filter micron rating has reduced to a level where the majority of contaminants in the fuel are being targeted, what has not occurred is a reduction in the specified fuel cleanliness level being filled into the asset fuel tank; it remains at ISO18/16/13. The answer employed by the OEM is to simply change out the engine fuel filter sooner. What should change is the cleanliness of the fuel entering the fuel tank in the first place.

Filters are not intelligent devices; they remove contaminants within the fuel as they are presented to the filter media. Reducing or maintaining the level of contaminants being presented to the engine filter each hour will increase its service life; this is an undeniable fact. It is, therefore, difficult to understand why engine OEM’s are not advising their customers of this.