CEN/TR 16982:2016

SLOVENSKI STANDARD
01-november-2016
Dizelske mešanice in goriva - Vprašanja glede hladnega filtriranja
Diesel blends and fuels - Cold filterability issues
Dieselkraftstoffe und Mischungen - Kaltefiltrierbarkeit Problematiik
Combustibles et blends pour moteurs diesel (gazole) - Problems avec filtrabilité en
temperatures bas
Ta slovenski standard je istoveten z: CEN/TR 16982:2016
ICS:
75.160.20 7HNRþDJRULYD Liquid fuels
2003-01.Slovenski inštitut za standardizacijo. Razmnoževanje celote ali delov tega standarda ni dovoljeno.

CEN/TR 16982
TECHNICAL REPORT
RAPPORT TECHNIQUE
September 2016
TECHNISCHER BERICHT
ICS 75.160.20
English Version
Diesel blends and fuels - Cold filterability issues
Combustibles et blends pour moteurs diesel (gazole) - Dieselkraftstoffe und Mischungen - Kaltefiltrierbarkeit
Problems avec filtrabilité en temperatures bas Problematiik

This Technical Report was approved by CEN on 8 July 2016. It has been drawn up by the Technical Committee CEN/TC 19.

CEN members are the national standards bodies of Austria, Belgium, Bulgaria, Croatia, Cyprus, Czech Republic, Denmark, Estonia,
Finland, Former Yugoslav Republic of Macedonia, France, Germany, Greece, Hungary, Iceland, Ireland, Italy, Latvia, Lithuania,
Luxembourg, Malta, Netherlands, Norway, Poland, Portugal, Romania, Slovakia, Slovenia, Spain, Sweden, Switzerland, Turkey and
United Kingdom.
EUROPEAN COMMITTEE FOR STANDARDIZATION
COMITÉ EUROPÉEN DE NORMALISATION

EUROPÄISCHES KOMITEE FÜR NORMUNG

CEN-CENELEC Management Centre: Avenue Marnix 17, B-1000 Brussels
© 2016 CEN All rights of exploitation in any form and by any means reserved Ref. No. CEN/TR 16982:2016 E
worldwide for CEN national Members.

Contents Page
European foreword . 3
Introduction . 4
1 Scope . 5
2 Background to this Technical Report . 5
3 Issues in specific European markets . 5
3.1 UK experience . 5
3.2 Sweden . 7
3.3 Italy. 11
4 Cold operability rig tests . 12
4.1 Infineum Freezer Rig . 12
4.2 PSA Filter Rig . 15
5 Filterability test developments . 17
5.1 Total contamination test (EN 12662, WG 31) . 17
5.2 CS-FBT (WG 31) . 18
5.3 Cold FBT (Energy Institute SC-B-5) . 18
6 Other experiences . 21
6.1 Afton investigations . 21
6.2 Argent experience with distilled TME . 24
6.3 Diesel fuel cold operability correlation (WG 34) . 26
7 Discussion and next steps . 28
Bibliography . 29

European foreword
This document (CEN/TR 16982:2016) has been prepared by Technical Committee CEN/TC 19 “Gaseous
and liquid fuels, lubricants and related products of petroleum, synthetic and biological origin”, the
secretariat of which is held by NEN.
Attention is drawn to the possibility that some of the elements of this document may be the subject of
patent rights. CEN shall not be held responsible for identifying any or all such patent rights.
At the plenary meeting in June 2015, CEN/TC 19 took Decision 45-2015 for new work under WG 24 to
produce a Technical Report titled “CEN/TR Diesel blends - Cold filterability issues” with the scope to
capture the key points raised in the presentations and discussions at the WG 24 Filter Blocking
Workshop held on 1 June 2015. Consequently, this Technical Report documents the findings,
interpretations and opinions of those involved in presenting the information, and these should not be
considered as the opinion of WG 24.
Introduction
During recent winters, a wide range of vehicles has been affected in specific European countries and
there is a possible link with FAME composition, base diesel quality, cold flow additives and oxidation
stability effects. In order to solve these issues, some countries have introduced new additional
requirements in their national fuel quality specifications or “best practice” market agreements:
• In the UK, a clear correlation between low temperatures and increased vehicle filter blocking was
reported, with ambient temperatures below 3 °C thought to be critical. The introduction by fuel
suppliers of a voluntary Filter Blocking Test limit of 2,52 in February 2014 seems to have improved
the situation, but has not solved the problem.
• In Italy, ENI recommended that ASTM D2709 could be an alternative method for fast evaluation of
contaminants in FAME. ENI also suggested, as an intermediate solution, a filtration step in
refineries or terminals to improve FAME quality if needed. In ENI’s experience, implementing this
quality control “best practice” in Italy, in collaboration with their biofuel suppliers, has resulted in
no further vehicle filter blocking incidents being reported in the last two years.
• In France, to solve the diesel fuel filter plugging when the decrease in temperature continues slowly
over several days, the saturated methyl ester content in FAME was limited in winter to a maximum
of 16 % (m/m) and in summer to a maximum of 30 % (m/m) in national law.
st
CEN/TC 19/WG 24 organized a workshop on the 1 of June 2015 in order to clarify the issue, to gather
relevant data and to propose recommendations to CEN/TC 19 with respect to changes to the EN 590
(regular B7 diesel), EN 16734 (B10), EN 16709 (B20/B30) and EN 14214 (B100) standards to protect
the market from filter blocking.
At the end of the workshop, it was agreed that a CEN Technical Report should be produced
documenting the WG 24 Filter Blocking Workshop held on 01 June 2015 (i.e. this report). It therefore
lays down the status-quo of the evidence on filter blocking issues in the European market at that point
in time. It should be read as such and later information will still be valuable for CEN/TC 19 specification
drafting.
1 Scope
This Technical Report provides the latest thinking described during a workshop on 1 June 2015 by
national experts involved in the investigations, and proposes possible solutions to solve the diesel fuel
filter plugging issues in these countries.
NOTE For the purposes of this Technical Report, the terms “% (m/m)” and “% (V/V)” are used to represent
respectively the mass fraction, µ, and the volume fraction, φ.
2 Background to this Technical Report
A filter blocking workshop was organized by CEN/TC 19/WG 24 on 01 June 2015 in London in response
to an increasing number of diesel vehicle filter blocking occurrences in several European countries
(Italy, Sweden and the UK), particularly during the winter period. The purpose of the workshop was to
share experiences and learnings from each affected country, identify common links and discuss possible
solutions. The workshop also considered the development status of the various rig and laboratory tests
designed to investigate and prevent low temperature filter blocking. The ultimate aim of the workshop
was to make recommendations to WG 24 with respect to changes to the EN 590, EN 16734 (B10),
EN 16709 (B30) and EN 14214 standards to protect the end user.
In his introductory comments, the WG 24 convenor advised that a wide range of vehicles is being
affected in several European countries and that there is a possible link with FAME composition, base
diesel quality, cold flow additives and oxidation stability effects. He also underlined the importance of
ensuring that the CEN diesel fuel specifications are robust and protect the consumer.
The workshop included a number of technical presentations on the topics that are described in the
Clauses 3 to 6 (order of presentation is followed). Publication of this Technical Report was one of the
agreed actions from the workshop (see Clause 7).
3 Issues in specific European markets
3.1 UK experience
A summary of the diesel vehicle filter blocking trends in the UK over the past few years was provided.
The monthly “baseline” level of diesel vehicle breakdowns due to filter blocking since 2009, as reported
by the Automobile Association (AA), was around 200. However, during the past three winters, filter
blocking breakdowns had risen to 5 times this level, with most of this increase occurring in the regions
of Northeast England, Central Scotland, East Anglia and Southeast England.
A clear correlation between low temperatures and increased vehicle filter blocking was reported, with
ambient temperatures below 3 °C thought to be critical. The introduction by fuel suppliers of a
voluntary FBT limit of 2,52 for Bx diesel in February 2014 seemed to have improved the situation, but
has not solved the problem. The UK experienced ~19 % reduction in the number of vehicle breakdowns
due to filter blocking in winter 2014/15 compared to the previous winter, despite winter 2014/15
having many more cold nights below 0 °C in the most impacted regions (the minimum nightly
temperature averaged for London, Glasgow and Middlesbrough was below 0 °C for 7 nights in winter
2013/14 compared to 30 nights in winter 2014/15).
Data from an extensive UK-wide retail diesel sampling program conducted by a major fuel retailer were
presented. In addition, test results from a UK Department for Transport nationwide retail diesel
sampling program undertaken between January 2015 to March 2015 were also presented. Data
reported by region from both sampling programs included FBT (both ambient by procedure B and cold
soak), Total Contamination, FAME content, saturated FAME content and particle counting. To provide a
broader European context, FBT results from another retail diesel survey conducted across 8 European
countries by a major international fuel retailer were shared.
A number of observations were made on the data from these fuel sampling programs:
a) Ambient FBT and Cold Soak FBT tests gave very similar results.
b) FBT results show several excursions above the UK voluntary limit of 2,52. In particular, the timing
of one cluster of excursions corresponded with a higher number of vehicle failures due to filter
blocking.
c) FBT results for diesel in other European countries are lower than in UK diesel (see Figure 1). Out of
111 samples, no FBT results were measured above 1,7 and most were below 1,1.
d) It was noted that the level of imported diesel fuel into the UK had increased considerably over the
time period. It was also highlighted that base diesel fuels can have an impact on FBT results.
e) FAME content of UK diesel was fairly static over the period 2011 to 2015 at an average of 3 % (V/V)
to 4 % (V/V). Regional differences exist with generally higher levels of FAME in Southeast England
and East Anglia, and lower levels in Scotland. It was also noted that FAME blending levels were not
the main trigger of occurrence of problems.
f) FAME used in the UK generally contains between (20 to 25) % (m/m) saturated FAME, however it
is sometimes up to ~40 % (m/m)
g) The highest saturated FAME levels in EN 590 diesel were found in samples from Southeast England
and East Anglia; whilst the lowest levels were measured in fuels from Scotland and the Midlands.
h) Particle count levels were high in some samples from the South East and East of England.

Figure 1 — Average and maximum FBT (right y-axis) by IP 387 Procedure B, average Total
Contamination in mg/kg and average FAME content in % (V/V) (left y-axis) from 111 samples of
diesel fuel collected from service stations in 8 European countries
The UK investigation also included analyses of deposits from blocked vehicle fuel filters and from the
Infineum freezer rig. The analysis of residues from blocked vehicle filters in winter 2012/13 found the
material to be mainly saturated mono-glycerides (SMG). In winter 2013/14, polyethylene, and in some
cases polyamide, was detected on blocked vehicle fuel filters. These components were also isolated
from some of the fuel samples taken from the tanks of failed vehicles. Analysis of material recovered
from blocked vehicle fuel filters in winter 2014/15 indicated high levels of ethylene vinyl acetate (EVA)
which is one of the main ingredients in cold flow improver additives.
Filter deposits from fuel samples run through the Infineum freezer rig as part of the industry
investigation were also isolated and analysed. Two sets of fuel from the same supply locations were
tested – the first set was sampled in winter 2013/14 and the second set was sampled in winter
2014/15. When the rig test filters were analysed after testing three of the winter 2013/14 fuels, SMGs
were detected, and in one fuel, there was also evidence of polyethylene and polyamide. Two of the rig
test filters analysed after testing the winter 2014/15 fuels showed the presence of SMGs, but there were
no indications of polyethylene or polyamide in any of these fuels. It was also reported, although not
seen on any of the rig test filters, that Drag Reducing Additive (DRA) had been observed on some
service station diesel filters.
In their concluding remarks, the UK experts stated that there has been no apparent correlation between
vehicle filter blocking and the Total Contamination test. As this is a gravimetric test they felt that it may
not protect the market from fuel with high organic particulate loading. They also commented that the
high particle counts measured in some market diesel fuels were not reflected in the Total
Contamination test results, but were reflected in the FBT results. It was concluded that, in their opinion,
no correlation existed between FBT and Total Contamination. For these reasons, it was explained that
the UK is considering the introduction of an FBT requirement with a maximum limit of 2,52 in the
national annex of BS EN 590 to provide improved market protection, rather than introducing a lower
Total Contamination test limit.
It was recognized by the UK experts that the ambient FBT or the existing Cold Soak FBT test might not
fully protect the market from all the potential root causes, therefore they are continuing to investigate
the possibility of a “Cold FBT” test in which the filtration step is conducted at a low temperature just
above the cloud point of the fuel (i.e. somewhere in the range −3 °C to +3 °C). They also plan to continue
with the freezer rig testing program at Infineum to identify the root cause(s) of the UK filter blocking
issue and requested that other European countries support the work by providing representative
market fuel samples.
3.2 Sweden
An overview of the recent diesel vehicle filter blocking problems in Sweden was provided from three of
the key stakeholders investigating the issue: BIL Sweden (representing the OEMs), SPBI (Swedish
Petroleum and Biofuels Institute) and Perstorp (Swedish FAME producer).
The BIL Sweden presentation started with a brief history of the issue. As is the case in the UK, this
appears to be a winter problem in Sweden which only affects diesel cars. The problems started in
winters 2011/12 and 2012/13 with some vehicles needing filter changes due to plugging. The situation
worsened in winter 2013/14 with a number of OEMs starting to report increased filter changes on
vehicles due to plugging. The problem was escalated to an industry issue which saw some initial
discussions between BIL Sweden and SPBI during summer and autumn 2014. By winter 2014/15 there
were more than 2 000 reported cases of fuel filter blocking with at least seven OEMs being affected. The
filter change rate per 1 000 vehicles for some models has been higher than for corresponding models in
the UK market (see Figure 2), although it has not been confirmed whether the design of the vehicle fuel
systems are identical in both countries. In response, a “Deposit Investigation Task Force” was formed in
spring 2015 by the Swedish Transport Agency, BIL Sweden and SPBI.
BIL Sweden shared some fuel analysis data which included FBT data for diesel samples collected from
seven service stations that were linked to blockages of actual vehicles’ filters. Whilst there were no clear
trends linking all the fuels analysed, attention was drawn to elevated SMG content in one of the fuels,
elevated steryl glucoside content in two fuels, high FBT (Ambient and Cold Soak) in two fuels and low
oxidation stability (Rancimat) in one of the fuels. It was noted that the high SMG content result of
74 mg/kg was unusual in the context of the low total monoglyceride content of 344 mg/kg, especially
when the FAME had been identified as Rapeseed Methyl Ester (RME), which is highly unsaturated.

Figure 2 — Fuel filter replacement rate per 1 000 vehicles for one OEM (2 different model years)
In addition, blocked filter deposit analysis data generated by three OEMs was shared. In most cases, the
filter deposits were dark and sticky. One of the OEMs reported that it seemed to be oxidized biodiesel.
Attention was also drawn to the high levels of zinc and silicon measured on many of the filters.
However, another OEM reported no indication of aged biodiesel and low levels of silicon and zinc in
their filter deposits. When asked about where the zinc could be coming from, BIL Sweden responded
that some of the impacted models use fuel tanks with zinc-containing coatings. However, it was also
that no fuel injector deposits have been reported as might be expected in fuels containing elevated
levels of zinc.
A common link between many of the filters analysed by the three OEMs was the high proportion of
steryl glucoside deposits relative to SMG deposits. BIL Sweden felt that this was the opposite to the UK
experience citing the example of one of the OEMs who had measured high SMG but low steryl glucoside
deposits on a UK vehicle filter that was blocked in 2013.
In their concluding remarks, BIL Sweden stated that the root causes to the problem are not yet
understood, however they offered a few of their own thoughts on what could be contributing to the
problem:
— 2011/12 was the first winter when the filter plugging problems started to appear, and the problem
has increased each year after that. This coincided with the first B7 diesel blends that were
introduced onto the Swedish market in 2011. Over the next couple of years, some diesel containing
7 % (V/V) FAME + HVO up to 35 % (V/V) was put on the market. Therefore, BIL Sweden wondered
whether a contributing factor to the filter blocking problem might be lower fuel solubility towards
FAME impurities since Swedish MK1 diesel has worse solubility characteristics compared to typical
EN 590 diesel and HVO reduces the solvency (aromaticity) of the fuel even further.
— Until recently there have been diesel fuels in the Swedish market that do not contain performance
additives. In BIL Sweden’s experience, a majority of the “problem fuels” seem to have been B7
containing no performance additives and sometimes containing HVO.
BIL Sweden called upon the workshop delegation for the urgent development of a reliable performance
test on fuel filterability, which they would like CEN to include as a requirement in the EN 590 standard.
In their presentation, SPBI added to the information presented by BIL Sweden. It was noted in their
opening remarks that the vehicles from all seven OEMs that are experiencing fuel filter blocking are
equipped with suction fuel pumps and fuel filter heaters that activate when the temperature drops to
approximately −3 °C. Placement of the main fuel filter on the suction side of the pump is known to make
filters more sensitive to blocking.
It was explained that the FAME blended into diesel in Sweden is RME with a typical Cloud Point of
between −4 °C to −5 °C. No used cooking oil or tallow based FAME is used. RME has the best cold
properties of all commonly used FAME feedstocks. SPBI member companies also have to meet
additional FAME quality requirements that are more stringent than EN 14214, such as lower water
content and lower total monolgyceride content limits.
SPBI had observed three types (cases) of filter blocking in Sweden:
1) Case 1: filters from two vehicle brands blocked due to a black sludge deposit at approximately 0 °C.
Interestingly, the filters did not block in these vehicles in the lower temperatures experienced in
the far north of Sweden. When analysed, the black sludge typically contained oxidized FAME
components, silicon (which could either originate from antifoam additive in the fuel or silica
contained in the filter housing) and ethylene vinyl acetate (cold flow improver additive). High levels
of zinc were also found in the deposit. Analyses of a new fuel filter housing of the type used in the
vehicles affected by Case 1 failures showed that the surface is made from zinc and iron. The
presence of zinc in diesel fuel is known to accelerate the degradation of FAME, which can
contribute to filter blocking.
2) Case 2: fuel filter blocking reported on diesel cars with diesel particulate filters (DPF) that use a
cerium / iron containing additive to facilitate the regeneration of the DPF. Under the control of the
vehicle’s ECU, the additive is automatically injected into the vehicle fuel tank each time the vehicle
is refuelled. In the case of these blocked filters, cerium, iron and oxidation products from FAME
were found on the fuel filter. SPBI conducted experiments to demonstrate the effect of the DPF
additive on the fuel’s oxidation stability. This showed that oxidation stability, as measured by the
Rancimat and PetroOxy tests, reduced significantly with increasing DPF additive concentration (see
Figures 3 and 4). The effect was far more significant for B7 than B0 diesel. SPBI commented that a
frequent start-stop driving pattern seems to exacerbate the problem. It was noted, however, that
this DPF additive is used extensively in other European countries markets (e.g. France) with no
reported problems.
Figure 3 — Effect of DPF additive in B7 diesel on Rancimat induction period
Figure 4 — Effect of DPF additive in B0 and B7 diesels on PetroOxy induction period
3) Case 3: other cases of filter blocking with limited information into possible causes.
Results of filterability tests on several lab blends of EN 590 and Swedish MK1 diesels with varying
FAME contents were reported by SPBI. In this study, all FBT and Total Contamination test results were
very low. SPBI also sampled a number of service stations in the areas surrounding three fuel
distribution terminals. The fuels had FAME contents ranging between 5 % (V/V) and 7 % (V/V). The
samples were tested for Cloud Point, FBT, Steel Corrosion and Rancimat. The oxidation stabilities of all
the fuels were well above the EN 590 minimum Rancimat limit of 20 h. All FBT results were below 2 and
most were very low. The reason for not introducing a voluntary FBT limit is due to that the final fuel for
delivery to service stations are blended at the loading rack. The possibility to control the FBT of the final
blend is very difficult given the large number of deliveries to service station.
In conclusion, SPBI called upon the CEN experts to investigate whether standard diesel detergent
additive has an effect on degradation of FAME in the presence of metals, and to investigate the causes of
filter blocking and propose possible solutions.
Perstorp presented slides on a series of experiments they have conducted on precipitates from fuel
samples and analyses of blocked fuel filters. In one experiment, Perstorp prepared blends of distilled
Tallow Methyl Ester (dTME) and Rapeseed Methyl Ester (RME) in Swedish MK1 at concentrations of
5 % (V/V), 7 % (V/V) and 10 % (V/V). The blends were stored in glass bottles for several days at −20 °C.
After 6 days, all the dTME blends contained a white precipitate which was present in the largest
concentration in the B10 blend. In contrast, none of the RME blends showed signs of any precipitates
after 3 days (see Figure 5).
An experiment to evaluate the impact of antioxidant and cold flow improver additives on the
precipitation of saturated monoglycerides (SMGs) in 7 % (V/V) blends of RME in Swedish MK1 was also
performed by Perstorp. In this experiment, the fuel blends were stored at −20 °C for two months before
shaking and cold filtering them under vacuum through a 1 µm glass fibre filter. The conclusion drawn
from this experiment was that antioxidant and cold flow improver additives have little effect on SMGs.
Various studies conducted by Perstorp have shown a large number of different components contained
in diesel fuel precipitates and isolated on test filters which may contribute to blocking vehicle filters.
These included glycerol, steryl glucosides, saturated methyl esters, SMGs, lubricants, free sterols,
polyethylene, ethylene-vinyl acetate copolymer, residues of synthetic polymers, citric acid, phosphoric
acid in addition to sand, metal flakes and undefined trash. But interestingly, Perstorp never found SMGs
on any blocked vehicle filters.
In conclusion, Perstorp felt that with proper handling, biodiesel works well in Nordic climates. To
ensure problem-free operation: good quality FAME with low amounts of less soluble components
should be used (RME is optimal); compatible construction materials should be used; fuel additives
should be carefully chosen and there should be elaborate procedures for introducing them; long periods
of extreme low temperatures should be avoided.

Figure 5 — The effects of prolonged low temperatures (−20 °C) on blends of different FAMEs at
various concentrations in Swedish MK1 diesel: dTME (left) and RME (right)
3.3 Italy
ENI presented a summary of their diesel vehicle filter blocking experiences in Italy. Vehicles typically
experienced driveability problems or complete power loss caused by fuel filter blocking. Some vehicles
were more significantly affected than others.
During their investigation, various FAMEs were sampled and analysed. Impurities were collected by
centrifugation (ASTM D2709). In some cases they found polymer compounds (identified as Nylon and
polyethylene) with variable sizes up to 100 µm and in other cases they found steryl glucosides. Other
parameters such as the CFPP, cloud point and total monoglycerides were measured in order to calculate
the SMG content according to the Italian National Annex to EN 14214. For all FAMEs tested, the SMG
content was within the national limit for SMG.
The presence of polymer compounds (Nylon and polyethylene) in the FAME correlated with their
findings from analysis of material that they collected on the surface of diesel filters from vehicles
affected by filter blocking in the field. In ENI’s view, the problems experienced by vehicles were caused
by the presence of these compounds in the FAME that was blended into diesel. However, it had
previously been noted in WG 24 that nylon has a melting point above 300 °C and the pre-treatment
processes at FAME production plants do not reach this temperature.
ENI performed studies to identify lab tests that could detect the presence of polymers and minor
components in FAME. They found that the following tests were effective to a greater or lesser extent:
1) Separation by centrifugation (ASTM D2709) – although the test was originally designed to detect
water and sediment in fuel, ENI found that it was effective as a fast screening test to check for the
presence of polymers and small amounts of other insoluble components in FAME. For FAME
samples, if water content is within the EN 14214 maximum limit of 0.05 % (m/m), the water
doesn’t come out of solution during the test. ENI has started checking FAME batches they purchase
using this test as they feel that it could correlate with FAME showing poor filterability
characteristics.
2) Total Contamination (EN 12662) – if the FAME contained polymers and small amounts of other
insoluble components it proved difficult to filter, even under vacuum. These FAMEs were typified
by very long filtration times. This is consistent with the findings of CEN/TC 19/WG 31 during their
inter-laboratory study to generate precision for the test – some tallow and used cooking oil methyl
esters were difficult to filter. However due to inconsistencies in the current method text, it is not
clear whether filtration times in excess of 30 min constitute a test failure. This should be addressed
by WG 31 when they complete their current revision of the test method. ENI found a good
correlation (R = 0,868 7) between the EN 12662:2014 method and the ASTM D2709 centrifuge
test for FAME samples.
3) FBT (IP 387, Procedure B) – ENI found a reasonable correlation (R = 0.7306) between IP 387 FBT
results and ASTM D2709 centrifuge test results for FAME samples, however the FAME had to be
pre-diluted in order to flow through the FBT test equipment. It should be noted that when ENI
originally presented this finding to WG 24 in November 2013, WG 24 requested that WG 31 should
evaluate ASTM D2709 in its filterability pilot tests. WG 31 concluded that the centrifuge tests
carried out on their samples did not provide useful data and consequently ASTM D2709 was not
pursued any further.
In conclusion, ENI stated that it is important to introduce into EN 14214, a predictable and fast test to
measure the filterability of FAME. They felt that since the cold soak FBT test is still under development,
a final decision about its applicability is needed. They recommended that the ASTM D2709 centrifuge
test could be an alternative method for fast evaluation of contaminants in FAME. ENI also suggested, as
an intermediate solution, a filtration step in refineries or terminals to improve FAME quality if needed.
However, they stated that the best solution, in their experience, is to implement the EN 14214 standard
to ensure that FAME has optimal filtration characteristics. Implementing quality control “best practice”
in Italy, in collaboration with their biofuel suppliers, has resulted in no further vehicle filter blocking
incidents being reported in the last two years.
4 Cold operability rig tests
4.1 Infineum Freezer Rig
Infineum presented an overview of the low temperature “freezer rig” that they had designed and
constructed on behalf of the BSI taskforce investigating the UK diesel vehicle filter blocking issues. The
aim was to mimic the failure mode experienced in the field by designing a rig using a main fuel filter and
fuel pump from actual models of vehicles that had been impacted by filter blocking in the UK. The rig
represents the low pressure fuel system of a typical modern diesel vehicle and is set-up in suction mode
(i.e. the filter is on the vacuum side of the pump). The fuel tank and fuel system is located inside a cold
box which allows the temperature of the rig to be adjusted to the desired level (see Figure 6).
During the development phase of the rig and the test protocol, different conditions were investigated in
order to determine the best conditions for discrimination between fuels and to reproduce the filter
blocking issue seen in the field. After trying various conditions, the following test protocol was agreed:
— 50 l of fuel are used for each test.
— The chamber is cooled gradually to −3 °C over a period of 12 h.
— The chamber is held for 8 h at −3 °C.
— The fuel flow rate is set at 80 l per hour at the start of the test by adjusting the pump.
— The fuel is recirculated around the rig and through the filter for 4 h at −3 °C or until the filter is
blocked (note: the temperature of the fuel is approximately −1.5 °C).
Figure 6 — Schematic of the Infineum Freezer Rig
Under these test conditions, the rig is able to discriminate between fuels’ filterability performance.
Diesel fuels from the most affected area of Northeast England had significantly worse filterability
performance than the other control samples from less affected areas of the UK (see Figure 7).

Figure 7 — Example of typical delta pressure curves produced by the Infineum freezer rig when
testing fuels with good and bad low temperature filterability
A couple of interesting findings were noted by Infineum. Firstly, there appears to be an “induction
period” or lag time for some fuels before onset of delta pressure increase across the filter. In some cases
the delta pressure only started to increase after 2 h of the test (i.e. after the entire 50 l of fuel had been
recirculated more than 3 times through the rig). This observation has not yet been explained. Secondly,
for all the fuels that caused filter blocking, the delta pressure across the filter returned to normal values
as the fuel was allowed to gradually warm-up after the test was complete. In all cases the filter
unblocked rapidly once the fuel temperature had reached approximately +2 °C to +5 °C.
Several retail diesel samples collected from the loading racks of several fuel distribution terminals in
both the affected and unaffected areas of the UK were tested using the rig. In addition, a number of lab
blends using different FAME and base diesel combinations from samples provided by various UK fuel
suppliers were tested on the rig. Results showed a variety of filterability performance between different
fuels (see Figure 8) and a correlation with field experience. However, it was not possible to correlate the
filterability performance with the FAME or base diesel quality on the basis that a combination of several
factors appears to be involved in causing vehicle filter blocking.

Figure 8 — Infineum freezer rig results for various retail diesel samples and lab blends
Since it was not possible to correlate the observed filterability performance with either FAME or base
fuel quality, the BSI taskforce have designed an experiment to investigate whether the combination of
more than one potential causal factor could provide a better understanding of the rig and field
filterability performance. The experimental design includes a matrix of different lab blends using two
different base diesel fuels (B0), two different distilled FAMEs (distilled tallow and distilled used cooking
oil methyl ester) and four different impurities (saturated monoglycerides, water, polyethylene and
steryl glucosides). The aim of the experiment is to provide an understanding into the effects of the base
diesel, the saturated FAME content and the combination of different impurities.
The rig and laboratory testing on the various blends in the experimental study is still underway,
however Infineum presented some preliminary data. This showed that the saturated monoglycerides,
FAME and different base diesels all seem to contribute to the filterability performance of the finished
fuel (see Figure 9).
In conclusion, Infineum stated that a filterability rig based on the low pressure fuel system of a light
duty diesel vehicle has been successfully developed. The rig correlates with the field issues and can be
used to investigate filterability performance of fuels at low temperature (but above the cloud point of
the fuel). The filterability performance of diesels in the UK market is complex and is likely the
combination of several vehicle and fuel factors. An experimental study is underway to understand their
relative contribution.
Figure 9 — Preliminary Infineum freezer rig data on experimental lab blends
4.2 PSA Filter Rig
PSA presented their studies into FAME composition impacts on cold flow properties of diesel fuel. In the
introductory slides, a comparison was made between the number of complaints of PSA vehicle filter
blocking in the UK, France and Germany from January 2011 to May 2013. This showed a marked
increase in UK complaints in winter 2012/13 compared to France and Germany. PSA felt that this was
related to the absence of a UK limit on the amount of FAME produced from waste (e.g. used cooking oil
and animal fat / tallow) allowed in diesel, whereas in France there is a maximum limit of 0,35 % energy.
This led PSA to conduct a study into the impact of FAME composition on cold flow properties.
In their studies, PSA found that there was no correlation between total monoglyceride content and
Cloud Point. However, they found a strong correlation (R = 0,96) between saturated methyl ester
content and Cloud Point of the FAME (see Figure 10). This is logical since the monoglyceride content
depends on the FAME production process, whilst the Cloud Point depends on the feedstock (% of
saturated methyl esters).
Figure 10 — Correlation between FAME Cloud Point and saturated methyl ester content
This finding prompted PSA to develop a filter rig test to investigate the impact of varying the saturated
methyl ester content in B7 and B10 diesel fuel. The rig test uses a 4 l fuel tank linked to a vehicle fuel
distribution system with pressure sensor on the fuel filter. The apparatus is enclosed within a climatic
chamber and the following test cycle is used (see Figure 11):
a) Equilibrate the fuel for 2 h at 20 °C in the rig tank;
b) Decrease the temperature of the climate chamber at a rate of 1 °C per minute to reach −15 °C after
35 min. At the same time as the climate chamber cooling commences, activate the fuel pump to
circulate the fuel through the filter for 10 min;
c) Deactivate the fuel pump for 30 min;
d) Thereafter, continue to activate / deactivate the fuel pump following this cycle:
1) 10 min on, and
2) 30 min off.
Figure 11 — Photograph of PSA low temperature filterability rig (left) and test cycle (right)
PSA have found that the results they have obtained using the rig test correlate with their customer
complaints of vehicle filter plugging. They also obtained similar results with different diesel fuel filters
tested in their rig. Only when they tested fuels with a saturated FAME content below 10 % (m/m), did
they find that variations in the saturated monoglyceride content had an impact on the delta pressure
measured across the filter.
Based on the selection of different FAME types used in the study (and noting that not all classes of
FAME given in EN 14214, Table 3 were tested), PSA concluded that a maximum limit of 12 % (m/m)
saturated methyl esters in FAME is suitable for diesel during winter in order to avoid vehicle filter
blocking (this limit in B100 corresponds to a limit of 0.84 % (m/m) saturated FAME in B7). In summer,
this limit could be relaxed to 20 % (m/m) saturated methyl esters in FAME (see Figure 12).
In conclusion, PSA stated that the CFPP test is not sufficient to cover all the situations encountered by
vehicles in the field. Further, they found no correlation between the total monoglyceride content and
the fuel filter plugging. They felt that CEN needs to take additional action to prevent customer issues
due to saturated methyl esters in the diesel which cause problems when the temperature falls gradually
for long periods (e.g. over one or more nights). In these conditions, the impact of saturated FAME on
fuel filter plugging is more important than the level of monoglycerides.
Figure 12 — PSA rig test results obtained on B7 samples with different levels of saturated FAMEs
As a result of their findings, PSA called upon CEN to include a note in EN 590 and EN 16734 strongly
recommending fuel suppliers to control the level of saturated FAME in diesel during the winter. They
also proposed that the EN 14214 standard should limit the total saturated methyl ester content of the
FAME to a maximum of 12 % (m/m) in winter and 20 % (m/m) in summer.
During the discussion that followed, some WG 24 members highlighted that the PSA proposal to restrict
the saturated FAME content does not correlate with the experiences in other countries. Additionally it
was noted that the EN 14214 climate tables are already incorporated into the standard to tackle long-
term cold soak issues, whereby limitations are made depending on the climate of the country.
5 Filterability test developments
5.1 Total contamination test (EN 12662, WG 31)
A brief update was provided on behalf of CEN/TC 19/WG 31 on the progress of the taskforce towards
improving the EN 12662:2014 “Total Contamination” test method. A taskforce was formed after several
issues were highlighted by the industry when the test is applied to FAME (B100). The main activity of
this taskforce is to revise the procedure for the determination
...