FEFCO Standards Committee
The FEFCO BCT Round Robin Test 2001
In the work programme for the FEFCO Standards Committee are included Test Rounds, i.e.
investigations on test methods for corrugated board regarding their precision and the deviation
in results received on identical material by different laboratories.
At the meeting No. 48 of the FEFCO Standards Committee in autumn 2000 was decided to
perform the first Round Robin Test regarding BCT, the test method for determination of the
compression strength of corrugated board boxes.
On behalf of SWIF, the Swedish national member of FEFCO, Lars-Erik Eriksson at
Packforsk, Sweden, undertook to administrate this BCT Test Round and to co-write the
Jósef Kuminka at the Dunapack Company in Hungary prepared the test material (box blanks)
which was distributed to the participating laboratories.
The chairman of the Standards Committee, Urs Ernst of EMPA laboratories in Switzerland,
participated in the BCT Round in the calculation of the results, in the preparation of the result
diagrams, the co-writing the report and printing the reports.
Through the members of the Standards Committee, laboratories were invited to participate in
the investigation. Fifteen laboratories intended to participate in the Test Round.
The testing was performed during spring 2001 and the first preliminary results were presented
at the Standards Committee meeting No 49 in April 2001. A detailed discussion took place at
the 50th meeting September 2001 and the 51 st meeting March 2002.
The authors want to express many thanks to everyone who has been involved in this FEFCO
Round on testing BCT of corrugated board boxes.
In December 2001 / March 2002
Lars-Erik Eriksson Urs Ernst Jósef Kuminka
Packforsk, Sweden EMPA, Switzerland Dunapack, Hungary
FEFCO Standards Committee: BCT Round Robin Test 2001 page 2 of 12
1. Aim of the investigation
Aim of the FEFCO investigation on box compression test, BCT, was to:
1.1 Evaluate deviations in BCT test results (within and between the laboratories)
1.2 Compare results received by testing according to the FEFCO testing method No.
50 (unsealed boxes) with results received by testing of sealed boxes with fixed inner
1.3 Look at the influence of different testing speeds (exercise the SUW Lifetime
2. Participating laboratories
The below listed laboratories participated according to the instructions in the FEFCO BCT
Round Robin Test.
Company / Laboratory Address
AIDIMA Paterna (Valencia)
Packaging and Transport Dept. Spain
AssiDomän Italia Lunata
BVI-IBE Belgian Packaging Institute Zellik
Dentas Packaging and Paper Industry Inc. Denizli
EMPA Eidg. Materialprüfungs- und Forschungsanstalt St. Gallen
LNE, Laboratoire National d'Essais Trappes Cedex
Kappa Packaging, Development Center Hoogeveen
LGAI Technological Center Bellaterra (Barcelona)
Campus de la UAB Spain
Packforsk Kista (Stockholm)
PRI Paper Research Institute Budapest
SCA PKG Market Development Group Hinckley
Stora Enso Packaging Lahti mills Lahti
wfp, Institut für Wellpappenforschung- und Prüfung Darmstadt
ÖIV Österreichisches Institut für Verpackungswesen Wien
One laboratory performed the BCT testing only according to the test method UNE EN 22872.
These results are not included in this report.
FEFCO Standards Committee: BCT Round Robin Test 2001 page 3 of 12
3. Test material
Three different board grades, one single wall of B-flute, one single wall of C-flute and one
double wall of BC-flute, with the following material composition were used for the
- B-flute: TL 140 / B-Wellenstoff 110 / TL 135 g/m2
- C-flute: TL 140 / B-Wellenstoff 110 / TL 135 g/m2
- BC-flute: TL 140 / B-Wellenstoff 110 / Schrenz 120 / C-Schrenz 120 / TL 135 g/m2
The test material was prepared by the Dunapack Co in Hungary and sent to the participants by
Mr Kuminka. Each participating laboratory received 40 blanks for corrugated board boxes of
the configuration FEFCO 0201 in the size 400 x 300 x 300 mm (LxBxH) of each board grade.
The participants were asked to apply a preconditioning procedure as laid down in EN 20187
in order to reduce the moisture content of the board material below 5 %.
Before the BCT testing all the boxes had to be conditioned according to EN 20187 in the
atmosphere 23±1 oC and 50±2 % R.H. according to detailed instructions.
The instructions for the testing (see Annex) included requirements for the following tests:
5.1 Compression test methods
Compression tests were performed on each board grade
A) according to the FEFCO testing method No 50 (no gluing or fixation of flaps of the
erected box) → FEFCO 50
B) according to an alternative method, ASTM D 642-94 or similar, prescribing sealed boxes
and fixed inner flaps → Flaps fixed
An additional BCT test method was used by the Kappa Packaging Development Center in the
Netherlands. This method includes a dummy to be placed inside the tested box in order to
force the panels of the box to bulge outwards only → Kappa special.
5.2 Compression test speed
Two speeds were laid down for the compression testing:
a) the ordinary speed of around 10 mm/min
b) one higher speed at 100 mm/min.
5.3 Testing of moisture content
Testing of the moisture content of the material in the compression-tested boxes according to
the procedure described in ISO 287 or similar test method was recommended.
FEFCO Standards Committee: BCT Round Robin Test 2001 page 4 of 12
6. Test results
The test results are presented as tables and diagrams calculated by the statistical computer
programme used by the CEPI-CTS organisation in its control programme for round robin tests
for paper testing apparatus.
The calculation and reporting is concentrated on the force results for the BCT measurements
(BCT values). However, for some test series the deflection results are also shown.
Every individual BCT force result for all boxes tested at the participating laboratories are
implemented into the computer programme. The averages for each laboratory as well as the
standard deviation and coefficient of variation within and between laboratories were
calculated together with the grand mean for each corrugated board material.
For the results for the box compression deflection, the same information has been calculated,
but only for testing according to the FEFCO Test Method No. 50.
Fourteen laboratories have reported results that could be taken into the assessment.
The Kappa laboratory made some extra tests using the method developed by the Kappa
Packaging Development Center. Main difference to the other methods is the use of a dummy
inside the box in order to secure outward bulging of the box panels.
6.1 FEFCO testing method No. 50
In the diagrams 1 – 3 are given the compression force results in kN, for the FEFCO testing
method No. 50 (unsealed flaps) for all participating laboratories when using the ordinary
testing speed of 10 mm/min and for the box materials B, C and BC, respectively.
Diagrams 4 – 6 show the results for the BCT deflection in the test series using the FEFCO
testing method No. 50 in for the participating laboratories.
In the tables beside the diagrams, are shown figures for the number of samples tested, the
mean values and the coefficient of variation for each laboratory, as well as the calculated
grand mean, the standard deviation and the coefficient of variation for both within and
between the laboratories.
6.2 FEFCO testing method No. 50 (unsealed boxes) versus testing of sealed boxes
In the diagrams 7 – 9 are given the compression force results in kN, for both the FEFCO
testing method No. 50 and the alternative method with fixed flaps for all participating
laboratories when using the ordinary testing speed of 10 mm/min and for the box materials B,
C and BC, respectively.
In the diagrams 10 – 12 are separated the coefficients of variation within laboratories for the
force results in the test series using the FEFCO testing method No. 50 in comparison with
results for the Fixed flap method for the participating laboratories.
Diagrams 13 – 15 show the results for the BCT deflection in the above test series.
FEFCO Standards Committee: BCT Round Robin Test 2001 page 5 of 12
6.3 Influence of the compression speed
In the diagrams No. 18 – 20 are given the compression force results in kN, for the FEFCO
testing method No. 50 when using the higher testing speed of 100 mm/min for all
participating laboratories and for the box materials B, C and BC, respectively.
In the diagrams 21 – 23 are shown comparisons between the test results received by the
ordinary speed of 10 mm/min and the higher testing speed of 100 mm/min. The results for
laboratories using a higher speed other than 100 mm/min, are not included in the diagrams.
As well are excluded all results at the higher speed showing lower BCT values than the
ordinary testing speed. The reason for this exclusion is discussed in the section 7, below.
6.4 Moisture content
In the diagrams 24 – 26 are given the values for the moisture content, in %, for the
conditioned material in the tested boxes when BCT was tested, in relation to the BCT force
recorded. Values are given for the speed 10 mm/min, both when the FEFCO test method No.
50 and the alternative method with fixed flaps were used. Only nine of the participating
laboratories have reported the moisture content of the materials tested.
6.5 Kappa special method
This test has only been performed by the Kappa laboratory. In the diagrams No. 16 – 17 are
given the compression force results and the deflection results for the FEFCO testing method
No. 50, the alternative method with fixed flaps and the Kappa special method for the Kappa
laboratory when using the ordinary testing speed of 10 mm/min and for the box materials B,
C and BC, respectively.
7. Discussion and comments on the results
7.1 General considerations
The analysis in this FEFCO Test Round on the BCT test method for corrugated board boxes
has concentrated on the maximum compression forces (BCT-values) expressed in kN.
The reasons for this concentration of the evaluation on the force measurements were
- Primarily that the force in general is seen as the most interesting information coming out
from the BCT testing. The BCT force measurements are e.g. used for estimating the load
that boxes can stand in practice.
- Secondarily that the spreading of the results between the laboratories concerning the
deflection was extremely large (see item 7.4)
According to information received by the participating laboratories and to the overall results,
the test material prepared by the Dunapack Co, was suitable for the investigation and arrived
to the laboratories in good condition.
FEFCO Standards Committee: BCT Round Robin Test 2001 page 6 of 12
7.2 FEFCO testing method No. 50
7.2.1 BCT force
An important aim of the investigation was to get statistical figures for the existing FEFCO test
method No. 50 regarding the spreading of the results within a laboratory and between
different laboratories running the same test.
With reference to the diagrams 1 – 3, the variation of the reported results were:
- Variation coefficients within the laboratories: almost within the range 2 % to 8 % for the
individual laboratories. The calculated coefficients of variance within all laboratories were
for the 3 board types tested in the range of 4 % to 5 %.
- Variation between the laboratories: The coefficient of variance between the laboratories
was for the 3 board types tested in a range of 6 % to 7 %.
7.2.1 BCT Deflection
The results of the compression deflection of the boxes at the maximum force (BCT value) are
given in diagrams 4 – 6. As it is clearly visible from these diagrams, the results measured by
the laboratories differ extremely large. Differences of 300 % up to more than 450 % between
the maximum- and minimum mean values of the laboratories have been reported. This results
in a coefficient of variance between the laboratories of 35 % to 45 %!
But also the variation of the results in the individual test series of a laboratory and
consequently the variation within the laboratories is very large with figures between 11 % and
7.3 Comparison BCT-values FEFCO testing method No. 50 - the method using fixed flaps
An important aim of the investigation was to compare results received by testing according to
the FEFCO Test Method No. 50 with results received by testing according to a method
prescribing sealed boxes and fixed inner flaps.
The FEFCO Test Method No. 50 specifies only erecting the blanks to the box shape without
sealing the box or fixation of inner flaps before the box is loaded in the compression tester.
Other BCT test methods often specify sealing and fixation of the flaps.
With reference to the diagrams 7 – 9 can be observed that the BCT-values received when
using the alternative test method with sealed boxes and fixed flaps and the normal test speed
of 10 mm/min, were mainly reported smaller; in average 5.1 % lower than the results for the
FEFCO test method No. 50. The difference of the mean BCT force values between the two
methods did not differ much regarding the board grade. The differences found in the results
reported by the laboratories were from 6.4 % higher to 29 % lower values for the fixed flap
A compilation of the comparison between the FEFCO Test Method No. 50 and the fixed flap
results, based on the average figures, is shown in diagram 31.
FEFCO Standards Committee: BCT Round Robin Test 2001 page 7 of 12
7.4 Analysis of the force measurement deviations: Methods FEFCO No. 50 and Fixed
When comparing different test methods, most often the spread for the results are of vital
interest. There are many reasons for the deviations at testing, e.g. the tested material itself, the
conditioning of the material, the test method used, the testing equipment, and so on. It is
essential to, as far as possible, reduce the deviations caused by the test procedure and
The spread is often defined as the statistical “standard deviation” for the test series. The
variation coefficient, expressed in %, is the standard deviation related to the actual average
value. These values can be calculated for the test series both within each laboratory as well as
between the laboratories.
With reference to the diagrams 10 – 12 and to a compilation of the comparison between the
FEFCO testing method No. 50 and the fixed flap results, based on the average figures
summarised in diagram 31, the variation coefficients within each single laboratory were for
both test methods used almost within the range 2 % to 8 %. In average the FEFCO method
gave somewhat lower deviation within laboratories, 4.4 % in comparison to 4.9 %.
Regarding the deviation between the laboratories the FEFCO method No. 50 gave somewhat
higher deviation at the single wall boards (approx. 6,7 % in comparison to 5.8 %), but
somewhat lower deviation at the double wall board (7.1 % in comparison to 8.3 %). In the
average of the whole test round the deviations between laboratories were at the same level
(6.8 % in comparison to 6.6 %).
When comparing these coefficients of variation found with the corresponding figures for
other corrugated board properties, reported in earlier FEFCO Test Rounds, e.g. ECT,
Puncture, etc, the conclusion must be that the BCT test shows a deviation in the same range as
for most physical testing of corrugated board.
7.5 Analysis of the deflection measurement
Very wide variations were found for the compression deflection of the boxes at the maximum
force (BCT value), as can be seen from the diagrams 4 – 6 and 13 – 15, which are related the
test method FEFCO No. 50 and Fixed flaps, speed used 10 mm/min.
This situation can be observed in all tests series run, that means for both methods (FEFCO
No. 50 and Fixed flaps), for all the boards used (B, C and BC) and for both crushing speeds
used (10 mm/min and 100 mm/min), but only the results of the upper conditions are plotted
The results reported by the laboratories are slightly higher for the Fixed flaps method
compared with the FEFCO No. 50 method. They varied from 3 to 11 / 4 to 17 mm (mean
values for method FEFCO No. 50 and Fixed flaps, respectively) for board B, from 4.5 to 14
mm / 4.7 to 17 mm for board C and 5.5 to 25 mm / 6.3 to 36 mm for board BC.
The variation coefficient within each single laboratory varies from a few percent to more than
20 % deviation within the laboratories, some variation coefficients close to 30 % can be
observed. Looking at the average variation within the laboratories, the figures range from
FEFCO Standards Committee: BCT Round Robin Test 2001 page 8 of 12
10 % to 20 %. For the variation between the laboratories the corresponding figures are in the
range 35 % to 50 %.
However, it can be seen especially from the diagrams 13 and 15 (board B and BC), in a
tendency also from diagram 14 (board C) that there is a focussing of many laboratories at a
comparable lower level and that there are only a few laboratories that report very high
deviation results. A plausible explanation for this fact is a difference in the “zero setting” of
the deflection in the beginning of the test: FEFCO testing method No. 50 prescribes a “zero
setting” at 5 % of the BCT force or maximum 200 N respectively. The FEFCO Standards
Committee questioned, at the meeting No. 50, whether all laboratories had applied the zero
setting as prescribed in the FEFCO testing method No. 50.
Viewing these facts no other detailed analysis has been carried out for the deflection results.
7.6 Influence by testing speed
In this Test Round was planned for an investigation based on the SUW Lifetime Estimation
method. (LTE = Life Time Estimation method, developed by the Scandinavian Development
Group for Corrugated Board, SUW). This method was originally using three different
compression speeds, but can be used also for only two compression speeds. The instructions
for the Test Round recommended an ordinary speed about 10 – 14 mm/min (10 mm/min
preferred) and an additional speed about 100 mm/min.
The background for the LTE method was that practical experiences of physical strength
testing of paper and paperboard material have indicated that an increased testing speed will
result in higher values. Theories for the relationship between force and speed at testing have
been developed. Such a theory by Caulfield1 was the basis for the development of the LTE
method within SUW.
According to that theory, the slope of the line connecting the compression forces received by
testing at different compression speeds for BCT, contains information on the time which the
boxes can stand a certain load during storing. In the SUW Lifetime Estimation computer
programme (Atle), these slope figures are seen as a material property and used for calculation
of the relationship load/storing time for the boxes. Unfortunately the slope figures were
hidden in the programme and there were no resources in this FEFCO Test Round to get access
to the slope figures.
As no slope figures were available, the presentation is simply focussed on the measured BCT
values for the two speeds used, 10 mm/min and 100 mm/min (diagram 21 – 23). The
diagrams 18 – 20 show the results of all laboratories at a higher speed. However it has to be
recognised that 2 laboratories run the tests at a significantly lower speed than 100 mm/min.
Two other laboratories reported consistently lower BCT values for the test series with the
higher speed. Such results are against both the theories and general experience. They can
nether be explained by statistical means based on normal deviation.
In contact with suppliers of compression testing equipment, a probable explanation was
mentioned. The force recording device of the machine is not suitable, or adjusted, for the ten
times higher speed than ordinarily used. Thus testing with the higher speed may lead to false
FEFCO Standards Committee: BCT Round Robin Test 2001 page 9 of 12
Both laboratory groups, the one showing lower BCT values for the higher testing speed and
the other with a higher speed lower than 100 mm/min, have been excluded in the calculations.
The increase of the BCT value received by using the higher testing speed 100 mm/min, is
given in the table below for the accepted test series, expressed in % of the value for the
ordinary speed 10 mm/min.
Increase of BCT-value due to the higher speed at 100 mm/min (base: 10 mm/min)
FEFCO Test Method No. 50 Alternative test method with fixed flaps
Boxes B 13.4 15.4
Boxes C 15.7 15.9
Boxes BC 7.3 7.4
Average 12.1 12.9
The average figures for the influence by the speed are close to each other for the two test
Looking at the individual results for the test series in the laboratories, very wide variations
can be seen, from nearly no influence of the speed up to more than 25 % higher BCT values
for 100 mm/min.
It can also be seen as an interesting observation, that boxes made of single wall corrugated
board indicated higher influence by the compression speed on the BCT value than the boxes
made of double wall.
7.7 Influence by moisture content
In order to get an indication regarding the conditioning of the test material, the participants
were asked to determine the moisture content of the board when testing BCT.
Obviously there were differences in the measured moisture content in the corrugated board
material when BCT was tested at the participating laboratories. As seen in the diagrams 24 –
26, the span was in the range of 6.3 % to 8.4 %. The different laboratories showed
consistently almost the same order for the moisture figures for the different test series. This
indicates a repeatable measuring procedure and more or less constant test climate conditions
in the laboratories.
General experience is that the moisture content of the board influences the BCT force: the
higher the moisture content, the lower the BCT force. This tendency is in most diagrams
visible, even when it has to be stated that the correlations are not very good and the influence
obviously small. However the question arise whether the differences in the BCT values are
the result of different moisture content and as such of different test conditions (relative
humidity in the air of the test room) or whether they are usual variation of BCT tests.
FEFCO Standards Committee: BCT Round Robin Test 2001 page 10 of 12
EN 20187 allows a variation of the humidity of ± 2 % RH at a level of 50 % RH. Data
reported by Statens (S) and used by CEPI-CTS for the laboratory climate check show that 1 %
RH change in the test climate corresponds to approx. 0.11 % moisture content in the board.
As it can be seen from the diagrams 27 – 29 the variation of the reported moisture contents is
much bigger than the influence by the test conditions. The allowed variation in the test
conditions according EN 20187 can therefor not be the reason for the differencies in the BCT
values between the laboratories.
An approach to assess the influence of the moisture content measured on the BCT values can
be made according to the general experience described above: Assuming that the grand mean
of the reported moisture contents corresponds to the correct one, the BCT values of each
laboratory can be compensated by a factor related to the difference of the measured moisture
content to the mean moisture content. A BCT value related to a low moisture content will
therefore be reduced (to dry board → to high BCT value) and vice versa. A possible formula
is given by the following formula:
overall mean moisture content actual moisture content
Force compensatd Force actuel (1 k )
overall mean moisture content
Such an approach should be allowed taking into account that all laboratories reported a test
climate according to EN 20187 that means at a relative humidity of the air in the range of 50
% RH. Data given in the Statens Technical Report 2 say that at this air humidity level the
slope k of the correlation BCT-force to water content is approximately 0.5, while Markström3
proposes approximately 8% change in BCT force due to 1% change of moisture content.
The original results and the compensated results are given in the diagrams 24 – 29. With the
same formula the influence expected by the influence of the humidity variation according EN
20187 on the BCT value has been calculated and included in the diagrams. A summarisation
of the results is given in diagram 30.
The following interoperation of the comparison of the measured and the compensated BCT
values can be made:
The compensation of the measured BCT forces according to the difference “Measured
moisture content – grand mean moisture content” decreses the range of the laboratory
By the above compensating procedure the standard deviation between the laboratories gets
lower at the single wall boards B and C, but not at the double wall board BC.
The influence of the allowed rel. humidity range in the test climate (± 2 %RH according
EN 20187) on the BCT-force is small, approximately ± 1.5 to 2 % BCT-force.
Summarised it can be stated that the correlation between the measured BCT -forces and the
moisture content measured in the board can explain only a minor part of the force
deviations between the laboratories. The spreading of the results between the laboratories
must maily be explained by the BCT test procedure and the test material itself.
7.8 Kappa special method
The method developed by Kappa Co., which force the panels of the box to bulge outwards
during the BCT test by using a dummy inside the box, was only performed by the Kappa
laboratory. Diagrams 16 (BCT force) and 17 show the results of the Kappa special tests
compared with the FEFCO No. 50 method and the Fixed flaps method. The comparison can
be done only for the Kappa laboratory.
FEFCO Standards Committee: BCT Round Robin Test 2001 page 11 of 12
The results regarding the mean force values did not differ between the methods FEFCO No.
50 and Kappa special. The Fixed flaps method gave similar results for the boards B and C, but
lower ones for the board BC.
The variance of the result series were lower at the Kappa special method for the boards B and
C, but of the same size for board BC. The variation found for the Fixed flaps method was
higher than for the Kappa special method for all three boards tested.
7.8.2 BCT deflection
The mean deflection reported is of similar size at board B, but higher at boards C and BC for
the Kappa special method compared with method FEFCO 50.
Summarised, the conclusions from this FEFCO round robin test on compression strength on
corrugated board boxes (BCT) are the followings:
The FEFCO testing method No. 50 Determination of the Compression Resistance of
corrugated fibreboard containers, edition 1997 is a useful test method and provides good
and reliable measuring results regarding the compression forces on corrugated board
The average variation coefficient found for the BCT force value when testing according to
the FEFCO testing method No. 50 is in the range of 4 – 5 % within the laboratories and 6
– 7 % between the laboratories. These variations found are comparable to deviations
received for most physical testing of corrugated board.
The FEFCO testing method No. 50 shows no disadvantages in force results compared
with a test method prescribing a more complicated sample preparation by sealing the box
and fixing the inner flaps. In average the BCT force was found 5 % lower, the variation
coefficient higher within the laboratories, but in the same range between the laboratories
for the fixed flaps method in comparison with the FEFCO testing method No. 50.
A crushing speed of 100 mm/min gave in average 12 – 13 % higher BCT force values
than the ordinary speed of 10 mm/min.
The crushing speed of 12.5 ± 2.5 mm/min prescribed in the FEFCO testing method No. 50
should not be increased.
Only a minor part of the differences in BCT force between the laboratories can be
explained by the influence on the BCT force caused by different moisture content in the
board. The influence on the BCT force due to the allowed variation in the test condi-
tioning referred to in the FEFCO testing method No. 50 is estimated maximum ± 2.0 %.
The FEFCO testing method No. 50 shows no disadvantages when compared with the test
method ”Kappa special” (Inside dummy forcing outward bulging of all side panels and
closing the box by sealing and fixing the inner flaps). However this conclusion is based on
the results of only one laboratory.
Whenever the determination of the deflection is relevant, care has to be taken that the
recording procedure is done strictly following the description in the test method.
FEFCO Standards Committee: BCT Round Robin Test 2001 page 12 of 12
9 Final decision
With reference to the FEFCO Standards Committee’s decision at the meetings September 25,
2001 and March 20, 2002, the FEFCO testing method No. 50 Determination of the
Compression Resistance of corrugated fibreboard containers, edition 1997 is to remain
Diagrams 1 – 31
Deatil information regarding the participant laboratories
FEFCO testing method No. 50
1. D.F. Caulfield: A chemical kinetics approach to the duration-of-load problem in
wood; Wood and Fiber Science, 17(4), 1985
2. Statens Skogsindustrier AB, Sweden: Statens Technical Report No. 2:
Deterioration of case strength during transport and storage.
3. Håkon Markström: Testing Methods and Instruments for Corrugated Boards,
chapter 11: The importance of the test atmosphere for the strength properties;
Lorenzen & Wettre, Sweden, 1999