Agriculture Reference
In-Depth Information
in Table 10.8 and Fig. 10.18. A significant
shortening by the medium separator MA can
be seen, while the influence of the fine separa-
tor FA on length is nearly negligible. As with
the width distribution, the steam explosion
causes a drastic shortening of the fibre
bundles.
that it yields a high statistical accuracy based
on a relatively small number of tested samples.
For cotton, Warrier and Munshi (1982) showed
in a single-element test that 300 measurements
were needed to achieve a level of accuracy of
5% at a significance level of 95%.
The same level was achieved in the hemp
collective tests on sample K1/RS after 60 and
for sample K1/GADO after 35 measurements.
The mean number of necessary tests was 80.
By decreasing the level of accuracy down to
10% at a significance level of 95%, the number
of collectives to be tested was 22 on average.
As shown in Fig. 10.19a, the collective
strength decreases with each separation step from
RS to GADO. It has to be investigated whether
this decrease is influenced by damage to the fibres
or can be explained by differences in the number
of elements in the collectives. This effect is
described by Suh
et al
. (1994) for cotton: the col-
lective strength decreases dramatically with an
increasing amount of cotton fibres in a collective.
As explained by Suh
et al
. (1994), the
number of elements in a collective has an
important impact on the results. To make the
results more comparable, it is necessary to
know the number of single elements in a
10.4.3 Strength
Fibre samples were taken at different positions
in the separation line and tested for their
strength properties. The fibre bundles were
tested as collectives and as single elements. The
results are presented in the following sections.
Collective test
The results of the collective test with the
Stelometer device are presented in Fig. 10.19.
From the results presented in Fig. 10.19b, it
can be seen that the collective strength values
for mean and median are equal and the meas-
ured values are almost normally distributed.
A general advantage of the collective test,
in comparison with the single-element test, is
Table 10.7.
OFDA results from the samples K1/RS - K1/GADO.
Width at
1% of the
measured
elements
(
Coefficient
of
variation
(%)
Level
of
confidence
95% (
Amount
<30
Amount
>100
x
50
(LNV)
(
Mean
(
Standard
dev. (%)
Median
(
μ
m
μ
m
Sample
Number
m)
m)
m)
(%)
(%)
m)
m)
μ
μ
μ
μ
μ
RS
12,477
54.3
59.4
109.3
2.6
30.2
49.9
15.9
266.1
32.8
GA
22,470
47.0
49.4
108.8
1.5
26.9
53.6
12.0
233.2
30.4
MA
20,777
47.3
49.0
104.9
1.6
27.2
53.2
12.2
231.4
30.7
FA
22,745
46.8
47.7
102.9
1.3
27.9
52.5
11.6
228.4
30.7
GAD
44,941
16.8
12.5
74.5
0.3
13.5
93.7
0.4
69.1
14.5
GADO
62,166
16.2
10.2
63.0
0.2
13.5
94.8
0.2
57.7
14.3
Table 10.8.
Almeter values of the tested hemp fibre bundles.
Sample
Mean (Q) (mm)
CV (Q) (%)
L(Q) <25 mm (%) L(Q) 50% (mm)
L(Q) 1% (mm)
GA
47.8
78.4
37.7
36.0
180.8
MA
40.9
68.4
37.3
31.9
123.5
FA
38.2
80.2
45.0
27.7
156.5
GAD
13.0
66.4
92.8
10.6
45.5
GADO
13.1
65.8
93.4
10.8
39.8
