fiber property measure

1. This article was downloaded by: [Dalhousie University]
On: 05 January 2013, At: 02:49
Publisher: Taylor & Francis
Informa Ltd Registered in England and Wales Registered Number: 1072954
Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK
Journal of The Textile Institute
Publication details, including instructions for authors
and subscription information:
http://www.tandfonline.com/loi/tjti20
Determination of the 50%
and 2.5% Span Lengths and
the Uniformity Ratio with the
Fibrograph
R. S. Krowicki
a
, J. M. Hemstreet
a
& K. E. Duckett
b
a
USDA, ARS, Southern Regional Research Center, New
Orleans, USA
b
The University of Tennessee, Knoxville, USA
Version of record first published: 30 Mar 2009.
To cite this article: R. S. Krowicki , J. M. Hemstreet & K. E. Duckett (1997):
Determination of the 50% and 2.5% Span Lengths and the Uniformity Ratio with the
Fibrograph, Journal of The Textile Institute, 88:3, 167-173
To link to this article: http://dx.doi.org/10.1080/00405009708658542
PLEASE SCROLL DOWN FOR ARTICLE
Full terms and conditions of use: http://www.tandfonline.com/page/terms-and-
conditions
This article may be used for research, teaching, and private study purposes.
Any substantial or systematic reproduction, redistribution, reselling, loan, sub-
licensing, systematic supply, or distribution in any form to anyone is expressly
forbidden.
The publisher does not give any warranty express or implied or make any
representation that the contents will be complete or accurate or up to date. The
accuracy of any instructions, formulae, and drug doses should be independently
verified with primary sources. The publisher shall not be liable for any loss,
actions, claims, proceedings, demand, or costs or damages whatsoever or
howsoever caused arising directly or indirectly in connection with or arising out
of the use of this material.

2. Determination of the 50% and 2.5% Span
Lengths and the Uniformity Ratio with the
Fibrograph
R.S. Krowicki*, J.M. Hemstreet*, and K.E. Duckett^
*USDA, ARS, Southern Regional Research Center, New Orleans, USA
University of Tennessee, Knoxville, USA
Received 17.12.1992 Received in revised version 18.9.1996 Accepted for publication
/H.9.I996
Length measures of cotton fibres on a Fibrograph will appear about 6% short because of
crimp. Fibrograph measurements must be started at a distance from the comhs and this
requirement causes the 50% span length measurement to be increased by half this distance.
The 2.5% span length is increased by a negligible amount. The resulting uniformity ratio can
he increased hy as much as 25%. An arithmetic determination of the average length of fibre
held behind the front edge of the comb teeth is approximately 0.44 cm. This holding length is
dependent upon cotton fibre rigidity, thereby causing a variahle shortening of the measured
lengths. Corrected parameters from theoretical Hbrograms generated from array data
compared well with measured data presented in the literature in four of the five cases
examined.
I. INTRODUCTION
Fibre length is a very important physical measure in the textile industry. Length measures
and measures of variability in fibre length have been used in establishing settings on
processing equipment. Fibre length measures obtained al various phases of processing
when compared to initial measures also indicate fibre breakage or loss of fibres during
processing.
Methods of fibre length measurement, unfortunately, are subject to sources of error.
These sources of error and their relation to span length and uniformity ratio determined
by using the Fibrograph are of primary consideration. These errors, however, do not occur
in the array method. They only occur whenever a clamp, comh. or other holding device is
used vi'hen scanning fihres to determine length parameters. These errors, if not corrected,
cause the measured lengths to deviate from the actual length values.
Hertel and Lawson (1964) studied several factors affecting fibre length-scanning
measurements and concluded that fibre crimp, fibre-end taper, and lack of random clamping
were the important causes for length measurement errors. We suspect that the factors
affecting fibre length measures are fibre crimp, lack of random clamping, start of scan
(Krowicki and Ramey, 1984), lens width (Krowieki, 1986), fibre taper, and holding length.
It is felt that fibre taper causes a proportionate but minimal effect tending toward longer
lengths since more fibres end close to the holding element than towards the edge of the
beard. Lens width also has an affect, tending to increase the measured length at the end of
the beard. This increase in the measured 2.5% span length is minimal.
J. Text. Insl.. IW7, m I'eiri I. No. _
-
f © Textile Imuliile 167
Downloaded
by
[Dalhousie
University]
at
02:49
05
January
2013

3. Krowivki, Hemslreel. und Duckell
2. EXPERIMENTAL
Five commercial cottons, representing ditferent Upland varieties, were used in relating
array data to Fibrograph span length and uniformity data. Table I indicates the codes and
other general identification used for these varieties. Arealometer (ASTM Designation
D1449) and Stelomcter (ASTM Designation D1445) data are shown in Table II. Note that
Sample 3 has an average perimeter, but is immature (high D). Hertel and Lawson (1964),
in the early 1960s, used a modified array technique to determine the fibre length measures
for these varieties. These previously unpublished data are used as a basis in this study.
Samples were prepared initially according to ASTM method D 1440-55 for placement in
a Suter-Webb sorter. These samples were weighed and then suhdivided prior to placing
them on slides coated with petroleum jelly and mineral oil. The fibres on each slide were
straightened and their images projected to obtain high resolution 24 magniUcation. The
projected lengths were measured with a calibrated wheel map measuring device and the
results were converted to actual length arrays accurate to the nearest 0.01 in (0.03 cm).
We used these arrays to generate simulated theoretical fibrograms (Krowicki et al., to be
published; Landstreet, 1961) from which span lengths and uniformity indices were
determined. These determinations were obtained from theoretical fibrograms, theoretical
fibrograms corrected for holding length, and theoretical fibrograms corrected for holding
length and start of scan. A correction for fibre crimp was also applied to the last corrected
data.
Table 1
Cotton Sample Codes
ID
1
2
3
4
5
ID
I
2
3
4
5
A
mm '
447
438
521
436
376
Bale No.
167522
101149
649953
736080
D
mm '
29
25
45
24
21
P
49
48
49
47
52
Other ID
Middling 1 -3/32 - Rio Grande Val ley
Strict low plus 1-1/16-Centr.il Delta
Strict low bright 29/32 - t.ubbock. Texas
Middling 1 - Central Texas
Rowden 1953
Table II
Fibre DaU
W CR T,
jig/in. gf/tex
42 O'*34 16.0
4.2 0.922 15.9
3.6 0.934 18.8
4.2 0.936 17.6
5.3 0.933 17.0
E,
7.6
7.4
7.9
8.3
7.3
A - Arealometer Specific Area in miii'/mm'
D - Arealomeier Specific Area difTerencc lor two compressions
P— Periniclcrcalcuhiled from Arealometer data
W - Linear density calculated from Arealometer data
CR - Crimp ratio determined with the crimp meter
= I - beard exiension/gaugc length
T| - Stelometcr tenacity at 0,3175- cm gauge
E| - Stelometcr elongation at break for 0.3175- cm gauge
The Fibrosampler (ASTM Designation D1447) consists of a partial cylindrical plate
with 1.59-cm. diameter holes through which samples are caught by (he comb. Assuming a
randomized sample, the average projected length is about 0.79 cm. Fibres caught and held
(Krowicki and Thibodeaux, 1990) will have two ends e.Ktending from one or more adjacent
needles so that, on the average, the length of fibre held is 0.40 cm. The average measured
168 J. Text, hist., 1997, 88 Pan I. No. J © Textile Institute
Downloaded
by
[Dalhousie
University]
at
02:49
05
January
2013

4. Oi'tcnuiimutm of the 50% and 2.5% Span Lengths and the Uniformity Ratio with Ihe Fibrograph
needle diameter was 0.03 cm. One quarter of the needle perimeter plus one radius adds
0.04 cm to the length of fibre held to approximate a holding length of 0.44 cm., which we
use in our calculations.
Crimp was determined from a beard prepared with the Fibrosampler by using a crimp
meter designed by Hertel and Lawson (1964). This device applied a 100-gm force to a
portion of the beard. A crimp factor was defined as the normalized ratio of the beard
extension to the original gauge length. Tbe crimp ratio is then (1 - the crimp factor). This
ratio (Table 11) is used as the final correction to tbe span lengtbs and uniformity ratios
because the measure was made on a beard similar to tbat presented to the Fibrograpb for
measurement.
3. RESULTS AND DISCUSSION
Fig. 1 is typical of tbe combined number array data considered in this discussion. Fig. 2 is
the theoretical fibrogram generated by a double accumulation of tbe data from the fibre
length distribution by number (Krowicki et al., to be published; Landstreet, 1961). Fig, 3
compares a theoretical fibrogram to a theoretical fibrogram with a ().44-cm holding length
applied and to a theoretical fibrogram where the combination 0.44 cm holding length is
applied and tbe 0.38 cm start of scan position is used to adjust the relative weight level of
the fibrogram. Table III compares 2.5% span lengths and uniformity ratios determined
from values measured on tbe Fibrograph witb corresponding parameters determined witb
array data for each variety as well as from array data corrected for holding length, start of
scan, and crimp ratio. Measured and corrected array values were divided into the respective
initial array values to obtain the ratios shown in Table IV for the five varieties.
0.000
Fig. I Fibre length distribution by number from 5317 single fibre measures - Sample I
!n their work, Herte! and Lawson (1964) calculated an average ratio of 1.18 for the
theoretical-to-measured upper half mean. Tbe 2.5% span lengtb is an approximation to
tbe upper balf mean and the average theoretical to the measured 2.5% span lengtb sbould
be of the same order. The average calculated ratio of these 2.5% span lengtbs as shown in
J. Text. Inst.. 1997. Hti Purl I. No. 3 © Textile Institute 169
Downloaded
by
[Dalhousie
University]
at
02:49
05
January
2013

5. Kro^vicki. Hemsireet, and Duckell
o.e-
0.6-
0.4-
1 2 S 4
(om)
Fig. 2 Theoretical Fibrogram generated from the fibre length distribution by number - Sample I
Fig. 3
1 2 3
Length (cm)
llworeticBi -Hold ..Hofd-Start ot scan
Comparison of Fibrograms: theoretical, theoretical with holding length applied, and theoretical with
holding length and stan of scan appl ied - Sample 1
Table IV is 1.15. The average ratio of the theoretical value for 2.5% span length to that
array value corrected for holding length is 1.14. The start of scan (involved in all tneasures
where held fibres are presented to an instrument for measurement) combined with holding
length changes the average ratio to 1.12. The additional adjustment for crimp changes the
average ratio to 1.20.
Examining Ihe ratio value of the theoretical 2.5% span length to the 2.5% span length
corrected for holding length, all but Sample 3 gave a ratio relative to the theoretical 2.5%
.span length which is less than the iheorelical-to-mea.sured ratio. This indicates ihat the
holding length is too large for Sample 3 and may or may not be too large for other samples.
The start of scan adjustment will not change unless the instrument is improperly calibrated.
The ratio of theoretical span length to the theoretical span length adjusted for holding
170 J. Te.Ki. lii.u.. , fi8 Pan 1. No. 3 © Textile Institute
Downloaded
by
[Dalhousie
University]
at
02:49
05
January
2013

6. Oetemiinalion of the 50% and 2.5% Span Length.^ and Ike Uniformity Ratio with lite Fibrograph
Table [II
Theoretical, Measured and Corrected Lengths
Correcied Array D.ita
Array Measured Hold +Slart +Ciiiiip
ofScan
Cotton
ID
2.5% Span Length {cm)
I
2
1 .
4
5
3.10
2.92
Uniformity Ratio
1
S
S
3
4
5
39.7
^.r
40.2
37.8
38.3
2.64
2.57
2.11
2.64
2.51
•
45.2
43.6
48.2
44.2
48.5
2.74
2.69
1.98
2.67
2.57
38.9
36.8
39.7
39.0
38.6
2-79
2.74
2.03
2.72
2.62
47.3
46.3
50.0
47.7
47.6
2.62
2.54
1.90
2.57
2.44
46.6
46.0
49.3
47.5
47.9
Table IV
Ratios'" of Lengths and Uniformity' Ratios
Con-ected Array Data
Cotton
ID
2.5% Spun l.fiifill
1
2
3
4
5
Avg
Uniformity Ratio
1
2
3
4
5
Avg
Measured Hold +Start
ofScan
+Crimp
<cm)
1.17
1.19
1.11
1.14
1.16
1.15
0.88
0.84
0.84
0.86
0.79
0.84
1.13
1.13
1.18
1.13
1.14
lai
MS
1.11
1.12
1-18
1.20
1.23
1.18
1.^
1,14 1.12 1.20
1.02
1.00
1.01
0.84
mo
on
0.85
em
am
0.W
1.00
•Ratios of theorctical(array) lengths to measured and correcicd k
ratios of theoretical to measured and corrected uniformity ratios.
. and
length and start of scan will lower the ratio due to adjusting for holding length alone.
These ratios again indicate tbat tbe holding lengtb for Sample 3 is too large. An additional
adjustment for estimated crimp sbows an increase in all the ratios. Referring again to
Table III, length values determined for Samples 1 and 2. witb all corrections differ from
the measured samples by 0.03 cm or less. (We use the laboratory standard deviation of
0.038 cm (Worley and Krowicki. 1968) for measured length.) Satnples 4 and 5 differ by
0.07 cm. Adjusted Sample 3 differs from the measured length by 0.21 cm. Tbis is greater
than five standard deviations wbicb is outside the normal distribution of deviations about
the mean. All indications are tbat the corrected value for Sample 3 was obtained by using
loo large a holding length and or too large an estimate of crimp.
The uniformity ratios (UR) in Table III are changed slightly from theoretical values by
J. 'I'e.xl. In.si.. 1997. 88 Part I. No. S © Textile Institute 171
Downloaded
by
[Dalhousie
University]
at
02:49
05
January
2013

7. Krowicki. Hemstreet. and Duckett
correcting tbe array data for holding length. The average ratio of the theoretical value to
the value adjusted for holding length is 1.00, indicating no difference. Additional correction
for start of scan tends to increase the UR by about 20% for these data. This increase has
been explained (Krowicki and Ramey. 1984) as primarily due to an increase in tbe 50%
span length. Adjustment due to estimated crimp changes Ibe UR sligbtly.
Consider tbe possibility of determining an estimate of ibe UR which is a better
approximation of the theoretical UR. Each span length (SL) measure is dependent upon a
lengtb to start of scan (LS). (LS is defmed as the distance from the outer edge of the comb
teeth to the centre of tbe measuring lens.) The measured 50% SL (SL^,,^^) is overestimated
by 50% of LS wbile the 2.5% SL (SL,,_J is overestimated by 2.5% of LS. Length data are
now generally presented as SL,^^ and calculated UR. Since:
• - (1)
3,,n, , , J (2)
A corrected 50% SL (SL^^^) can be determined for use in calculating a corrected UR
.u , . . (3)
and
UR =100SL,JSL^^ (4)
Assume that LS = 0.38! cm. The correction to the 50% SL is then -0.190 cm and to tbe
2.5% SL is -0.00952 cm. Corrected UR values for the five varieties are compared witb
theoretical and measured values in Table V. The UR. are a much better estimate of
theoretical UR tban the UR .
m
Table V
Comparison of Uniformity Ratios*
Cotton
ID
I
2
3
4
5
Array
39.7
36.7
40.2
37.8
38.3
Measured
45.2
43.6
48.2
44.2
48.5
Difference
+5.5
+6.9
+8.0
+6.4
+ t0.2
Corrected
Measured
38.5
36. t
39.2
37.0
40.9
Difference
-1.2
-0.6
-1.0
-0.8
+2.6
•This is a comparison of values determined from array data with values determined from
Digital Fihrograph data and with values determined by correcting Ihe Digital Fihrograph data.
4. CONCLUSIONS AND SUMMARY
The effects on span length and uniformity ratio of holding length, stan of scan, and fibre
crimp have been examined. Tbey have led lo the following conclusions.
Holding length and fibre crimp, in conjunction with start of scan, can account for the
major proportion of the shorter lengtb measured by tbe Digital Fibrograpb.
Holding length and fibre crimp shorten the measured length while tbe start of scan
increases the length.
The start of scan increases tbe 50% span length by one-balf tbe distance from the edge
of tbe clamp to the start of scan line, but only increases tbe 2.5% span lengtb by 0.025
limes tbis distance. This causes the uniformity ratio determined from measured values to
be excessive, with tbe largest increases occurring in shorter varieties.
A crimp estimate of Ihc order of 6% appears to be a good approximation for the varieties
172 J. Ten. hist.. 1997. f(S Pan I. No. J C Textile Institute
Downloaded
by
[Dalhousie
University]
at
02:49
05
January
2013

8. Dficniiiniiiion of the 50% and 2.5% Span Lengths and the Uniformity Ratio with the Fibrograph
examined.
Holding length is generally of tbe order of 0.432 cm (0.17 in.) for samples obtained
using tbe Fibrosampler. but may differ owing to the rigidity of the fibres.
REFERENCES
American Society for Testing and Materials. Philadelphia. PA, USA. ASTM Designalion: D 1440-55.
American Society for Testing and Materials, Philadelphia. PA. USA, ASTM Designation: D 1445.
American Society for Testing and Materials. Philadelphia. PA. USA. ASTM Designation: D 1447.
American Socieiy for Testing and Materials. Philadelphia. PA. USA, ASTM Designation: D 1449.
Hcdel. K.L.. and Lawson. R., 1964. 'Factors Affecting Fiber Length-scanning Measurements'. Text. Res. J..M.
866.
Krowicki. R.S.. 1986. The Effect of Lens Width on Digital Fibrograph Length Measurements*./ Text. Insi..77,
22.1.
Krowicki. R.S.. and Ramey, H.H..jun.. 1984. 'An Examination of the Digital Fibrograph Length Unifonnity index'.
Crop .Sci.. 24. 378.
Krowicki. K.S.. and Thibodeaux. D.P., 1990. 'Holding Length: Effect on Digital Fibrograph Span Length". Text.
Ri:s../.. 60. 383.
Krowicki. R.S., Hemstreet. J.M.. and Duckett. K.E. 'A Different Approach to Generating the Fibrogram from Fiber
Length Data. Par! I: Theory'./ Text. Ins!., accepted for publication.
Landsireet. C.B.. 1961. 'The Fibrogram: ils Concept and Use in Measuring Cotton Fiber Length'. Fcvf. Hull..S7.,
54.
Worley. S.. jun.. and Krowicki, R.S.. 1968. 'Quality Control in Fiber Testing". 7c,v;. BiilL. 94, No. 4. 32.
J. Te.-t. hist.. 1997. 8S Part I, No. 3 © Textile Institute 173
Downloaded
by
[Dalhousie
University]
at
02:49
05
January
2013