💾 Archived View for gemini.spam.works › mirrors › textfiles › science › htt.txt captured on 2020-10-31 at 02:39:09.
-=-=-=-=-=-=-
Tension testing of four different twines
By Derek Voll
EM 307
4/30/92
Abstract
In this experiment I pulled apart cotton, jute, hemp and nylon twine to
test their ultimate strength. I used a standard tension testing machine
equipped with a load versus displacement plotter. I could not calculate
strain and was thereby limited by the lack of theory to back up my
observations and make descriptive numerical calculations. I did repeat the
tests to produce an average value of ultimate load for each twine group. I
used this value to make a rough stress calculation. Nylon is the strongest,
cotton the weakest and jute and hemp are about equal in strength. I had quite
a bit of difficulty with the nylon specimens because of their high strength
but the others worked out all right.
Introduction
For my independent project I choose to test the strength of four
different kinds of twine, cotton, jute, hemp, and nylon. There are many
factors in choosing the right twine of the job it will be used for, cost,
temperature to be used at, availability, creep and fatigue characteristics.
These and other parameters could be analyzed in future studies to find the
best twine but my project will focus on tensile strength. I think the results
will be meaningful to someone buying twine and the twine producers. In fact
the suppliers of the hemp twine were quite interested in my report and would
like some copies; they would like more scientific information on hemp since
there is so little scientific investigation or research concerning hemp.
My procedure was to obtain twines with similar dimensions, pull them
apart using a standard tension testing machine, collect load versus
displacement plots for each of the specimens and then compare and analyze the
data. I used a tension testing machine with a capacity of 1000 pounds which
had a load versus displacement plotting machine connected to it. By wrapping
the twine around the round spool three times, I relied on the large friction
force to hold the twine in place. This force was not large enough for the
nylon twine and I used the pneumatic grips instead. The pneumatic grips
provided more friction which I needed to hold the nylon twine in place.
OBSERVATIONS AND RESULTS
I would like to start with some general observations of the experiment.
First, the tests of the cotton, jute and hemp twines proceeded with few
problems and their failure occured in the middle of the specimen, which is
desirable in any tensile test since the experimenter can more easily observe
the fracture area and disregard any stress concentration at the clamp-twine
connection. however, for my first two samples of nylon I tried to use the
same clamps that I had used for the other twines but in both cases the twine
overcame the clampUs friction force before failure but after some stretching
(the twine was pulled out of the clamps). Therefore, I switched to the
pneumatic clamps but the nylon still slipped some, wearing the surface of the
twine and causing stress concentrations. The nylon broke at this worn area
near the clamps. Stress concentrations are the very tiny notches and
imperfections in a material that produced a high localized stress. Also, I
did not have enough nylon so I used the same specimens that I had used in the
other clamps and one new specimen. The first two specimens broke sooner and
under less load (see Fig #XX) and this was expected since they had already
undergone some plastic deformation and recovery. I did not realize how strong
and difficult to test the nylon would be. I know my procedure and the
following results for nylon are not accurate but it should be obvious that the
nylon is definitely the strongest of the four twines. I have graphed each of
the specimens together in their respective group (Figs XX- XX) to show the
variances between the individual specimens; nylon has the greatest variance in
displacement and ultimate load as expected but it should be noted that hemp
twines show the second largest variance in ultimate load (all hemp specimens
are from Hungary but the specimens with the lower ultimate load were from a
different supplier than the other three). From these graphs we see that
cotton is the weakest and nylon is the strongest. I would like to point out
that these graphs do not tell the whole story and a better indication of
strength would be a stress versus strain plot, which was impossible to make
since stains could not be calculated because we did not have access to an
extensometer. However, if you look at the sample calculations in the appendix
you will see that the hemp twine had a slightly thicker cross section and the
corresponding stress was comparable to that of the jute twine. Even with this
fundamental calculation we must realize that each twine was probable woven
differently and all their diameters were slightly different.
Looking closely at the graphs for the cotton and nylon specimens you will
see that there are little ridges and drop-offs before fracture; these points
are where the rope must have been slipping in the grips. The curves reach a
high point and then drop off suddenly, the high point is the ultimate strength
point. This high point can be considered the failure point too but I would
like to point out the sharp rises after this point. These sharp rises occur
in the jute, cotton and hemp twine and represent the few fibers that did not
snap at the ultimate strength point. These last fibers stretched a little
further and then snapped under a lesser load. This is different from the
characteristic necking and fracturing that we have learned about in class
where we mainly dealt with metals.
CONCLUSION
In conclusion, I have learned more about tensile testing and the
improvements needed in different applications involving the fundamentals of
stress, strain and fracture mechanics that we have learned in class. I think
that with more accurate tests the results I have found would hold up. I
observed that nylon is indubitably the strongest, cotton the weakest and hemp
and jute are about even. There was some slipping, variance in cross sections
and some amount of error attributed to operator inexperience and the overall
measuring procedure.