Note : Les descriptions sont présentées dans la langue officielle dans laquelle elles ont été soumises.
17
18
1 9 -
Introduction
United States Patent 4,194,114 issued March 25, 1980
~1 to the assignee of the present application describes an improved
22 b#*~ique ~or producing m~lded th rmcplastic ~icles that ha~e levels
23 of electrical conductivity that are useful for many purposes.
24 The articles are molded from plastic pellets that contain a
core of metalized glass fibers. The glass fibers are metalized
26 as they are drawn from a melt of glass and the metalized fibers
27 are formed into a roving. In the subsequent manufacture of the
28 plastic pellets, the roving and a thermoplastic are dxawn from
29 a plastic extruder with the ro~ing forming the core of an
extrusion that is chopped into pellets. When an article is
PO9-77-015 -1-
, .,, i ~
.
,~ ~
- .
1 molded from these pellets, the fibers of metalized glass are
2 distributed through the article and make it electrically
3 conductive.
4 As one example of an application for a conductive
plastic, an article that has been molded from a plastic
6 that conducts only slightly can be yiven an electrical
7 charge and can then be painted by electrostatic painting
8 techniques. A more demanding requirement for conductive
9 plastics is presented by enclosures (covers) for electronic
apparatus that requires electromagnetic shielding. The cover
11 isolates the apparatus from electromagnetic radiation that
12 could otherwise produce spurious signals in the circuits of
13 unshielded apparatus, and it similarly prevents the apparatus
14 within a shielded cover from transmitting signals to interfere
with other nearby apparatus.
16 The shielding capability of a conductive plastic is
17 measured by first molding a sample plaque of the material to
18 be tested. The plaqu~ has standard dimensions and is mounted
19 in the window of a metal box that contains a radio transmitter.
The strength of the radio signal outside the box is measured
21 at various frequencies with the window open and with the window
22 covered by the test plaque. The ratio of the signal measured
23 when the window is open to the signal measured when the window
24 is closed is formed to express the attenuating effect of the test
plaquP in decibels (db's).
26 SUMMARY OF THE INVENTION
27 Conductive plastics are not conductive to the degree that
28 metal conductors are conductive. The metalized glass fibers
29 in the molded article averaye about a quarter of an inch in
length and they provide conductivity to the Pxtent hat they
31 fortuitously touch or very nearly touch. Electromagnetic
PO9-77-~15 -2-
1 shielding may also be attributed to capacitive and inductive
2 coupliny between isvlated fibers.) In an ideal situation,
3 the fibers would have random positions and random orientations
4 so that electrical pathways would extend in three dimensions
from each individual fiber. The fibers can be seen readily
6 in special test samples that are molded of clear plastic,
7 and on the surface of the samples the fibers appear to have
8 this random organization superimposed on the general fill
9 pattern that occurs in the surface of the plastic as it
flows into the mold. However, when samples are disected
11 for further analysis it can be seen that within the body
12 of the sample the fibers line up parallel to each other in
13 the direction of the flow of the heated plastic into the mold.
14 The contact between fibers that lie parallel to each other is
much reduced from the contact that would be expected from the
16 pattern of random orientation of the fibers that are visible ~
17 at the surface of a sample.
18 The random orientation of the fiber at the surface of a
19 plaque probably comes about because friction between the
surace of the mold and the flowing plastic causes turbulence
21 in the plastic near the surface. The surface plastic hardens
22 firstl entrapping the randomly oriented fibers. Similarly,
23 when a test plaque mold is only partially filled, the leading
24 edge of the plastic partial plaque shows a random orientation
of the fibers that is probably caused by turbulence ir~ this
26 region of the fIowing plastic mass. The inner portion of the
27 plaques remain hot and molten and produces less turbulence
28 and the fibers in this region are aligned in t~e general
29 direction of plastic flow during molding.
According to this inventio~ a plastic molding pellet has
31 a core of metalized glass fibers that are arranged to p~rtly
32 disperse as individual fibers
33 and to partially remain as a clump that has an appreciable
PO9-77-015 -3-
1169
1 cross section as compared with the cross section of an
2 individual fiber. The fibers in a clump have numerous
3 points of contact and thus provide good electrical conductivity
4 throughou~ the clump. These fiber clumps also line up in the
direction of flow of the plastic in a molded article, but
6 they have sufficient width in the direction across the
7 direction of flow to significantly raise the amount of
8 bridging that occurs between fibers in a molded article.
9 Fibers can be arranged in the core of a molding pellet
in various ways that will cause the fibers to partially clump
11 in the molded article. For example, the core can be formed
12 of braided roving o metalized glass fibers. Alternatively,
13 a roving of twisted fibers provides a particularly advantageous
14 technique for producing pellets that cause clumps of fibers
in the molded article. To produce a twisted roving, the
16 glass fibers are first formed into sub-bundles and several
17 sub-bundles are twisted about a central sub-bundle in a way that
18 resembles some wire rope. On the other hand, coiling of the
19 fibers, which occurs in some glass making processes, does
not provide clumping, probably because the coils are
21 straightened out during pellet manufacture by the tension that
22 is applied to the roving of metalized glass in pellet manufacturing.
23 We attribute the clumping effect of twisted fibers to the
24 fact that the fibers within the twisted roving do not wet well
to the plastic during the pellet manufacture. During a
26 subsequent injection molding operation with these pellets, the
27 previously unwetted fibers are not mixed as well with the plastic
28 and thus tend to remain in their original form as a sub-bundle
29 of the metalized glass roviny. Thus, we contemplate that other
PO9-77-015 -4-
1 techniques for isolating a sub-bundle of fibers from the
2 mixing action of the injection molding apparatus will
3 pxovide good clumping results.
4 Although the invention adds a step to the glass manu-
facturing operation, it simplifies the subsequent handling
6 of the metalized glass roving by permitting the usual creeling
7 step to be eliminated.
8 The following examples will suggest other features and
9 advantages of the invention.
The Examples
11 Introduction
12 As described later, a number of test plaques were
13 molded at a uniform size of 3" x 6" x 1/4" thick. These
14 plaques are smaller than the plaques used for the tests that
ar~ reported in our United States Patent 4,195,114 ~12" x 12" x V4") and
16 were tested in a smaller test chamber, and as a result the
17 shielding results are numerically lower than the results
18 reported in the related application. To show the comparison
19 between the clumped fibers of this specification and the
non-clumped glass fibers of the related application, sample
21 1 was molded from pellets made from a roving of non-clumping,
22 metalized glass fibers. In all of the samples the pellets and
23 the molded product contained approximately 25 Wt% metalized
24 glass fibers~ The pellets were 3/4" in length and the compo-
sitions of the glass and the metal coating were identical
26 in each example. The genexal conditions of the tests were
27 comparable, and we attribute the improved rPsults of the
28 pellets containing twisted or braided roving to the clumping
29 effect that these pellets produce in a molded test plaque.
The following table shows the shielding levels for
31 samples at frequencies from 12 tc 100 megahertz.
POg-77-~15 -5-
L69
1 Control Example 1 Example 2 Example 3
2 Frequency Sample 1 Sample 2 Sample 3 Sample 4Sa~ple 5
(megahertz)
3 12 16 19 21-22 18 18
~ 20 16 19 20-21 18-19 18-19
15-16 18-19 21~23 18 19-20
6 70 18-19 23-24 22-25 18-21 21-22
7 100 29-32 31-34 23-25 25-26 33
8 The Control SamPle - Sam~le 1
9 Three test plaques of structural roam were molded of
polycar~onate contalning 25 -~,~7t~ metali~ed glass fibers in the
11 form of a straight rovlng, as described in United States Patent 4,195,114.
12 The shielding remained below 20db th ough 70 megahertz. The
13 shielding of the test sample rose to about 30db at 100 megahertz
14 The test was stopped at 100 megahertz. The conductivity of
these test samples is suitable for many purposes when general
16 conductivity is desirable, and the shielding capability is
17 suitable ror many shielding applications. ~owever, it would
18 be desirable to have shielding levels of about 20db from
19 the 1/4" test samples.
Exam~le 1
-
21 Metalized glass fibers containing a binder coating as
22 described in the related applicatio~ were formed into sub-bundles
23 contalning 100 fibers that were essentially straight without any
24 twisting. Eight sub-bundles ~ere woven into a braid having
800 fibers. The braid is the pattern co~only used for the
26 outer cylendrical conductor of coa~ial cables. The braid
27 had eight cross-overs per inch, but we contemplate that any
28 suitable number of cross overs per inch can be used~ T~o of
29 these braids were used to mold pellets of polycarbonate con-
taining 1600 fibers to produce about 25 ~t~ of metalized
31 glass fibers. On the surface, these plaques display large
PO9-77-015 -6-
69
1 swirl patterns showing the path of the plastic flowing
2 into the mold. A plaque was buffed to improve the contrast
3 between the metalized fibers and the plastic, and at the surface
4 the fibers appeared to have a random 3-dimensional orientation
that was generally independent of the swirl patterns.
6 Samples were subsequently fractured approximately
7 parallel to the direction of flow and across the direction
8 of flow to observe the distribution of the metalized glass
9 fibers. The sample contained both clumps of fibers and
individual fibers. The clumps generally retained essentially
11 their initial length and individual fibers appeared in a
12 full range of lengths from dust particles to approximately
13 the length of the original pellets. The clumps appeared
14 to consist of generally parallel, densely packed, fibers and the
braid pattern did not appear in the molded test samples. Both
16 the individual fibers and the clumps appeared to be aligned
17 in the direction of flow of the plastic through the mold. The
18 clumps appeared to be scattered at random with approximately
19 1/4" (about the length of a clump) separating nearby clumps.
Sample 2
21 In sample 2, two test plaques were molded of a mixture of
22 pellets having 50% pellets manufactured with two braids of
23 metalized glass fibers as described in this example and 50~
24 metalized glass fibers having a straight roving as described
for the control sample. Thus, the test plaques had 12.5 Wt%
26 of metalized glass fibers in the form of the braid and a total
27 of 25 Wt% glass fibers. As the table shows, a useful improvement
28 in shielding was measured.
PO9-77-015 -7-
llG116g
1 Sample 3
2 In sample 3, three test plaques were molded and tested
3 according to the procedure described for sample 2 except that
4 the pellet mixture had 75% pellets containing ~he braid of
metalized glass and 25% pellets containing the straight un-
6 twisted metalized glass fibers. Thus the molded plaques
7 had 18 Wt% metalized glass fibers ~n the form of a braided
8 roving and 7% metalized glass fibers in the form of the
9 straight roving. The plaques had a shielding level of more
than 20db across the frequency spectrum of the test.
11 Example 2
12 Metalized glass fibers were formed in the way already
13 described into thirteen sub-bundles that each contained
14 100 metali ed glass fibers. One of these bundles formed a
central member of a twisted roving. Six of the sub-bundles were
16 wound clockwise about the central sub-bundle and the remaining
17 six of the sub-bundles were wound counter clockwise about the
18 inner seven sub-bundles. Thus, the roving had a total of
19 1300 fibers. The twisted roving was used for manufacturing
pellets having the roving as a core, in the way already
21 described. The pellets had about 25 Wt% metalized glass fibers.
22 The sub-bundles wexe twisted together on laboratory
23 apparatus o~ the type used for similarly twisting textile
24 fibers. The roving had about one quarter of a turn per inch.
Although this is only three quarters of a turn along the length
26 of a pellet, the fiber sub-hundles showed interlocking at a cut
27 end of a roving and did not readily come apartO Commercial
28 equipment for ~imilarly twisting textile fibers is available
29 and we contemplate that about eight turns per inch will be
desirable. A minimum tension is required on metalized glass
31 fiber~ to prevent linting w~ich occurs when tension is too
32 low, probably because relati~e motion of the fibers of a
PO9-77-015 -8-
1 sub-bundle occurs under low tension and the fibers abrade
2 each other. A maximum tension is established by the ten~ion
3 at which the roving will break. It appears that within these
4 limits the tension on the sub~bundles is not critical except
that the sub-bundles are given equal tension.
6 Sample 4
7 Three test plaques were molded of pellets manufactured
8 according to this example (without other pellets). Thus the
9 plaques had about 25 Wt% of metalized glass in the form of
a twisted roving. The plaques wexe tested with the results shown
11 in the table. Although the test results are lower at some
12 frequencies than the samples of example 1, the results are
13 sufficiently better than the control test of sample 1 that
14 for most applications the twisted roving will be the preferred
roving form for pellets for the improved shielding of this
16 invention.
17 Test plaques were fractured at approximately right angles
18 to the direction of flow and approximately parallel to the
19 flow and the fibers and clumps of fibers were observed. The
pattern of clumps and the individual clumps were essentially
21 indistinguishable from samples molded with the braided roving.
22 The clumps appeared to correspond to the original sub-bundles
23 of about 100 fibers.
24 Subsequently, the plastic molding pellet~ of Example 2
were used for molding a multipart cover for a terminal device,
26 and the product tested satisfactorily.
27 Example 3 - Sa~ple 5
28 Three test plaques were molded from a mixture of pellets
29 having 75% pel~ets containing the twisted roving described in
Example 2 and 25~ pellets containing the braided roving of
PO9-77-015 -9-
~L1~ 9
1 Example 1. The plaques were tested with results that are
2 generally comparable to the tests of sample 4. A test plaque
3 was then fractured and the brok~n edge was inspected with a
4 microscope. The plaque showed both individual fibers and
clumps of fibers. There was no apparant difference among
6 the clumps that would show whether clumps arose from the braided
7 roving or from the twisted roving. These results appear to
8 support our belief that the twisted roving will be preferred for
9 most applications.
10 OTHER EXAMPLES
11 The pellets of the preceding examples were extruded from
12 polycarbonate but we contemplate that the invention will be
13 used with other plastics that are commonly used ~ith unmetalized
14 glass fibers for both foam and solid plastic molded articles.
A variety of techniques for giving the fibers clumping
16 characteristics will be apparent from the well developed
17 arts of twisting and braiding or otherwise uniting textile
18 fibers, fine metal wires, and the like. For example, outer
19 sub-bundles can be wrapped around an inner sub-bundle with
many turns per inch as compared with the twisted sub-bundles
21 of Example 2. Similarly, the inner bundles can ~e tied together
22 with ~n outer net-like arrangement of a few fibers. ~owever,
23 it is an advantage of the invention that the roving can be made
24 ~y simple techniques such as ~wisting that should not add
significantly to the manufacturing cost of the pellet.
26 From the preceding description of several sepcific
27 examples of the in~ention, tho~e skilled in the art will
28 recognize a variety of modification within the scope of the
29 claims.
PO9-77-015 -10-
