Printer Friendly

Reducing warp in thermoplastics with bilobe glass fibers.

Warp reductions of 20% to 50% have been achieved in crystalline polyesters and their blends with no loss of mechanical properties compared with materials containing standard chopped glass fibers.

Glass fibers have long been used to enhance the physical properties of crystalline resins(1). Unfortunately, the anisotropic nature of the glass-fiber thermoplastic matrix, coupled with uneven shrinkage, can lead to stress in the part, often resulting in warp. The causes of warpage in plastic parts are complex. Material type, part design, and processing conditions all play a role(2). In this article, glass-fiber cross section is shown to have a considerable effect on the warp of injection molded crystalline thermoplastics.


The ingredients were gently tumble blended, with glass fiber added last, and extruded on a vented-barrel, 30:1 L/D, 2.5-inch single screw extruder at 450|degrees~F to 530|degrees~F and 100 rpm. All components were throat fed. Specific formulation data are listed in the Tables.

The pelletized product was dried for more than 4 hrs at 250|degrees~F and injection molded on an 80-ton, 3-oz machine at 450|degrees~F to 500|degrees~F, with a 150|degrees~F mold temperature, and a 30-sec total cycle. Samples were allowed to set for at least one day before testing. Tensile tests were conducted on large, side-gated 7.5- x 1.8-inch bars. Physical properties were determined under the appropriate ASTM procedures.

Measurement of warp can be a complex issue. The method used here was a simple side-by-side comparison of the distortion of 4- x 1/16-inch injection molded discs. A disc is held down on a flat surface and the maximum height from the surface to the bottom of the disc is measured as millimeters of warp(3). Some TABULAR DATA OMITTED TABULAR DATA OMITTED TABULAR DATA OMITTED test samples were 7- x 1/8-inch discs, and their warp values are given in centimeters. Each warp value is an average of at least three samples. Warp measurements were also made on discs annealed for 0.5 to 1 hr at 120|degrees~C to 190|degrees~C. Although the effect of fiber cross section on warp of parts more complex than discs would be of interest, such an evaluation was beyond the scope of this investigation.

Glass Fibers

Chopped glass fibers of non-round cross section have been prepared by several methods in various geometries(4). The fibers evaluated in this work had bilobe and trilobe configurations. As with standard round fibers, bilobe and trilobe fibers can be made in various cross-sectional areas and in different aspect or modification ratios.
TABLE 3. Effect of Coating on Warpage in 30% Glass Filled
 Warp, strength,
Fiber(b)/coating cm MPa
419/"Polyester" 1.40 126
419-418/50:50 1.49 103
418/"Nonbonding" 1.38 63
861/Bare glass 1.28 97
737 1/32-inch 0.28 45
milled fibers
a GE Plastics Valox 300.
b All fibers were K-filament.

The bilobe and trilobe fibers used were experimental samples prepared by Owens-Corning Fiberglas Co. (OCF). The OCF 492 bilobe fibers have a cross-sectional area equivalent to a G-filament round glass fiber of about 10-microns diameter, and an aspect ratio of approximately 2.5. The OCF 492 trilobe fibers have a similar cross-sectional area and a modification ratio of 2.5 to 3.0. The OCF 492 coating is an amino-silane polyurethane type. The BC-01, 02 bilobe fibers have an OCF 189 coating optimized for polyesters(5) and an aspect ratio of 3.0. Other types of round-fiber coatings evaluated include OCF 408 and 419 (designed for thermoplastic polyesters); OCF 418 (a nonbonding glass coating that shows little reactivity with polyesters); and an uncoated fiber, OCF 861. Both K-filament and G-filament round fibers of about 14- and 10-micron diameters, respectively, were used in control experiments. All fibers are composed of borosilicate E-glass.

Crystalline Thermoplastics

The comparison of bilobe, trilobe, and round fibers in polybutylene terephthalate (PBT) and other crystalline resins showed interesting differences. While there are only minor changes in mechanical properties, as reported by others(6), there is a very large reduction in warp using bilobe vs. round or trilobe fibers.

In Table 1, a standard 10-micron-diameter round fiber is compared to a bilobe fiber and a trilobe fiber with an identical coating and similar cross sections in PBT. Mechanical properties showed small variations; however, warp is reduced 30% to 40% using the bilobe fiber. As expected, subsequent annealing further increased warp. The bilobe fiber compositions still showed the best warp control. Analysis of the fibers in molded parts did not reveal any significant differences in fiber length, nor did microscopy show any major differences in fiber orientation. The bilobe fiber geometry is clearly effective in reducing warp without affecting other properties.

In limited experimental work, trilobe fibers did not significantly reduce warp. Apparently, the higher aspect ratio and flatness make the bilobe fiber more effective for warp control. The bilobe fiber combines the plate-like form of a mica filler with the reinforcement of long fibers. In contrast, the trilobe fibers used in this work were similar to round fibers and not as effective in reducing warp.

To determine the effect of glass fiber cross-sectional area on warp, a series of round fibers from 3- to 23-micron diameter were evaluated at 30% loading in PBT. A trend is shown towards lower warp with finer-diameter fibers; however, it is not as significant as the effect of bilobe geometry. As previously observed, because of the greater number of filaments present in a given weight of glass, tensile and Izod impact strength are improved with thinner-diameter fibers(7).

Use of different coatings on round fibers did not have a major effect on warp. Warp and tensile strengths of 30% glass-filled PBT were compared using a bonding coating (419), a nonbonding coating (418), and uncoated glass (816) on K-filament fibers. The nonbonding glass showed reduced strength and no change in warp. The uncoated glass showed slightly reduced warp, but this may be due to increased fiber breakage and more lower-aspect-ratio fibers. Such fibers provide less orientation and, hence, lower differential shrinkage and lower warp. Milled glass fibers controlled warp but did not enhance strength, as do higher aspect fibers.

Thermoplastic Blends

Bilobe fibers were also effective in reducing warp in lower-modulus materials. Bilobe-fiber-reinforced copolyether-polyesters, copolyetherimide-polyesters, and their PBT blends all showed reduced warp compared to similar round-glass compositions. Warp reduction was also observed in compatibilized PBT-polyphenylene ether blends, and in polyester-polycarbonate and polyethylene terephthalate-PBT blends. Bilobe fibers appear to reduce warp in a wide variety of crystalline resins and their blends.

To further improve dimensional stability, blends of polyesters were prepared with round, bilobe, and trilobe fibers in combination with mica, which is known to reduce warp in crystalline resins(3,8). The glass/mica-reinforced blends all showed reduced warp. Again, the bilobe fiber reduced warp more effectively than similar loadings of round and trilobe fibers.

TABLE 6. Warp in 30% Glass Filled PBT/Polyphenylene Ether
 Glass Type(b)
Properties K-Round G-Round Bilobe
@ 676 psi, |degree~C 218 219 216
Notched Izod, J/m 69 69 69
Unnotched Izod, J/m 603 737 657
Tensile strength, MPa 125 139 126
Flexural strength, MPa 188 202 190
Flexural modulus, MPa 7894 8308 9811
Specific gravity 1.436 1.435 4.433
Warp, mm
As molded 15.4 15.1 9.1
1 hr
@ 180|degrees~C 21.3 20.9 16.7
a 60/40 GE Plastics functionalized Valox, PPE blend.
b Fibers: OCF 408 K-round, OCF 189 G-round, and BC-01 bilobe.



1. B. Lauke, B. Schultrich, and W. Pompe, Polym. Plast. Technol. Eng., 29, (7&8), 607 (1990).

2. R. Shaefer and R. Sherman, Plastics Technology, June 1991, pp. 80-83.

3. S. Naik, M. Fenton, and M. Carmel, Modern Plastics, June 1985, pp. 179-184.

4. L. J. Huey and P.D. Beuther, U.S. Patents 4,622,054 and 4,636,234, Owens-Corning Fiberglas Co.

5. A. M. Delvin, R. R. Gallucci, K. N. Gray, and R. M. Harris, U.S. Patent 4,990,549, General Electric Co.

6. K. Shioura, S. Yamazaki, and R. Yoshiyama, 45th Annual Conf. Composites Institute, SPI, Feb. 1990, Session 18-D, pp. 1-6.

7. F. Ramsteiner and R. Theysohn, Composites Science and Tech., 14, 231 (1985). A. D. Wambach, U.S. Patent 4,123,415, General Electric Co.

8. D. L. Phipps and A. D. Wambach, U.S. Patent 4,124,561, General Electric Co.
COPYRIGHT 1993 Society of Plastics Engineers, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1993 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:Additives
Author:Gallucci, R.; Naar, R.; Liu, N.I.; Huey, L.; Schweizer, R.
Publication:Plastics Engineering
Date:May 1, 1993
Previous Article:Changing lifestyles.
Next Article:Society of Plastics Engineers.

Terms of use | Copyright © 2016 Farlex, Inc. | Feedback | For webmasters