Printer Friendly

Direct pelletizing with rotating-oscillating kneaders.

Direct Pelletizing With Rotating-Oscillating Kneaders

Screw pumps and gear pumps have traditionally been used to generate discharge pressure for pelletizing with rotating-oscillating kneaders. The kneader screw is mechanically synchronized to simultaneously rotate and oscillate. The screw interacts with stationary pins in the barrel, resulting in material transport, self-wiping, and distributive and dispersive mixing. During one rotation of the screw, one cycle of oscillation occurs, resulting in a pulsating output of the kneader discharge.

The feasibility of direct strand pelletizing without a secondary discharge device, the pulsating output of the kneader notwithstanding, is explored in this article. Preliminary screening work indicated that this method was best suited for polymer systems with low die swell and good melt strength. Three glass-fiber-reinforced polymer systems were compounded. The polymers, selected for their diverse rheological characteristics, were nylon 66, polypropylene, and polycarbonate.


The evaluation was carried out on an MDK/E-46 rotating-oscillating kneader with a screw diameter of 46 mm and a process length of 11 L/D. An overview of the machine configuration is illustrated in Fig. 1; the screw design used for all three compositions is shown in Fig. 2. The polymers were metered with a gravimetric feeder into the first inlet, where the sequence of screw elements in the first 4 L/D provided solids conveying into a kneading section, where melting occurred. A restriction ring, located at the end of the first kneading section, provided a distinct melting zone and a dynamic melt seal.

The chopped glass fiber was metered gravimetrically into the second inlet, where the second 4 L/D was designed for melt conveying in the initial stage and fiber distribution and wetting in the latter. A restriction ring of lesser severity aided glass fiber incorporation and provided a dynamic melt seal for vacuum venting in the final section. The final 3 L/D was composed primarily of conveying elements for generating discharge pressure and vacuum venting. A short mixing section was placed at the end of the screw to fully homogenize the devolatilized melt.

The strand die body was a circular-to-slot design; it was flanged directly to the end of the kneader. The strands were in a linear pattern, 30 [degree] down from the horizontal. The pressure transducer was mounted in the die body immediately behind the die plate.

The die plate had six 4-mm orifices with a cylindrical length-to-diameter ratio of 1.6 and a centerline spacing of 11.8 mm. Two of the orifices were plugged to simulate a four-orifice die for certain experiments.


An important aspect of this research was the determination of the interaction between the rotating-oscillating screw and the pelletizing die. This was accomplished by high-speed measurement of the oscillating die pressure while simultaneously measuring the point in time when the screw was in the full-forward position.

The oscillating melt pressure upstream of the die plate was continuously monitored with a digital storage oscilloscope. The stored data was also recorded on a high-speed printer. The pressure signal was generated with a pressure transducer having a range of 0 to 20.7 MPa. An indicator produced a 0 to 10-VDC signal, which was fed to the oscilloscope. These voltages, plotted as a function of time, were correlated directly to pressure.

An inductive proximity switch, located on the gearbox end of the machine, was used to correlate the axial position of the oscillating screw with the resultant die pressure. This was measured on the second channel of the oscilloscope simultaneously with the pressure trace. The melt temperature of the strands exiting the die was measured with a platinum RTD.


The three glass-fiber-reinforced compositions are given below, with weight percentages of their respective components. Compound B contained 20% of a coupling agent, an acrylic acid-grafted polypropylene.

Compound A. Nylon 66: Du Pont Zytel 101-NC010, 67 wt%. Glass fiber: PPG 3540, 1/8 in, 33 wt%.

Compound B. Polypropylene: Himont Profax 6523, 40 wt%. PP (g-acrylic acid): BP Performance Polymers Polybond 1001-40 MF, 20 wt%. Glass fiber: Owens Corning OCF457BA, 3/16 in, 40 wt%.

Compound C. Polycarbonate: Mobay Makrolon 2600-1000 Nat, 80 wt%. Glass fiber, PPG 3090, 1/8 in, 20 wt%.

The output and kneader screw speed were varied with each composition while constant barrel, kneader screw, and die temperature settings were maintained. The test program was generally carried out with two die configurations - four orifices and six orifices, both with 4-mm diameter. Portions were also repeated with the die body removed (open discharge). The temperature settings for each compound follow.

Nylon 66: screw 280 [degrees] C; zone-1 300 [degrees] C; zone-2 280 [degrees] C; die 310 [degrees] C.

Polypropylene: screw 170 [degrees] C; zone-1 190 [degrees] C; zone-2 170 [degrees] C; die 190 [degrees] C.

Polycarbonate: screw 200 [degrees] C; zone-1 240 [degrees] C; zone-2 200 [degrees] C; die 270 [degrees] C.

Process data for the three compounds are shown in the Table.

Table : Process Data for the Three Experimental Compounds.
 Specific Melt Die type, Peak die
Output, energy, temp., no. orifices pressure,
kg/hr rpm kwh/kg [degrees] C x diam., mm MPa

Compound A:
20 100 0.090 292 4 x 4 4.4
20 100 0.091 300 6 x 4 4.4
20 200 0.135 309 4 x 4 2.5
20 200 0.131 309 6 x 4 2.7
20 300 0.185 322 4 x 4 1.2
20 300 0.195 322 6 x 4 1.7
40 200 0.174 309 4 x 4 5.6
40 200 0.160 298 6 x 4 4.7
40 200 0.155 300 None -
40 300 0.188 320 4 x 4 3.7
40 300 0.187 321 6 x 4 3.3
40 300 0.176 312 None -
60 300 0.192 319 4 x 4 5.7
60 300 0.192 307 6 x 4 4.8
60 300 0.183 306 None -

Compound B:
20 100 0.075 207 4 x 4 4.9
20 100 0.065 198 None -
20 200 0.140 213 4 x 4 3.1
20 200 0.115 217 None -
20 300 0.190 232 4 x 4 2.4
20 300 0.165 231 None -
40 200 0.130 218 4 x 4 5.3
40 200 0.123 208 None -
40 300 0.165 227 4 x 4 3.9
40 300 0.153 225 None -
50 300 0.176 235 4 x 4 4.7
50 300 0.162 216 None -

Compound C:
20 100 0.175 269 4 x 4 8.6
20 100 0.165 274 6 x 4 7.2
20 100 0.150 270 None -
20 200 0.240 292 4 x 4 6.2
20 200 0.225 285 6 x 4 5.0
20 200 0.205 292 None -
20 300 0.260 310 4 x 4 4.5
20 300 0.290 - 6 x 4 3.8
20 300 0.260 307 None -

Results and Discussion

Die pressure as a function of time. The general form of this relationship, illustrated in Fig. 3, shows a nearly symmetrical rapid increase in die pressure to its peak value and an equally rapid decrease to an essentially pressureless region. The pressure trace indicated that measurable die pressure existed for approximately on half of the time, while near-zero pressure was observed during the remaining time.

Die pressure as a function of axial screw position. The peak die pressure occurred once for each screw rotation and was found to coincide with the full-forward position of the rotating-oscillating screw. This was determined by measuring the time between peaks and back-calculating the screw rpm.

Once the screw reaches the full-forward position at peak die pressure, its axial motion changes from a forward to a reverse direction, and die pressure begins to diminish. Die pressure and positive material flow are sustained for a portion of the reverse motion of the screw, as evidenced by the similarity of the pressure increase to the peak and its subsequent decrease (Fig. 3). Eventually, the rearward axial motion of the screw exceeds the melt-conveying capability, and the die pressure drops toward zero.

Based on the operating principle of the kneader, the midpoint in time between two peaks corresponds to the rear position of the rotating-oscillating screw and, Fig. 3, the approximate midpoint of the pressure valley. Once again, the axial direction of the screw changes back to forward. However, the die pressure does not begin to build immediately, probably because material must be replenished before additional die pressure can be generated.

Average die pressure. This was estimated from the area under the pressure plot averaged over one screw rotation and found to be 20% to 30% of the peak die pressure.

Influence of screw speed on die pressure. The data in the Table show that at fixed output, for all three compounds, die pressure decreased significantly with increasing screw speeds. At a fixed output, as the screw speed is increased, a lesser volume of material is extruded during each revolution at a greater frequency. In addition, specific energy and melt temperature increase with higher screw speeds (see the Table), leading to a decrease in polymer viscosity. Each effect should act to lower die pressure. Changes in flow rate are shown later to have the greater influence on die pressure.

Influence of output on die pressure. Data in the Table show that at fixed screw speed, the peak die pressure for all three compounds increased in nearly linear fashion as the output increased. At fixed screw speed, the frequency at which material is extruded is constant. Therefore, as the output is increased, a larger volume of material must be extruded for each screw rotation, and peak die pressure is increased.

Influence of melt temperature on die pressure. The data points for Compound C, plotted in Fig. 4, indicate that this is not a simple relationship. Similar results were found for the other compounds.

As previously discussed, die pressure is strongly influenced by screw rpm and output. These factors determine the volume of material extruded per screw rotation. In order to properly understand and optimize the process, a correlation was sought to help explain the effect of die back pressure (or resistance to flow) on melt temperature rise. Figure 4 shows a correlation between die peak pressure and exit melt temperature as a function of shear rate at constant fill factor. A proportional relationship between average shear rate and melt temperature is observed. Because this data is for an individual orifice, it strongly suggests that melt temperature could be reduced simply by using a die with more orifices.

Figure 5 compares direct pelletizing with a conventional kneader system that uses a crosshead extruder. Major differences are noted in residence time and melt temperature, with the "direct" system exhibiting only half the total residence time in the molten state. The excess temperature generated during the last 3 or 4 sec of kneader length is attributed to increased residence time or back-filling caused by the die pressure. Another 10 sec at temperature occurs in the die adapter and manifold.

Stranding performance. The majority of the samples were strand pelletized through a water bath with ease, particularly at medium and high outputs. A few of the samples with low outputs from the six-orifice die suffered from low strand speeds and inadequate pelletizer power.


Direct pelletizing was successfully implemented for glass-fiber-reinforced polymers with diverse rheological characteristics - nylon 66, polypropylene, and polycarbonate.

The die pressure oscillated so rapidly, and with such regularity, that a quasi steady-state flow through the die resulted in a stable stranding operation.

Discharge melt temperature could be reduced, as evidenced in Fig. 4, by increasing the number of orifices.

Product quality may be improved by this process through a reduction in residence time and elimination of fiber attrition from a secondary discharge device.

PHOTO : Figure 1. Overview of the equipment for direct pelletization.

PHOTO : Figure 2. Kneader screw design.

PHOTO : Figure 3. Very rapid and regular oscillations in die pressure created a quasi steady-state flow that enabled a stable stranding operation.

PHOTO : Figure 4. The effect of die back pressure on melt temperature as a function of shear rate.

PHOTO : Figure 5. Time-temperature comparison of direct-pelletizing vs. conventional pelletizing.
COPYRIGHT 1989 Society of Plastics Engineers, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1989 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Schneider, Loren T.; Sedlack, Thomas E.
Publication:Plastics Engineering
Date:Dec 1, 1989
Previous Article:U.S. machines recouping market share from imports.
Next Article:SPE Automotive Division hosts record awards event.

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