What You Need to Know About Melt-Temperature Measurement in Single-Screw Extruders

To be profitable, a single-screw extruder must operate at the maximum rate while discharging at a specified pressure and temperature. Measuring the rate and discharge pressure from an operating extruder is easy and straightforward. Measuring the discharge temperature, however, is not so simple, especially using thermocouples positioned in the transfer line just upstream of the die. This configuration is used widely for commercial extrusions. The difficulty occurs due to the high thermal conductivity of the surrounding metal and the low thermal conductivity of molten resins.

For example, a variable-depth thermocouple positioned in a transfer line and through a flange is shown in Figure 1. The thermocouple measures the temperature at the junction at the tip of the probe. The temperature at the tip of the probe depends on thermal conduction and convection in the local region. The sheath of the probe is typically made from stainless steel while the transfer line is constructed from carbon steel. The thermal conductivity for stainless steel is 17 W/(m°C) and for carbon steel it is 52 W/(m°C). The molten plastic flowing in the transfer line, however, has a thermal conductivity of about 0.25 W/(m°C).

Thus, the thermal conductivity for the surrounding metal is between 70 and 200 times higher than that for the molten resin. Because of this wide difference in thermal conductivities, the junction of the thermocouple is highly influenced by the transfer line temperature and to a lesser level from the molten resin.

This measurement problem is clearly identified with a series of experiments. These experiments were performed using a 1.25-inch diameter single-screw extruder connected to a 25-mm diameter transfer line. The conditions of the extruder were held constant with a rate of 15 lbs/hr. of LDPE at a screw speed of 60 rpm. The transfer line pipe was maintained at either 183°C or 220°C.

The temperature profile of the flowing resin in the transfer line was measured using a plastic bridge constructed using a high-temperature resin (not shown). The bridge was positioned across the flow stream, and it was designed to eliminate energy conduction through the thermocouple device. It was positioned in the middle flange in Figure 1. Thus, the bridge device contained several thermocouples and measured the actual temperature of the flow, and virtually eliminated the thermal conduction problem.

The bridge was too fragile to be used in commercial operations. The transfer line carbon steel pipe was controlled at a temperature of 183°C. The radial temperature profile from the bridge device is shown by the “actual profile” line in Figure 2. Here the profile is parabolic with the lowest temperature being at the wall at 198°C, and the temperature at the center of the pipe at 233°C. This profile occurred because the extruder was discharging at a temperature near 233°C, and the transfer line was in a cooling mode with the pipe temperature at 183°C. The flow velocity in the downstream direction was parabolic and symmetric to the pipe axis.

Next, a variable-depth thermocouple was positioned in the transfer line as shown in Figure 1. The temperature at the junction was measured as a function of the insertion depth, as shown by Figure 2. Here, the measured temperature increased as the probe was inserted deeper into the transfer line. The maximum temperature was at an insertion depth of 22 mm at 232°C. Commercially, melt temperature measurements are obtained using probes that are flush mount to the wall. For this experiment, the temperature near the wall was measured at 198°C. This measurement is clearly in error as the bulk of the material is at a temperature near 233°C.

Moreover, the actual temperature from the bridge device at 22 mm into the stream was measured at 213°C. A high level of thermal conduction through the sheath of the variable-depth thermocouple, however, provides an incorrect measurement at 232°C at the probe junction. At 22 mm into the stream, the conduction through the sheath, however, provided a good estimate of the bulk temperature. Insertion of a thermocouple 90% across a flow stream is commercially impractical, especially for larger diameter pipes. The forces from the viscous flow can be high enough to bend the probe.

The transfer line pipe temperature was then increased and controlled at 220°C. The operation of the extruder was unchanged and thus the extrudate should be at the same temperature as before. The actual temperature profile was measured using the bridge device and it is shown in Figure 3. The inside wall temperature was 229°C, and the actual temperature profile was parabolic with the maximum temperature of 234°C at the center. The profile is considerably flatter than that shown in Figure 2 with a pipe temperature of 183°C. It is obvious that the thermal gradients are very small. A pipe temperature of 220°C did not induce the high level of cooling that was observed for a pipe temperature at 183°C.

The transfer line should not be used as a method to decrease the discharge temperature from an extruder.

The variable-depth thermocouple measurements were nearly linear from 229°C at the wall to a maximum temperature of 235°C at a depth of 22 mm. For a commercial transfer line with a flush-mount thermocouple, the temperature would be reported at 229°C, a temperature close to the bulk temperature of 233°C. A variable-depth thermocouple that would be inserted 4 mm into the flow would report a temperature of 230°C. The probe experiences less thermal conduction from the transfer line pipe, and it provides a better measurement of the resin flow.

The transfer line should not be used as a method to decrease the discharge temperature from an extruder. As shown in Figure 2, a considerable level of thermal gradients was developed in the flowing resin during cooling. The gradients will affect the viscosity of the resin and possibly distort the shape or thickness of the product coming out of the die. Instead, the transfer line should be controlled near the bulk temperature of the extrudate, minimizing thermal gradients at the die. Because the bulk temperature of the extrudate is typically not known, it should be occasionally measured using a handheld thermocouple in the extrudate exiting the die. The transfer line pipe temperature should be controlled near this temperature.

Measuring the melt temperature using a thermocouple positioned in a transfer line is standard in the industry. Even though the method can provide measurement errors, it is used because it is simple, low cost and safe. Many times, the melt temperature can be measured using a handheld thermocouple probe by sticking it into the extrudate stream. It can take a minute or more to get the probe at the temperature of the extrudate. Sometimes the downstream equipment such as rolls can prevent the safe measurement of the extrudate. In this case, the extrudate temperature should be measured using in infrared (IR) gun.

About the Author: Mark A. Spalding is a fellow in Packaging & Specialty Plastics and Hydrocarbons R&D at Dow Inc. in Midland, Michigan. During his 39 years at Dow, he has focused on development, design and troubleshooting of polymer processes, especially in single-screw extrusion. He co-authored Analyzing and Troubleshooting Single-Screw Extruders with Gregory Campbell. Contact: 989-636-9849; maspalding@dow.com; dow.com.