Rubber testing for injection molding.

The process of injection molding rubber compounds runs most efficiently when operated continuously. Setup costs for injection molding are usually higher than for other molding processes, but operating costs can be significantly lower if the process operates continuously (ref. 1). Economic benefits from injection molding depend on successful processing of each batch to maintain continuous operation. Rubber testing helps optimize molding conditions, predicts the processing potential of each batch and predicts or measures the quality of cured products.

The injection molding process consists of an injection machine, mold and a rubber compound. Each component influences the success of every molding cycle. Batch variation of a rubber compound particularly affects the success of each molding cycle. Rubber compounds contain many ingredients and require a complex mixing and preparation process that can cause significant variation from batch to batch. Recently introduced tests (refs. 2-4) and injection molding control systems (ref. 5) can improve molding efficiency by measuring compound variation more accurately and by adjusting the process if necessary.

Injection molding process

Injection molding of vulcanizing rubber compounds combines components of simpler forming and molding processes into an integrated feed and mold system. The most basic component of this process is similar to compression molding. Other components include transfer channels, transfer reservoirs, feed pistons and feed extruders.

In the compression molding process, a sample is preweighed and pre-formed so as to fill the mold cavity quickly and easily when the mold is closed. The mold is held closed with the sample under pressure until sufficient time has elapsed for vulcanization. The mold is then opened, the sample removed and the mold cleaned and reloaded for the next cure. Cavity pressure is maintained by slightly overfilling the mold and by forcing the mold components together, usually with a hydraulic press. The heat for curing is provided by electric heaters and/or hot water or steam.

Transfer molding is a form of injection molding. Transfer molding systems hold the compound to be cured in a heated reservoir. At the beginning of the cure cycle, the amount of rubber necessary to fill the mold is transferred through a runner system and into the mold cavities by means of a piston. Pressure is maintained on the rubber in the mold by the piston and a mold closure system. The mold must be hotter than the transfer reservoir so the rubber in the mold is cured, but the rubber in the reservoir is not cured.

Injection molding combines transfer molding and preform equipment into one system. In the transfer mold system, a ram or screw type extruder is used to mass the rubber into a preform. The preform is a volume of rubber designed to fill the reservoir for transfer to a mold under pressure. In an injection molding machine an extruder is mated with a reservoir and mold.

The extruder is designed so that a piston or a reciprocating screw can be advanced under pressure to precisely fill the runner and mold cavities. While the rubber in the mold cures, fresh material is prepared for the next injection cycle. By continuously feeding the injection extruder, molding cycles may be run without interruption. The injection molding machine is usually designed to operate with minimal operator interaction by automatically ejecting cured parts and automatically feeding stock to the injection extruder. Process interruptions occur if the pans are improperly cured or if the rubber cures before injection. Improper cures include: underfill of the mold, excessive flash, air voids in cured parts, undercured parts, overcured parts and unsatisfactory part performance. These process interruptions may be caused by variations in the injection molding press conditions or by variations in the rubber compound.

Injection molding compounds

Compounds used in compression or transfer molding may not be optimized for injection molding (tel. 1). Injection molding compounds must flow through the nozzle and runner system and fill the mold within the injection pressure range available. The stock must not cure before filling the mold, should cure in a cycle time that allows efficient continuous operation, and the cured physical properties of the injection molded part must meet desired levels. Other restrictions may also apply, such as elimination of molding voids and requiting sufficient compound strength to permit automatic ejection of cured parts.

A high viscosity compound may not fill the mold properly or may generate too much viscous heat when injected, leading to scorch. A low viscosity compound may not generate-desired viscous heat, leading to incompletely cured parts. If scorch time is too short, the compound may cure before the mold is filled and interrupt the process. If cure time is too long, the compound may be incompletely cured. Adjusting the cure cycle time to compensate for a long cure time may increase the heat history for the stock in the reservoir and lead to scorch. Limits for each of these stock properties define a "process window" for a specific compound and a specific process.

It is useful to define the process window for a compound and match the press conditions to the compound. The process window may be found by conducting experiments on the injection mold press, but this can be expensive and time consuming. Laboratory tests conducted on rubber compounds can help define the process envelope for a typical compound and identify batch to batch variations that might cause problems.

Rubber tests for injection molding

A number of rubber tests are commonly used to monitor and control the injection molding process: Mooney viscometer, capillary rheometer and curemeter.

Other tests that may also provide useful information include stress relaxation and dynamic mechanical theological tests.

Mooney viscometer

Mooney viscometers measure low shear rate viscosity at processing temperatures and viscosity and scorch time at curing temperatures. Viscosity tests are usually run at 100[degrees]C and 121[degrees]C using ML-1+4 test conditions according to ASTM Test Method D-1646. Mooney scorch tests are typically run at 121[degrees]C and 135[degrees]C, also following ASTM D-1646.

The shear rate of the Mooney viscosity test is 1.5 [s.sup.-1] at the edge of the rotor, and zero at the center of the rotor. Shear rates in injection molding range from 100 [s.sup.-1] to 10,000 [s.sup.-1] The large difference in shear rates between the Mooney viscometer and injection molding limits the usefulness of Mooney viscosity data in predicting mold flow variations. Rubber stock temperature also covers a wide range in injection molding, and it is useful to measure viscosity and scorch at more than one temperature.

Capillary rheometer

Capillary rheometers can model the flow conditions in an injection molding machine. As mentioned before, shear rates in injection molding are much higher than in a Mooney viscometer. An example where Mooney viscosity cannot detect differences that may be significant at high shear rates is given in table 1. In this example, 16 batches of an SBR compound were mixed. Mooney viscosity values were not significantly different among the 16 batches. Flow resistance measured by a Monsanto Processability Tester (MPT) capillary rheometer varied significantly between batches, especially at 1000 [s.sup.-1]. Variations reported by the MPT in table 1 might cause significant variation in injection mold filling.

Curemeters

Rubber curemeters measure many compound characteristics that influence injection molding. There are two types of curemeters, oscillating disk rheometers (ODRs) and rotorless rheometers. The Monsanto ODR 2000 rheometer is an example of an ODR. The Monsanto MDR 2000 rheometer (MDR) is a rotorless curemeter. The MDR has a more isothermal test chamber than an ODR, and the MDR measures both elastic and viscous torques, while an ODR only measures elastic torque responses (ref. 6).

Curemeters measure the change in stiffness of a rubber compound as heat is applied over time. Stiffness is measured as resistance, or torque, responding to an applied oscillation of a rotor or die.

Figure 1 shows a typical MDR rheograph. Elastic torque (S') usually drops to a minimum value (ML) as the specimen reaches the test temperature, and then the torque rises as the vulcanization reaction takes place. Elastic torque will reach a maximum (MH) or marching cure state when the specimen is fully vulcanized. Scorch time (TSI) is measured as the time to reach a rise of one dN.m above minimum (ML). Cure times (TC50 and TC90) are typically measured as the times for the torque to rise to 50% or 90% of the difference between ML and MH. Scorch and cure times are usually shorter for MDR tests than for ODR tests (ref. 7), especially at high mold temperatures used in injection molding.

Stress relaxation

Many instruments have been designed to perform stress relaxation tests. Mooney viscometers are now capable of performing a stress relaxation test at the end of a viscosity test (tel. 3). Stress relaxation tests provide information about the viscoelastic character of a compound in addition to the viscosity measured by standard tests. Both viscous and elastic properties influence the flow of rubber through the nozzle, runners and gates of injection molding systems.

Dynamic mechanical rheological testers

Dynamic mechanical rheological testers (DMRTs) have been available for many years, but have been limited to measuring dynamic properties of cured rubber, or viscoelastic properties of uncured rubber at low shear rates. A new type of DMRT, the Monsanto RPA-2000 Rubber Process Analyzer (RPA) has recently been introduced (ref. 4). The RPA uses the same sealed test cavity as the MDR, but has the added capability to vary frequency, amplitude of oscillation and temperature during a test. Because the test cavity is sealed under pressure, a wider range of test conditions can be tested than with other types of DMRTs.

Dynamic mechanical rheological tests can measure similar properties to those reported by capillary rheometers and stress relaxation tests. In the case of the RPA, cure results can also be measured in one comprehensive test.

Example use of rubber tests for injection molding

Example injection molding process

To demonstrate how rubber tests can help control injection molding, example test results for the NR formulation in table 2 are applied to a hypothetical example molding process. The example process uses a 1.33 MN (150 ton) horizontal reciprocating screw injection molding machine. The extruder barrel and nozzle are heated by a circulating fluid system. An electrically heated mold requires a shot size of 85 [cm.sup.3], the reservoir has a volume of 221 [cm.sup.3] and the nozzle is 5.56 mm in diameter.

Injection molding setup involves a number of conditions that interact with certain characteristics of the rubber stock. Stock properties that are desirable to know include:

* Scorch time at more than one temperature;

* cure time at more than one temperature; and

* flow properties at multiple temperatures and multiple flow rates.

Scorch times help estimate the state of scorch before the mold is filled. Cure times estimate the state of cure at points in the mold based on the predicted temperature history of the stock. Flow properties aid in predicting injection times at a given temperature and injection pressure.

Desired injection molding machine information includes:

* Stock temperature in the reservoir for a barrel temperature setting;

* temperature rise and injection time for an example injection condition; and

* thermocouple heat history for the mold surface and the coolest part of the mold in an example injection molding.

Rubber stock and injection molding machine information can be combined to form a model of the process. If this information is unknown, it must be estimated to create an injection molding setup. Estimated values may be significantly different from the optimum cure condition.

Example test results

Test results used for the molding example are listed in tables 3-6. Table 3 contains Mooney viscosity and Mooney scorch data at 100[degrees]C, 121[degrees]C and 135[degrees]C. Table 4 contains ODR data at 150[degrees]C to 190[degrees]C, in 10[degrees]C increments. Table 5 contains MDR data at the same temperatures as in table 4. Capillary rheometer data from a Monsanto Processability Tester (MPT) are listed as apparent shear stress vs. apparent shear rate at typical process temperatures and shear rates in table 6. Note that scorch and cure times measured by the MDR are significantly shorter than for the ODR in this example.

Example use of lab tests for setup

In the molding example, an "optimum" cure condition must meet the following goals:

* The coolest part of the mold must reach a 50% cure, based on estimated state of cure;

* the surface of the molded part must reach at least 90% cure, and not exceed 1.5 times 90% cure time, based on estimated state of cure;

* estimated scorch of the stock to fill the mold must be less than 1/2 the scorch time, based on estimated state of cure;

* the time for a cure cycle shall be minimized as long as the specified cure and scorch requirements are met; and

* the difference in state of cure between the coolest and hottest parts of the molding shall be minimized.

Scorch times are set at 50% of the scorch measured by test instruments as a safety factor. The need for a scorch safety factor is based on possible localized hot spots that might cause scorching. Experience has shown this safety factor to be a useful practice (ref. 8). MDR data are used for scorch and cure times to establish the cure objectives in this example. The more isothermal test temperature of the MDR results in more accurate cure times compared to ODR tests (ref. 9).

The cure objectives expressed as numerical limits for cure times equivalent to 170[degrees]C MDR test results are:

* Mold core cure > MDR TC50 = 124 seconds;

* mold surface cure > MDR TC90 = 170 seconds;

* mold surface cure < 1.5 x MDR TC90 = 255 seconds;

* cure after mold fill < 1/2 MDR TSI = 62/2 = 31 seconds;

* minimum cure cycle time; and

* minimum difference of (mold surface cure - mold core cure).

Injection time

The time to fill the mold depends on the rheology of the rubber stock. Rubber flow rates vary with temperature, injection pressure and nozzle and mold geometry. Mold flow design programs (tel. 10) can aid in the design of molds for complete fill and uniform cure. These mold design programs usually require rheological information obtained from capillary rheometer tests, such as shear stress at specified values of temperature and shear rate. Mooney viscometer data may also be used but their correlation with mold flow behavior is more qualitative than quantitative due to differences in flow conditions.

Capillary rheometer data may also be used to predict mold fill times for an existing mold. It has been shown (ref. 11) that the injection rate at other pressures and temperatures can be predicted with relatively small errors by correlating a single injection trial to capillary rheometer data.

Equation 1 describes apparent shear stress as a function of temperature and apparent shear rate, based on the MPT capillary rheometer data in table 6:

[sigma]a = (38.885 x 106) (T-3.0564) ( [gamma a.sup.0.5873]) (1)

where: [signa]a is the apparent shear stress, kPa; T is temperature of the rubber stock, [degrees]C ; and [gamma]a is the apparent shear rate, 1/seconds ([s.sup.-1]).

An experimental injection trial in the hypothetical mold at a barrel temperature of 115[degrees]C and a mold temperature of 170[degrees]C takes 10 seconds at an injection pressure of 52.5 MPa. The stock temperature at the nozzle is estimated by:

Nozzle temperature = [T.sub.n] = ([T.sub.b] + 2 x [T.sub.m])/3 = 151.7[degrees]C (2) where: Tn is the nozzle temperature, [degrees]C ; Tb is the barrel temperature, [degrees]C ; and [T.sub.m] is the mold temperature, [degrees]C.

With a nozzle diameter of 5.56 mm, the calculated apparent shear rate for the example injection averages:

[gamma]a = (32 x V)/([Pi] x [D.sup.3] x I) z (32 x 85,000)/([Pi] x [5.56.sup.3] x 10) = 5037/10 = 504 [s.sup.-1] (3)

where: V is the injection shot volume, [mm.sup.3]; D is the nozzle diameter, mm; and I is the injection time, seconds.

The apparent shear stress at an apparent shear rate of [504.sup.s-1] and a temperature of 152[degrees]C is 324.5 kPa, from equation 1.

In the injection molding trial, the injection pressure of 52.5 MPa may be used in equation 4 to calculate an effective L/D value for this hypothetical nozzle and mold combination:

[sigma]a = 324.5 kPa = IP/(4(L/D)) = 52,500/(4(L/D)) (4) where: IP is the injection pressure, kPa, L/D is the effective length/diameter ratio for the injection molding system and L/D = 40.4 in equation 4.

Combining equation 1 for apparent shear stress vs. apparent shear rate with equation 2 for stock temperature at the nozzle, equation 3 for apparent shear rate vs. injection time and equation 4 for apparent shear stress vs. injection pressure, the relationship between injection pressure and injection time for the molding example is expressed in equation 5:

I = [(IP x [T.sup.n.sup.30564])/(938.67 x [10.sup.9])][sup.-1.7027] (5) Table 7 lists predicted injection times for the hypothetical example injection molding process, using equation 5 and capillary rheometer data in table 6.

Temperature rise during injection

The temperature increase as rubber stock passes through the nozzle and fills the mold is a key parameter affecting scorch and state of cure in injection molding. For a given rubber stock and nozzle, the temperature rise should be proportional to injection pressure (tel. 8). Experiments have verified this for a number of examples, with the observation that the temperature rise is negligible below a minimum flow rate (ref. 8). Ideally, the temperature rise should be measured at two injection pressures using an air shot. In this test, the stock is injected with the mold open, and collected in an insulated container for temperature measurements. Table 8 indicates that for the hypothetical example, at a barrel temperature of 115[degrees]C, the temperature rise is 9[degrees]C at an injection pressure of 70 MPa and 3[degrees]C at an injection pressure of 35 MPa. Temperature rise at other conditions in table 8 is calculated from these values, assuming a linear rise in temperature with an increase in injection pressure.

Variation of stock viscosity from batch to batch, and variations in barrel temperature over the normal operating range have small effects on injection temperature rise, since work energy due to injection pressure is the main contributor to temperature rise in the injection process (ref. 12). Compound formulation and polymer type can influence temperature rise significantly (ref. 12), and an estimate of temperature rise should be made for each situation. In the example, injection temperature rise is assumed to vary only with pressure.

Cure time calculations

To estimate the state of cure as the stock moves through the injection molding process, a time-temperature profile may be calculated. Figure 2 is the temperature profile calculated for the "start" cure cycle in table 9. In this setup, the cure time in the mold is set at 228 seconds (the ODR TC90). Injection time varies with the stock and injection conditions. The time required to open the mold, remove the cured part and close the mold is assumed to be 20 seconds. The shot size of 85 [cm.sup.3] is 1/2.6 of the barrel and reservoir volume, so the rubber stock will be in the barrel for 2.6 times the combined cure, injection and mold open times. Refill of the reservoir is assumed to take place during the beginning of the cure time, and is not included in the cycle time calculations. Thus, for the "start" setup in table 9, the following time interval calculations apply:

C = cure time -- 228 seconds

I = injection time = 11.2 seconds

O = mold open time = 20 seconds

R = reservoir time = 2.6 x (C+ I + O) = 2.6 x (259.2) = 673.9 seconds (5)

Time to mold fill = R + I = 685.1 seconds (6)

Time to mold open = R + I + C = 913.1 seconds (7)

Cycle time, the time required for each cure in a continuous process, is determined by the sum of the injection time, the cure time and the open time:

Cycle time = (C+ I +0) -- (228 + 11.2 + 20) -- 259.2 seconds. (8)

The temperature of the stock as it moves through the injection molding process may be calculated using information obtained from the injection molding machine and rubber tests, as previously discussed. In the molding example, the barrel and reservoir are assumed to be at the barrel control temperature for the reservoir time interval, R. The temperature rise during injection is assumed to take place over the injection time interval, I. The temperature of the stock in the mold is estimated at two locations. Stock at the mold surface is assumed to rise in temperature at a rate of 1[degrees]C/second until the mold temperature (T.sup.m]) is reached. The mold core is assumed to be at least 13 mm from the mold surface, and its rate of temperature rise is assumed to be ([T.sub.m]/1000) [degrees]C/second. These assumptions were used to calculate temperature conditions in table 9 and figures 2, 3 and 7.

The state of cure in a variable temperature process may be estimated by a variety of methods (tel. 9). The simplest and most prone to error is to use an arbitrary factor based on the curemeter TC90. More detailed methods calculate equivalent cure times based on a calculated rate of cure at different temperatures. An alternative is to program a curemeter to simulate variable temperature cure conditions and monitor torque as an indication of state of cure. The most desirable method is to use MDR tests to measure reaction rates, and reaction kinetics to calculate the activation energy. Once the activation energy is known, equivalent cure times for a time-temperature profile can be accurately calculated using the Arrhenius equation 9. Equivalent cure times in table 9 were calculated using an activation energy of Ea -- 95.7 kJ/mole (ref. 9).

([1/t.sub.2]) = ([1/t.sub.1]) x exp {-[E.sub.a(T.sub.1-T.sub.2)/(R x T.sub.1 x T.sub.2)]} (9)

Where: [t.sub.i], seconds, is the cure time at temperature [T.sub.i], [degrees]K; [E.sub.a] is the activation energy, kJ/mole, and R is the universal gas constant.

The "start" setup in table 9 ends with less than 50% cure at the core of the mold. The cure time for this setup was arbitrarily set at the 170[degrees]C ODR TC90. This cure condition would have been satisfactory for a thin molding, since the surface equivalent cure was between the TC90 for the MDR and the TC90 for the ODR at 170[degrees]C . When molding thick parts, the state of cure in the center of the part must also be calculated, and the molding time must be long enough for the center to reach a minimum state of cure.

In table 9, several setup conditions are shown as stepwise changes that improve the match to the molding objectives. In setup 1, the cure time of the "start" setup is increased from 228 to 360 seconds. The longer cure time increases the equivalent cure time of the core to equal MDR TC50 at 170[degrees]C, but the surface cure exceeds the overcure restriction. Setups 2, 3 and 4 reduce surface cure, cycle times and reduce the state of cure difference between surface and core of the molded part. Setup 4 meets all of the cure objectives.

Optimum cure conditions may be predicted by using a designed experiment to minimize the number of trials. Assumptions in the hypothetical molding example were used to generate trials according to the 3 factor central composite experimental design listed in table 10. Independent variables are barrel temperature, mold temperature and injection pressure. Cure times at each molding condition were calculated to achieve a cure equivalent to 170[degrees]C MDR TC50 at the core of the mold. Dependent variables are cycle times, equivalent cure at the surface, cure state after mold fill and the difference in cure state between surface and core.

Figure 4 is a contour plot for cycle time as a function of mold temperature and injection pressure at a barrel temperature of 120[degrees]C , using regression equations calculated from the designed experiment. Regions defining limits for scorch at fill (1/2 TSI) and overcure (1.5 x TC90) are shaded. According to figure 4. mold temperatures of 170[degrees]C and 160[degrees]C allow a range of injection pressures at 120[degrees]C barrel temperature that define setups that meet the cure objectives.

Cycle time contours are reduced in figure 4 by increasing mold temperature and injection pressure. Contours for surface minus core cure. shown as dashed lines. are reduced by decreasing the mold temperature. In this case. the process engineer must choose between the lowest cycle time and the lowest surface minus core cure difference in selecting mold temperature.

At 170[degrees]C mold temperature, contours for cycle time as a function of barrel temperature and injection pressure are plotted in figure 5. As in figure 4, regions defining scorch and overcure limits are shaded, and contours for surface minus core cure are added. The shortest cycle time in figure 5 occurs at 170[degrees]C mold temperature, 120[degrees]C barrel temperature, and 87.5 MPa injection pressure. This is setup 4 in table 9.

Testing for batch variation

Once a setup has been established for a compound, the process must deal with batch to batch variation. Experience may be used along with test measurements of each batch to detect stock properties that present process problems. Test information can also be used to predict the effect of compound variations and possible corrections. This is examined in the following sections using the model of the example process (setup 4 in table 9) to predict the effect of variation in test results.

Viscosity variation

A batch with a viscosity five Mooney units higher than a nominal batch would be expected to increase the mold fill time. Injection pressure could be increased to maintain the same fill time, and the temperature rise due to injection would likely increase. The amount of these changes can be estimated from experience or by trials in an injection molding machine.

A more precise prediction of the consequences of a viscosity change can be made from capillary rheometer data. For example, a 10% viscosity increase at all shear rates measured by the MPT in table 6 alters equation 5 by increasing the number 938.67 x 109 by 10%. The new flow equation is:

I = [(IP x [T.sub.n.sup.3-0564])/((1.1) x (938.67 x [10.sup.9]))][sup.-1.7027] (10) At an injection pressure of IP -- 87,500 kPa, and a nozzle temperature of [T.sub.n] = 153.3[degrees]C, the injection time I = 4.6 seconds using equation 10. Thus, for a 10% increase in viscosity, the injection time is predicted to increase from 4.0 seconds to 4.6 seconds for the hypothetical injection molding process. A longer injection time adds to the heat history of the compound.

Cure time variation

Batch variation is usually measured by changes in the rheometer cure curve. In our example, the effect of an envelope of cure curve variations surrounding a nominal cure curve represented by 170[degrees]C data in tables 4 and 5 is considered. Variations in MDR S' ML and S" @ML are similar to variations in Mooney viscosity in their effect on injection time. Variations in MH and S" @MH indicate changes in physical properties of the cured part. Cure time variations can also have a significant Effect on the injection molding process.

Figure 6 indicates a range of MDR cure curves where scorch times vary [+ or -] 10%, and cure times vary by [+ or -] 5%. In setup 4 in table 9, the cure time was selected to achieve a 50% equivalent cure in the mold core. This 50% cure was based on the nominal cure curve. As figure 6 indicates, some batches can have TC50 times shorter or longer than nominal. Since the first requirement of the cure was to reach at least 50% cure in the core, the setup must be adjusted to accommodate the longest TC50 time expected. If the 170[degrees]C MDR TC50 is increased by 5%, from 124 to 130 seconds, the cure time in setup 4, table 9, should be increased from 233 seconds to 238 seconds, based on equivalent cure calculations. This modified setup is configuration 5 in table 9.

Increasing cure time to create a safety factor for core cure increases the equivalent cure times for scorch at fill and surface cure. Based on a 170[degrees]C equivalent scorch time at mold fill of 25.1 seconds in setup 5, no batch can be accepted with an MDR TSI of less than 2 x 25.1 = 50.2 seconds vs. the nominal 62 seconds in table 5. Allowing a [+ or -] 5% range for cure times, and using setup 5 in table 9, a time-temperature profile is drawn in figure 7 showing critical cure points based on cure objectives and equivalent cure calculations.

It is interesting to note that the difference in cure times between the MDR and the ODR would lead to significantly different cure predictions if ODR data were used instead of the more accurate MDR data. In addition, the MDR has been shown to be more repeatable and more sensitive to compound variation than the ODR (ref. 13). The ODR is less capable of measuring a [+ or -] 5% change in cure times than the MDR.

Molding conditions for batch variation

Another way to use batch variation test results to optimize injection molding is to modify the mold conditions to accommodate stock variation. This could be done by calculating the optimum setup for each cure curve and altering the cure cycle as required to meet the cure objectives. This can be done if the calculations are written into a computer program. The next step in optimizing the process is to combine cure model calculations with test data and the control of the injection molding machine.

Injection molding control systems

Control systems for injection molding machines have been designed in recent years to allow input of test results from rubber tests and to change the molding conditions as required. In one control system (tel. 5) calculations similar to those discussed in this article are performed by the control system based on setup conditions and rubber test information for each batch of stock. Curemeters equipped with microprocessor data collection can send test results electronically from the curemeter to the injection molding machine.

Summary

Rubber tests measure compound characteristics that can significantly affect an injection molding process. Several examples of how test information can help in estimating the flow properties and time-temperature history of a rubber stock in an injection molding system have been shown. Models of injection molding enhance the usefulness of rubber test information, helping to set limits on acceptable variations from batch to batch, or identifying changes in molding conditions to compensate for batch variation. Test instrument capabilities and control systems available for injection molding machines have improved significantly in recent years and show promise of further automating the injection molding process.

[TABULAR DATA OMITTED]

References

1. "Injection molding of tubber," M.A. Wheelans. Halstead Press, John Wiley & Sons, NY. 1974, pp 16-18, 78.

2. "Improved cure testing," J.A. Sezna, P.J. DiMauro, H.A. Pawlowski, Rubber & Plastics News, Technical Notebook, April 18, 1988.

3. "MV-2000 Moone.y viscometer-Mooney relaxation measurements on raw and compounded rubber stocks," H. Burhin, J.A. Sezna, presented at a meeting of the Rubber Division, ACS, Oct. 89, Paper No. 74.

4. "A new dynamic mechanical tester designed for testing rubber," J.S. Dick, H.A. Pawlowski, presented at a meeting of the Rubber Division, ACS, May J92, Paper No. 70.

5. Curetrac injection molding control software, REP Corporation, P.O. Box 8146, Bartlett, IL 60103.

6. "New test results from rotorless curemeters," J.A. Sezna, H.A. Pawlowski, presented at a meeting of the Rubber Division, ACS. Oct. '89, Paper No. 78.

7. "Correlations of results from curemeters of different designs, "J.A. Sezna. Rubber World. Vol. 205, No. 4, Jan. '92.

8. "Some fundamental aspects of injection molding of elastomers," D.A.W. Izod, G.D. Skam, Injection Moulding of Elastomers, Ed. by W.S. Penn, Gordon and Breach Science Publishers, New York, 1969, pp 1-7.

9. "Thick article cure prediction," J.A. Sezna and W.C. Woods, presented at a meeting of the Rubber Division, ACS, Oct. '90, Paper No. 79.

10. "Moldflow," by Moldflow PTY LTD, Melbourne, Australia.

11. "Processability testing of injection molding rubber compounds, "J.A. Sezna and P.J. DiMauro, Rubber Chemistry & Technology, VoL 57, pp. 826-842, 1984.

12. "Injection molding of natural rubber'," M.A. Wheelans, Injection Molding of Elastomers, Ed. by W.S. Penn, Gordon and Breach Science Publishers, New York, 1969, pp 82-127.

13. "The use of rheometers for process control, "J. A. Sezna and J.S. Dick, presented at a meeting of the Rubber Division, ACS, Oct. '91, Paper No. 44.