Increasing output and quality by updating the mixing room equipment.

A highly competitive marketplace is forcing producers to operate rubber mixing rooms at peak capacity. Machine manufacturers are responding with equipment designs capable of operating 24 hours per day, seven days per week for several years with a minimum of maintenance. An appropriate material handling system, a modern process control, variable speed drive and/or temperature control unit are important. Equipment manufactures have realized that and are now able to supply turn-key mixing room installations (ref. 1). But the internal mixer is still the heart of the mixing room. In this article we concentrate on the internal mixer itself and hot feed extruders. For existing machines, major improvements in terms of quality and output can be made by modernizing components, that need regular replacement anyway, against new, high technology parts.

Mixer body

Rotor design

During the past few years, differing approaches have been applied to tangential rotor designs. The common goal is to increase the output of the mixer while improving, rather than compromising, the mix quality. Starting with the standard, four-wing rotor, there are two approaches.

The first approach increases the ratio between the lengths of the long and short wings. As die long wing is lengthened, the wing generates more pressure. The rotor becomes more aggressive, resulting in higher torques and increased energy input. The second approach deals with the arrangement of the long and short wings. W&P's ZZ-2 rotor utilizes this design (ref. 2).

Regardless of the design, the rotors can be run at the same speed (even speed) or at a friction ratio. Even speed can optimize some of the mixing characteristics like feeding and discharge behavior, others might be reduced. Overall, the mixer becomes more specialized.

The ZZ-2 rotor design (ref. 3) is now working in both the tire and the technical rubber goods area worldwide. The major difference, compared to the standard four-wing rotor design, is that the long wings originate from opposite sides of the rotor (figure 1). A short wing follows the long wing, originating from the same side of the rotor. The short wing cleans the rotor end plate. The angle of the long and short wing, a patented feature, is between 35 and 45 degrees. This arrangement balances the forces on the rotors and establishes a uniform shear rate and very low thrust on the rotor. Even and homogenous working of the compound results.

The long wings do not extend to the end of the rotor. Thus, an additional flow path is created between die wing and the rotor end plate. The additional flow path further optimizes the mixing behavior since material cannot lodge in a corner formed by the rotor body and end of the wing. Areas of extremely high pressure do not exist at the dust stops. Instead, the material undergoes extra shear in the flow path, and more importantly, additional distributive mixing is achieved. The starting edge of the long wings functions as a flow divider. It divides the material flow into one part that undergoes shear between the rotor tip and the chamber side. The other part is kneaded and folded. Tbis accounts for the distributive mixing and results in increased homogeneity. Also, the material is in contact with the rotor end plates over a longer period of time resulting in additional cooling.

Table 1 compares the mixing behavior of die ZZ-2 rotor and a standard four-wing rotor in a tire application. The fill factor could be increased because of the fact that the long wings.do not exercise as much pressure. Therefore, more material can be mixed in the chamber without preventing the weight from reaching the lower position. The throughput is also increased because of the higher fill.factor and the better mixing performance. The higher throughput does not require higher energy input because the mixing is done more precisely and evenly. The full 4-wing geometry, for example, is bringing in more energy but this results mainly in high material temperatures. Since the ZZ-2 rotor is able to hold the temperature lower, high shear stresses can be induced over a longer period of time. Also, even distribution of curing systems is much faster. The thermal characteristics are optimized significantly, as a result of reduced heat generation and a very effective cooling system. The mixing efficiency was increased, which means that because of the excellent distributive mix it is possible to distibute the curing system much faster.

  Table 1 - comparison of a ZZ2 and a standard
4-wing-rotor in a tire application

                                  ZZ-2      4-wing rotor
Application                       Tire          Tire
Fill ratio/volume                 +++            ++
Throughput                        +++            ++
Energy consumption                 +             ++
Thermal behavior                  ++            0 (-)
Mixing efficiency                 ++             +
Feeding and discharge behavior    ++            +++
                                Two-piece    One-piece
                                  rotor        rotor

+++ Excellent/++ Good/+ Satisfying/- Not sufficient

Wear protection

Productivity of a mixing line is diminished by time lost during major maintenance and down time required to exchange certain wear parts. Rotors, chambersides and rotor end plates are the parts that require the largest time losses during exchange. For example, rotor changes typically require five days. However, certain mixers in the marketplace require only one day as a result of maintenance friendly, modular design (ref 4). The GK-type mixers feature a three-module design, consisting of the base, the body and the hopper. Each module can be turned against another by 180'. The body features horizontally-split bearing housings and end frames, resulting in easy maintenance.

Long service life of all parts is essential to high productivity. To improve service, these parts are generally protected with some form of wear resistant coating. Current methods include chrome plating and hard surface welding.

Chrome plating is usually 50 to 200 microns (0.002 to .008 inches) thick. Chrome plating is easily damaged by hard pieces of foreign objects passing through the mixer. There is no possibility of in-process repair (without dismantling). Commonly used hard surfacing materials are iron-, cobalt- or nickel-based materials. The hardness is achieved by means of carbides formed during the welding process. Different hardsteel welding techniques are used in 3 to 5 mm (.118 to .197 inches) thicknesses and with hardnesses from 30 to 60 Rockwell C.

For chambersides, Stellite #1 is used in many applications. This cobalt-based material has the advantage of hardness above 55 HRC and relatively high tensile strength (> 600N/m[m.sup.2]). It possesses highest abrasion resistance; however, the welding process is difficult leading to frequent quality problems. The material cannot be applied crack-free. Those cracks usually penetrate into the base material. This allows corrosion of the base material beneath the coating, resulting in deterioration of die hard steel surface layers over time.

For a long life, the mixing chamber should be of strong design, tough and hard in order to resist abrasive materials. Most important for long life is a crack-free, high-hardness, corrosion-resistant surface. To achieve long life, the surface coating must be at least 50 HRC, with a tensile strength exceeding 60ON/m[m.sup.2], and possess a matrix (no carbides) with a chromium content exceeding 12.7%.

Such a material has been developed recently (ref. 5). Numerous trials have been made, including a field test which has been running for more than two years, achieving excellent results. The new hard-steel coating material, called WP 53 M, results in a crack-free surface with a hardness of higher than 50 HRC, a tensile strength of approximately 700N/m[m.sup.2] and a chromium content in the matrix of more than 13%. The abrasion resistance is excellent and service life is increased greatly compared to currently available materials. The time between changes should increase, therefore resulting in less time and materials required for maintenance, and increasing the overall output and productivity of the mixer.

A good portion of the energy put into the compound during mixing is transferred through the mixing chamber walls and the rotors as heat. For a chamberside, the thermal behavior depends on the peripherally drilled cooling channels. To avoid cracking into the cooling channels, the chamberside used to be designed with much heavier wall thicknesses than required (from strength considerations). Here we see an important advantage of the crack-free hard coating. The coolinq channels can be positioned closer to the crack-free surface. For each reduction of one millimeter between the channels and the surface, an improvement of 5% to 7% is made in the cooling effect.

Improved cooling leads to lower batch temperatures, shorter mixing cycles, and therefore higher outputs. It also allows a lower compound temperature during the mixing process (figure 2). At lower compound temperatures, the mixing viscosity increases and the specific energy input increases (without increasing rotor speed). The higher rotor torque with higher compound viscosity results from increased shear and strain stresses in the mixing chamber. Shear and strain are most important for dispersion of particles like carbon black. Therefore, the higher level of stress reduces the time needed to achieve a certain dispersion level. That means that the mixing time is reduced.

Thermocouples

To take advantage of improved cooling, accurate measurement of the batch temperature is necessary to fine-tune the mixing process. Thermocouples protruding into the mixing chamber must be extremely tough and wear resistant in order to survive the mixing actions. Such a device is contradictory to proper thermocouple design - the ideal thermocouple should contain as little mass as possible. As the probe mass increases, the response time decreases; this means that any temperature change is registered with a greater time delay. As the mass of the probe increases, its heating time increases, further increasing the time delay. All this leads to the optimal design requirements for a thermocouple: Very tough material; excellent wear protection; and small thermocouple mass.

There is yet another problem: The thermocouple is typically mounted into a cooled housing, such as the doortop. The probe is heated on one side by the compound in the mixer and cooled on the other side by the housing. As the cooling effect of the housing grows compared to the heat transfer to die probe by die material in the mixer, the temperature indicated is lower than the actual compound temperature. The answer is to restrict the cooling effect of the housing as much as possible in order to get a reading as close to the real temperature as possible. A fourth requirement is established:

*Restricted heat transfer. A thermocouple design which fulfills these requirements is shown in figure 3. The design features are:

*Extremely tough material;

*Special, thin hardcoating for wear resistance;

*Thin wall design;

*Thermal isolation to the shaft and housing.

The reaction time of this new thermocouple is excellent compared to the standard probe. The graph in figure 4 compares the two sensors in an oil bath of 150 [degrees] C. The t-90 time, which is the time required to show 90% of a temperature change in a measured medium has decreased by more than 50%.

More accurate temperature measurement can result in increased output and quality because the mixing steps can be controlled more precisely.

Hopper assembly

When the correlation between dispersion, mass temperature and ram pressure (figure 5) is observed (ref. 2), the advantage of reproducible ram pressure becomes apparent. Using a pneumatic ram, pressure fluctuations of the air supply causes ram pressure variations from stroke to stroke within one batch, or from batch to batch, causing variations in mix quality.

In addition, there are problems in the maintenance and operation of a pneumatic ram. The hopper cylinder and the built-in seals are not easily maintained. Changing these seals is a major undertaking. When the ram is down, the piston rod is not readily accessible; therefore when the ram sticks, it is difficult for the operator to free it.

These problems and the high cost of pneumatic energy led to development of hydraulically-operated rams. The force on the ram is applied through a horizontal crossbar, moved by hydraulic cylinders located outside the feed hopper. The cylinder location prevents hydraulic oil from entering the hopper and/or mixing chamber if leakage occurs. When cylinder maintenance is needed, the external location is readily accessible. One or more of the cylinders are quickly exchanged, in minimum time. The removed cylinder can be repaired while the mixer continues operation. This design allows easy access to the main rod and installation of additional material feeding ports in the rear as well as through die sidewalls often required in the case of automated material handling systems. Hydraulic pressure is produced by a dedicated hydraulic power unit, located near the mixer, assuring constant pressure, which can be easily controlled during the entire mixing cycle.

The ram design combined with a well-designed hydraulic control system ensures that the action and reaction of the ram is similar to that of the pneumatic ram. The hydraulic ram floats during the mixing cycle.

The energy consumption of hydraulic and pneumatic rams are compared in tables 2 and 3. Table 2 illustrates a sample calculation of energy savings and table 3 contains annual savings for a variety of assumptions. Depending on the number of rain movements and the energy cost, the savingsare-computed for any application. In many cases, there are also shorter mixing times because of the fast travel of the hydraulic ram compared to the pneumatic ram and because the ram pressure remains constant during the whole mixing cycle.

[TABULAR DATA OMITTED]

  Table 3 - annual cost savings for hydraulic ram
at various cost and operating levels (in $)

Cost of 116 psi pneumatic air per 1,000 cubic meters

Energy
cost*       $18      $21      $25      $27     $30
($/KWH)
  .05     20,457   24,897   30,817   33,777   38,217
  .06     19,220   23,660   29,580   32,540   36,980
  .07     17,984   22,424   28,344   31,304   35,744
  .08     16,747   21,187   27,107   30,067   34,507
  .09     15,511   19,951   25,871   28,831   33,271
  .10     14,274   18,714   24,634   27,594   32,034

Cost of 85 psi pneumatic air per 1,000 cubic meters

  .05     15,507   19,122   23,942   26,352   29,967
  .06     14,270   17,885   22,705   25,115   28,730
  .07     13,034   16,649   21,469   23,879   27,494
  .08     11,797   15,412   20,232   22,642   26,257
  .09     10,561   14,176   18,996   21,406   23,784
  .10      9,324   12,939   17,759   20,169   23,784

Assumptions: 620 liter doubles strokes, 3 strokes per batch, 240
batches per shift, 3 x 250 shifts per year
Note (*): Cost of energy to operate the hydraulic ram

Updating hot feed extruders with rotary pushers

A large number of hot feed extruders, of various sizes, operate in mixing rooms worldwide. The overwhelming majority of these extruders are equipped with a single or double pusher. These pushers have been a maintenance problem ever since their inception. The main disadvantages are:

*Complex and expensive design;

*Excessive number of high wear parts;

*Material leaks into the pusher housing;

*Possible contamination of subsequent batches;

*Lubricants easily enter the extruder;

*Inspection and cleaning is difficult.

Yet, the advantages of the pusher concept are strongly desirable:

*Material is forced into the extruder;

*Material is pressurized in the hopper cylinder.

A new design, maintaining the advantages while avoiding most of the problems of a linear pusher is the rotary pusher (ref.5). The pusher no longer performs a linear movement in a housing. Instead, the movement is a rotation. A shaft guided by radial bearings holds the pusher body. Hydraulic cylinders, mounted externally on the chute and the pusher, apply the force. Sealing lips prevent material from escaping the pressurized area. When material leaks past these seals, it immediately falls outside the pusher and does not contaminate subsequent batches. As a result of an optimized design and specific materials of construction, lubrication of the parts contacting the compound is unnecessary. Perhaps the most important detail is a service position that allows easy access to the contact areas by simply extending the cylinder stroke (figure 6). The number and cost of wear parts are reduced significantly.

This design is easily applied to existing extruders. In figure 6, a single pusher for a 15" x 18" extruder is shown. The hopper cylinder stays unchanged while the linear pusher is replaced by a rotary pusher.

In-line quality control

Of course, quality has to be produced. But quality documentation becomes more and more important, also. Important properties are the viscosity of the compound and the distribution of different ingredients throughout the compound. Today this is analyzed off-line in a laboratory using small material samples. The results of these tests can be delayed up to hours. Very often only one sample per batch is reviewed. Therefore, in-line quality control systems are needed that give instantaneous results and provide information about the whole batch rather than a small sample. Two systems that are tailored for this are currently under development.

The Relma system was developed to characterize the homogeneity of a rubber slab (refs. 6-8). The method is based on remote laser-microanalysis (Relma). The light of a pulsed UV-laser is focused on the rubber surface producing a hot bright plasma. The optical plasma emission is detected spectroscopically with high time resolution. Compound element concentrations are measured in a fraction of a second. RELMA can be applied to an open mill or a batch-off with a scanning unit. That results in in-line quality control covering large areas of the rubber slab surface.

The second system, named Rheomill, is a sensor system which allows the measurement of the flowing behavior of rubber mixtures when they are still being worked on the roll mill (ref. 8). During the rolling process, a wedge is switched on the rubber slab. The wedge gap measures through sensor units the power/pressure ratio of the wedge gap flow, as well as the temperature and the swelling behavior of the mixture. The temperature is measured with an infrared sensor and the thickness of the rubber sheet by a laser. Tests were made in which the measured values of the tangential force correlate very well with the Mooney viscosity.

References

[1] A. Limper, W Hader: "The mixing room - system aspects" presented at a meeting of the Rubber Division, American Chemical Society, Detroit, Michigan, October 8-11, 1991. [2.] W.M. Wiedmann and H.M. Schmid, "Optimierung tangierender und ineinandergreifender rotorgeometrien von gummiknetern. Kautschik + Gummi, Kunststoffe 34 (1981)6. [3.] W.M Wiedmann and H.M. Schmid, "Innenmischer zum kneten von plastichen massen, insbesondere von Gummi," Patent No. DE 28 36 940 C2. [4.] J. Pohl and A. Limper, "Modular mixers cut down-time." European Rubber Journal, 12 1992). [5.] J. Pohl and A. Limper, "New angles in rubber mixing technology," Tire Technology International, 93, 1993. [6.] F. Leis, W. Sdorra, J.B. Ko and K. Niemax, "Basic investigations for laser microanalysis," Mikrochim. Acta (1989) no. 2, p. 185. [7.] C.-J. Lorenzen, C. Carlhoff, U. Hahn and M. Jogwich "Applications of laser-induced emission spectral analysis for industrial process and quality control," J. of Analyt. Atomic Spectroscopy (1992), pp. 1029-1035. [8.] W. Hader and C.-J. Lorenzen, "Relma: On-line quality control in the rubber mixing room using laser microanalysis," Technische Mitteilungen Krupp 2/1992, pp. 97-102. [9.] A. Limper and W. Hader, "Verfahren und vorrichtung zum messen der viskoelastischen eigenschaften von polymerem material," Patent No. DE 4032375.