Advanced technology in internal mixers.

Mixer development program

A high quality mix from an energy efficient process is the number one goal of any rubber compounding operation and as such, dispersion and energy efficiency are the two limiting factors to mixer sysrtem development. FroM the viewpoint of both mix quality and energy efficiency, it is widely believed that rotor design has the greatest effect on mixing efficiency, while rotor speed is the single most important factor influencing the productivity of internal mixers. As such, KSB's mixer development program has been largely concentrated in these two areas.

Kobelco has had an on-going rotor design program involving both experimental and theoretical work. The experiemental work is split into three phases. Phase 1 includes a two dimensional "sectional profile" study. Phase 2 involves a "blade dispostion" program, carried out using a three dimensional experimental mixer. Phase 3 verifies Phase 1 and 2 results through actual mix trials carried out using laboratory and production size internal mixers.

Rotor sectional development and design

Four-wing rotors were origInally introduced in the 1960s to improve the productivity of internal mixers. Unfortunately, they resulted in poorer dispersion, especially that of sulfur, than with two-wing rotors. Later, attempts were made to overcome this problem but the results were relatively unsuccessful. As a result, an empirical research and design program was carried out in collaboration with the tire industry to determine the best rotor design for intensive mixers and to optimize the mixing process.

Phase 1 - two dimensional study

This "rotor sectional profile" study was carried out using a two-dimensional model mixer (straight bladed rotors) with actual tire compounds. Time, temperature, carbon black dispersion and Mooney viscosity breakdown were investigated, enabling the "shear work" to be studied for various rotor profiles (figure 1). Two, four and six-wing rotors were tested having a variety of tip widths, tip clearances, operating angles and "sections." It was discovered that rotors which had "slimmer" sections than the traditional rotor profile produced a superior mixing effect.

Four-wing rotors were found to have much improved productivity with respect to two-wing rotors of the same profile. ReductioN in usable chamber volume associated with six-wing rotors was so large that productivity was greatly reduced.

Phase 2- three dimensional study

The rotors of this laboratory mixer are fabricated from thin oval section plates, the configuration of which may be easily changed. The mixer chamber is made from clear plexiglass, making it possible to visually study the "mix flow," both between the rotors and along the rotor axis. Effects of blade orientation, "helix angle" and "long to short wing ratio" on mixing performance can therefore be easily studied. Good "mix flow" is crucial to dispersion of vulcanizing agents for successful final mixing. It was found in the case of four-wing rotors that by increasing the long to short blade ratio form 2:1 to 4:1 and decreasing the helix angle of the short blade, the "black pressure" effect of the short blade was decreased (figure 2). Accordingly, the longitudinal pressure difference between the ends of the rotors has the effect or increasing material flow axially along the length of the chamber and also from rotor to rotor, since the pressure will tend to equalize between the two rotors by material transfer.

Phase 3 - production size mixer studies

After rigorous analysis of the results of Phase 1 andPhase 2, several types of trial rotors were produced and tested in 3.4 and 240 liter internal mixers. From these trials, also carried out using actual tire compounds, it was possible to study "production" parameters, such as Mooney reduction (table 1), carbon black and sulfur dispersion, energy consumption, fill factor and temperature rise. It was also possible to make a comparative study of the productivity of different type of four and two-wing rotors and futher, the mixing performance of the various types of four-wing rotors.

Increasing the number of wings from two to four gives a four-wing rotor a 30% higher mixing efficiency than a two-wing rotor.

And, although the net volume of the mixing chamber is reduced by 10% in the case of the four-wing rotor, the fill factor can be 6-7% higher than with the two-wing rotor. The net effect of these differences is that the productivity and mix efficiency is increased by approximately 25% in the case of a four-wing rotor (table 2).

In the case of the traditional four-wing rotors, these improvements are at the expense of a reduction in sulfur dispersion. However, in the case of KSB's four-wing H swirl rotor, the improved sectional profile design and rotor disposition provides a marked improvement in sulfur dispersion (table 3), eliminating the problem of poor sulfur dispersion inherent to the earlier models of four-wing rotors.

Four-wing H swirl rotors are fully interchangeable with current four-wing traditional rotors. Although they are also dimensionally interchangeable with two-wing rotors, four-wing rotors generally require 25% more power than two-wing rotors at equal rotor speeds. In the case where it is not possible to increase motor power, four-wing H rotors can be run 25% slower with equal productivity and still realize the benefits of improve ingredient dispersion and mix quality.

Optimum orientation - even speed mixing

In the early 1980s, Kobelco began an investigation to determine whether there was any technical reason, other than tradition derived from the two-roll mill, for running the front and rear rotors of internal mixers at different speeds (friction ratio).

It is common for the front and rear rotors of internal type mixers to have a rotational speed difference of approximately 15%. It was supposed that this speed differential provided the mixing action with an additional distribution effect. However, this frictional effect introduces a temperature gradient between material in the two rotor chambers (figure 3) which in turn has an adverse effect on within batch and batch to batch uniformity. In addition, cyclical rotor orientations cause the "pull-in" action of the rotors to be non-uniform allowing for further mixin variations. Once it was determined that it is beneficial to operate rotors at even speed, KSB then carried out an in depth "even speed" experimental program to determine the effect of varying relative orientation of the rotor wings, since orientation remains constant under even speed conditions. It was found that there is an optimum orientation for the faces of the front and rear rotors, defined at the center of the mixer.

In certain cases, zero degree orientation provided optimum performance; however for general purpose, single pass and final mixing, results were most often optimized with 90[degrees] orientation.

Even speed operation at the '0[degree] orientation" theoretically offers the following advantages:

* The temperature gradient between material in the two rotor chambers is eliminated.

* When the blades are in the vertical position, material may be quickly fed (pull-in effect of rotor blades due to large opening area), thus the ram down time may be reduced and reductions in total mix time may be achieved.

* When the rotors are in the horizontal position< they work so as to spread the material evenly throughout the chamber and distributioN is improved.

Due to the strong pull-in effect, however, peak power draw is higher which causes a more rapid temperature rise.

Even speed operation at 90[degrees] orientation offers the following advantages for general purpose mixing:

* The temperature gradient between material in the two rotor chambers is eliminated.

* Material transfer between the two rotors is more active and consistent, improving dispersion and overall mix quality.

* Peak power is reduced due to less pull-in effect, therefore, material temperature is kept lower.

In both cases, the orientation of the rotor faces is constant,

      Table 1 - Mooney reduction efficiency (breakdown
                     mixing)
                 4 Wing H swirl rotor
Rotor                                  [DELTA] ML . kg/kWH
2W - traditional                                  1.0
4W - traditional                                 1.08
4WH swirl                                        1.25
              Table 2 - productivity
                  4 Wing H swirl rotor
       Mixing      Net      Fill     Output    Reg.D
        eff.     volume    factor              power
2W       1.0       1.0       1.0       1.0      1.0
4WH      1.3       0.9      1.07      1.25      1.25
    Table 3 - dispersion of sulfur (standard
                deviation of FMAX)
                            4 Wing H swirl rotor
Rotor/compound       Soft    Semi-tough     Tough
                    rubber      rubber      rubber
2W - traditional     1.0          1.0         N/A
4W - traditional     10.4         3.1         1.0
4WH swirl            0.75         0.5         0.9

hence the position of the power peak during initial mixing is constant and accordingly constant from batch to batch. Material uniformity within batch and from batch to batch is therefore improve. Likewise, the shape of the power curve is constant for all batches, providing a further increase in material uniformity, (a phenomenon which is particularly useful in cases where mix control is by "power curve program," to be discussed in greater detail later).

Rotor design re-evaluation

At this stage, KSB re-evaluated its four-wing rotor design data, hoping to be able to further improve material flow and dispersive efficiency. This was in fact found to be possible by increasing the long wing helix angle. The larger long wing helix angle considerably increased the axial flow of material iN the mixer chamber (figure 4 and table 4). These rotors (four-wing H Mark II) may be used to replace all types of two-wing and four-wing rotors and provide increases in both productivity and product quality, regardless of rotor friction ratio or orientation. Currently, KSB has over 100 sets of our-wing H and four-wing H Mark II rotors in operation around the world.

High speed mixing

As mentioned previously, rotor speed is the single most important factor influencing the performance of internal mixers. Productivity increases proprotionally with rotor speed, but is limited by the associated material temperature rise. Increasing rotor speed causes a dramatic temperature rise,

      Table 4 - performance of 4WH Mark II
         16lMixer, even speed, M.B. to F.M.
   Parameter       Mark II      Original 4 wing H
  Productivity      1.08               1.00
C.B. dispersion     92%                85%

[TABULAR DATA OMITTED]

which may only be reduced by increasing rotor tip clearance. This in turn reduces the mixing effect by reducing the shear stress at the rotor-chamber interface. It was theorized that the reduction in "high shear stress" mixing at the rotor tip is more than compensated for by an increase in mixing at the rotor face due to high speed rotation and the great increase in pass volume. To test the theory, a "high speed - optimum orientation" versus "traditional" mixing trial was carried out. It was concluded that a high speed optimum orientation system provided superior results in every aspect when compared with a traditional system (table 5).

Computer program control

Power pattern following is a new mixing theory which has been tested and proven using 16 and 76 liter machines. During mixing of nearly all rubber/carbon black systems, the torque/time profile follows a generalized pattern. The characteristic peaks and valleys are equivalent to particular "states of mix" in the mixing process. For the case of this simple carbon black masterbatch, the mix signature may be described as follows: at first the cold rubber is forced into strong contact with the rotors by the actionof the ram. As the rubber is sheared, rubber to metal friction causes a rise in compound temperature. This in turn, results in a drop of compound viscosity and likewise the torque decreases as material breaks down. Carbon black is then gradually dispersed throughout the rubber, reinforcing the rubber and once again the mix viscosity begins to rise. Continuing temperature rise counteracts this effect and the torque begins to decrease once again, equivalent to "simple mixing." Steady state is then reached concluding the mix and the batch is discharged. For some of the more complex "tire compound" mix stages, the mix signature is less easily explained. However, a characteristic of the mix process can be defined.

MP controller

Time, temperature, instantaneous and integrated motor power reflect the state of the rubber mix. By monitoring any of the above, the mix condition can be followed. Mix steps can be set according to a particular time, temperature and/or integrated motor power. In the case of the MP (multiple program) controller, a combination of "and/or" conditions are set with these parameters resulting in high quality repeatable and consistent batch mixing.

PPP controller

In the case the PPP (power pattern program) controller, the optimuM mix signature for each step of the mixing process is first determined by recording and averaging repeated trials. From this average, the "optimum" power pattern can be determined, and loaded onto floppy disc via the display and keyboard. For subsequent batches the optimum power pattern is entered from floppy disc and the mixing parameters iteratively adjusted to follow this idealized motor power mix signature. The simple theory being that if the process of energy input (volume-timing) to the batch is the same as that for the optimum mix of an equivalent batch, the mixing conditions (and therefore material properties) will be the same. This is achieved in practice by varying rotor speed, ram pressure and machine temperature, so that the current mix pattern follows that of the optimized pattern for that particular recipe. Power input is very sensitive to rotor speed, while the influence of ram pressure and temperature are relatively small. Therefore, variation of only the rotor speed, using a DC motor, may be sufficient for production purposes in most cases.

Mini oil dust stop

In addition to the great care taken with rotor design and mixer control systems to achieve optimum energy efficiency productivity and mix quality, extensive design work has been carried out with the specific aim of further reducing down time and operating costs associated with internal mixers. One of these developments is mini oil dust stops.

Years ago, KSB started producing water cooled dust stops in order to eliminate lubrication oil contamination of the mix and reduce lubrication and process oil consumption. By replacing conventional dust stops with those of a water cooled design, rubber producers were able to virtually eliminate contamination and reduce process adn lubrication oil usage by up to 50%.

In addition to contamination and high oil consumption, traditional dust stops required frequent replacement. Mini oil dust stops often last as long as the mixer chamber, requiring changeout only during chamber rebuilding. This is possible because the lubricating oil's high viscosity (and lubricating capability) is maintained by water cooling the dust stop and because super hard bearing metals are used on the rotating frictional surfaces. The main features of the mini-oil dust stops can be summarized as follows:

* Low frictional heat generation.

* Simple construction; interchangeable with conventional YH types.

* Super hard bearing metals.

* Active water cooling system.

KSB predicts an 80-90% savings in lubrication and process oil with respect to traditional dust stops, three or more years of service life and mixes which are free of lubrication oil contamination. These results are being verified by actual field installations.