Full-Scale Example

A process is described [224] in which an exothermic reaction takes place in a semi-batch reactor at elevated temperatures and under pressure. The solid and liquid raw materials are both toxic and flammable. Spontaneous ignition is possible when the reaction mass is exposed to air. Therefore, the system must be totally enclosed and confined in order to contain safely any emissions arising from the loss of reactor control, and to prevent secondary combustion reactions upon discharge of the materials to the atmosphere. Further, procedures and equipment are necessary for the safe collection and disposal of solid, liquid, and gaseous emission products. 

The process is run in a semi-batch mode, and multiple reactors are used. There are several possible causes for a loss of control such as insufficient heat removal and loss of agitation. Overpressurization leading to the bursting of rupture discs takes place several times per year, indicating both the clear need for containment but also a need to consider design and control improvements. The reference describes the autoclave rupture disc assembly, procedures for replacement of the discs, the cleaning of the containment vessels, and the routine maintenance procedures for the containment vessels. 

---

Accuracies of the flow meters discussed herein are specified as either a percentage of the full-scale flow or as a percentage of the actual flow rate. It may be convenient in some appHcations to compare the potential inaccuracies in actual volumetric flow rates. For example, in reading two Hters per minute (LPM) on a flow meter rated for five LPM, the maximum error for a 1% of full-scale accuracy specification would be 0.01 x 5 = 0.05 LPM. If another flow meter of similar range, but having 1% of actual flow rate specification, were used, the maximum error would be 0.01 x 2 = 0.02 LPM. To minimize errors, meters having full-scale accuracy specifications are normally not used at the lower end of their range. Whenever possible, performance parameters should be assessed for the expected installation conditions, not the reference conditions that are the basis of nominal product performance specifications. 

New product introductions are generally heavily supported by the technical service function. Many customers using chemical feedstocks to produce multicomponent products for the consumer market require extensive on-line evaluations of new raw materials prior to their acceptance for use. An example of this would be the use of a new engineering polymer for the fabrication of exterior automobile stmctural panels. Full-scale fabrication of the part foUowed by a detailed study of parameters, such as impact strength, colorant behavior, paint receptivity, exterior photodurabiHty, mar resistance, and others, would be required prior to making a raw materials change of this nature. 

Herbst et al. [International J. Mineral Proce.ssing, 22, 273-296 (1988)] describe the software modules in an optimum controller for a grinding circuit. The process model can be an empirical model as some authors have used. A phenomenological model can give more accurate predictions, and can be extrapolated, for example from pilot-to full-scale apphcation, if scale-up rules are known. Normally the model is a variant of the popiilation balance equations given in the previous section. 

For non-New tonian fluids, viscosity data are very important. Every impeller has an average fluid shear rate related to speed. For example, foi a flat blade turbine impeller, the average impeller zone fluid shear rate is 11 times the operating speed. The most exact method to obtain the viscosity is by using a standard mixing tank and impeller as a viscosimeter. By measuring the pow er response on a small scale mixer, the viscosity at shear rates similar to that in the full scale unit is obtained. 

If the yields are accepted without full-scale testing, questions can and should arise as to how much contingency exists in the yields (since after all, they were obtained by correlations of similar coals, or perhaps by small-scale tests for your coal, for example). For at least one study, initially presented yields of this sort were found to represent a conservative case and upon request, yields were revealed that were closer to licensor expectations with no contingency. Some design contingency must be provided, but to do this intelligently, any yield contingency must be identified. 

Chemical reaction hazards must be considered in assessing whether a process can be operated safely on the manufacturing scale. Furthermore, the effect of scale-up is particularly important. A reaction, which is innocuous on the laboratory or pilot plant scale, can be disastrous in a full-scale manufacturing plant. For example, the heat release from a highly exothermic process, such as the reduction of an aromatic nitro compound, can be easily controlled in laboratory glassware. Flowever,... 

Pilot plant experiments represent an essential step in the investigation of a process toward formulating specifications for a commercial plant. A pilot plant uses the microkinetic data derived by laboratory tests and provides information about the macro kinetics of a process. Examples include the interaction of large conglomerates of molecules, macroscopic fluid elements, the effects of the macroscopic streams of materials and energy on the process, as well as the true residence time in the full-scale plant. 

The conditions for model experiments can be explained in the following way. The governing equations are made nondimensional in the full scale and in the reduced scale used in the model experiments. For example, the velocity in the room is divided by the diffuser velocity in the room, and the velocity in the model is divided by the supply velocity in the model in order to normalize all velocities. The two sets of equations are identical and they describe the same solution provided that requirements f, 2, and 3 mentioned at the beginning of this section have been met. 

The assumption of a self-similar flow (Reynolds number-independent flow) simplifies full-scale experiments and is also a useful tool in the formulation of simple measuring procedures. This section will show two examples of self-similar flow where the Archimedes number is the only important parameter. 

Like any other effective test, your company s PSM pilot should be neither too difficult nor too easy. A "successful" pilot test is one that (1) thoroughly exercises the PSM system you have developed so that you can identify its strengths and weaknesses, and (2) is sufficiently well-designed and -executed that it can act as a model for full-scale implementation. For example, testing a management of change system in a way that ranks changes would be appropriate. In short, consider the pilot test as a microcosm—and treat it with the same attention you would devote to a companywide installation. 

Dimensional analysis techniques are especially useful for manufacturers that make families of products that vary in size and performance specifications. Often it is not economic to make full-scale prototypes of a final product (e.g., dams, bridges, communication antennas, etc.). Thus, the solution to many of these design problems is to create small scale physical models that can be tested in similar operational environments. The dimensional analysis terms combined with results of physical modeling form the basis for interpreting data and development of full-scale prototype devices or systems. Use of dimensional analysis in fluid mechanics is given in the following example. 

After the dominant independent variables have been brought under control, many small and poorly characterized ones remain that limit further improvement in modeling the response surface when going to full-scale production, control of experimental conditions drops behind what is possible in laboratory-scale work (e.g., temperature gradients across vessels), but this is where, in the long term, the real data is acquired. Chemistry abounds with examples of complex interactions among the many compounds found in a simple synthesis step,... 

As illustrated by the examples above, the possibility of removing the generated heat from the reaction zone decreases with an increase in reactor size. As proven above, it can happen that the temperature of the reaction mixture in a full-scale reactor becomes higher than in the laboratory flask reactor. If multiple chemical reactions of distinctly different temperature sensitivities take place, differences in yields and selectivities between small and large reactors will be observed. This has a large influence on safety also. The laboratory reactor might still show satisfactory performance, while the industrial reactor might even explode. 

The variance diagram obtained for the example discussed before is quite simple. Clusters of pure variables are found at 30 degrees (var = 0.5853) and at 300 degrees (var = 0.4868) (see Fig. 34.36). The distance from the centre of the diagram to each point is proportional to the variance value. Neighbouring points are connected by solid lines. All values were scaled in such a way that the highest variance is full scale. As can be seen from Fig. 34.36, two clusters of pure variables are found. The... 

The following two project descriptions incorporate examples of completed, full-scale applications of pump-and-treat technology to MTBE-contaminated sites. 

Another instability mode of interest is due to the flow regime itself. For example, it is well known that the slug flow regime is periodic and that its occurrence in an adiabatic riser can drive a dynamic oscillation (Wallis and Hearsley, 1961). In a BWR system, one must guard against this type of instability in components such as steam separation standpipes. The design of the BWR steam separator complex is normally given a full-scale, out-of-core proof test to demonstrate that both static and dynamic performance are stable. 

---