Page 29 - Chemical Process Equipment - Selection and Design

6 INTRODUCTION
closer appraisals of energy utilizations by complete processes are With regard to specific types of equipment, the safety factor
being made, from the standpoints of both the conservation laws and practices of some 250 engineers were ascertained by a questionnaire
the second law of thermodynamics. In the latter cases attention is and summarized in Table 1.4; additional figures are given by Peters
focused on changes in entropy and in the related availability and Timmerhaus (References, Section 1.1, Part B, pp. 35-37).
function, AB = AH - TOAS, with emphasis on work as the best Relatively inexpensive equipment that can conceivably serve as a
possible transformation of energy. In this way a second law analysis bottleneck, such as pumps, always is liberally sized; perhaps as
of a process will reveal where the greatest generation of entropy much as 50% extra for a reflux pump. In an expanding industry it is
occurs and where possibly the most improvement can be made by a matter of policy to deliberately oversize certain major equipment
appropriate changes of process or equipment. Such an analysis of a that cannot be supplemented readily or modified suitably for
cryogenic process for air separation was made by Benedict and increased capacity; these are safety factors to account for future
Gyftopolous [in Gaggioli (Ed.), Thermodynamic Second Law trends.
Analysis, ACS Symposium Series No. 122, American Chemical Safety factors should not be used to mask inadequate or
Society, Washington, DC, 19801; they found a pressure drop at careless design work. The design should be the best that can be
which the combination of exchanger and compressor was most made in the time economically justifiable, and the safety factors
economical. should be estimated from a careful consideration of all factors
A low second law efficiency is not always realistically improv- entering into the design and the possible future deviations from the
able. Thus Weber and Meissner (Thermodynamics for Chemical design conditions.
Engineers, John Wiley, New York, 1957) found a 6% efficiency for Sometimes it is possible to evaluate the range of validity of
the separation of ethanol and water by distillation which is not measurements and correlations of physical properties, phase
substantially improvable by redesign of the distillation process. equilibrium behavior, mass and heat transfer efficiencies and similar
Perhaps this suggests that more efficient methods than distillation factors, as well as the fluctuations in temperature, pressure, flow,
should be sought for the separation of volatile mixtures, but none etc., associated with practical control systems. Then the effects of
has been found at competitive cost. such data on the uncertainty of sizing equipment can be estimated.
Details of the thermodynamic basis of availability analysis are For example, the mass of a distillation column that is related
dealt with by Moran (Availability Analysis, Prentice-Hall, directly to its cost depends on at least these factors:
Englewood Cliffs, NJ, 1982). He applies the method to a cooling
tower, heat pump, a cryogenic process, coal gasification, and par- 1. The vapor-liquid equilibrium data.
ticularly to the efficient use of fuels. 2. The method of calculating the reflux and number of trays.
An interesting conclusion reached by Linnhoff [in Seider and 3. The tray efficiency.
Mah (Eds.), Foundations of Computer-Aided Process Design, 4. Allowable vapor rate and consequently the tower diameter at a
AIChE, New York, 19811 is that “chemical processes which are given tray spacing and estimated operating surface tension and
properly designed for energy versus capital cost tend to operate at fluid densities.
approximately 60% efficiency.” A major aspect of his analysis is 5. Corrosion allowances.
recognition of practical constraints and inevitable losses. These may
include material of construction limits, plant layout, operability, the Also such factors as allowable tensile strengths, weld efficiencies,
need for simplicity such as limits on the number of compressor and possible inaccuracies of formulas used to calculate shell and
stages or refrigeration levels, and above all the recognition that, for head thicknesses may be pertinent.
low grade heat, heat recovery is preferable to work recovery, the When a quantity is a function of several variables,
latter being justifiable only in huge installations. Unfortunately, the
edge is taken off the dramatic 60% conclusion by Linnhoff’s Y = Y (Xl, x2, . . .I,
admission that efficiency cannot be easily defined for some
complexes of interrelated equipment. For example, is it economical its differential is
to recover 60% of the propane or 60% of the ethane from a natural
gas?

1.8. SAFETY FACTORS
Some relations of importance in chemical engineering have the form
In all of the factors that influence the performance of equipment
and plant there are elements of uncertainty and the possibility of
error, including inaccuracy of physical data, basic correlations of
behavior such as pipe friction or tray efficiency or gas-liquid whose differential is rearrangable to
distribution, necessary approximations of design methods and
calculations, not entirely known behavior of materials of con-
struction, uncertainty of future market demands, and changes in
operating performance with time. The solvency of the project, the
safety of the operators and the public, and the reputation and that is, the relative uncertainty or error in the function is related
career of the design engineer are at stake. Accordingly, the linearly to the fractional uncertainties of the independent variables.
experienced engineer will apply safety factors throughout the design For example, take the case of a steam-heated thermosyphon
of a plant. Just how much of a factor should be applied in a reboiler on a distillation column for which the heat transfer
particular case cannot be stated in general terms because cir- equation is
cumstances vary widely. The inadequate performance of a
particular piece of equipment may be compensated for by the q = UAAT.
superior performance of associated equipment, as insufficient trays
in a fractionator may be compensated for by increases in reflux and The problem is to find how the heat transfer rate can vary when the
reboiling, if that equipment can take the extra load. other quantities change. U is an experimental value that is known