Page 222 - Applied Process Design For Chemical And Petrochemical Plants Volume II

Distillation 21 1

Overall Tray Pressure Drop: municate with the final designer/manufacturer and inter-
pret the significance of the final results, and (2) be capa-
h, = hh + hl = 1.77 + 2.33 = 4.3 in. liquid ble of preparing approximate designs for preliminary
information and to develop calculated results to compare
Froth Height: with the final designs of others. The designs developed by
the methods/procedures presented here are considered
h, = 1.15 in. liquid, (see calculation above) reliable for these purposes, and even as final designs, pro-
vided there are actual process data and experience to
Using F-Factor, determine p and I$ from Figure 8-149. compare with.
Capps [ 1881 compares valve and sieve tray performance
at F,, = 1.04, then: p = 0.61 and I$ = 0.22
as related to capacity and flooding. Also see sieve tray sec-
tion presented earlier in this chapter.
hl = 2.53 in. liquid. (see calculation above)
Capps El881 examines sieve and valve tray capacity per-
Calculate froth height, hr: formance and Figure 8-151 [188] k offered for preliminary
column sizing or for determining whether a debottleneck-
hf = hI/@ = 2..53/0.22 = 11.5 in. ing study is justified. The correlation for flooding, tray rat-
ing, and design of a tray are all based on the capacity factor,
Klein E2011 refers to Thorngren [206,207] but suggests CT, equation (Souders and Brown [68] by Capps [ 1881).
that this proposed valve tray flooding is reasonably At total reflux (L/V) = 1) Capps found several points in
involved, although considered useful. the FRI data that corresponded with this [241] definition
of ultimate capacity, i.e., the liquid and vapor load at
Proprietary Designs which any increase in either liquid or vapor would induce
flooding by at least one of the following mechanisms:
The design engineer cannot adequately design a valve
tray that includes the operating valves and expect to have 1. Figure 8-151 shows capacity factor (Souders-Brown
reliable performance. The proper approach is to assemble velocity) versus system factor (pressure, in this case
all of the required system/column operating performance for hydrocarbons) with L/Y as a parameter. In Figure
requirements and then turn the problem over to a manu- 8-151 the predicted ultimate capacity for a hydrocar-
facturer who has tested its own valve designs and is capa- bon is obtained by reading the capacity factor at
ble of predicting reliable performance. The manufacturer incipient flood for a given pressure at a given reflux
can then provide a hydraulic design for the tray, as well as ratio, L/V. This Souders-Brown velocity then can be
the expected performance of the entire column/tray sys- used to predict the maximum load achievable for a
tem. The major manufacturer/designs are Nutter Engi- given column diameter, or, the minimum tower area
neering, HdrSCO Corporation, [ 2041 ; Koch Engineering required for a given load [188, 2411.
Co., Inc. [203]; Glitsch, Inc. [202], and Norton Chemical 2. Flood factor is the usual design safety factor (e.g.,
Process Products Corporation [233]. 80% of flood, Fflood = 0.80
There are other manufacturers and engineering com-
panies that are capable through good computer programs % = mwp (24/S)0-’/[A~ (hap (Pliq - ~vap))~.’l (8-320)
of designing competitive distillation designs, and it is not (8- 321)
the intent of the above listing to omit any reliable organi- Vload = mbap/ [ (P~xP) (Pas - PbXp) 1 ’”, ft3/sec
zation, but to simply list the generally considered major Or, CT = vload/(AT) (24/S)0.5, ft/SeC
suppliers in the U.S.
One important point to consider is whether or not the Or, % = [6*238 mvap,fl?oa/(Fflood) (%,flood) (hap
- ~vap) s)0.310.3
organization has obtained commercial sized data on (8-322)
equipment designed and fabricated to their designs, and Capacity factor, CT = Lrlwad/.4~
how the two results compare. The respective design pro- \’load = mMp/ hap -
cedure as set forth in each company’s design manual will P,,,) f$/sec (8 - 323)
not be outlined in this text, as there is too much detail Using the Souders-Brown factor:
necessary to produce a reliable tray performance design,
and this is included in the manuals. The overall purpose CT = m,p (24/s)’.’/[A~ [pbZp (Pliq - (8-324)
of the information presented in this text is to allow the
designer to (1) become knowledgeable in the component or,
details necessary for a proper design and be able to com- C+ = Vload/(At) (24/F1)~.~, ft/sec (8- 325)