Gs
with hL taken from equation (23).
Downcomer Capacity
The liquid head in the downcomer required to provide uniform distribution of liquid to the tray distributors is obtained from a conventional pressure balance:
Gs
where hL = static head on the tray = w LZc, mm
and hd = head loss through tray slots, mm
hcoll = head loss for the phase mixture flowing through the collector elements, mm
hdisp = pressure loss through the liquid feed dispersion elements, mm.
All head terms are expressed in mm of liquid under tray conditions. Values of hdisp may be calculated from the conventional orifice equation:
Gs
For distributor pipe holes, Litz, 197213 recommends an average value of Cv = 0.61, but for the larger holes normally used in the cocurrent tray, a coefficient of 0.80 is more appropriate. However, the exact value of Cv depends on the particular design of the liquid entry dispersion system.
EXPERIMENTAL RESULTS
The test system is located at the Pickle Research Campus of the University of Texas at Austin. A flow diagram of the distillation portion of the system is shown in Figure 5. The circular column is 0.55 m inside diameter, is fully instrumented, and is computer-controlled to provide quality research data. For these tests the column was operated under total reflux conditions. The test mixture was cyclohexane/nheptane and head pressures of 33.3, 106, 166 and 414 kPa were used. The test mixture is the same as that used by FRI, and more information on these mixtures is contained in the various publications of that organization.
Three sets of tests of the cocurrent tray were made, and in the following discussion the sets are designated by the years in which they were made, 1988, 1993 and 1998. Key dimensions of the trays for the three sets are given in Table 1. This paper emphasizes the 1998 results, with 1988 results being used for comparison. The tray design for the 1993 tests was non-optimal, but the results for 1993 are available2 . Operations at a pre-determined set of conditions were continued until trend displays on the control house monitors showed steady conditions (usually after 90 minutes), and three samples taken at 45 minute intervals showed little or no variation. Analyses were made by gas chromatography.
EFFICIENCY RESULTS AND MODEL CONFIRMATION
The model described above was used to compare measured efficiencies with those generated by the model. Results of the most recent tests are given in Table 2 and Figure 6. Efficiency data for the 1988 tests are included for comparison in Figure 6; it is clear from Figure 6 that the tray has a limited turndown ratio, and the onset of weeping and dumping is apparent from the plots. However, comparisons with conventional sieve trays tested with the same column, test mixture and operating pressures show the co-current tray to have a much higher capacity and a higher efficiency, based on point efficiency determinations of the two types of device.
Figure 6 indicates an improved efficiency at relatively low loadings; note especially Figure 6a where the two ‘stray’ points may actually be correct. The mechanisms for such action have not yet been investigated.
The calculated efficiencies shown in Table 2 and Figure 6 include both gas and liquid phase resistances. Except for the vacuum data, a value of the dispersion factor b = 1.65 was found to give a reasonable fit to the measurements. For the vacuum tests, b = 1.50 was found to be more suitable. It appears that under the low liquid flow conditions of the vacuum tests, quality of the dispersion suffers to some degree, but at high loadings a value of b = 1.65 would be suitable. The ‘effective’ contacting height might be adjusted to allow for additional mass transfer that undoubtedly occurs as the liquid droplets are separated from the gas in the collector zone. For the present analysis the simple contacting heights shown in Table 1 were used.
For all of the tests, a ‘flood point’ was not clearly indicated. One would expect the upper operating limit to be defined by excessive loading of the collector elements, or excessive backup in the downcomers. For the highest liquid load studied (at 414 kPa), the calculated backup in the downcomers is 277 mm clear liquid, less than half the height of the downcomers. For the 414 kPa tests, it was not possible to take the tray to flood because of capacity limits on the reboiler (see Table 2). It is estimated that the tray could have been carried to at least an F-Factor of 3.0 m/s(kg/m3)0.5. 大容量精馏塔塔盘英文文献和中文翻译(5):http://www.youerw.com/fanyi/lunwen_12881.html