Mcnulty Flexipac

Figure 9.3 indicates that at this flow parameter, 1" IMTP

is the most likely choice. This will be used as a first trial.

Capacity Factor of sieve trays at 18 inch tray spacing =

Capacity Factor of 1-inch IMTP « 0.22 (Figure 9.3) STEP 2 Mccify the above capacity factors as per Appendix 9A.

(1) Correct sieve tray capacity factor for the smaller tray spacing. From Figure 5.10, this correction for the capacity factor is roughly (at a flow parameter of 0.16) 0.21/0.23 ■ 0.91

(2) Allow a greater safety factor for the packed column as p«r assumptions (ii) and (iii), Appendix 9A. This will allow for the considerations in assumption (iii) in Appendix 9A, and also for some capacity reduction because of the low surface tension (as per assumption (ii) in Appendix 9A). Let the safety factor be a conservative 0.8.

(3) Assume no additional corrections are needed for support rings, downcomer size and hole area.

(4) Apply a surface tension correction to the tray capacity factor as per assumption (vii) in Appendix 9A.

(5) Apply a correction factor for the lower visccs.zy zz the packed column capacity factor. As a rough approximation, assume (based on.Ficure 7.6) that capacity is proportional to , and that Figure 9.3 is based or. a viscosity of "unity.

(6) Calculate the corrected capacity factor

For trays: correction factors in (1) and (4) apply

Cg, corrected » 0.18 x 0.91 x 0.86 = 0.14 For packings: correction factors in (2) and (5) apply Cg, corrected 0.22 x 0.8 x C.83 = 0.15

(7) As a first approximation, it appears that column capacity with 1 inch IMTP will be of the same order as the existing trays.

STEP 3 Calculate the HETP for trays and packings using Figure

9.4. For trays, at a tray efficiency of 59% and 18 inch tray spacing, HETP ■ 30 inches (Figure 9.4). Correct fcr a tray spacino of 15.75 inches by multiplving 30 x 15.75/18"» 26.25.

For packings, HETP »18 inches (Figure 9.4). Carreer fcr the loss of height to distribution equipment, ie 1S/0 . 93 « 18.4 inches.

A= a first approximation, the application of 1" IMTP can therefore be expected to increase the number of theoretical trays by about 26.3/18.4 « 1.43, or by about 30-50%.

STEP 4 From the above rough analysis, it appears that replacing trays by packings may be attractive, and may boost the number of stages for the separation by about 30-50% at roughly an equivalent capacity. From this rough analysis it can be concluded that replacing trays by packing in the above application is feasible, and warrants further study; and that the studies should concentrate on the application of 1 inch high-efficiency packing (eg IMTP, Pall rings). Note that the analysis is far too rough to permit a more specific conclusion to be drawn.

Footnote: The above is based on a case study reported in Reference (21). 1 inch IMTP was used to replace the trays, with a 50-55% boost in the number of theoretical stages.

Best results for replacing trays with packings are obtained at low values of flow parameters, i.e. low liquid loads and low vapor densities, such as those experienced in vacuum and atmospheric pressures, particularly if the trays operate at a relatively low efficiency (Figure 9.3 and 9.4). Another place where replacing trays with packings is effective is systems which tend to foam (Section 7.5.1), as packings can handle these better than trays.

Poor results with this type of revamp may be expected at high liquid rates, in high pressure systems, and where sieve trays are capable of achieving high tray efficiencies. In addition, poor results may be : achieved with this type of revamp if the system contains solids or in large diameter columns, where maldistribution is a significant factor.

TRADING OFF CAPACITY AND EFFICIENCY Column loadings often vary from one section of column to another, and sometimes throughout the same section. When this occurs, packings have a distinct advantage as a revamp tool, because changing packing size can be used to easily convert spare capacity into additional number of stages; similarly, it can convert a spare number of stages into additional capacity. This is somewhat similar to the technique of revamping a trayed column by changing tray spacing; however, tray spacing changes often require substantial amount of welding of tray and downcomer supports to the column shell, and this may be expensive, if at all possible.

A very popular revamp technique is to pack the section of column which operates closest to maximum loading with a packing large enough to ensure that maximum capacity is achieved, even though some separation stages may be lost, and to make up for this loss (if any; alternatively, to increase the number of separation stages) by using a smaller packing in the unloaded section. This is perhaps the most common technique used for replacing tray with packings.

9.2.5 Proprietary Devices

There is a multitude of proprietary devices catering for specific types of revamps. Two of the more popular types are discussed below.

LINDE MD (MULTI DOWNCOMER) TRAYS These trays are particularly suitable for high liquid rate, high pressure applications. They are best applied for capacity revamps, but by reducing the tray spacing and increasing the number of trays their capacity advantage can be traded off for a greater number of stages, and therefore energy savings.

The unique features of these trays include:

(i) A greater effective length of outlet weir due to the large number of downcome'rs. This reduces liquid load.

(ii) The downcomer terminates above the tray, and the liquid issues from the downcomer through spouts; these are used to distribute the liquid to the tray below. This eliminates the downcomer apron on the tray and increases the area available to vapor and liquid flow.

(iii) Each successive tray is rotated by 90", thus minimizing vapor and liquid distribution problems.

Additional information is available in reference 1 and in the manufacturer's literature (16,17).

STRUCTURED PACKINGS These packings have been discussed at length in Section 7.6. Their main application has been in vacuum operations, although they can also be used for atmospheric and pressure operations. They are effective for capacity, energy savings and pressure drop reduction revamps. Their main disadvantage is their high cost compared to either trays or random packings, and the scarcity of information on their performance.

The analysis which compared trays and packing performance in Section 9.2.4 is extended in this section to include structured packings. As stated in Section 9.2.4, this analysis is crude and is suitable only for deriving very general guidelines. The limited amount of data available on structured packings adds to the uncertainties already discussed in Section 9.2.4 and Appendix 9A, and renders the extension of the previous analysis to structured packings even less reliable.

Results of the capacity comparison are shown in Figure 9.5. Flexipac was selected to represent structured packing capacity because it is the only structured packing for which published flooding data was adequate for developing a diagram similar to Figure 9.5. It is anticipated that the capacity characteristics of other corrugated-layer type packings will generally resemble Flexipac's r

FIGURE 15 CflPftCITy COMPARISON OF FLEX!PRC V* SIEVE TRAyS AND RANDOM

FIGURE 15 CflPftCITy COMPARISON OF FLEX!PRC V* SIEVE TRAyS AND RANDOM

FLOW PARAMETER =

(Figure 9.5), while the capacity of wire-mesh type packings will be lower.

The values compared are the design values of C based on total column cross-section area. The curves for IMTP and sieve trays are identical to those shown in Figure 9.3. The curves for Flexipac were obtained from the McNulty and Hsieh Flexipac flooding correlation (18) .

In the course of the comparison, several assumptions needed to be made, similar to those made in deriving Figure 9.3. These assumptions, as well as the procedure used to obtain the curves shown on Figure 9.5, are spelled out and discussed in Appendix 9B. Should the user wish to use this chart for any purpose other than drawing general guidelines, such as those described below, he is strongly encouraged to derive a diagram similar to Figure 9.5 using the procedure outlined in Appendix 9B and a set of assumptions which fits the case study considered. The Appendix also provides a checklist of the factors that affect the difference in capacity, which should be considered when revamping with structured packings.

Figure 9.5 shows that the capacity of Flexipac Number 2 is slightly lower than that of 2 inches IMTP. Therefore, comments made in Section 9.2.4 about replacing sieve trays at 24-inch tray spacing by 2 inch IMTP also apply to replacing these trays by Flexipac Number 2. Further, little capacity can be gained by replacing 2 inch Pall-rings or IMTP by Flexipac Number 2 or vice versa.

Figure 9.5 also shows that replacing 2-inch Pall rings, 2-inch IMTP or sieve trays at 24-inch tray spacing with Flexipac Number 3 or Number 4 will enhance column capacity. Replacing 2-inch IMTP by Flexipac Number 3 will enhance capacity by about 20 to 25 percent throughout. Replacing sieve trays at 24-inch tray spacing by Flexipac Number 3 will increase capacity by about 40 to 50 percent at low flow parameters, and by 30 to 40 percent at high flow parameters. At the high flow parameters, however, the margin between Flexipac Number 3 and sieve trays can be reduced to as little as 5 to 10 percent by using multi-pass trays, because liquid load can be lowered by increasing the number of passes on a tray (ie, lower flow parameter).

The efficiency of structured packings can be compared to the efficiency of trays and random packings by means of a diagram similar to Figure 9.4. As in the case of random packings, the efficiency curve will be a straight horizontal line. Comments made in Section 9.2.4 regarding allowance for uncertainty and for distribution and support equipment also apply to structured packings. With structured packings, two additional precautions apply:

(1) As shown in Section 7.6, HETP for structured packings increases with the F-factor. Therefore, the HETP figure used must correspond to the design vapor and liquid load. For instance, if the column is designed for a capacity of 80 percent of flood, the HETP used must be evaluated at 80 percent of flood.

(2) As discussed in Section 7.6, HETP1s for aqueous systems, high viscosity systems, and high surface-tension system are likely to be much higher than HETP values .shown in Section 7.6. It is therefore important to ensure that the adequate HETP data is used in the comparison.

Some efficiency data for various structured packings are presented in Section 7.6. Staying with Flexipac as in Figure 9.5, some efficiency data have been presented for the chlorobenzene-ethylbenzene system in Figure 7.22 (250.Y Mellapak is almost identical to Flexipac Number 2). Assuming this data is applicable to the sample case considered, , the expected HETP at high F-factors (at about 80 percent of flooding) j is somewhere between 13 and 16 inches. A horizontal line can be j drawn at this ordinate, and can be used to compare the efficiency of | Flexipac Number 2 with that of trays and random packings. This line ■ j falls below the line for one inch packings, and suggests that quite a j substantial improvement in efficiency can result from replacing j either trays (at 24-inch spacing) or 2-inch random packings with j

Flexipac Number 2, with the gains being largest when low-efficiency trays at 24 inches tray spacing are replaced by Flexipac Number 2.

Overall, it can be seen that structured packings can enhance column capacity, efficiency or both when used to replace either random packings or trays. When used to replace trays, the gains are most substantial at low flow parameters (low L/V ratio, vacuum) and where tray columns achieve a low efficiency. However, these capacity and efficiency improvements are not always achievable, and several pitfalls exist. The main pitfalls have been outlined above and in Section 7.6, and must be carefully considered when planning to revamp a column with structured packings.

9.2.6 Internal Hardware Revamps Summary

Figure 9.6 is a general application chart for column internals revamps. This chart by no means covers all possible options, and is only designed to provide very broad application guidelines.

HIGHER CAPACIiy

PORpOJE Of ftcvRMr

ENERGy

ifl age