Upset trays

Steam is ineffective because of upset trays. At the design steam rale, check the pressure drop across the trays. For a typical four-tray stripper, the pressure drop should be roughly 1 ft of water. If the pressure drop is very much less than this, the stripping trays are likely to be damaged. 

Reduced ASTM gap Flooding Upset tray decks Pumparound rate too high Top trays corroded Water in reflux Downcomer Installation... 

Fractionation problems same as crude unit Water pockets upset trays Check steam under relief valves... 

If the observed pressure drop across the trays is small (in Equation 1) the cause of the temperature inversion is upset trays. On the other hand, if an increased reflux rate enhances a temperature inversion between trays, flooding may be predicted with confidence. 

Problem Absorber bubble caps were replaced with sieve trays. A severe flow upset dislodged the new trays. 

On the other hand the tower should be able to operate at changing vapor and liquid loads wthout serious upset. In this tjpe of tray the designer has a selection of holes, in this case For the top select 1,100 to 1,500 for the bottoms, select 1,300 to 1,750, and still expect acceptable performance. 

However, if there is a sudden drop in cooling water temperature (as can occur during a thunder shower or blue norther ), the pressure in the column can fall rapidly. This can cause flashing of the liquid on the trays, will upset the composition and level controls on the column, and could even cause the column to flood. 

Figure 8-53 shows a propylene-propane fractionator controlled at maximum boil-up by the differential pressure controller (DPC) across the trays. This loop is fast enough to reject upsets in the temperature... 

A vertical vessel (drum), as shown in Fig. 7.1, can be used to separate vapor from liquid following equilibrium flash vaporization or partial condensation. A reasonable estimate of the minimum drum diameter Dr, to prevent liquid carryover by entrainment can be made by using (13-7) in conjunction with the curve for 24-in. tray spacing in Fig. 13.3 and a value of Fha = 10 in (13-5). To absorb process upsets and fluctuations and otherwise facilitate control, vessel volume Vv is determined on the basis of liquid residence time t, which should be at least 5 min with the vessel half full of liquid. Thus... 

Guidelines for sump residence time are scarce in the published literature. Those given by Wheeler (420) for chimney trays are the only detailed list available (Table 4.1). Wheeler (420) did not state the residence time definition on which his guidelines are based. The author believes that their application with the above definition is reasonable for chimney trays, bottom sumps, and side-draw drums. Caution is needed when these guidelines are applied, particularly when the main consideration is to buffer upsets. 

In some instances, installing automatic stripping steam shutoff trips on high bottom level has been highly successful and saved many trays in refinery vacuum columns (296). Unfortunately, this technique is only suitable for the few services where cutting out steam, boilup, or bottom feed is unlikely to greatly upset the column. 

One case history contributed by J. A. Hatfield (1515) describes a methanol-water column, reboiled by live steam injection, which used a non-MB control scheme. The scheme was the same as those in Fig. 16.4a or 5, but there was no automatic temperature control and both steaun injection and reflux were flow-controlled. The operators manually adjusted the set point of these flow controllers to suit a tray temperature. The system performed poorly and the column expereienced upsets and off-spec products. Changing the system to scheme 16.4d eliminated the problems and gave excellent control. 

Tower trays were repeatedly upset due to pressure surges resultii from water accumulating in the tower during short unit outages. The source of water was condensation of purge steam used under the cohmm relief valves to prevent their inlets from plugging. 

Even a perfectly assembled valve tray may become leaky onstream. Usually, a small slug of water enters the fractionator, flashes, and produces a pressure surge. The trays are bumped but not upset. If a product drawoff has lost capacity after a brief unit outage (such as a short power failure), water is the probable culprit. 

Most often, tower packing supports are dislodged, or distillation trays upset, when a tower is operated with an excessive bottom liquid level. Forcing heat into a tower when the liquid level is several feet above the bottom tray deck often results in dislodging the bottom few trays. Occasionally, trays are mis-assembled during a turnaround. The results of either of these misadventures is diminished fractionation efficiency. 

Effective steam stripping is indicated by a 30°F-50°F temperature drop between the flash zone and the tower bottoms. A lower AT means that the steam is not effectively contacting the resid. This can be due to upset or corroded trays or tray flooding. Often, poor level control in the bottom of the tower will permit resid to back up and flood the stripping trays. 

Stripping steam tcx low Stripping trays upset Adjust steam to jets... 

FIGURE 17-3 Malfunctioning temperature trip upset tower trays. 

A major problem with rerunning wet hydrocarbons is unexpected slugs of water that appear in furnace feed. Such slugs of water are only partially vaporized in the preheat furnace. Upon entering the hot fractionator flash zone, the water suddenly flashes. Pressure surges, powerful enough to upset the fractionator trays, are a frequent consequence. 

In some services, corrosion to the overhead reflux condensers contaminates the reflux flow with particulates. Alternately, caustic may back up into the reflux drum through the distillate product line. This can happen during a unit upset. Either way, the tower top tray can become plugged while all the trays below remain open. 

Say, "We have upset the trays in the fractionator," not "You have a problem, as I have determined that the trays are upset." Avoid you and I concentrate on using the pronouns we and our. 

---