THIS MONTH’S PUZZLER
The atmospheric tower (see figure online at http://goo.gl/ErGQki) of our crude distillation unit performs adequately but we’d like to improve its efficiency and capacity as well as eliminate a few nagging problems. We’ve run a few experiments to justify simulation runs. Gamma scans have shown flooding in the sieve trays at and above stage 8 in the heavy gas oil (HGO) section of the column when we tried to increase crude feed. We also see some channeling in the structured packing in the light gas oil. In addition, we’re concerned about whether the type-304 stainless steel used for the structured packing in the naphtha pump-around and elsewhere is adequate now that we’ve increased the ratio of heavy to sweet crude; samples in that section show higher H₂S concentration than ever before. We have caustic injection quills downstream of the desalters to neutralize HCl but we’re worried about downstream sodium. Other problems include a slightly higher pressure between the outlet of the air condenser and the vent discharge to the sour gas knockout drum. Occasionally, we’re seeing dry trays between stages 13 and 20. Also, the control valve for the HGO steam stripper sometimes runs 100% open, restricting tower throughput. What should we include in future analyses and what do you recommend to improve tower performance?

CONSIDER A VARIETY OF STEPS
Let’s address the operational problems first: 1) heavy crudes put a strain on the (bottom) stripping section and middle of a crude distillation tower; 2) a high pressure in the condenser overhead reduces the efficiency of the upper stages in the rectifying section, reducing the purity of the overhead; 3) the flexibility of the crude distillation unit to changes in the crude composition largely is won or lost in the desalter; 4) controlling pH in the rectifying section with caustic is limited by increase in sodium, which shows up in the residuals from the vacuum tower and atmospheric tower — most refiners discontinue caustic injection while processing higher-than-normal concentrations of residue; and 5) “dry trays” in a distillation tower mean there isn’t enough liquid to interact with the vapor, thereby reducing the efficiency of the particular stage.

Consider increasing the efficiency and stages in the HGO section by decreasing the tray heights to make room for more trays and replacing the pump-around trays with structured packing. You also may want to look at the efficiency of the HGO stripper. Regarding the steam stripper flow, check the size of the control valve.

A high pressure in the overhead could stem from a number of things: a fouled exchanger, a blocked vent or mesh pad, or high downstream pressure. On p. 335 in “Distillation Troubleshooting,” Henry Kister quotes Smith’s famous three questions: “Is it clean? Is it vented? Is it drained?” Kister notes that the first question covers most problems.

Regarding the desalter, caustic addition has its limits and heavy crudes test those limits. This is especially true with sour water recovered for re-use in desalters. Some refiners are adjusting crudes to a more neutral pH (~5.5) to reduce amine reaction with overhead HCl that can cause condenser fouling and to promote better separation in the desalter; calcium naphthenate, a product of reaction of calcium with naphthenic acids, stabilizes emulsions in desalters. Weak acids, such as acetic and lactic, have been injected into (sour) wash water prior to its addition to desalters. This might address material concerns in the tower.

A dry section in a tower always is serious. I noticed that the desalted vapor feeds into the top of this zone. Perhaps there’s a pressure control problem in your crude flash drum; that’s where the vapor and liquid feeds to the crude tower are split. Tune the loop before doing anything drastic involving mechanical modifications. You may want to run simulations on relocating this stream, moving it further up the tower or maybe splitting it so that it doesn’t disrupt the vapor/liquid equilibrium. Another possibility could be a leaky tray caused by corrosion or fouling of a sieve. You should run a gamma scan to identify the tray and replace it during turnaround.

As for the turnaround, look at improving your desalter operation to reduce your reliance on caustic soda, pretreat the sour water you’re using and consider adding de-emulsifier injection quills to your desalter piping. An inspector should take a careful look at the type-304 stainless steel trays and structured packing during the turnaround. If additional gamma scans reveal leaks in trays, then replace the column internals in a staged fashion over the next few years: start at the top, where the HCl, water and H₂S are worst.

When looking inside the tower, your inspector will check for rust lines that shouldn’t be there in stainless steel. They may appear where the plate is welded together and chromium has been depleted: all stainless reverts to carbon steel. Consider type-316 stainless or iron-nickel-molybdenum-chromium alloys, especially for smaller components of sieve trays.
Dirk Willard, senior process engineer
Ambitech Engineering, Joliet, Ill.

AUGUST’S PUZZLER
We’re expanding our plant and are planning to add two heaters to our steam system, which sometimes can barely keep up with demand as it is now. Figure 1 shows a rough sketch of the process flow with the steam usage from the data sheets. The plant barometric pressure is 14.66 psia. I’m concerned the new loads will be too much for our steam piping. Our plant superintendent feels the boiler makeup water pump is a limitation. How should we upgrade the steam system?

Send us your comments, suggestions or solutions for this question by July 11, 2014. We’ll include as many of them as possible in the August 2014 issue and all on ChemicalProcessing.com. Send visuals — a sketch is fine. E-mail us at ProcessPuzzler@putman.net or mail to Process Puzzler, Chemical Processing, 1501 E. Woodfield Rd., Suite 400N, Schaumburg, IL 60173. Fax: (630) 467-1120. Please include your name, title, location and company affiliation in the response.

And, of course, if you have a process problem you’d like to pose to our readers, send it along and we'll be pleased to consider it for publication.