(55a) Custom Design and Evaluation of Oldershaw Columns for Pilot Plant Operations

Pilot plant distillation operations are typically challenged to provide realistic feedback on separations capabilities and recycle stream compositions by mimicking commercial scale operations as closely as possible.  Matching commercial scale operating conditions to provide information for scale-up can be especially difficult for columns with unique hydraulic loads.  Since columns represent a significant capital expenditure and can be long lead time items impacting the project schedule, it is imperative that the column perform as expected at startup to avoid additional expense and lost time due to rework.

The challenge for pilot plant designers is that little data and design methodologies are available for columns at the smaller scale.  Design correlations have been studied and published extensively for commercial scale equipment, but these correlations have often been untested or underreported at smaller scales.  Oldershaw column manufacturers typically offer off-the-shelf designs that have been proven to work in typical applications over the years, but they offer little assistance in evaluating the columns for a specific application prior to purchase.  Standard offerings may also be a poor choice for unique process conditions such as high or low L/V ratios, fouling services, etc. 

MATRIC (Mid-Atlantic Technology, Research and Innovation Center) has developed a methodology for evaluating Oldershaw column designs based on process simulations and to custom design Oldershaw trays to meet specific process needs.  This methodology has been developed through a combination of extrapolating commercial scale correlations, testing in MATRIC labs, and experienced based observations on the pilot scale.

An outline of the sizing and evaluation methodology is presented along with a case study demonstrating the use of the methodology to size a pilot scale column with a unique L/V ratio and fouling characteristics.  The adaptation of design correlations to the pilot scale are discussed and the methodology is followed to yield a custom tray design.  Project execution considerations for custom designed columns and validating the tray efficiencies at startup are also discussed briefly.