Big Chemical Encyclopedia

Chemical substances, components, reactions, process design ...

Articles Figures Tables About

Strategy of Performance Testing

General. It is best to carry out a performance test engulfing the whole unit. Individual testing of columns, one at a time, increases the total effort and time consumed and reduces the reliability of measurements. Testing the entire unit provides several material balance cross checks, and enables better identification of erroneous meters and lab analyses. For instance, if the column feed analysis is off, the column component balance may not be sufficient to point out which analysis is suspect but if data from a component balance on upstream and downstream equipment is also available, the incorrect analysis can easily be identified. [Pg.409]

A shortcut to the above recommendation may be acceptable when the plant inlet rate and compositions have not significantly changed since the last plant test and the problem areas are well-known. In such cases, it may suffice to test only the specific column area (429a). [Pg.409]

Another shortcut is often acceptable when the column is near the end of a processing train which 3delds reasonably pure products. Product analysis and metering tend to be far more reliable than intermediate stream meeisurements, and there is usually more to be gained from cross checks with downstream measurements than with upstream measurements. In such cases, it may suffice to restrict the testing to the column area and downstream equipment. [Pg.410]

Duration. It is best to carry out a performance test over a 2- to 3-day period (2) and add another 1 to 2 days for coliunn pinch testing. If shorter periods are used, variations in plant conditions may introduce serious errors. Over a period of 2 days, errors are averaged out. Further, column control problems may make it difficult to obtain a sufficiently long period of stable operation if the test is short. Over a 2-day period, the column should be running under stable conditions at least for some of the time. [Pg.410]

Timing. The best time to carry out a performance test is when the plant is stable. In most plant situations, weekends are ideal, as changes due to fluctuations in upstream units are minimized. [Pg.410]


See other pages where Strategy of Performance Testing is mentioned: [Pg.409]   


SEARCH



Performance tests

Test Strategies

© 2024 chempedia.info