When a AP transmitter is used for measuring base level, it is common to locate it above the upper level nozzle and to purge at least the lower impulse line, and often both impulse lines. Generally speaking, purges are intended to accomplish either or both of two functions:
1. Sealing, which is simply the isolation of the transmitter from a clean (no solids, nonplugging) process liquid. Low-velocity laminar flow is usually adequate for sealing, but as we shall see later, may contribute to dynamic measurement errors. For these applications gas purges are usually used.
2. Flushing or sweeping, which may be required where solids in the process liquid tend to plug the impulse line. Gas purges often are not adequate. But for liquid or gas, turbulent flow through the exit diptube is recommended. This usually requires a vastly higher purge flow rate than is common practice.
Now let us look at some sources of level measurement error due to purge systems.
When the level transmitter span is small, one must be careful to ensure that both purge systems are equalized—that is, that they have the same flow rate and same length of same diameter impulse or purge line. Failure to accomplish this equalization may result in an offset in transmitter reading that will not be detected by conventional calibration procedures. This consideration is particularly critical for specific gravity measurement by A P.
If purge rate is low enough, a rapid increase in liquid level will back liquid up into the lower impulse line. The transmitter will only gradually reflect the level change as the purge flow slowly dsplaces the liquid (Figure 11.12).
A rapid increase in vessel pressure can cause a different kind of error if purge rate is low enough. In this case liquid is backed up into the lower impulse line and the transmitter indicates a false low level (Figure 11.13). Again the error disappears as the purge flow displaces the liquid. It is probably apparent that either of these phenomena can play havoc with a level control system.
Static pressure variations on the process side cause changes in the true volumetric flow rate of gas purges. This has been particularly troublesome with vacuum columns where the change in volumetric purge flow rate from startup conditions to normal operating conditions may be 100:1 or greater. The result
Was this article helpful?