Level measurements may be used to manipulate the flow of material through the plant, such as in a sewage treatment plant where a high level in the primary collection tank would increase the flow through the entire treatment plant.

Level measurements may be used to determine the interface between two materials. In some applications, the location of a liquid/liquid interface can be important to ensure that the separation occurs properly.

Most level measurement applications may seem routine but they can become relatively difficult to implement when additional factors are considered. For example, there are many level measurement systems that can be used to measure the level of a liquid in a vessel. However, the number of viable measurement systems decreases significantly when the vessel is agitated and operates at 200 degrees Celsius and 40 bar. Foaming and the effects of filling and emptying the vessel can further limit the viability of many level measurement technologies. Level measurement applications are rarely "cookbook" and typically require careful engineering.

Excerpted from The Consumer Guide to Non-Contact Level Gauges.

A common "rule of thumb" is to purchase the capillary tubes of the same length for both the high-pressure and low-pressure sides of the transmitter.  This would seem to make (mechanical) sense for flowmeters because the flowmeter element taps are generally close to one another, so the distances from the transmitter to each tap is similar. 

In level applications, the nozzles can be in quite different locations.  For example, a transmitter may be located 1 meter from the lower nozzle (near grade) and 8 meters from the upper nozzle (near the top of the vessel).  The "rule of thumb" would stipulate that the transmitter have (say) 10 meters of capillary tubing on both sides of the transmitter.  This would seem reasonable for the upper nozzle and excessively long (and costly) for the nearby lower nozzle (where the capillary tubing would typically be coiled for convenience).  However, this analysis only takes physical dimensions into consideration. 

The capillary tubing contains liquid that transmits the pressure from the diaphragm seal to the transmitter.  The density of this liquid changes with its temperature so the installation should be designed to maintain the liquid in both capillaries at the same temperature.  This is similar to the concept that was discussed for steam flow measurement transmitters. 

However, the liquid in the capillary tubing is different from the steam flowmeter seal in the sense that the liquid in the capillary tubing is captive.  Therefore, expansion of the liquid in one capillary will cause the liquid volume to change and affect the differential pressure measurement.  By its nature, the differential pressure transmitter will approximately cancel the expansion and density affects for capillary tubes of equal length (and liquid volume) when the capillaries are at the same temperature.  However, unequal lengths of capillary tubing (and unequal liquid volume) on each side of the transmitter can affect the measurement because the different capillary lengths (and volumes) result in different amounts of expansion.  Therefore, it is desirable for the capillary tube lengths to be not only equal, but also at the same temperature.

This article originally appeared in Flow Control magazine. 

1. 3 percent
2. 1.7 percent
3. 1 percent
   
4. None of the above

The performance of the flow measurement system should be inferior to the performance of the flow element itself.  Answer C could be viable (given "approximate" in the wording of the question).  However, the magnitude of the performance of the other two components (1 percent) causes Answer C to be incorrect. 

It is not often that all of the components in a measurement system are in error by the maximum amount in the same direction at the same time.  Therefore, the errors for the three components in this system would not be mathematically added to obtain 3 percent.  Answer A is not correct. 

The overall accuracy of measurement systems is typically calculated by taking the square root of the sum of the squares of the effect of each component.  For this system, this would be calculated as the square root of (12 + 12 + 12), or 1.73 percent.  It might appear that Answer B is correct, however further investigation would show that the flowmeter element has a percentage of rate error whereas the transmitter and indicator have percentage of full scale errors.  This means that the transmitter and indicator exhibit flow errors in excess of 1 percent at flows below 100 percent of full scale flow.  Therefore, the calculation that results in 1.73 percent performance is only correct at full scale flow.  Answer D is correct.

Additional Complicating Factors

Definitive calculations to determine the overall accuracy of a measurement system entail analysis of many factors that affect the measurement including dimensional, physical, process, operational, calibration, ambient and measurement considerations. 

This article originally appeared in Flow Control magazine.