DAVID W SPITZER'S E-ZINE (November 2025)

The transmitter sends a radar frequency pulse down one of the probes. The dielectric constant of gas above the material is close to 1, so the radar energy does not disperse and travels down the probe. When the pulse reaches the material, some of its energy is dispersed into the material. Some of the dispersed energy is sensed by a second probe in the assembly that is located parallel to the first probe. The sensed energy then travels on the second probe to the transmitter. In a sense, the energy traveling on the second probe could be considered a reflected signal from the surface of the material.

In effect, the transmitter sends a pulse down one of the probes and receives a reflected signal on the other probe. The reflected signal takes more time to return to the transmitter when the probe assembly is less covered. The transmitter measures the amount of time that the pulse takes to travel on the two probes and across the gap between the probes. The measured time, speed of the radar pulse, and probe geometry are used to calculate the covering of the probe. Mechanical dimensions can then be used to determine the level in the vessel.

The ability to utilize this technology is dependent upon the dielectric constant of the material. Materials with sufficiently high dielectric constants allow the energy to be absorbed into the material where it can reach the second (measurement) probe. In general, materials with higher dielectric constants tend to have stronger reflected signals that result in more reliable and more accurate measurements.

Materials with low dielectric constants do not absorb radar energy, so the radar energy tends to be confined to the first probe and cannot reach the second probe where the return signal is measured. Such materials are generally not appropriate applications for this technology.

Excerpted from The Consumer Guide to Capacitance and Radar Level Gauges

With the advent of distributed control systems (DCS) and programmable logic systems (PLC), the analog signals could be connected to a DCS or PLC input that would convert the signal to a digital number for display and/or totalization.  Remembering that the development of DCS and PLC technologies had their roots in analog and discrete control respectively, so it would not be surprising to find different that DCS and PLC analog inputs are of different quality. 

On a recent project, the accuracy of two PLC analog input cards from the same manufacturer used in the same application were 0.1 and 0.35 percent of full scale.  This is a wide divergence in performance.  Using the latter specification, the error attributable to the analog input is greater than the error attributable to a flowmeter with 0.5 percent of rate accuracy operating below approximately 70 percent of full scale flow.  It might seem counterintuitive that the analog input card would contribute more error than the flowmeter at such a high flow rate... but check the math for yourself. 

In contrast, a DCS analog input card might exhibit an accuracy of 0.03 percent of span.  Assuming that this performance is typical of PLC and DCS analog inputs, it is not surprising that the DCS (with a legacy of continuous analog control) exhibits better analog performance that the PLC (with a legacy of discrete control).  Conversely, given these legacies, one would expect to find that PLCs have better digital capabilities than a DCS. 

What is the quality of your analog input? 

 This article originally appeared in Flow Control magazine.

The affinity laws for centrifugal pumps are given as: 

• Flow is proportional to the pump speed
• Pressure is proportional to the square of the pump speed
• Brake horsepower (energy input) is proportional to the cube of the pump speed

Noting that the first affinity law states that the flow is proportional to pump speed, it would seem logical that the flow would be 60 liters per minute at 60 percent speed.  Therefore, Answer B would appear to be the correct answer. 

Not so fast.  The pump operates at 60 percent speed so it will generate a pressure of approximately 36 meters of water column.  If the application involves pumping water into a process vessel located 40 meters above the pump, the flow rate will be zero --- not 60 liters per minute.  In this application, no flow will occur until the pressure of the pump exceeds 40 meters of water column, which is at approximately 63 percent speed.  In short, there is not enough information about the process and installation in the problem statement to determine the flow at 60 percent of full speed, so the correct answer is Answer E.

Additional Complicating Factors

Presuming that additional process and installation information is given, this problem would be more complicated if the vessel is operated under pressure.  This is because different vessel pressures will require different pump pressures and hence, different pump speeds.

This article originally appeared in Flow Control magazine.