Chemical Processing magazine has a great article, Be Levelheaded About Surge-Tank Control. I shared it with Emerson’s James Beall asking for his thoughts. He noted that surge tank level control is a common application in many of the process industries. Often a surge tank helps to absorb disturbances between units such as between distillation trains or between a reactor effluent tank and the refining train.
He also shared that the bottoms level in a distillation column could be controlled much like a surge tank in the case where the bottoms level controller manipulates the bottoms flow as the feed to the next column. While the objective is to keep the level above the heat exchanger tubes and below the reboiler entry line into the column, the volume between these points can be used to “absorb variability”.
It’s more important to reduce the variability in the bottoms flow as long as the tuning keeps the bottoms level between the specified limits. In other words, we want the tuning to be as “slow” as possible while still “fast” enough to keep the level within limits. This will provide maximum “filter affect” to reduce the variability of bottoms flow due to upstream disturbances.
James noted that you could think of a surge tank as a filter on your process. Depending on how aggressively you tune the level loop determines the size of the filtering action. The tuning can increase or decrease the effective “filter affect” of the tank. Slowing down the tuning will increase the amount of filtering capability.
For instance, if the flow rate into the surge tank cycles in a range of plus or minus 10%, then slow tuning may reduce the cycle on the outflow to a range plus or minus ½%. Conversely, if the tuning is more aggressive than required, the 10% cycle on the input might be 5% on the output!
James noted that level control is an integrating process. For some level control applications, the objective is to hold the level steady. For many other applications, the article’s author highlights:
The purpose of a surge vessel is to provide averaging or smoothing so that changes in discharge flow are less rapid and less extreme than changes in feed flow. Tight control of level is counterproductive — very quickly translating any change in feed flow to a change in discharge flow. Instead, vessel level is allowed to vary “within reason,” that is, as long as it doesn’t cause process trips on high and low level.
James highlights two things a control engineer should think about. One is that the improper tuning CAN cause or amplify any oscillations. If the integral tuning is too fast, the tuning will cause oscillations on its own!
The second is that for surge control applications, the strategy is to use tuning parameters on the PID loop that are as “slow” as possible, while keeping the level within an allowable deviation range. Internal Model Control based Lambda Tuning yields non-oscillatory tuning for integrating processes and allows the user to select the proper speed of response.
What James and the team of process control consultants would suggest is to define a maximum allowable variation and understand what the maximum load disturbance possible would be. For instance, if the 50% flow rate into the 5000-gallon surge tank is 100 gallons/minute (GPM), an inflow disturbance of plus or minus 50 GPM is possible, and the maximum allowable variation is plus or minus 30%, then the tuning would be:
- Proportional Gain = 0.81
- Reset Time = 7270 seconds
- First order PV filter less than 36 seconds.
See James’ Emerson Exchange presentation, Process Dynamics and Advanced Loop Tuning, beginning on page 14, for the formulas used to calculate these tuning parameters.