Industrial dryers — no matter how effective they are in drying solid materials — are incapable of maintaining product target moisture content (MC) at the lowest level practicable when evaporative-load changes to the dryer occur. Why? One big reason is that most dryers are controlled using moisture-content data obtained too late to be effectively used in a control system. Long dead times (time to detect an evaporative-load disturbance entering the dryer) produce proportionally wider moisture-content variations.

What does this mean for most dryer operators? Lack of effective moisture-content control forces operators to run at lower target moisture-content levels to avoid producing wet product. This is a costly practice that results in less production, higher energy costs and poorer product quality.

3 Most Common Problems with Moisture Sensing and Control

After viewing many drying operations over nearly 30 years, my experience has shown there are three common problems with currently used (PID-based) systems for moisture sensing and control. These problems prevent recovery of the above-mentioned productivity losses. They are:

• Long dead time due to a lack of timely, accurate and reliable moisture-content data upon which to make control decisions.
• Inadequate evaporative-load sensors, which result in problems sensing the moisture content of wet feed.
• Lack of a simple, precise method for continuously calculating a new process-variable setpoint to maintain the target moisture content following an evaporative-load change to the dryer.

Each of the three main problems and potential solutions will be discussed.

Inadequate Access to Timely Moisture-Content Data

Long dead time is one of the three major problems that prevent effective moisture-content control. This is not a new problem. It has existed since the beginning of feedback-type control. There is a dead-time barrier at the dryer exit that prevents significant improvement in currently used moisture-content sensing and control systems.

The only way this barrier can be breached is to move the moisture-content sensor closer to the feed point inside the dryer. Unfortunately, most inline moisture-content sensors are incapable of operating in the harsh environment of a dryer.

Currently, three methods commonly are used for obtaining moisture-content data from the dryer.

• Inline moisture-content meter.
• Laboratory samples taken from a location downstream of the dryer exit.
• Use of the exhaust temperature as a measure of the moisture content.

Inline Moisture-Content and Laboratory-Samples Methods. Both the inline moisture-content meter method and the laboratory-samples method are located remotely from the product infeed and, therefore, produce long dead times.

Figure 1 depicts a dryer with two methods for supplying moisture-content data for control. It takes 20 minutes for an evaporative-load disturbance entering the dryer to be detected by an inline moisture-content meter and control action to commence. The laboratory sample requires 30 minutes for detection. Because dead time and moisture-content variation are directly proportional, control of this dryer is essentially no control due to the long dead times. This, however, is the case for a large proportion of dryers operating today. This causes wide moisture-content distributions that are costly in terms of production, thermal energy and quality.

A feedforward signal for the evaporative load often is suggested for use in reducing dead time. However, this fix has limited success in dryer moisture-content control systems due to the difficulty of measuring the wet-feed moisture content. Other methods suggested are complex and often too much for the average plant control personnel to maintain in many plants.

Exhaust Temperature Control Method. The third method is using the exhaust temperature as a measure of moisture content. Exhaust temperature control sometimes is used for moisture control in rotary and flash dryers. It offers an inline moisture-content sensor with a lower dead time; however, this improvement can be negated by the tenuous moisture-content vs. exhaust temperature relationship.

Using the exhaust temperature for moisture control is effective only as long as there are no evaporative-load disturbances. As soon as an evaporative-load change enters the dryer, the exhaust temperature control method requires a new setpoint. But, there is no precise, mathematical method available for calculating a new setpoint. In most installations using this method, the operator has placed the controller in manual mode. It is continually searching, but seldom finding, the proper setpoint that maintains the target moisture.

Other Challenges to Effective Dryer Moisture-Content Process Control

The lack of a simple, effective evaporative-load sensor also affects dryer moisture-level control. As previously discussed, the devices currently used to determine the evaporative load have not been effective due to problems with sensing the moisture content of wet feed. One way to resolve this is to find an evaporative-load sensor that is simple, effective and can be installed inside the dryer closer to the feed point.

Likewise, a simple, precise method for continuously calculating a new process-variable setpoint is needed to maintain the target moisture content following an evaporative-load change to the dryer.

One in-dryer moisture-content sensor addresses these problems using the following equation:

MC = K₁(ΔT)^p – (K₂/S^q)

The model relates the moisture-content level of the product when exiting a dryer to two variables: the temperature drop (ΔT) of the hot air after contact with the product, and the evaporative load or dryer speed (S). The calculation of constants (K₁ and K₂) and exponents (p and q) are not required for the system to work properly.

The in-dryer moisture-content sensor addresses each of the common problems with moisture-content control in dryers. The model relates moisture content to a temperature drop (ΔT) process variable that is easily measured. The sensor can be installed inside the dryer (figure 2). The sensor nearly eliminates the dead-time delay for dryers. It consists of two ordinary temperature sensors and software. It can be applied to direct, indirect and batch dryers. Figure 3 illustrates some of the applications for direct dryers using hot air or superheated steam as the drying medium.

The evaporative-load moisture-content soft sensor eliminates the need for moisture-content and mass-flow meters on the feed prior to entering the dryer. It has been mathematically shown that the evaporative load may be sensed in terms of the (ΔT) process variable simply by substituting it for moisture content in the in-dryer moisture-content sensor and software. The sensor solution uses a mathematical method for calculating the (ΔT) setpoint needed to maintain the target moisture content following evaporative-load changes to the dryer.

The in-dryer moisture-content sensing and control system has been shown to deliver results. In testing, the target moisture content is maintained with at least 30 percent less moisture-content variation than that achieved using alternative dryer moisture-content control systems. When two dryers are operated in series and each independently controlled with an in-dryer moisture-content sensing and control system, the possible percent reduction in standard deviation of the final product is 51 percent. Actual reductions in standard deviation have ranged from 45 to 47 percent in company testing.