The crash earlier this year of an Ethiopian Airlines Boeing 737 Max that took the lives of all passengers and crew aboard appears to have been caused by the combination of an angle-of-attack sensor that was sending an incorrect signal and control software that reacted to this signal, not allowing the pilot to override it.
With consequences that are fortunately not as drastic as this, with thousands of sensors and control loops that are in every pulp and paper mill, similar combinations of failed sensors and control reactions can often occur in the manufacturing process.
One common type of sensor that can send a false signal is a pH meter. Not only do they tend to drift over time, but if a deposit forms on the surface of the meter, it results in a false reading, and it can be difficult to remove the deposit. Some mills install a second pH meter in the same place to have a back-up signal, but if the chemical conditions are right for deposition, both meters will be affected. If a control loop is in place to control the pH to a constant value using acid or alkali addition, not only should a procedure be in place for regular calibration and maintenance of the meters, but an alarm should be triggered if the pH value or acid/alkali addition rate goes out of normal range.
Another type of sensor that is often neglected in a paper mill is a Venturi effect flow meter. Very simple in principle, the readout depends on the difference in pressure before and after a restriction, measured as the difference in height of an attached water column. Over time, due to changing conditions such as erosion, deposits, and air bubbles in the water column, these values can become inaccurate. At one paper mill, I tried to map the steam flow all around the mill, and the numbers from all of the flow meters never quite added up to the flow from the main header. I had to go and find the flow meters that appeared to be the most erroneous. Sometimes I could correct the reading by bleeding the air bubbles out. Other meters were completely disconnected, or I discovered a steam leak that was the source of the problem. Steam flow meters don’t get a lot of maintenance attention because steam is usually used as a variable to control temperature, and control loops generally don’t care about the absolute value of the steam flow. If you’re trying to improve energy efficiency, though, it’s hard to manage what you can’t measure accurately.
One solution to failed sensor signals that can prevent a control loop from overreacting is to generate a soft sensor. This is a signal that simulates the sensor value (such as a pH value) based on a combination of related values (such as caustic or stock flows). If the soft sensor and physical sensor disagree significantly, then you know it’s time to look at your sensor for calibration, maintenance or repair. In the meantime, the control loop can follow the soft sensor instead of the failed signal, preventing a system “crash”. Maybe this kind of approach is possible for Boeing’s control loop problem as well!
Martin Fairbank has worked in the forest products industry for 31 years,
including many years for a pulp and paper producer and two years with
Natural Resources Canada. With a Ph.D. in chemistry and experience in
process improvement, product development, energy management and lean
manufacturing, Martin currently works as an independent consultant,
based in Montreal. He is also an author, having recently published
Resolute Roots, a history of Resolute Forest Products and its
predecessors over the last 200 years.
Martin Fairbank Consulting
- Pulp and Paper Technology
- Materials Recycling
- Biorefinery Development
- Government Subsidy Programs
- Technical Writing
. White Papers
. Grant Applications
. Explain technical concepts
- Scientific Editing
. Review of articles for publication
- Project Assessment
. Evaluation of Technologies
. Project evaluation for funding agencies
- Pulp & Paper
. Conventional and emerging technologies