Chapter 4
Removing the Process
Variability Barrier with
Automatic Control Systems
In every manufacturing process, there are variables that are continuously
controlled in real time with hardware that is referred to as instrumentation,
which is the generic term for the components within a “control loop”: sen-
sors, limit switches, transmitters, indicators, and controllers. An example
control loop is shown in Figure4.1, which is used in process and instrument
diagrams (P&IDs) like the example shown in Figure8.2.
Point 3 in our Model Vision states that our process is monitored and con-
trolled with “modern instrumentation.” One definition of modern instrumen-
tation is a system that has control loops that can be interfaced to the site’s
computer network and supervised by a personal computer-based system like
. In addition to improved control, this enables process vari-
ables and trends to be displayed on any workstation in the plant and stored
in history. Then, management can see for themselves what kind of variabil-
ity the plant is generating. To get an idea what this system might look like,
see Figure4.2.
Does this mean you have to update all the control systems in a plant just
to get started on the main thing? Surprisingly, my answer would be no, but
it is certainly desirable. In my opinion, you do not have to have the latest
42 ◾  Removing the Barriers to Efficient Manufacturing
Figure 4.2 Process and instrument diagram.
Set Point
Figure 4.1 Instrumentationcomponents within a “control loop.
Removing the Process Variability Barrier with Automatic Control Systems ◾  43
electronic controls systems to have a stable system. Even quality pneumatic
systems like Foxboro and Moore from the 1970s can work well if they are
in good condition. However, maintaining them in that state may prove to be
uneconomical and not worth all the effort. If you decide to take the plunge
and upgrade everything, understand that, while electronic instrumentation
has come down considerably in price over the years, the installation costs
are still formidable.
Regardless of which path you take, the end point has to be same. A
single operator must be able to sit at a desk and oversee everything going
on in an operation with a personal computer, and the system must be visible
on a network.
Getting Started
The process starts by you surveying every process variable you currently
control with real-time closed-loop controllers. Some typical parameters are
temperature, pressure, speed (or frequency), voltage, current, humidity, pH,
and maybe concentration. You will need to have a team verify, by monitor-
ing with data-logging equipment, that each and every variable is stable—free
of oscillation and able to track step changes accurately. If your employees
are not able to do this on their own, then you will need to contract to have
it done. Spend whatever time and money it takes to do this survey accu-
rately because everything following depends on it.
If during the survey any process variables are found to be erratic, here
are some potential causes:
1. Uncalibrated or deteriorating sensors
2. Faulty controllers and interfaces
3. Improper control tuning
4. Excessive play or looseness in final control elements (valve positioners
and servos)
5. Loose or corroded wiring connections
6. Leaking actuator cylinders and diaphragms or connecting tubing
7. Improperly sized control components (either too big or too small)
8. Improperly sized “prime movers” (motors, pumps, actuators)
Let us look briefly at each one:

Get Removing the Barriers to Efficient Manufacturing now with the O’Reilly learning platform.

O’Reilly members experience live online training, plus books, videos, and digital content from nearly 200 publishers.