The fundamentals of troubleshooting in industrial automation

Industrial automation insights

• Maintaining industrial equipment can seem like a complex process — and it sometimes is. However, many people can resolve simple tasks if they know the fundamentals of automation and use their senses.
• Operators also can help themselves by reading the provided documentation and start from a basic level of understanding. From there, engineers can understand and diagnose the problem with industrial automation so they can take the appropriate action.

To maintain and troubleshoot industrial equipment, several disciplines or fields of study are required. While it is not necessary to have an engineering level of understanding, there is a level of mechanical and electrical knowledge that is required.

Mechanical powertrains, pneumatics and hydraulics are some of the most basic building blocks of industrial machinery. More complex systems include servomechanisms, robotics, and machine vision.

Understanding programmable logic controllers (PLCs) and computers are important in the control of machines, along with the operation of sensors and different types of actuators. Different types of software are used to access programs and different data types used in presenting numerical and text information.

There also are classifications of machinery that can be useful to know about. There are many different types of packaging machines, robots, pumps, conveyors, and material handling machines that have similar concepts behind them. Vibratory bowls, escapements, dial tables and indexers are other general classifications of equipment. Learning about these general categories as they apply to equipment and assets in use can be helpful.

Observation and analysis

Most machinery has certain things in common. It is usually used to move, process or change a product. There is usually a place or places where material enters the machine, and a place where product exits.

Watching a machine in operation is one of the best ways to learn how it works. How does the product enter and exit the machine, and what is done to it as it is processed? How fast does the machine or system run in parts or volume of material per minute?

Does a machine operator have to load or unload parts or material, and does their speed affect machinery operation? Do they have to fill hoppers, part feeders or dispensing equipment?

Does the machinery stop operating because of machine faults? If so, what must be done to get the machinery operating again? Is there a human-machine interface (HMI) or display to help diagnose problems?

Consider taking notes while observing the machinery in operation. It is best to learn about the machinery and process while it is in operation, before something goes wrong and repair is required. The time to learn machinery is before a problem emerges.

Use senses to detect a problem

Visual signs of problems may be the easiest to find. Leaking fluids, metal shavings or black sooty powder are signs that machinery needs attention. Burnt and blackened components on circuit boards, warped or melted plastic and wear points on tooling and components can all be indicators. Look for anything that out of the “ordinary” or normal condition.

Machines can be video recorded to analyze operations; motion can be slowed to examine functionality in detail. Cameras also can be strategically placed around and even inside a machine for full time monitoring.

Thermal imaging also can be used to identify hot spots in an electrical cabinet, such as determining if connections or terminations are tight. Follow proper safety procedures when working with exposed electrical devices.

Listen to the sounds a machine makes as it operates. Sounds will change as problems appear, grinding or high-pitched squealing can be a sign that something is wearing or misaligned. Knowing the sound of your machinery when it is operating correctly can help identify problems when the sound changes. You could even record the sound so that you have a reference.

The sense of touch can be helpful in identifying warmth or vibration in equipment. Of course, this can be dangerous. If you suspect something is hot, approach it with your hand slowly, you can often sense heat without actually touching the object.

Many components vibrate even when they are operating correctly, knowing how much vibration is normal can be helpful for comparison. Sensors also can be used to monitor vibration or temperature; trends are useful to identify wear.

Smells can also indicate problems. Burning or melting rubber has a distinctive smell, as does hot lubricants or metal. Again, knowing the smells of machinery when it is operating correctly can be helpful for comparison.

Read the provided documentation

Ideally, equipment will have documentation accompanying it. Usually, it will be provided by the manufacturer and may even have a troubleshooting guide.

Recommended maintenance instructions will often help identify areas where problems can occur in the machinery. Lubrication, belt tensioning and replacement of worn tooling are examples of maintenance procedures supposed to be performed on a periodic basis.

Where recommended maintenance is to be performed also can help to identify problem spots on a machine. Since these are items that move or wear out, they also are areas where things can go wrong.

Replacement parts lists or a bill of materials also often list components such as sensors or other electrical components that may need to be replaced.

Exploded parts views can show the internal workings of elements of machinery. Sometimes these are included as part of the maintenance or component replacement instructions. These can be very educational in the examination of machine operation and help to identify potential failure points.

Electrical schematics show the detailed wiring of the machine or system, and often have a bill of material attached. The first few pages can be scanned to identify the major sections and components of the machine. Power distribution identifies larger subassemblies by name; variable frequency drives, robots, controllers, servos and auxiliary equipment can be seen as an overview.

Piping and instrumentation diagrams (P&IDs) also may be a part of available documentation, especially if it’s a process control application.

The PLC program can be examined as a guide to the system layout. Looking at the hardware configuration can show where communication-based remote input/output (I/O) nodes are located, and programs are often written with subroutines that correspond to different operational parts of the machine or system.

Seven troubleshooting fundamentals to consider

What is meant by troubleshooting? Troubleshooting is a systematic approach to problem solving that is used to find and correct issues with complex machines, electronics and software. It is often applied to machinery that has stopped working or is not operating as expected.

Determining how a machine is supposed to operate involves the observation and documentation-reading skills mentioned earlier but may also require talking to the person who spends the most time with the machine, the operator.

There are some important concepts to understand when trying to determine the cause of a problem. One important relationship is between correlation and causation, just because one event occurred after another event does not mean the first event caused the second. It could be a coincidence, or some other unseen event could have caused both. This is often stated as “correlation does not imply causation.”

There are three elements to solving problems.

1. Identify and document what the problem is.

2. Determine why it happened.

3. Determine a solution to prevent it from happening again.

Consider these seven techniques when it comes to troubleshooting systems and equipment:

1. Start with the simple

It is often useful to examine the simplest or most obvious explanations first. People often complain when technical support people or manuals first ask the user if the receptacle has power or if the device is plugged in. It can save a lot of time to check for simple explanations like power, circuit breakers or fuses before moving on to more difficult or less obvious causes.

Look the machinery over and see if there are lights showing or obvious jammed parts. Keep in mind the jam could have been caused by something else and there may be other sources of power that power the devices. Ask what has changed since the last time the machine worked properly?

2. Begin from a known good state

Starting a machine from a known good state, such as a home position with no parts, can help identify problems. A well-known example of this is rebooting a computer; this clears the current RAM memory of the processor and re-starts the operating system.

This is especially important with assembly machinery. Ensuring a good part runs all the way through all of the steps of the process makes sure the tooling, sensors and actuators are all positioned correctly and their speeds are correct.

3. Substitute components

Sometimes called “shotgunning,” troubleshooters could check each component one by one, substituting known good components for each suspected bad part. This is certainly not the most efficient way of solving the problem, and there is a risk the thing that originally caused the failure also can make the new good component fail. While it can be effective, substitution should be a last resort because you could end up destroying multiple expensive components.

4. Checklists and flowcharts

Creating a checklist, flowchart or table of procedures in advance can help. This creates an organized sequence that technicians or operators can follow when troubleshooting equipment. These lists can be kept on a computer or even on the machine HMI so they can be easily accessed.

Keeping records of previous problems and their solutions can help in creating these lists. The lists should be updated as new events occur.

Dependencies can be used to create these lists or charts also, often a component depends on another system to operate correctly. Identifying these can help when creating flowcharts (Figure 1).

Figure 1: Example of a troubleshooting flowchart. Courtesy: Automation LLC

5. Reproduce symptoms

If the error can be recreated, it becomes easier to isolate and resolve its cause. If the problem can be reproduced consistently, this works well. Many problems can be intermittent in nature though, making them difficult to reproduce. Identifying environmental causes such as heat or humidity at the time of the occurrence can be useful here.

6. Split the system

A technique called “half-splitting” can be helpful in limiting the choices for bad components. Dividing a series of connections or nodes in half can identify where a voltage or communication signal is lost. It also works well on systems that are made of a series of sequential functions.

Figure 2: Example of a half-split. Courtesy: Automation LLC

7. Root cause analysis

Root cause analysis (RCA) is the process of discovering the origin of problems in a system. The purpose is to identify solutions for the base problem, rather than just treating symptoms and putting out fires. To this end there are templates and techniques that can be used for this.

There are many subjects that need to be learned to be a well-rounded machine troubleshooter, technicians usually attend vocational schools or colleges to learn the basics of machinery and electricity. There also is a vast amount of information available online or in catalogs that can help supplement this knowledge.

Observing machinery operating is one of the best ways to understand the normal functions of a machine. Locating all manuals and component documentation prior to machinery breakdown can save a lot of time also.

There are several well-known troubleshooting techniques that are important to know when approaching a troubleshooting problem. These include various diagrams and checklists that it may be useful to print out ahead of time.

Learning from the experience of others who have worked with the same or similar machinery is one of the most important ways to get a head start on your troubleshooting journey.

– This has been edited from the “Maintenance and Troubleshooting in Industrial Automation” book by Frank Lamb, the founder and owner of Automation Consulting LLC and a member of the Control Engineering editorial advisory board.

MORE ANSWERS

Keywords: industrial automation, troubleshooting

CONSIDER THIS

What skills and fundamentals do you use to diagnose a troubleshooting problem?

Do you have experience and expertise with the topics mentioned in this content? You should consider contributing to our CFE Media editorial team and getting the recognition you and your company deserve. Click here to start this process.