top of page
  • Writer's pictureDavid Peterson

What Makes 'Network Problems' so Terrifying?

Of all the problems that can haunt an industrial technician, network problems can top the list of most intimidating.

The main reason for this is actually pretty simple, so we'll answer that question first.

--- There are very few observable symptoms that point to the cause! ---

Think about that - you devote your mind and learning to industrial troubleshooting, and the process is consistent:

- Look for physical symptoms (lights, burn marks, smells, sounds, vibrations, etc)

- Build a mental list of most likely causes based on those symptoms

- Access the possible solutions and fix the problem!

Now, imagine if that first step is nearly impossible...

Not only do you have zero information to start with, but no real tools or knowledge enough to explore possible solutions.

So then what are you supposed to do? Give up? Never.

When something is difficult to observe, it doesn't mean you can't go to the next step, it simply means that you need to let your brain start the first step - perhaps your five senses aren't enough.

A little bit of background in networking concepts can help to rule out a few possible problems. Hopefully, this will help locate the actual cause just a little bit faster.

Learn the Network Structure

First, it helps to get a bit of an idea of how 'networks' are actually built.

Most modern networks are considered 'multi-drop' which means all the devices connect to the same set of wires interconnecting everything.

This is really handy because devices can be added and removed to any port where those wires are exposed, without affecting anything else.

Alternative systems are called 'serial' or the common term 'daisy-chained' where one device must connect to the next, which connects to the next, etc. If any device must be added in the middle, it will disconnect everything downstream.

Now, those 'exposed wires' are never just a bundle of copper wires in a loose junction box, but rather a specific device that connects multiple devices.

The simplest is the 'unmanaged network switch' with as few as 5, or 8, or more ethernet ports. Usually, it's a small blue or black rectangular unit that needs a 12-volt power plugin to work.

Every network consists of AT LEAST one device connected to one other device, and usually a lot more than that. The switches, along with routers, ensure that all the devices are supplied with a proper connection.

Troubleshooting a switch may be as simple as watching the green and amber LEDs blinking near the cables. Any port that has an active device should be blinking - if not, check the cable and device connected to it.

Addressing the Devices

There are two 'addresses' that are useful to understand in a typical network.

First, no matter the network, every device must have a unique ID address. In an Ethernet network, this is called an IP address. Since every device is connected to the same network, each one must be unique. If they were motors, you need to make sure you ONLY turn on the right motors at the right time!

Devicenet, Modbus, Profinet, and other industrial protocols also have unique addresses, but the IP address is the most likely one to find in modern equipment.

Every technician should have access to two things:

  1. The ability to read the IP address of any device. Sometimes, it's displayed on a screen on the front. Sometimes, you can access parameters that show the address. Sometimes you access it from the software. Regardless, it helps to know how to find the address.

  2. A map of what devices should have what address. Perhaps an IT staff set up the network, and the danger is that if just one number changes inside a machine, it can affect not just that machine, but others. And worse, the effect may not happen for days or weeks until one new device is turned on...

The WHOLE purpose of troubleshooting is knowing what should be correct, and then locating where the system is not correct.

Having access to the addresses ensures that you can make sure every device is identified as it should be.

The second kind of address is called a 'subnet mask'. Many times, a large Wide Area Network is divided into smaller Sub-networks, or Local Area Networks.

A subnet address allows devices to communicate with other addresses on only a local subnet, and not share information with the wider network. This allows a large facility to use several smaller networks for various areas in the facility. Each local network may rely on its own addressing scheme and never worry about address conflicts with any other subnetwork.

In effect, a 'mask' selects which parts of the IP address MUST be identical for two devices on a local network to communicate.

Once again - it's important to understand what this subnet mask SHOULD be, and be able to read the current value in the device.

Even if you don't understand the concept, you should still be aware of what numbers are correct.

If one number in the subnet mask changes, it may be unable to talk to a device next to it, even if they are physically connected to the same network switch.

Status Indicators

Sometimes we can find some physical indicators about network operation.

Almost every network port has LEDs that show a network status.

A common color scheme is to allow a green LED to either blink or solidly illuminate if the link is broken, or connected respectively.

A red or orange light may mean failures or faults, but simply seeing a green light often is not enough to ensure connection.

Check datasheets to be sure, but sometimes these visual indicators can at least point out the suspect device.

If a link is not established where one existed before, the cause is most likely a bad cable or a change of network values in the device.

It could also be caused by a device which has power, but does not have a proper controller connected to it.

Learn Concepts, not Details

Although networking is good to know, every one is different.

It never helps to memorize how a network is set up since you probably won't solve that exact same problem ever again.

Rather, devote your time to learning the idea of why networking concepts are used, and not exactly how it's accomplished.

Instruction manuals can always show you HOW to adjust values, but understanding WHY they exist in the first place can be a huge benefit to you and your company.

58 views0 comments

Recent Posts

See All


bottom of page