Whether you are designing a new machine control system or retro-fitting an older machine into the 21st century, there is a looming question:
"What's coming next?"
We never know what future upgrades may be necessary.
Perhaps your current model of controller will become obsolete too quickly and must be changed.
What if we must add a second identical machine beside the existing one?
To all of these questions - the answer is the same:
"Design your system with the future in mind."
Sure, we can't predict the future with certainty.
But we can absolutely make researched assumptions that will prevent us from being totally caught off guard when inevitable changes arrive.
There are three main areas of consideration we must make when designing a control system for a machine.
Obviously, no two design scenarios are alike, but if we have considered opportunities for future expansion or replacement, we can at least have some peace of mind when the machine is commissioned into service.
1. Adding more Input/Output Devices
Expansion of I/O is perhaps the simplest upgrade for a control system. The best part is that with isolated I/O terminals (like in a PLC), you can add more devices without affecting the existing ones.
But there is a caveat - the controller must be capable of adding more terminals.
In control system design, there is a tendency to opt for the least expensive controller which will perform the task.
Some PLCs are considered 'Micro' or 'Pico' form factors, which usually entails a limited number of I/O terminals and small size, but usually includes free programming software.
This certainly sounds great, but you must be 100% sure that this will be a sufficient number of terminals.
There are two solutions in this case -
Most micro PLC models can be selected with a variable number of terminals. The ones with more terminals will be more expensive, but the cost may be acceptable if greater production can be achieved in the future. The picture above shows two nearly identical PLCs, except for the number of I/O points.
It might actually be useful to consider a 'modular' PLC instead. They accomplish the same task, but with removable I/O modules. In almost every case, you can add more Inputs and Outputs as needed, and just attach them to the side right next to the others. The picture below shows a module with the connector at the bottom - ready for an additional module if needed.
Regardless of the two options, here is a factor to consider: Are you comfortable with allowing only the bare minimum number of connections? If you want to upgrade in the future, you might want to consider a slightly bigger PLC system.
2. Network Capabilities
For some industries, a stand-alone machine is not even an option - everything transfers data on Ethernet or similar types of networks.
But for smaller machines, it may not be needed.
Once again, it's going to come down to cost - controllers without network ports will almost always be cheaper.
But in today's world, can you expect to succeed when your machines cannot communicate with each other? Maybe. Again, no two scenarios are the same...
The cost for controllers with and without network ports is really not as extreme as you might think.
If I were to make a recommendation, I would say "Never install a PLC in a modern machine that does not have an Ethernet/IP communication ability".
This VFD below does not have an Ethernet port, so we would need to use an adapter if we planned to use it in a modern network.
Will you always need the network ability? No.
Is there a chance you will need it someday? Yes.
I would go a step further to say it's not just a chance, it's almost a sure bet that you can improve your system by allowing multiple machines to share data.
This might be as simple as a PLC talking to VFD nearby and controlling the speed of a motor.
It might be a little more advanced, such as counting production levels and entering the data on a building-wide intranet to track production data.
No matter your industry, it's hard to make an argument against including networking capabilities.
3. Modular Connections
Not every PLC is considered a 'modular' form factor. But the terminal connections can always be designed in a way that makes them easy to change.
Once again, we may need to change a controller, or replace a sensor, or perhaps add some new devices.
If these options have not been considered, we might just find that one day we need to spend hours unscrewing every wire, labeling it, and pushing it aside to make room for the new controller to be installed, then wires must all be reconnected in the same manner.
Instead, we turn to devices like removable terminal blocks (RTBs) or quick-connect fittings that allow quick changes.
As with everything before, this will add some extra expense upfront, and it might take a little longer, but it will save a LOT of time in service later.
Most PLCs have standard types of terminal blocks, but for some of the micro-size devices, it might be useful to use DIN rail terminal blocks nearby or use plastic wire end connectors if terminal blocks are not possible.
This turns every group of devices into its own 'module' which will never have a big bundle of individual wires tucked off to the side of the cabinet.
This allows a new controller to be per-wired with all the appropriate connections, then when the timing is right, it's easy to replace.
Turn off the power, unplug all the wires and terminal blocks, plus them into the new PLC, and turn the power back on.
It still takes a little time, but nothing compared to individual wire terminals.
Summary
There are plenty of modifications that can help to future-proof a control system, but these three main categories of consideration can help to select the right design.
I/O Device Expansion, Networking Options, and Modular Connections can create a system that will stand the test of time and relieve stress for you or anyone else who needs to maintain the system in the future.