Twenty four years ago, in my native country of Canada, I worked for an original equipment manufacturer (OEM) that made conveyors, palletizers and depalletizers for the food and beverage industry. Having graduated from college in 1988, I was still in the early years of my career, and this was my first job where my primary focus was programming and commissioning packaging equipment.
My first project with this OEM was to re-write the software application for a two-car shuttle-car palletizer that unitized bags of dog food. If that wasn’t daunting, the president of our company, upon meeting me a week after I was hired, casually mentioned that the last time a rewrite was attempted, it just about broke the company.
My only previous experience authoring a PLC program was on a product once offered by Allen-Bradley called a modular automation controller (MAC). It was basically a printed circuit board with spaces on the circuit board where one could plug in input and output modules.
The project was a conveying line that brought field product in, washed it and conveyed it through various processes with the result being a bottle of pickled onions. To go from this to rewriting the controls algorithm for a major packaging machinery manufacturer was nothing short of terrifying for a young controls engineer.
Perseverance is essential in a controls engineer, and, as you might have guessed, I did manage to rewrite that program, the company did not go broke, and I have made a career out of this fascinating line of work.
The controls in packaging equipment of that vintage came in the form of an Allen-Bradley PLC-5 with 1336 variable frequency drives. For those who may not remember, the PLC-5 was a huge boat anchor of a processor utilizing the 1771 I/O system.
The 1332/3/5/6 drives were equally large by today’s standards and were often mounted on the outside of the control cabinet because they generated so much heat that mounting them inside the enclosure would cause issues for the other components in the panel. Commands and status for the variable frequency drive were entirely digital, via terminals on the control board of the VFD.
As one can imagine, being restricted to digital signals to and from the variable frequency drive limited the capabilities of the control system. Hardware providers of the day expanded the capabilities of the VFD by adding analog input to the drive command structure; and, since parameters in the drive had to be entered manually prior to operation, they added terminals to give more options to the control of the drive.
Typical selections via the terminal interface included multiple speed selections by using binary coded input and a second set of accel/decel commands that could be engaged while the drive was in motion.
The flexibility of variable frequency drives was greatly enhanced by the introduction of machine-level network protocols. With the advent of Modbus, Profibus, ControlNet, DeviceNet, SYSMAC and others, the wiring of drives was simplified to just the enable/disable input, and the rest of the control could be accomplished via the network connection.
These early connections basically duplicated the digital interface in a memory block transfer format. PLC programming in those days was memory-based, unlike the tag-based programming of today. A typical drive interface would have four integers, two for input and two for output. The input would be in the form of a status register and feedback register, while the output would be a command register and an output frequency.
Remembering back to my first big programming project, we really did a lot with a system that would be considered archaic by today’s standards. Our primary mover on a multiple car palletizer was a shuttle car that moved back and forth to stop in multiple positions underneath a stationary palletizing tower.
All major motions were accomplished by a variable frequency drive and creative use of flags and proximity sensors to tell the control system where the various axes of motion were. To move shuttle car positions, we would ramp up to a running speed, sense the presence of a flag on the shuttle at the desired stop position, ramp down to a slower speed and stop when the stop sensor on the shuttle was triggered by a stop flag at the desired stop position.
This use of a variable frequency drive really hasn’t changed all that much in the ensuing years. I was pleased to meet up with my former colleagues, who like to say you can’t spend your whole career with the same company, when my current employer bought a shuttle car palletizer from that same company that I worked for so many years ago.
There is an expression: If it isn’t broke, don’t fix it. That same control approach from 24 years ago, as it turns out, is still a great way to control a shuttle car today.
As many readers know, I like to take a look at where we came from in order to have a better understanding of how the technology of today has evolved. While the first half of this piece seems to suggest that the use of a variable frequency drive hasn’t changed much in the past 25 years, nothing could be further from the truth.
The transition to machine-level networks was really just the start of the journey. Since then, hardware manufacturers enhanced the PLC to VFD interface by including additional blocks of memory into the data exchange. The advent of tag-based PLCs has opened up the potential interface to seemingly limitless possibilities.
Now one might ask why we need an expanded interface. While the control of the variable frequency drive might seem unchanged for many years, there is much more going on in the background.
A VFD, even in those earlier days of automation, has many parameters that can be manipulated to more finely tune the performance of a drive-motor combination. Motor technology has continued to advance over the years and the driving technology has matched pace with it.
Today’s VFD might have 400 or more parameters that can be manipulated. With the ability to access parameters via the machine network, today’s VFDs are usually configured using that network interface, using the programming software application itself in many cases.
As with all other sectors of automation, more performance in a smaller package, combined with the constant progress of technology, has moved servo and stepper motors into the forefront of automation projects. One of the primary motivations for moving in the direction of servos was the open-looped nature of VFD and stepper control.
As the control method described earlier suggests, much can be done with just sensors and some control over speed and acceleration, but, without feedback, open-loop control can only do so much. With the presence of the VFD on the machine-level network, the ability to use a linear or rotary encoder to provide positional feedback has added a way to partially close the loop. While not quite the same thing as true closed-loop control, being able to use an encoder instead of digital sensors makes for a more finite way to use a VFD to accurately provide motion control.
In the past couple of years, a new method of control has been introduced for a variable frequency drive. Some manufacturers have closed the gap between a VFD and a servo drive by producing a variable frequency drive that accepts encoder feedback directly into the drive. Now, motion is controlled at the drive level, instead of using the PLC algorithm to approximate closed-loop control.
These drives are less costly than a servo drive solution but are actually controlled using the same commands as a servo. Not to be outdone, some manufacturers of servo drives offer the option to additionally control a VFD from that same servo drive.
The benefit of these latest developments is purely in the hands of the designer and programmer: accurate control of motor-driven components for less money and with a complete tool chest of parameters to clearly define the behavior of the motor application.
Further sweetening the pot, today’s VFDs also make use of integrated safety features that provide a safe torque off (STO) function. With STO, the drive is prevented from producing any torque-generating energy.
When this feature—a physically wired, dual-channel connection—is used in conjunction with a safety relay, there is a means by which to guarantee that a motor is disabled when a safety device is tripped.
Finally, with all things related to automation technology, today’s VFD is a much smaller package that its ancestor. A 10-hp drive in that older technology would occupy a space that was 10 inches wide by 18 inches tall by 9 inches deep. The equivalent version today would be 5 inches wide by 10 inches high by 8 inches deep. One half the footprint and generating far less temperature during operation means the same control system can be in a much smaller electrical enclosure.
The great part of these advances in technology is the designer doesn’t necessarily have to invest in a servo drive to get servo-like behavior. For the maintenance team, the new machine doesn’t need someone with servo knowledge to troubleshoot a machine failure. Even if, ultimately, a servo is not needed, the ease with which today’s VFD can be commissioned and operational makes it an excellent choice for control of a motor. What’s driving your decision?
