This architecture had a number of problems. First, it was hardwired, resulting in large volumes of wire with intermediate connections, making installation and maintenance a time-consuming nuisance.
In addition, the system had evolved significantly over time, making it unwieldy. "If we had to add a brake or gate, a technician would just wire it in with a photo eye," Rock explains. "But the electrical schemes were not well documented. Soon, it was impossible to know how loaded individual circuits were. All in all, it was a sprawling, hardwired mess."
As documentation slipped, so did other practices. Many of the photo eyes were connected in the conveyor trough with wire nuts or similar hardware -- not an ideal, textbook routine. And the photo eyes themselves were problematic. Because they were of the "timed" variety, each eye had a dial that technicians could set to manipulate the time between it seeing a carton and sending a message. Essentially, all the logic was contained in the individual eyes. That made adjustments easy , sometimes too easy. Over time, the photo eye settings were neither consistent nor well documented.
All of these factors lead to an increasingly unstable system. Downtime was on the rise. And when equipment did malfunction, troubleshooting was anything but trouble-free. A maintenance person would head to the plant floor, locate the affected area, pinpoint the problem and deduce the cause. "At this point, we had to hope that the problem wasn't in the 120 volt circuit, because the wiring was so intertwined," Rock recalls. One failed sensor could exhibit symptoms in multiple places."
Because of the hardwired snarl, the time it took to get the conveyor back in working order varied wildly. A technician could locate and fix the problem in five minutes. Or it could take well over an hour. There was no consistency and Lipton confronted problems with fingers crossed.
"There never was a breaking point, just an accumulation of recurring headaches," Rock says. "We knew it shouldn't take more than an hour to find and solve problems. And this was occurring on a weekly basis."
With the obstacles defined, Lipton set several goals for the project. In brief, the company wanted to simplify the wiring scheme, centralize the logic and control for the field devices, and, in the process, reduce downtime.
Before the upgrade was underway, Lipton knew it wanted to build the new conveyor system around a programmable controller and digital communications. So Rock set his sights on technologies that were modern, but proven.
"We didn't want hardware with single-digit serial numbers installed in the plant." Rock adds. "At the same time, legacy equipment was out of the question." After considerable research, Lipton found the necessary balance with an Allen-Bradley ControlLogix programmable controller and open DeviceNet networks.
Today, individual SMC solenoid blocks contain 12 to 20 photo eyes (inputs) and eight to 16 outputs, each controlling conveyors supporting two machines. The solenoid blocks feature several input adapters that allow Lipton to connect the photo eyes using simple twist-on cords. When a block receives a signal from one of the attached eyes, it turns on a stream of air that drives a nearby cylinder , connected to the block with poly-flow tubing. In turn, the cylinder pushes a squeeze brake (similar to an automobile brake) that holds and/or directs the boxes from the carton former.
Lipton ties the solenoid blocks to the ControlLogix controller with a DeviceNet network. Because of its open, digital design, DeviceNet can link more than 60 blocks to the controller. And the controller can support multiple DeviceNet connections. Along with the twist cords and poly tubing, DeviceNet has eliminated traditional wiring from this section of the facility. There are power supply cabinets and one cabinet for the controller. That's it.