If you’ve been in controls long enough, it’s happened to you: a project that seemed to be going well suddenly hits a wall that was unforeseen and whose effects threaten to derail it irreversibly. My Waterloo came with a project to automate a press that made abrasive bars for honing pipe welds, compressing sand-like material into rectangular “sticks” that were then baked to produce the final product.
The automation portion was reasonably straightforward, applying mechanical and controls equipment that would feed the raw material into the die cavities, compress it and automatically offload the pressed sticks to small pallets. The usual risks were identified and hedged: subcontractor performance issues, equipment delays, personnel scheduling, etc. What doomed the project was my failure to heed a small, offhand comment made by the client even before the project began about how difficult the raw material was to work with, and how its properties changed from day-to-day and even hour-by-hour.
Previously, operators ran the press manually feeding the die cavities by hand using a small shoe holding the raw material. If the raw material was wet or fed poorly into the dies, the operator could agitate the shoe more vigorously. If it fed smoothly, a quicker motion was used.
Our machine functioned flawlessly in dry trials. Hopper feeders fed. Feed shoes indexed. Presses pressed. Manipulators picked and placed the “ghost” sticks perfectly. Conveyors conveyed. We were ready for our dress rehearsal.
At least we thought we were ready. The raw material ignored the law of gravity. It obstinately refused to fall into our die cavities as commanded. Voids were the rule, not the exception, and irregular sticks crumbled when picked up. We made adjustments and, just as we managed to get things working, the properties of the raw material changed as it dried and we were thrown back to square one. All the while the client’s remark about how stubborn the material could be haunted us. In retrospect, it was clear we should have run offline trials with the range of materials we would encounter in actual operation. We were too trusting in gravity.
Things eventually turned around and we ended the project on a high note. We managed to make the system work throughout the range of materials and properties presented to us, albeit after several redesigns of the feed system, multiple delays, and a tired project team. It was a hard lesson, but one from which we all learned and have used since to prevent a replay.
The moral of the story: Listen carefully to the process owners in the early stages of a project. They know their processes and have seen the full gamut of problems with equipment, materials, and processes. Problems can be overcome through the intelligent application of engineering, but only if one fully comprehends them, plans for them, and avoids going too far down the wrong path.
This post was written by Brad Ems. Brad is a project manager at MAVERICK Technologies, a leading system integrator providing industrial automation, operational support and control systems engineering services in the manufacturing and process industries.