ABSTRACT

In this article, we are going to discuss a case study in which we conducted an investigation to track down an elusive cratering problem involving the inside coating (IC) of aluminum cans, and discovered a surprising set of circumstances that led to a successful resolution of the problem.

In the autumn of 1994, a can manufacturing plant had a significant problem with cratering of the inside coating of aluminum cans.  We will review the analytical work that was performed, the identification of the cause, and the corrective action taken at the plant to prevent this defect from occurring in the future.

For this discussion, we are using the following definition for metal container cratering: “a defect in the interior coating film, where the coating is dewetted in a circular pattern, and a pit is present in the center of the crater.”  In many of the defect cans inspected, there were two pits (double pit) in the center of the crater.

In the two-piece can industry, the presence or absence of craters is detected via a quality control, the metal exposure “quick test” after the inside spray coating has been applied and cured.

Early in the investigation, we observed that there was consistency in the location of the craters; that is, they were all on the interior side walls between one and two inches from the bottom dome of the can.  Also, we did not see any craters in the bottom dome area of the can, or where there was a lack of metal deformation.  There was also an absence of craters on the upper two-thirds of the side walls.

BACKGROUND

This case study occurred at a Ball Packaging International North American multi-line aluminum can plant which supplies, mainly, beverage customers.  Each production line had: (1) its own press to produce cups from aluminum coil stock, (2) bodymaker machinery to produce the drawn and ironed cans from the cups, and (3) a can washer system to remove residual oils and clean the aluminum for subsequent ink and coating application.  In this plant, a new Quaker Chemical bodymaker lubricant/coolant was installed earlier in the year, supplying seven machines which were running for approximately nine months when the problem occurred.  Up until this point, there were no previous unexplainable occurrences with high metal exposure readings.

During the cratering problem, the plant was able to identify that the source of the problem was the bodymaker lubricant/coolant system.  They were able to take cans which were made from the Line 1 coolant system and divert those cans into the Line 2 washer, decorator (ink label application), and through the inside spray coating and curing ovens.  The defect followed the cans from the Line 1 bodymakers through Line 2 interior coating ovens.  The next experiment was to take cups that were feeding Line 1 and to divert them into the Line 2 bodymakers.  The defect did not follow the cups into the system.

Another experiment, run by the plant to help eliminate the washer system as a potential cause, was to take some of the cans from Line 1 bodymakers, wash the cans only with an organic solvent, IC spray and cure the coating, and again test for metal exposure.  The defect also occurred on the solvent washed cans.

In chronological order, we present a brief history of the cratering defect occurrences:

• Day 1 – Sporadic high metal exposure readings, primarily in the afternoon.

 

• Day 2 – Low readings in the morning.  Higher readings in the afternoon, which continued into the night.  Numerous changes were made in an attempt to eliminate this defect.

   

• Day 3 – Persistently high readings throughout the day; this was when the plant determined conclusively that the lubricant/coolant system was the source of the problem.  The decision was made to dump the lubricant/coolant system, clean the tanks, pumps and piping, and recharge with fresh material.

   

• Day 4 – Lubricant/coolant system was cleaned out with a caustic cleaner and recharged with fresh material.  One maintenance item that was found and repaired was a leaking heat exchanger gasket on the cooling tower water system.  The bodymaker lubricant/coolant was circulated through this heat exchanger.

   

• Day 5 – The plant restarted on the fresh lubricant/coolant and the cratering problem was eliminated.  All the metal exposure readings returned to normal very low levels.

Our initial theories were that some form of contamination was responsible for the cratering defect, either in the lubricant/coolant system or in the concentrated product bulk storage tank.  In meetings with plant officials after the problem was solved, we formulated a three-pronged approach to identify the key assignable causes.

• The first phase of the investigation was to take a careful look at the defect cans to analyze what chemical or physical changes had occurred on the surfaces.

   

• The second phase was to take a detailed look at the lubricant/coolant samples obtained during the height of the problem to see whether we could detect any unusual changes which may have occurred.

   

• The third phase in our investigation was done by the plant where they were going to study and analyze all their chemical processes to see whether any changes had taken place.