Easco Aluminum Corporation manufactures polyvinyl chloride extrusions which are used by its customers in fabricating prime replacement vinyl windows. In June, 1983, General Aluminum Products, Inc., contracted with Easco to buy Easco extrusions for use in vinyl windows that it planned to sell to building contractors. In August, 1983, General Aluminum began receiving complaints that the vinyl had bowed, allowing leaking in some of the windows sold by General Aluminum. When General Aluminum blamed Easco for these defects and refused to pay their bin, Easco sued and General Aluminum counter-claimed for breach of warranty. Following a four day bench trial, it was found that Easco had breached both express and implied warranties of merchantability by selling General Aluminum defective vinyl extrusion, and that the defective extrusion was the proximate cause of damages to General Aluminum. The district court awarded General Aluminum, $25,010.54.

Easco now appeals the decision, claiming that any defects in its vinyl extrusions were not the proximate cause of General Aluminum's damages. We affirm the district court.

The only issue on appeal is whether the evidence supports the District Court's conclusion that the "bowing of the Easco vinyl which was sold to General Aluminum occurred because of heat deformation attributable to poor manufacturing techniques employed by Easco." The standard for review of a district court's fact finding is well-established. See Davis v. Manfield Metropolitan Housing Authority, 751 F.2d 180, 1823 (6th Cir.1984). As the Supreme Court recently stated in reviewing a Rule 52(a) case Anderson v. Bessemer City, 470 U.S. 564, 573 (1985):

This standard plainly does not entitle a reviewing court to reverse the finding of the trier of fact simply because it is convinced that it would have decided the case differently. The reviewing court oversteps the bounds of its duty under Rule 52(a) if it undertakes to duplicate the role of the lower court. "In applying the clearly erroneous standard to the findings of a district court sitting without a jury, appellate courts must constantly have in mind that their function is not to decide factual issues de novo." Zenith Radio Corp. v. Hazeltine Research, Inc. 395 U.S. 100, 123 (1969).

Mindful of Rule 52(a)'s mandate limiting review of factual findings, we are unable to say that the district court's conclusion that heat deformation caused bowing of the extrusions in the windows was "clearly erroneous". Indeed, while the causation issue in this case is far from clear-cut, the record presents ample evidence supporting the district court's conclusion.

Two experts testified that heat deformation could have caused the bowing in the vinyl extrusions. Mr. Wallace A. Parker, a technical service engineer for rigid profile compounds at Georgia Gulf Corporation, performed a heat shrinkage test on samples of Easco's vinyl extrusion. The test showed "considerable increased bowing despite being such a short piece." Mr. Parker explained that a faulty extrusion process can put internal strains in the vinyl that are not immediately apparent. But when the PVC extrusion is "relaxed through heat, PVC has a memory so it will return to its original form. So, if you put a strain in, that strain will be removed out in the sunlight in a window frame, for instance, and it will return to its extruded condition."

Mr. Parker also refuted one of Easco's claims, that is, that it would be impossible for the extrusion to bow at temperatures less than 160 degrees: "The glass transition temperature from the PVC is 160 degrees F. Now, at that temperature, PVC tends to soften and release its stresses. It could also be done at 140 degrees, but it would take a longer time." Although the record does not make clear the minimum temperature that extrusions with inner stresses will bow, there is substantial evidence that it is well below the 160 degrees claimed by Easco.

The testimony of Mr. Charles Bastnagel, a former sales manager of the vinyl division of Easco, also supports the district court's conclusion. Mr. Bastnagel investigated the complaints of bowed vinyl extrusions. He testified that he saw examples of windows in the Midwest where he was sure that the bowing could have been caused by heat deformation. Mr. Bastnagel's testimony also refuted Easco's alternate theory of what caused the defects in the windows. Easco alleged that because no other customers complained about the problem, the defect must have resulted from faulty window fabrication by General Aluminum. Yet, Mr. Bastnagel named six other Easco customers that complained of similar problems with Easco extrusions. While it is true that Easco presented conflicting testimony, Rule 52(a) compels this Court to defer to the District Court's conclusions as to the credibility of the witnesses. A factual finding will be found clearly erroneous only when the reviewing court is firmly and definitely convinced that a mistake has been made. See Easter v. Jeep Corp., 750 F.2d 520, 52223 (6th Cir.1984).

General Aluminum also presented evidence that Easco's extrusion process was not up to industry standards. Mr. David Bartrum, a technical expert who is now president of a plastics company, testified that he observed during a visit to the Easco plant in 1983 that Easco was running the extruders at a "very low rate," and that the settings on the machines were "colder than the prescribed temperatures." And, Mr. Bartrum testified, in his total experience, he "had never seen a successful vinyl, rigid vinyl extrusion at the temperatures that Easco was producing."

Given the above evidence, we are unable to find the judgment of the district court clearly erroneous. Therefore, the judgment of the district court is affirmed.