Anderson v. General Grinding Wheel Corp. , 74 Ill. App. 3d 270 ( 1979 )

74 Ill. App.3d 270 (1979) 393 N.E.2d 9

LILLIAN ANDERSON, Adm'r of the Estate of John Anderson, Deceased, Plaintiff-Appellant,
v.
GENERAL GRINDING WHEEL CORP., a/k/a SGL Industries, Inc., Defendant-Appellee.

No. 76-505.

Illinois Appellate Court — First District (3rd Division).

Opinion filed May 23, 1979. Modified on denial of rehearing August 1, 1979.

Serpico, Novelle, Dvorak, Navigato & Hett, Ltd., of Chicago, for appellant.

Isham, Lincoln & Beale, of Chicago, for appellee.

Reversed and remanded.

Miss JUSTICE McGILLICUDDY delivered the opinion of the court:

Lillian Anderson, administrator of the estate of her husband, John Anderson, instituted a product liability action to recover damages resulting from the death of her husband. She named the General Grinding Wheel Company (now known as SGL Abrasives, a division of SGL Industries, Inc.) as defendant. At the close of the plaintiff's case, the defendant moved for a directed verdict. The court reserved ruling on the motion until the close of all the evidence. The defendant renewed its motion when all evidence had been introduced, but the court again reserved ruling on the motion pending the deliberation of the jury. When the jury was unable to reach a verdict, the court granted the defendant's motion for a directed verdict. The plaintiff now appeals that judgment, raising the sole issue of whether the trial court erred in granting the defendant a directed verdict.

The principal facts of the incident which resulted in John Anderson's death are largely uncontested. He was killed, on August 21, 1972, when a 30" x 2 1/2" x 12" reinforced resinoid snag grinding wheel ruptured. At the time of the accident, Anderson was employed as a snag grinder by the Commercial Stamping and Forging Company and was using the wheel to grind the flash off of a forging called a henschen wheel spindle.

The grinding wheel which disintegrated was 30 inches in diameter and 2 1/2 inches thick. At the center of the wheel was a hole, some 12" in diameter. Encircling this hole were two steel rings. These rings were to provide additional support for the wheel once it had become worn. The wheel primarily consisted of abrasive particles, bond material and filler. The purpose of the bond material was to hold the abrasive particles together; in this particular wheel, the primary bonding material was resin.

Since the wheel in question was a reinforced grinding wheel, embedded within the abrasive material were two pieces of fiberglass mesh. The function of the fiberglass was to hold the wheel together while the wheel operated at normal speeds should the wheel become cracked. In order to permit the fiberglass to adhere to the bonding material, it was impregnated with resin.

The wheel was mounted upon a double end floor grinder; the machine operated two grinding wheels and Anderson was using the wheel on the left side of the machine at the time of his death. There was an adjustable guard over the wheel and an adjustable work rest in front of the wheel. Adjustments made to the position of the guard altered the speed of the machine. An 1/8" gap had to be maintained between the surface of the wheel and the work rest. If the gap became too great the workpiece could become jammed between the wheel and the work rest.

The spindles, being ground by Anderson, were essentially conical in shape with the narrowest end being blunted. Approximately a third of the way up from the base a flange encircled the cone. Anderson was grinding the flash, or excess residue, off of the face of this flange at the time he was injured.

The assistant plant superintendent, Samuel Marich, testifying for the plaintiff by an evidence deposition, stated that the normal procedure before mounting a wheel upon a grinding machine was to "ring" test the wheel. This procedure was accomplished by standing the wheel vertically and striking it in two places, about 120 degrees apart, with a wood and rawhide mallet. The sound produced would indicate whether the wheel was good or defective. The wheel which disintegrated was mounted on August 19, 1972, by Lonnie Crawford, another snag grinder employed by Commercial Stamping. Testifying for the defendant, Crawford could not recall whether a ring test was applied to the wheel before mounting it; in fact, he did not know what a ring test was.

Crawford used the wheel for approximately 5 1/2 hours on August 19, 1972. The wheel was not used again until August 21, when Anderson began to grind the spindles. James Savage, the acting foreman in the grinding department on that date, testified that he assigned that task to Anderson. Savage set the work rest at 1/8", measuring the distance with a ruler. Savage also testified that normal operating procedures provided that a device, called a cup holder, be placed upon the work rest on each side of the grinding wheel. This device held the body of the spindle while the operator ground the flange and prevented the piece from jamming. However, Savage noted that on the machine which Anderson was using, there was a cup holder only on the left side. Savage watched Anderson as he ground the first few spindles. Anderson was using the left side of the wheel and the machine and wheel were operating normally. Savage then left. Approximately 15 or 20 minutes later he heard an "explosion." Employees at Commercial Stamping found Anderson unconscious. The pieces of the grinding wheel were scattered about and the work rest and the guard were broken off of the machine.

Following the incident, Marich gathered the pieces of the shattered wheel and reconstructed it. He determined that the wheel was 28" in diameter at the time of the accident. Marich also "tached" the grinding machine and determined that the wheel was operating at the proper speed.

Norris Yonker, a consulting engineer, testifying for the plaintiff, stated that in examining the reconstructed wheel he discovered that the wheel had delaminated in two places directly opposite from each other.

Savage estimated that before the accident Anderson had ground some 50 to 75 spindles and that this amount of work would have consumed approximately one-sixteenth to three thirty-seconds of an inch of material off of the wheel. He also stated that for this particular job, any distance over three-sixteenths of an inch between the wheel and the work rest was too wide a space from a jamming standpoint. Chester Pawlowski, the grinding supervisor at Commercial Stamping, stated that even if the distance between the wheel and the work rest is more than 1/8", the flange could not become caught.

Professor Serope Kalpakjian testified as an expert witness for the plaintiff. He is a professor of mechanical engineering at the Illinois Institute of Technology and holds a master of science degree from both Harvard University and Massachusetts Institute of Technology. His thesis for one of these degrees was on the subject of the mechanics of the grinding process. He also testified that he had been involved in approximately four research projects on the abuse and destruction of the grinding wheels. During cross-examination, he disclosed that his studies largely involved the cut-off type of grinding wheel rather than the "snag" type of wheel that Anderson was using. The cut-off wheel resembles a thick phonograph record and is used to cut such substances as pipe or concrete. In such wheels, the reinforcing material is placed on the outside of the wheel and, while the stress placed upon the cut-off wheel is essentially the same as the stresses acting upon the snag wheel, the cut-off wheels are subject to certain twisting and bending actions, which are not applicable to snag wheels.

Professor Kalpakjian was retained by the plaintiff to investigate the disintegration of the grinding wheel involved in this case. He reviewed correspondence, reports, photographs and data of other investigators. He also examined remnants of the wheel and the forging and visited the site of the incident. It was his opinion that when there is a change in the geometry of the wheel it will disintegrate. This change could be the result of an obstruction or a piece missing from the surface of the wheel, which, in turn, as the wheel spins, will cause it to jam. Kalpakjian was of the opinion that the disintegration of the grinding wheel which caused Anderson's death began when a section of the wheel delaminated along the surface of the fiberglass and flew out of the wheel while it was in operation. The space left by this piece then "caught" the spindle which Anderson was grinding, causing the jamming process to commence. The jamming process then resulted in the disintegration of the wheel. It was Kalpakjian's opinion that the wheel was defective because a section of the wheel delaminated and separated itself from the rest of the wheel, and this condition existed at the time the wheel left the manufacturer's hands.

On cross-examination, Kalpakjian disclosed that the precise nature of the defect was the weakness of the bond between the fiberglass reinforcing material and the abrasive material. He reached this conclusion because there existed such a large piece of the broken wheel which had split along the plane of the fiberglass material. He explained that:

"* * * when a piece like this breaks or a grinding wheel breaks there should be an equal opportunity for the material to break into all kinds of directions. In fact, if we looked at the rest of the pieces which comprised the original wheel, we'll find that there are five additional major pieces and from what I have seen in the photographs both black and white and in color, I do not see any of the other five pieces — in fact, the rest were, I think, larger than this particular piece (the large section which broke off) — none of them have separated in this manner. The other pieces left the wheel when the wheel exploded. They went all over the place and hit things, but they did not delaminate.

So, when I see a broken piece like this which, again, consistently on this plain [sic] it has separated, this is of great concern to me."

Kalpakjian stated that he did not make a chemical analysis of the bond; thus, he could express no opinion as to the materials which composed the bond. He also acknowledged that a grinding wheel could break, even in the absence of any defect, if it was subjected to a jamming incident. However, he rejected such an occurrence in this instance, primarily because he failed to find any indication on the remnants of the wheel of a force sufficiently strong to cause such an occurrence. He also did not believe that the large section could have flown out of the wheel in one piece and then delaminate upon impact with another object. He noted that the large section showed signs of secondary fractures and that the energy which the combined pieces would have had would have been absorbed in causing the secondary fractures which these pieces exhibited.

Kalpakjian also stated that the existence of the continuous band of deposits along the right side of the wheel was consistent with his theory. These deposits, he felt, could have been caused during the grinding process or when the wheel struck some other object. Moreover, he described the disintegration of the wheel as a "dynamic" process in which it was possible for the spindle upon which Anderson was working to have come into contact with the right side of the wheel.

Harold Van Orden, the manager of product engineering for General Grinding Wheel, testified for the defendant. He stated that during the summer of 1974 he examined the fragments of the wheel. He took two samples from the wheel and analyzed their chemical content. He also analyzed their density and their grain distribution. Van Orden indicated that the tests revealed that the wheel fell within its proper specifications. On cross-examination, Van Orden agreed that no tests were made to determine the quality of the bonding materials. He could not recall from which portions of the wheel the test samples had been taken. Van Orden also admitted that variations in composition within a 30" wheel were possible but indicated that such variations were less likely in a large wheel than in a small one.

Norris Yonker testified as an expert witness for the defendant. Yonker held a bachelor's degree in metallurgical engineering and, for over 10 years prior to trial, worked as a consulting engineer. He had been involved, only once, in a project involving the study of an intentional destruction of a grinding wheel. However, subsequent to his experience with the wheel in this case, he examined other wheels for the purpose of determining the reasons for their breakage.

In examining the pieces of the disintegrated wheel, Yonker found a continuous band of heavy metal deposit along the grinding face of the wheel, approximately 30" in length and along the right side of the grinding face. There was some metal deposit on the large section which separated from the wheel and delaminated. Yonker took a sample of the metal deposit and subjected it to microscopic and chemical analysis. The microscopic analysis revealed that the material was steel build-up. Chemical tests then indicated that the steel was "comparable" to the steel in the spindles which Anderson was grinding.

Yonker testified that this heavy metallic loading was indicative of jamming. He stated:

"The construction is such that the abrasive materials will break away and continuously expose new clean cutting particles so that you do not get a metallic buildup on the periphery of a wheel such as this unless there is some extreme jamming and loading condition which occurs over a very short period of time."

He concluded that the breakage of the wheel was the result of the operator permitting a work piece to become jammed into the wheel. He further testified that, in his opinion, the jamming occurred prior to the delamination of the section which separated from the wheel. The basis of this conclusion was, first, the strength of the fiberglass, when considered along with the uniformity of the bonding and the density of the wheel and the fact that the wheel exhibited no defects. Secondly, he stated that in a wheel of this nature, all of the stress to delaminate is practically nonexistent.

On cross-examination, Yonker stated that even if the reinforcing rings near the center of the wheel were eccentric prior to the accident, such a condition was not the cause of the disintegration of the wheel. He also indicated that, under normal grinding conditions, metal residue would be left upon the wheel, but only in "minute" portions. He felt that it would not be unusual to find heavy scoring on the inside of the machine guards from wheels which have disintegrated, but he rejected the notion that the heavy deposits found on the wheel in this instance came from the wheel striking the guards. He felt that the uniformity of the path of the loadings indicated that the wheel had to be intact when the loadings were made. He did not chemically compare the steel of the guards to the steel content of the loadings.

Yonker further revealed, during cross-examination, that the nominal carbon content of the spindles upon which Anderson was grinding was .45 and that the carbon content of the residue on the wheel was .63. He felt that the differences could have been attributed to the fact that the wheel was made of silicone carbide abrasive. He also indicated that the small size of the sample may have resulted in some margin of error.

John Gregor, the president of SGL Abrasives, testified for the defense. Gregor held a bachelor's degree in chemical engineering and had worked in the grinding wheel business since 1950. Gregor had designed the type of wheel involved in this case and developed the process by which the wheel was manufactured. In his work with the safety committee of the Grinding Wheel Institute, he had observed and analyzed the patterns of breakage of grinding wheels.

Gregor testified that, based upon the crack patterns the disintegrated wheel exhibited, the wheel was broken as a result of jamming. The cracks were not strictly radial but tended to bend slightly in the direction in which the wheel was running. Such a pattern, he stated, indicated that the wheel was jammed and stopped suddenly. Gregor also testified that the laboratory reports indicated that the wheel was composed of the proper material in the proper amounts. He believed that although the test samples were not taken from the piece which delaminated, the results were applicable to all parts of the wheel. Gregor also noted the special deposits on the wheel and stated that the continuous nature of the deposits indicated that they must have occurred prior to the disintegration of the wheel.

Gregor rejected Kalpakjian's theory that a large section of the wheel delaminated prior to the disintegration of the wheel for two reasons. First, if the wheel had cracked, the fiberglass would have been strong enough to hold the wheel together at normal operating speed. Considering the disparity between the weight of the section and the weight of the entire wheel, he did not believe that there would be enough force to permit the piece to tear the fiberglass. Secondly, he explained that the forces at work in a rotating wheel are tangential and radial; there are no forces present that would cause a wheel to break laterally. He also testified that an examination of the interface between the two delaminated pieces of the wheel disclosed that much of the abrasive material still clung to the reinforcing fiberglass, thereby indicating good adhesion. He noted that the resin coating on the fiberglass appeared uniform and evenly spread and that its amber color indicated that the resin had been properly cured.

Finally, Gregor stated that he noticed that the rings in the wheel were eccentric. However, tests which had been done in the company's laboratory indicated that even where extreme eccentricity existed, there was no significant loss of strength. He, therefore, concluded that the absence of concentricity had no effect upon the breakage of the wheel in question.

On cross-examination Gregor reiterated that assuming normal operating conditions, he could not account for the delamination of the large section of the wheel prior to the disintegration of the entire wheel, but he agreed that if such a situation occurred, a sufficient imbalance would result so as to cause the wheel to disintegrate. Gregor also agreed that the bonding materials in a grinding wheel have time-dependent properties. However, on redirect examination, he explained that the shelf life of the fiberglass is relative with respect to the conditions under which it is stored and that once the resin coating loses its freshness, the fabric will lose its flexibility.

George Reitmeier also testified for the defendant. Reitmeier was the executive vice-president of Birlair Structures, Inc., a firm engaged in the design and manufacture of special structures using modern materials. He obtained both a bachelor's and a master's degree in mechanical engineering and for several years had worked at the Cornell Aeronautical Laboratory. During his tenure at Cornell, he had been responsible for a research project designed to study the forces which tend to cause a grinding wheel to break. He subsequently engaged in extensive consulting work with respect to grinding wheels, approximately 75% of which related to the determination of the cause of breakage.

Reitmeier subjected the wheel to both chemical and microscopic analyses. From his microscopic analysis he determined that there was a good distribution of both bond material and abrasive grains and that the individual grains were well-coated by the bond material. He also found that some of the individual abrasive grains in the fracture surface were, themselves, fractured, indicating that there was a strong adhesion between the bond and the abrasive material. He also found that the fiberglass was still coated with bond and abrasive grains, including some fractured grains. Reitmeier's chemical analysis was performed upon two samples taken from the large section of the wheel. These tests showed that the wheel's composition was as specified.

Reitmeier stated that, in his opinion, the cause of the disintegration of the wheel was a jamming incident and there was no defect in the wheel. He reached this conclusion because he could find nothing wrong with the abrasive material, the mixture, the bond, or the fiberglass. The fracture pattern, moreover, further supported this conclusion.

Reitmeier rejected the theory that the large section delaminated from the wheel prior to the disintegration. First, he noted that neither the bond nor the fiberglass appeared defective. Secondly, this section was too small to break the fiberglass at normal operating speeds. If the piece did come off, and the workpiece went into the gap left behind, the next piece of the wheel should have indicated crushing and, in this instance, there was none. Moreover, if the wheel was defective, the fragment would probably have come out when the wheel was first used since it is at that point that the centrifugal force on the piece would have been greatest. The delamination which did occur, he believed, came as a result of secondary breakage following the disintegration of the wheel when the piece struck some hard object. Reitmeier also noted that during normal operation, there were no forces which would tend to cause the wheel to break into layers. Reitmeier did not feel that the amount of eccentricity of the rings had an effect upon the breakage of the wheel.

Reitmeier also stated his opinion as to how the jamming occurred. He believed that the workpiece became "caught in between the work rest and the guard, and the face of the wheel" causing the crushing of a portion of the wheel. He testified that a space of three-sixteenths of an inch between the work rest and the wheel was sufficient to allow the flange to jam.

On cross-examination, Reitmeier agreed that delamination could occur where the quality of the bond was defective, but he stated that for a wheel operating at normal operating speeds, such a defect would have to cover at least 50% of the wheel before that wheel was unusable. He explained that when delamination occurs it usually occurs along the plane of the fiberglass because that plane is of a lesser strength than any other in the wheel. He also indicated that while his chemical analysis was done upon a piece taken from the separated section, it was not taken near the interface of the two pieces of the section which separated from the wheel and delaminated.

Robert Markotan testified for the plaintiff as a rebuttal witness. Markotan held a bachelor's degree in physics and chemistry and in chemical engineering. At the time of trial he was a manufacturer of a limited line of abrasive products, particularly cut-off grinding wheels. He had previously worked for Acme Abrasives.

Markotan testified that it was his opinion that a grinding wheel composed of less than 20 parts per volume of resin had deficient resin content. He explained that normally delamination "appears as a relatively smooth plane with little if any abrasive grain fracture." In examining the section which separated and delaminated from the larger section, he found a kidney-shaped area, approximately 4 to 5 inches by 2 inches of delamination. He also examined the large section and noted that few fiberglass particles were clinging to its surface, which indicated to him an inadequate resin content.

On cross-examination he stated that he had made no microscopic examination of the wheel fragments. He also admitted that there were fractured abrasive particles in the area of delamination. He further stated that his opinion as to the low percentage of resin referred to the entire wheel.

The defendant sought to impeach Markotan with his evidence deposition in which he stated that he never produced wheels with less than 20 parts per volume resin. The defendant introduced documents which showed that when Markotan was employed at Acme, grinding wheels had been produced with less than 20% resin and that some of the formula cards used in the manufacture of those wheels had been prepared by Markotan. However, on redirect examination, Markotan stated that while working for Acme, he had recommended that at least 20% resin be used; however, his recommendations were not always followed.

As surrebuttal evidence, the defendant introduced the testimony of Robert Beebe, the president of Acme Abrasives. Beebe confirmed that during the time which Markotan was employed by the firm they produced grinding wheels similar to the one involved in this case with a content of less than 20 parts per volume of pure resin and that they continued to produce such wheels at the time of trial.

• 1 A directed verdict is properly entered only when all of the evidence, when viewed in its aspect most favorable to the opponent, so overwhelmingly favors the movant that no contrary verdict, based upon that evidence, could ever stand. (Pedrick v. Peoria & Eastern R.R. Co. (1967), 37 Ill.2d 494, 229 N.E.2d 504.) Thus, where there exists a substantial factual dispute, or where the assessment of a witness' credibility and the election between conflicting evidence may be decisive, a directed verdict is inappropriate. (Reynolds v. American Oil Co. (1975), 32 Ill. App.3d 905, 337 N.E.2d 403; Lovejoy v. National Food Stores, Inc. (1973), 12 Ill. App.3d 982, 299 N.E.2d 816.) However, the determination of the propriety of such a motion is to be made from a consideration of all of the evidence introduced. See St. Paul Fire & Marine Insurance Co. v. Michelin Tire Corp. (1973), 12 Ill. App.3d 165, 298 N.E.2d 289.

• 2 In order to recover upon the basis of strict products liability it was incumbent upon the plaintiff to establish that an unreasonably dangerous condition existed in the product at the time the product left the control of the manufacturer and that the unreasonably dangerous condition was the proximate cause of injury to the plaintiff. (Suvada v. White Motor Co. (1965), 32 Ill.2d 612, 210 N.E.2d 182.) Existence of a defect may be established by showing that the product failed to perform in the manner reasonably expected in light of its nature and intended function. (Gillespie v. R.D. Werner Co. (1978), 71 Ill.2d 318, 375 N.E.2d 1294.) The plaintiff's theory is that the wheel was defective because the bond between the abrasive material and the fiberglass was weak. This weakness permitted a section of the wheel to delaminate and leave the wheel while it was in operation. The plaintiff then asserts that the spindle upon which Anderson was grinding caught in the gap made by the section of the wheel breaking away from the wheel, initiating a jamming incident. The defendant, on the other hand, contends that the jamming incident occurred when the spindle became caught between the face of the wheel and the work rest. It suggests that Anderson was grinding on the right side of the wheel, without the aid of the cup holder, and that the absence of this device permitted the flange of the spindle to become caught in the gap between the work rest and the wheel. This inference was not proven since there was no direct evidence that Anderson was working on the right side of the wheel.

Both parties presented expert witnesses to support their explanations of the accident. Kalpakjian, the plaintiff's expert witness, stated that he concluded that a fragment of a section of the wheel delaminated prior to the disintegration of the wheel and that it was the only piece which delaminated. This conclusion was contradicted by Yonker's testimony that there were two areas of delamination found on the broken wheel, rather than the one section as stated by Kalpakjian.

Gregor and Reitmeier testified for the defendant that even assuming delamination did occur, the piece was too small in relation to the size of the wheel to come out while it was operating. This point would seem to be corroborated by the plaintiff's rebuttal witness who specified that delamination referred to a condition where there is a separation along a relatively smooth plane with little grain fracture and found an area which, at most, was 5" x 2" that exhibited such characteristics.

Both Van Orden and Reitmeier conducted chemical analysis of the wheel and established that the composition of the wheel was within the manufacturer's specifications. The plaintiff attempts to attack this evidence on the theory that during the manufacturing process a great deal of variation in composition may exist in the wheel and that since none of the test samples were taken from the fracture surface, the results are irrelevant. The plaintiff also attacked these results through Markotan's rebuttal testimony that, in his opinion, resin content of less than 20 parts per volume is too low. However, it was established that this view was not followed by his former employer, a competitor of General in manufacturing the same type of wheel involved in this case.

Reitmeier also conducted a microscopic examination of the fracture surface and found that rather than "wandering" through the bond material, the fracture also split the individual grains. This fact was also attested to by Markotan, the plaintiff's witness.

The defendant supported its theory that the flange became caught on the right side of the wheel with testimony that after grinding 50 to 75 spindles, there was a gap of sufficient size to allow the spindle to catch. In addition, there was no cup holder on the right side of the work rest at the time of the accident and there was a continuous band of metal deposits on the right side of the wheel. By chemical analysis, these deposits were shown to be at least "comparable" to the spindles which Anderson was grinding.

Although Kalpakjian admitted that a wheel could break as the result of a jamming incident, even in the absence of a defect, and he agreed with Reitmeier that delamination could occur if the portion of the wheel was struck by an external force, he rejected these theories because he found no evidence to support either possibility. In addition, he explained why the existence of the continuous band of deposits along the right side of the wheel was consistent with his theory.

• 3 In this case there is considerable conflict between the factual evidence and the expert witnesses' testimony presented by the plaintiff and the defendant. In cases of conflicting expert testimony, it is the duty of the trier of fact to assess the credibility of the witnesses and to assign the weight to be accorded their testimony. However, the weight to be accorded their testimony. However, where there is no factual basis for the expert's opinion and it is impeached by physical exhibits introduced into evidence, the opinion is entitled to little weight. Mullen v. General Motors Corp. (1975), 32 Ill. App.3d 122, 336 N.E.2d 338; St. Paul Fire & Marine Ins. Co. v. Michelin Tire Corp.

The defendant urges that the St. Paul and Mullen decisions control the outcome of this case.

In St. Paul, the issue was whether the cause of a tire blowout was an unreasonably dangerous condition existing in the tire when it left the manufacturer's control, or a penetration of the tire by an external object lying on the road. The plaintiff's case consisted primarily of the testimony of an expert witness who asserted that a defect existed in the manufacture of the tire. At the close of all of the evidence, the court held that judgment should have been entered in favor of the defendant and reversed a judgment for the plaintiff. The basis for the court's decision was the fact that portions of the plaintiff's expert's testimony was contradicted by the defendants' experts, and, of greater importance, that his opinion was severely impeached by the physical exhibits introduced into evidence.

The Mullen decision also involved a tire blowout. In this case the court also reversed a judgment for the plaintiff whose entire case depended on the opinion of her expert. The court concluded that the expert's opinion was thoroughly discredited because of ambiguous indicia upon which his opinion was premised and contradictions between his courtroom testimony and an earlier written report.

The facts in the instant case are distinguishable from the facts in St. Paul and Mullen. Even though there was conflict between the testimony of Kalpakjian and the defendant's witnesses, we do not believe that Kalpakjian's testimony was severely contradicted by the physical exhibits, nor do we believe that the defendant's explanation of the accident was supported unequivocally by the evidence. Therefore, we believe that it was the responsibility of the trier of fact to assess the credibility of the witnesses.

• 4 Considering all of the evidence in this case, we believe it was error for the trial court to grant the defendant a directed verdict. Accordingly, we reverse the trial court's order granting the defendant a directed verdict, and remand the case for a new trial.

Reversed and remanded.

McNAMARA and JIGANTI, JJ., concur.