U. S. Fibres, Inc. v. Proctor & Schwartz, Inc.

ORDER OF JUDGMENT FOR DEFENDANT KAESS, Chief Judge. This is a civil action for damages. Plaintiff has claimed fraud, breach of express warranties, breach of implied warranties, and negligence. Jurisdiction is based on diversity of citizenship. 28 U.S.C. § 1332. After plaintiff had completed the presentation of its evidence, the defendant moved for a dismissal, on the grounds that, upon the facts and the law, plaintiff had shown no right to relief. Rule 41(b) Federal Rules of Civil Procedure. On December 1, 1972, this court granted the motion and dismissed Counts I, II, III and IV of plaintiff’s complaint, 358 F.Supp. 449. The trial then continued only with respect to Count V, which involved the alleged negligence of the defendant. The facts and circumstances surrounding this case were set forth in the opinion and order of December 1, 1972 and will not be restated here. However, at the outset the court reiterates that at the time the parties entered into the contracts involved in this case, neither of them knew precisely what manufacturing parameters were necessary to produce dry resinated pads. Plaintiff thought it did; however, subsequent events have proven this belief to have been erroneous. I. Defendant Was Not Negligent. It is the court’s finding that the evidence as presented by the defendant is more worthy of belief and that, based on this evidence, the defendant was not negligent in the design or manufacture of the dryers involved in this case. Specifically, the court finds that the structures of the dryers were adequately designed from a strength standpoint, as well as from the standpoints of rigidity and deflection. The court bases this finding of fact on the opinion of defendant’s expert, Dr. Despres, whose opinion was based upon structural analysis utilizing accepted engineering practices and upon the physical examination of the subject ovens in the field, as well as the viewing and examination of the Globe machine in operation. The structural aspects of the jack support points are adequate. The bearing load is not supported on the fan side solely by two-inch longitudinal tubing. The total framework of the fan section *469 of the oven, including the heavy fan support brackets, roof members, jack line shaft support structures and motor mount support structures, act as a system bearing the load and keeping the total deflection at each jack point reasonably uniform. The evidence also established that the ovens utilized an independent adjustment procedure in the jack screws as a means of compensating for deflection in the structure. The adequacy of this procedure for maintaining close tolerance is clearly established by plaintiff’s ability to hold a level of .010" across the jack points (Exhibit 73). Further, plaintiff’s own report shows the jacks remain level from side to side during the raising and lowering of the compression conveyor frame. Turning to the allegations of weakness in the side plates of the compression frame, the evidence establishes that the maximum deflection that would be attained would be under 0.006 of an inch between jack points. The worse condition is experienced with no material between the conveyors. The introduction of material between the conveyors reduces the deflection by permitting the carrying conveyor structure to assume a portion of the load being supported by the jack screws. There is ample evidence to support a finding of fact that the design of the conveyor flights did conform to the design criterion set forth in the contract. Portions of each of the four conveyor assemblies were subjected to loading tests. The physical measurements taken at the girt and at the hinge, at both elevated and room temperatures, clearly show that under 30 lb/ft.2 uniform loading the deflections are well within ± Vs2 of an inch. Indeed, with loads of up to 50 lb/ft.2, the deflections are within the ± V32 of an inch (Defendant’s Exhibits 335, 336, 337 and 338). However, the loading conditions within the ovens while operating at 30 lb/ft.2 on the product are substantially less than the test loadings. Each conveyor section experiences approximately 15 lb/ft.2 force by virtue of its own weight. The conveyors are essentially flat in an unloaded condition and the deflection resulting from their own weight is negligible and of no consequence. The plates are designed to be flat under their own weight in an unloaded condition. During actual operation, a deflection near the design criteria limit could exist at the hinge of a single conveyor flight under 50 lbs/ft.2 loading with the hinge wire of too small a diameter. However, Shegda and Flaith both testified that hinge wires of .156 diameter were supplied for both lines. However, the installation of such hinge wires would reduce the deflection as observed during testing of the hinge point deflection (Exhibit 338). The finding of fact is inescapable that, under a uniformly distributed load of 30 lb/ft.2, the conveyors deflect substantially less than the contract criteria. The plaintiff has offered certain proofs to support its conclusion that negligent design is shown by the inability of the flights to take a thirty-pound load and hold the ± %2 of an inch tolerance when cold. This evidence takes the form of its loading tests conducted upon a single conveyor flight. The data obtained by plaintiff is invalid and misleading in several important respects. Primarily the testing of a single conveyor section ignores the contract description, and does not accurately reflect operating conditions within the oven. In operation, each conveyor is tied to the other through the piano hinge in a continuous belt. Each flight is designed so as to give strength to the adjoining flight as an integrated system. Of equal importance is the method chosen by plaintiff to load the single flight. A 50-pound three-foot long steel bar placed on a 2" x 4" piece of wood seven feet long was used in plaintiff’s test. Plaintiff claims that such a loading of a single conveyor flight at the hinge edge furthermost from the girt was equivalent to a uniform load of *470 about 10 lbs/ft.2. However, this is incorrect. A 50 lb. load spread over the area of the piece of wood beneath the three-foot steel bar is equivalent to an actual loading of 55.6 lb/ft.2. The next important factor rendering the test invalid is the placement of the load at a distance furthermost from the centroid of the flight. This placement of load creates a moment of force quite different than that experienced under a uniform load. The evidence established that the effect of such an off-center load was to create a comparable uniform load of several times the 55.6 lb/ft.2. The result was a twisting of the conveyor flight in addition to deflection, both of which were measured by Meldrum in giving his testimony that the single flight deflected over five times beyond the ± %2 tolerance. Thus the court finds that the conveyor flights were properly designed not to deflect more than ± %2 of an inch within the terms of the contract and in operation deflected within this design criteria when subjected to uniform working loads of 30 lb/ft.2. Furthermore, the evidence establishes that the level of residual stress induced by fabrication of the flight was minimal and of no consequence. The condition of the residual stress would be virtually relieved within two or three days of operation and thus there are no residual stresses existing in the flights after prolonged operation. Certainly, the natural phenomena of thermal expansion exists. Thus, nonuniform temperatures will cause nonuniform deflections of the ' conveyor flights. Plaintiff’s heat deflection tests are a clear example of this natural phenomena. In this test plaintiff used a heat sensitive crayon which was able to change color when heated, but could not change color when cooled. The testimony revealed that a heat sensitive crayon was capable of only changing color once to show when the heated member came up to temperature. The physical change in the crayon resulting in its color change prohibits it from returning to its original color when cooled. Plaintiff used a torch to heat the conveyor tested. The utilization of such a localized heat source as a torch makes it abundantly clear that an uneven heating of the flight resulted. The distortion of the flights experienced in each of the three tests performed by plaintiff is attributable to the uneven temperature induced in the test members, which was not present in the actual operation of the oven. The plaintiff claims that having the return run of the carrying conveyor beneath the subfloor will cause distortion in that conveyor. The concept is theoretically true, but the amount of distortion encountered is so minimal as to have no effect upon the commercial operation of the oven. The testimony clearly establishes that, in the operation of the oven, the carrying conveyor passes beneath the subfloor on its return run. The temperature beneath the subfloor will be less than that of the oven, but will approach an equilibrium condition with respect to the flights since the unit is entirely enclosed except at the ends. The temperature of the girts will be maintained in a reasonably stable condition. The plates will cool slightly. If the plate temperature dropped 50 °F., it would result in a deflection of 0.010 to 0.015 of an inch in the flight. The temperature of the plate, however, will rise rapidly upon entering the air-up compartment. The design adequately provides for the elimination of this temperature differential by subjecting the carrying conveyor plates to circulating heated air in the initial air-up compartment of the dryer. Thus, the plaintiff has not sustained his burden of proof that the conveyor flights are unable to withstand a deflection of ± Vs2 when heated under a uni-formerly distributed load of 30 lb/ft.2. Plaintiff attempted to establish a non-uniformity of temperature within the dryer. To establish this allegation *471 it presented evidence of bluing of certain parts of the oven. The condition of bluing of the rollers and bushings on the fan side of Line I is cited by plaintiff to support the conclusion that fan side operated at 560° or above while the apron side operated at 300°. Even the most rudimentary reasoning would tell an engineer that a bluing condition which existed on the chain rollers of the compression conveyor, fan side, and did not exist on the chain rollers of the carrying conveyor four inches away, was not induced by the circulation of air in the operation of the dryer. Plaintiff’s contention ignores the testimony concerning fires in the upper portion of the compression conveyor at a point where the rollers of the carrying conveyor and compression conveyor are physically separated. Furthermore, the evidence clearly shows that: (a) At 560° the cotton material would be seriously charred and decomposed from heat; (b) The oven was provided with temperature controllers set at 375°F in the first two zones and 325 °F in the last two zones (air down) (Exhibit 85) ; (c) These controllers determine the temperature of the air prior to its going into the conveyor space and . the only way the air temperature can be above that set on the controllers is by a malfunction of the controller. There is no evidence that the controllers malfunctioned, thus the only source of temperature for bluing of the rollers must be the fires as testified to by defendant’s and plaintiff’s witnesses. The court finds that the dryer design incorporates fan and plenum components experimentally, theoretically and practically developed to optimize uniformity of temperature. The dryers were equipped with plenums, distribution plates, fans and a circulating system that provided uniformity in temperature of air across the plenum width and consequently the conveyor system. The evidence establishes the research and testing involved in these designs, and the basic theoretical utilization of the plenum to provide uniform temperature of the air within the mixing chamber. Finally, the only person to have actually measured temperature within the oven during its operation was Shegda. His testimony outlined the procedure followed utilizing thermocouples and his observations that there was only a 5° temperature difference within the dryer across its width. The preponderance of the evidence dictates a finding of fact that the temperature within the dryers was reasonably uniform as set by the temperature controllers. Plaintiff claims that the ovens were not properly designed to allow for stresses induced by thermal expansion. Basically the ovens are divided into three main units: (1) the main body, (2) the compression conveyor assembly, and (3) the carrying conveyor assembly. The evidence clearly established that the levels of stress that are generated by expansion are not sufficiently high to cause significant distortion which will interfere with the commercial operation of the dryers. Expansion within a machine subjected to circulating heated air is a fact of life. This fact of life is compensated for through pre-heating the oven before operation to permit the anticipated expansion to occur prior to product being manufactured. The plaintiff claims the main body section bowed from lateral expansion, thereby bringing the two conveyors closer together. Exhibit 85 shows conclusively that just the opposite condition existed where during the second measured hour the conveyors actually grew apart. The basic main body design utilized by defendant had been used in many of single conveyor dryers at even higher temperatures without expansion problems. *472 The testimony at trial also established the free expansion characteristics of the oven and structural strengths which would not allow buckling to occur. The compression frame is wholly within the heat zone and if the conveyors are properly cleaned and maintained, is subjected to a constant temperature within each zone. The design is such as to permit and maintain expansion after preheating. The carrying conveyor was likewise properly designed to allow for expansion. Plaintiff made an attempt at putting an expansion joint in this structure (Exhibit 329) but it is clear that such an effort was. ineffective because the wear strip remained whole and bridges the gap in the metal. Any movement in the surrounding metal would have the result of causing waviness in the wear strip. At best, the cutting of the channel section served only to weaken the carrying conveyor structure. In summary, the defendant made proper allowances for thermal expansion in the dryers as shown by the following design features: (a) That the temperature will expand the subfloor as well as the concrete floor and the difference between these expansions is relatively small; (b) That the hold down bolts are smaller than the holes in the support angle feet and allow movement relative to each other; (c) That the structure, if completely restrained, will not be subjected to stress levels of a magnitude such that buckling would occur; (d) That the drive shafts and chain sprockets are located separately from the oven body within the delivery and feed and extensions and the take-up cylinders and sprockets maintain their locations independently of the temperature or expansions within the oven body. The evidence fully establishes that expansion was anticipated and fully provided for in the design of the subject dryers. II. Plaintiff was Contributorily Negligent. It is well settled that if a plaintiff fails to exercise ordinary care for his protection against injury, and his failure to exercise such care is a proximate cause of his injury, plaintiff is guilty of contributory negligence which will bar any recovery. Jones v. Michigan Racing Assoc., (1956) 346 Mich. 648, 78 N.W.2d 566; Neal v. Cities Service Oil Co. (1943) 306 Mich. 605, 11 N.W.2d 259. In the present case the court finds that all of the problems which plaintiff encountered with the dryers and with the non-uniformity of the finished product were due to plaintiff’s contributory negligence as shown by the following: (a) Overloading the dryers by exceeding their compressive capacity; (b) Lack of proper maintenance of the dryers; (c) Lack of control over the manufacturing process. 11(A) Overloading. There is no question in the court’s mind that plaintiff overloaded the dryers. The court has already found that on numerous occasions the plaintiff manufactured pads whose density was in excess of 4 lbs,/eu. ft. (See pp. 454-455 of this court’s order of December 1, 1972). Both the plaintiff’s and the defendant’s test results show that forces in excess of the 30 lb/sq. ft. limitation are required to produce pads which are of densities greater than 4 Ib/cu. ft. Extrapolation of plaintiff’s own test data clearly established that in order to produce padding of the density produced by plaintiff, forces far in excess of 30 lb/sq. ft. are required. Plaintiff’s data (Exhibit 287) clearly show the sharp rise in compressive forces as density in *473 creases. This data shows that as density increases, the compressive force increases, but not in a direct relationship. Thus, as density increases, the compressive force required compounds to the point where a small increase in density results in a substantial increase in force. This compound relationship between density and force was clearly shown to this court by the plaintiff’s demonstration on Exhibit 266 with the Measurematic. The first small increments of weights applied resulted in large decreases in thickness. However, as the pad was further compressed, the addition of heavy weights were required to achieve any thickness reduction. In the courtroom, plaintiff was unable to reduce Exhibit 266 to % of an inch with a force of 42 lbs/ft.2 Defendant was able to achieve a thickness of % of an inch with a force 47 lbs/ft.2, but it took a force of 125 lbs/ft.2 to further reduce said Exhibit 266 to of an inch (Test 8, Exhibit 320). These values were obtained after precompression of the Exhibit by plaintiff which the evidence showed would reduce the amount of compressive force necessary. It must be stressed that Exhibit 266 was prepared by plaintiff on equipment chosen by it with fiber it selected and was offered to this court to demonstrate typical batts utilized by plaintiff in its own compression tests. The test data submitted by the defendant only reinforces the court’s conclusion that forces far in excess of 30 lb/sq. ft. were employed by the plaintiff in manufacturing its products. This data clearly established that compressive forces in excess of 200 lbs/ft.2 had to be employed to produce padding of densities in excess of 8 lbs/cu. ft. and forces in excess of 250 lbs/ft.2 had to be used to produce padding of densities in excess of 9 lbs/cu. ft. (Exhibit _ 315). Furthermore, the evidence also establishes that the process used by plaintiff required “over compression” of the pads. That is, if a pad of % inch finished thickness was required, it was necessary to compress it in the dryer to a thickness of less than *4 of an inch. The evidence concerning over-compression took the form of a description of what the defendant described as “snap-back”. Snapback is the characteristic of the fiber, after compression and curing, of seeking to return to its original shape when compression is released. The presence of resin as a bonding agent restrains snapback. The number of fibers present seek to overcome the restraint of the resin. Thus, the higher the density of the product, the more snapback that can be expected if the resin content is relatively uniform. If the resin content varies, snapback is more pronounced in the areas of the batt where the resin content is least (Exhibit 310). The importance of snapback as a characteristic of the product is seen by the fact that to obtain a finished thickness of a certain dimension, the conveyor spacing must be set at a closer dimension. The testimony established that snapback appears above 3.6 lb/ft.3 density and increases in amount with an increase in density. Below 3.6 lb/ft.3, the product experienced no snapback or actually shrank as evidenced by Steuernagel’s tests in May of 1965 (Exhibit 6A). In his tests Steuernagel cured the product to the inch finished thickness by utilizing % inch thick wooden stops in the test tray, thereby compressing to the finished thickness desired. Steuernagel further testified that to manufacture product he would, as a matter of procedure, set the conveyors while empty to the finished thickness desired and then commenced production. Thus, snapback was encountered by Fibres in manufacturing products above a 4 lb/ft.3 density. Exhibit 211 was identified as being a sample of Carpet-mate Special having a specified thickness of ^4 inch. The physical measurement of said Exhibit showed the actual thickness to be nearly % of an inch. *474 This difference in thickness is due to the snapback of the denser material to a thickness greater than the inch conveyor setting prior to production. Further evidence of snapback is found in Exhibit 200, a test report prepared by General Motors Company on July 30, 1968. The subject matter of the test are two products, to wit: % inch 69.5 grams of a density of 7.35 lb/ft.3 and % inch 70 grams of a density of 4.94 lb/ft.3 The actual measurements of thickness by General Motors shows the minimum thickness of the % inch product to be nearly V?& of an inch greater than inch, with the thickest measurement being nearly lio of an inch greater than of an inch. The comparable measurement of the % inch product shows the thickness to be within %ith of an inch above and below the specified finished thickness. There are several conclusions which may reasonably be drawn from the contents of the General Motors Test Report (Exhibit 200). The conclusions are: (a) The thickness measurements (.282" to .314") of the 7.35 lb/ft.3 density product confirms snap-back of between %z of an inch to Via of an inch greater than the inch specified thickness: (b) The thickness measurements (.362" to .386") of the 4.94 lb/ft.3 density product demonstrates snapback to a much lesser degree, i. e. Vei inch beyond the % inch specified thickness; (c) It is reasonable to conclude that plaintiff would carefully control the uniformity of resin and fiber distribution within the samples submitted to General Motors. Both thickness measurements are over a range of ± Vsz which show the ability of the subject dryers to deliver a product with that thickness uniformity when uniformly fed. The foregoing evidence of overloading is borne out by the condition of the dryers from both Line I and Line II. The evidence supporting the finding of fact that Line II was overloaded to destruction is abundantly clear. The testimony and the physical evidence that Line II was overloaded to destruction includes : (a) The evidence of wear on both the fan side and apron side on the compression track (Exhibit 331). (b) Permanent deformation of the compression track angle on the apron side (Exhibit 332). Such a condition of permanent deformation could only be caused by an upward force against said track from overloading the oven far beyond 30 lb/ft.2 or from a jam such as that experienced in December of 1967 when Fibres negligently set the cutoff switch too low on the take-up cylinder (Exhibit 112). The plaintiff contends that the condition of permanent track deformation is attributable to the sagging of the upper conveyor support structure. Such a contention is not true for at least two reasons, to wit: the corresponding track on Line I is not so deformed, and the condition of deformation likewise exists on the fan side of Line II. (c) Permanent deformation of . the compression track angle on Line II fan side. This deformation was caused by subjecting the compression track to average working loads of 200 lb/ft.2 and greater, across the width of the batt. The plaintiff has offered no explanation for the deformation of the track on the fan side. (d) Steuernagel, as corroborated by Simon, testified to barrel-shaped batts being produced on Line II. This condition did not exist in Line I where the complaint was wedge-shaped batts (Exhibit 48). The barrel-shaped batt was obviously caused by a deflection of *475 the conveyor plates and girts across their 96-inch width. The plaintiff agrees that the batt from Line II was at least % of an inch thicker at the center than . at the edges. The evidence establishes that in July of 1968 when the flights were removed from Line II that the compression conveyors had a permanent bow. Such a bow is the direct result of deflection of the conveyor plate and girt beyond elastic limits caused by average uniform loading beyond 250 lb/ft.2 (Exhibit 334). Further corroboration is seen in the physical condition of Line I. The evidence supporting the finding of fact that Line I was subjected to overloading include: (a) The wear patterns present in the bushings (Exhibit 287) caused by continual contact at the sprocket while the dryer -was operating at forces about 30 lb/ft.2 (b) The wear patterns encountered on the compression track of Line I (Exhibit 330). (c) The necessity of replacing the compression wear track in Line I (Exhibits 324 and 330). (d) The broken bushings caused by water quenching from a fire hose after being subjected to fires. Steuernagel admitted the fires did exist and Stradling witnesses one such fire being extinguished. The quenching caused cracking in the bushings which was aggravated by overloading. The evidence is clear and convincing that plaintiff never manufactured the low density products tested in May of 1965, and that the products actually manufactured overloaded the subject dryers shortening the commercial life of the Line I dryer and resulting in the destruction of the Line II dryer. While plaintiff’s use of the dryers at compressive forces slightly beyond 30 lb/ft.2 shortened their commercial life, these minor overloads, while serious, are not fatal. The most serious damage occurred when plaintiff elected to run at weights much beyond those which were originally tested. These increases, coupled with variations in the uniformity of weight of fiber and resin present in the batt, were largely responsible for plaintiff’s difficulties. On many occasions pads with variations of weight were fed into the dryer with disastrous results. Consider the 4 lb/ft.3 density product of Exhibit 2. The approximate 10% weight variation encountered in the duoform operation (Exhibit 48) had the effect of increasing the density to 4.4 lb/ft.3 At that level of operation, the changes in density result in little change in compressive force. Between 3 lb/ft.3 and 5 lb/ft.3, the range of maximum compressive force is between 20 lb/ft.2 and 59 lb/ft.2 (Exhibit 315) However, consider the products which were actually manufactured. As the weight of the product was increased, the variation in the uniformity of the web from the duoform increased. In some instances, the weight variation was over 60%. Exhibit 208 is an 8.35 lb/ft.3 product. At that level of operation, any change in density results in a substantial change in compressive force. This substantial change in force is further compounded by the 50% variation which the record shows actually occurred. At the 8.35 lb/ft.3 density, a 60% variation -in density would raise the range to 13 lb/ft.3 density. Between these two densities, the maximum compressive force is between 190 lb/ft.2 to in excess of 400 lb/ft.2 Thus the operation of the dryers at the density ranges admitted to by plaintiff, when taken together with the variations in weight encountered in the duoform, resulted in massive overloading of the subject equipment. Plaintiff contends that defendant’s compression tests are false and misleading. Plaintiff bases this assertion on a quote from Exhibit 302, which states: “During the compression testing, it was noted that a certain amount of *476 the fiber extruded from the edges of the pressure plates. Therefore, all of the fibers included in the weight of the speciments were not actually compressed by the lid. It is recognized that this condition makes a small bias in the test results and that the error tends to make the test results density slightly greater than it actually should be.” (24) The plaintiff also asserts that the extrusion of the fibers into the perforated holes of the test plate likewise gave unreliable test results. The contention of the plaintiff is based upon a complete misunderstanding of the physical facts of nature involved in the compression tests. The remark quoted above must be taken in the context that it is made. During testing the density of a given sample is obtained by weighing the entire sample and then computing density from the finished thickness obtained. However, during compression of the sample, certain amounts of the fibers extrude from the edges of the pressure plates. The extrusion of these fibers removes them from their acting upon the compressive force. Accordingly, the amount of compressive force is reduced since a lesser amount of fibers and resin are actually compressed. The readings do have a bias, but the bias favors the plaintiff. The actual compressive forces recorded are the compressive forces reacting on less than the entire volume of fibers present in the given sample. Therefore, the compressive forces are actually lower than those which would be found in compressing 100% of all fibers and resin present in the sample. Accordingly, the density reflected in the test results is slightly greater than the actual amounts of fiber subjected to the test. A similar effect occurs with the extrusion of the fibers and resin into the holes of the perforated plates in the test fixture, as well as with any deflections that would exist in the test plates. Density in the tests are measured by the spacing between the solid portions of the perforated plates. Any extrusion of fibers into the perforations are the result of fibers and resin not being compressed to the finished thickness of the spacing between the solid portion of the plates. Therefore, the density actually being seen by the perforated plates is less than the actual density being measured in the test between the solid portions of said plates. The resulting compressive force is correspondingly less because the perforated plates are not compressing all of the fiber and resin to the same thickness. Therefore, the compressive forces recorded are those actually generated by a density less than that which is recorded. 11(B) Lack of Maintenance. The severe lack of maintenance is a contributing factor to the poor performance of the dryers, as well as to their rapid deterioration. This lack of maintenance included the rather singularly important failure to replace the replaceable wear shoes on the carrying convey- or. The evidence clearly reflects plaintiff’s failure to replace these wear shoes. Steuernagel testified to full knowledge of the purpose of said wear shoes and that they were designed to be easily replaced when they were worn. Steuernagel further testified that on numerous occasions prior to July 1966, he, as part of regular maintenance, replaced the wear shoes. Despite this admitted knowledge and practice in 1966, plaintiff thereafter chose to wholly ignore this maintenance requirement. The results of plaintiff’s negligence in maintenance are seen in Exhibit 327, which is an actual wear shoe found on Line I as it stood in the field on August 11, 1972. • The wear shoe had been permitted to wear completely through and fall from its bracket. Photographic evidence contained in Exhibit 297 records the condition of the wear guides as they existed in 1968. *477 These photographs taken by the plaintiff’s own photographer clearly show that the wear guides were permitted to wear completely away. The brackets to which the wear shoes were attached were in several instances themselves totally severed. When plaintiff did replace wear guides, they were replaced with material other than that which the design provided for (photograph 11222 of Exhibit 297). The condition of the totally worn wear shoes and worn guide brackets, as they existed at the conclusion of operations on Line I, is photographically depicted in Exhibit 328. The direct effect of the failure to replace the replaceable wear guides is readily seen in Exhibits 246 and 247. The carrying conveyor was permitted to travel sideways until its chain attachment came in contact with the carrying conveyor track resulting in wear on the conveyor itself. These Exhibits further show actual evidence of locking the conveyors together through contact of the button head bolts. As a result of this contact, the compression conveyor was forced to follow the path of the carrying conveyor which, in turn, caused contact between the compression conveyor chain extension and replaceable wear guide bracket (Exhibit 329). The indirect effect of allowing such conditions of wear to prevail is to cause additional working loads on the chains and drive mechanism. In addition, the conveyors now are in a position to hang up on the tracks within the oven, stretching the chains and increasing the torsional deflection in the drive shafts beyond design limits, thereby aggravating the tracking of the conveyors. The design adequately provides for reasonable ease in maintenance. The replaceable wear guides and wear strips are an integral part of the design concept. Additional evidence as to lack of maintenance and abuse of the units includes : (a) Plugging of the conveyor plate perforations (Exhibits 246 and 247) with fiber and cured resin. This plugging restricts the uniform flow of heated air within the oven. (b) Severe wear in bushings, rollers and pins. This (Exhibit 287) is a striking example of plaintiff’s complete inattention to the dryers, since for over three and one-half years it failed to even inspect the bushings for breakage or wear, much less replace any as needed. (c) Sprocket teeth worn away completely as testified to by Steuernagel. This shows plaintiff operated the dryers without any consideration for signs of wear and until failure occurred. (d) Plugging of guide rollers with fiber and resin as depicted in Exhibit 297 and Exhibit 326, which conclusively shows the failure on plaintiff’s part to properly clean the machine of material which would interfere with the normal lubrication of moving parts. Thus, the evidence clearly establishes that a lack of maintenance was a contributing cause for plaintiff’s being unable to consistently manufacture a commercially acceptable product, as well as shortening the commercial life of the dryers. 11(C) Lack of Control over the Manufacturing Processes. The evidence also clearly established that plaintiff had very little control over the manufacturing processes it employed in making its product. For example, plaintiff encountered substantial variations in the weight of the web coming from the duoform. The degree of variation actually experienced in the weight of the web coming off the duoform, both across its width and along the length of the web, is amply demonstrated in Exhibit 100, the measurements taken of the actual rolls of finished product supplied by plaintiff (Exhibits 237 to 245), the laboratory test report of Ruberoid Corporation (Exhibit 225) and the field engineering report of *478 Draving prepared after his visit for the purpose of correcting the side-to-side unevenness in the web coming off the duoform which was giving a non-uniformly thick finished product (Exhibit 38). The deviation shown in Draving’s report is 36.5%. This variation increases over the duration of the operation of Fibres to a high of 61% in Exhibit 245. Exhibit 225 contains a laboratory report from Ruberoid, one of plaintiff's customers, under date of February 12, 1968, which was submitted in support of a cancellation of further business. This laboratory report discloses Ruberoid found substantial variations in the weight of the product. The five rolls tested weighed from a high of 131 pounds per 1000 ft.2 to a low of 92 pounds per 1000 ft.2 This is a weight variation of 42.4%. A comparison of each square foot sample shows a maximum variation in weight of 66.5%. It should be noted that at the specification weight of 38 grams, 1000 square feet should weigh 83.77 lbs. of goods, it is no wonder that plaintiff was unable to make a profit. The variation in weight is also found in the production report of plaintiff (Exhibit 230) where a 37% variation appeared on January 19, 1968. A weight variation of 47% was present during 1967 when Maleno made his report (Exhibit 100). In contrast, the substantial weight variation did not exist in 1966 just shortly after the duoform had been cleaned and adjusted as set forth in Exhibit 48 and Exhibit 38. The maximum weight variation shown in Exhibit 48 is 11.5%. There were large weight variations in Exhibits 237 through 245. The weight variations within the rolls of manufactured batts were from 15.3% in Exhibit 238 to a high of 61.7% in Exhibit 245. Carpetmate samples supplied by plaintiff and contained in Exhibits 305 and 306 also contained substantial weight variations. Measurements taken within one foot squares selected at random from within the sample indicated that 4 inch squares within the one foot squares varied in weight up to 51.4%. Further evidence of actual variations heavier than specifications is found in the comparison of actual weights and thickness of products to actual specifications (Exhibit 214). The comparison is tabulated as follows: Exhibit 214 Carpetmate Maximum Weight Exhibit Product in grams/ft2 Number Special 55 g 211 Percentage Actual Weight Variation of Exhibit Over Maxi-in grams/ft 2 mum Weight 58.4 g 6% Special 55 g 207 104 g 89% Standard 68 g 212 88.4 g 30% Standard 68 g 206 132 g 94% Supreme 98 g 208 132 g 35% The evidence is abundantly clear that plaintiff encountered 40% to 60% variations in the weight of resin and fiber within any given batt, and that the variation existed both across and lengthwise in the batt. The dryer receives what is fed into it and cannot readjust the distribution of fibers and resin coming from the duoform. The plaintiff takes the position that since it could not consistently produce a pad which had a thickness tolerance Vzz of an inch, the dryer must have been negligently designed. When taken in the light of the evidence before this court, such a position is untenable. From a consideration of both contracts, as well as the surrounding facts and circumstances, the court concludes that it was the intent of the defendant to write its specifications in terms of deflection of the conveyor flights. *479 Stark v. Budwarker (1970), 25 Mich.App. 305, 181 N.W.2d 298; Wistrand v. Bese (1970), 23 Mich.App. 423, 178 N.W.2d 826. The tolerance in the contracts is plus or minus V&2 of an inch for each convey- or plate. Since there are two conveyors involved, the tolerance is duplicated for each conveyor. In any event, the tolerance was not guaranteed because the word “guarantee” was not used in the machinery description by agreement of the parties (see Opinion and Order of December 1, 1972, pages 456-458). At the time the contracts were entered into, plaintiff made no request that defendant guarantee the thickness tolerance of the padding to be produced. Steuernagel testified the machine description was Proctor’s and not in response to any requirement which he had stated. The Ford Specifications (Exhibit 2) carry no thickness tolerance. The products therein described are the products which plaintiff claims it purchased the dryers to manufacture. Plaintiff, however, admits it never produced any such product for the purpose stated in said Specifications. Furthermore, plaintiff was actually told that a tolerance of ± %2 was not able to be held by the dryer, thus it did not rely upon the fact that the dryer could or could not so perform. Moreover, the dryers did, when properly operated with a properly adjusted duoform and resin system, produce batts within ± 1/32 of an inch. This condition existed at the time of Shegda’s report (Exhibit 48) and the General Motors Laboratory Report (Exhibit 200). Furthermore, plaintiff acknowledged that it was manufacturing “ . . . within Vie of any thickness required throughout the pad” (Exhibit 68). The effect upon finished thickness of the batt of uneven resin distribution is clearly shown in Exhibit 310. The sample was especially prepared with a gradient of resin heaviest on one side diminishing to none on the other. After curing between parallel heated platens, the sample had a wedge shape with the thickest part of the wedge being in the area with the least resin. An examination of said Exhibit shows the resin to be entirely cured throughout the pad. There is ample evidence to support the finding of fact that there was uneven resin distribution in much of plaintiff’s product. In Exhibit 267, Henderson notes Clapp’s concern regarding inconsistency in firmness and is hopeful that Buchan can make a contribution in the area of even distributions of resins. Henderson’s concern was again voiced in Exhibit 274 where he commented that “ . . . the rolls contained air streaks which, due to our air laying process, are our primary nemesis.” Steps were taken to correct the problem of uneven resin distribution with the purchase and installation of BIF applicators (Exhibit 278). The testimony clearly established the effect upon the uniformity which resulted from moving the point of resin introduction from the intermediate doffer to the elephant trunk of the duoform. The problem of resin distribution also was present in Exhibit 222 where a customer of plaintiff complained of the “considerable variation in stiffness of the product.” This complaint of a lack of uniformity in firmness was likewise expressed in Exhibit 225 where the customer commented, “ . . . this material was highly inconsistent to such a point that half the sample had the appearance of a rigid board while the other half was ‘fluff’.” The facts as well as logic dictate that plaintiff cannot be correct in its contention that the alleged lack of uniform thickness in the finished product was caused by the negligent structural and mechanical design of the subject dryers. The plaintiff claims that its inability to manufacture a product which was commercially acceptable in terms of uniformity of thickness was intermittent and not continual during the three and one-half year period of operation. It is axiomatic that any deficiency in the de *480 sign of the subject dryers would be a permanent deficiency and not intermittent. For example, if the jack screw support structure was too weak, it would always be too weak and the dryer would never be capable of producing a commercially acceptable product. Accordingly, the intermittent commercially acceptable product must be the result of intermittent causes which are independent of the unchanging design of the dryers. The record is replete with the intermittent feeding of the duoform and innovative dry resin system in terms of substantial variations in the distribution of fibers and resin as fed to the dryer. The record is likewise replete with evidence that variations in fiber content and resin distribution directly affect the thickness of the finished product. The dryer receives what is fed into it. These facts further lead the court to conclude that it was plaintiff’s negligence which was responsible for its inability to produce a commercially acceptable pad, rather than any negligence on the part of the defendant. Therefore, judgment must be entered against the plaintiff. Defendant has counterclaimed against the plaintiff U. S. Fibres, Inc., for payment of an account stated. The evidence including Exhibits 322 and 322-A, clearly establish that defendant is entitled to recover $17,475.00 upon the account stated, plus interest from the date of filing suit, plus statutory costs and attorney fees. Defendant has also filed a claim against both U. S. Equipment Company and U. S. Fibres, Inc., based on fraud. In effect, defendant claims that they actively concealed the fact that it was overloading the dryers to induce the defendant to perform repairs without charge. Generally to constitute actionable fraud, the following elements must be established: 1. A material misrepresentation of an existing fact; 2. The misrepresentation was made with the intention that it be relied on; and 3. The misrepresentation is in fact relied upon and results in damage. Dur-ram Packaging v. Self-Seal, 18 Mich.App. 81, 85, 170 N.W.2d 473 (1969); Columbus Pipe and Equipment Co. v. Sefansky, 352 Mich. 539, 90 N.W.2d 492 (1958). Fraud can be either actual or constructive. To constitute actual fraud there must be an intentional deception. In the present case the evidence clearly established that there was no intentional deception on the part of either U. S. Fibres, Inc., or U. S. Equipment Company. Liability for constructive fraud may be based on a negligent or even an innocent misrepresentation. 37 Am. Jur.2d 22 § 4; Aldrich v. Scribner, 154 Mich. 23, 117 N.W. 581; Foster Machine Co. v. Covel Mfg. Co., 219 Mich. 455, 189 N.W. 228 (1922); Irwin v. Carlton, 369 Mich. 92, 119 N.W.2d 617 (1963). In Irwin v. Carlton, supra, the court stated, at page 97, 119 N.W.2d at page 619: “ . . . plaintiff was entitled to have the jury instructed, as he requested the trial judge to do, that defendant might be found guilty of actionable misrepresentation even though the representation was made in good faith believing it to be true.” In the present case there is no evidence which establishes that either company did, in fact, know that U. S. Fibres was misusing the dryers (e. g. overloading them) and nevertheless represented to the defendant that it was not, thus inducing defendant to repair the dryers at its own expense. Therefore, there can be no recovery based on an express, innocent or negligent representation. Under some circumstances, even if there were no active misrepresentations, there may still be liability for fraud. In Michigan, if one party to a contract or transaction has knowledge which is not within the fair and reason *481 able reach of the other party and which he could not discover by the exercise of reasonable diligence or means of knowledge which are not open to both parties alike, he is under a legal duty to speak, and his silence constitutes fraud, especially when the other party relies upon him to communicate to him the true statement of facts to enable him to properly act under the agreement. 37 Am.Jur.2d, Fraud & Deceit § 146, 148; Wolfe v. A. E. Kusterer & Co. (1934), 269 Mich. 424, 257 N.W. 729; Groening v. Opsata (1948) 323 Mich. 73, 34 N.W.2d 560; Nowicki v. Podgorski (1960), 359 Mich. 18, 101 N.W.2d 371. However, even based on this theory, defendant has failed to establish its claim of fraud. It is true that U. S. Fibres was overloading the dryers. Nevertheless, defendant has failed to prove that at the time this occurred either company knew that U. S. Fibres was overloading the dryers.. In fact, the evidence shows that U. S. Fibres did not know how much compressive force it was using during production of the padding. Moreover, defendant has failed to establish that it could not have discovered the overloading by the exercise of reasonable diligence, or that the means of discovering (this fact was not open to both parties alike. Thus, it has not been shown that there was a legal duty to speak and that silence constitutes fraud. Therefore, defendant may not recover on Count II of its counterclaim, which is based on fraud. In view of all of these considerations, it is ordered that judgment be and hereby is entered against plaintiff, U. S. Fibres, Inc., of no cause of action, and it is further ordered that a judgment be and hereby is entered for defendant, Proctor & Schwartz, Inc., on its counterclaim on an account stated for $17,475.-00 against plaintiff, U. S. Fibres, Inc., together with interest, and statutory costs and attorneys fees; and it is further ordered that a judgment of no cause of action be entered against defendant Proctor & Schwartz, Inc., on its counterclaim against plaintiff U. S. Fibres, Inc., and third-party defendant U. S. Equipment Company.