OPINION
W. KENNETH LAW, Chief Justice.
Tiburcio Ledesma, Jr. sued Ford Motor Company for damages in connection with a collision between Ledesma’s truck and two parked cars. The critical factual dispute at trial was whether the truck’s axle dislodged and caused the truck to hit the parked cars, or whether the axle dislodged as a result of the collisions. Finding that a manufacturing defect was a producing cause of the accident, and that Ledesma’s negligence was not a proximate cause, the jury awarded Ledesma $215,380 in damages for the cost of repairs to his truck, the loss of use of the truck, profits lost from the period the truck was out of service for his business, and damage to his [81]*81credit reputation.1 The trial court also awarded prejudgment interest.
Ford raises procedural and substantive complaints. It argues that the court erred by allowing Ledesma’s experts to testify because their opinions were not reliable, and by excluding some testimony by Ford’s expert on the basis that he was not a qualified accident reconstructionist. Ford also argues that the court submitted incorrect definitions of manufacturing defect and producing cause to the jury. Ford also contends that legally insufficient evidence supports the jury’s finding that Ledesma’s truck contained a manufacturing defect that was a producing cause of this accident.
We will affirm the judgment.
BACKGROUND
Ledesma bought a new 1999 Ford F-350 pickup truck in March 1999 for use in his construction business. The truck had dual rear tires housed in wheel wells that protruded from the sides of the truck. In June 1999, the truck had about 4,100 miles on it, had not been in any collisions, and had no mechanical problems. On June 5, 1999 at approximately 5 p.m., Ledesma had an accident while driving the truck on a dry, two-lane street on which several cars were parallel parked. At some point during the accident sequence, the truck’s rear leaf-spring and axle assembly came apart and the driveshaft disengaged from the transmission. The truck collided with at least two of the parked cars. The parties presented different accounts of the accident and different theories of causation. The jury resolved this dispute in Ledesma’s favor.
Ledesma’s version
Ledesma’s theory of the accident is that the rear axle displaced and the driveshaft disengaged and “pronged” into the street, causing him to lose control of his truck and crash into the parked cars.
Ledesma testified that he was driving the truck at a little more than thirty miles per hour. He felt a jolt while shifting into third gear. The truck slid to the right and its back end went into the air. Ledesma said he then turned the steering wheel to the left to avoid a Pontiac Firebird parked along the curb. The right back end of his truck struck the left back end of the Fire-bird. Ledesma said he was unable to control the truck because the rear end kept “bouncing.” The truck then hit a second parked car before coming to rest with its right rear tire up on the curb. Ledesma denied that the truck stopped on anyone’s lawn.
Geert Aerts, a mechanical engineer, testified about two possible causes of the accident. Both theories arose from a manufacturing asymmetry involving the u-bolt that clamps the axle in suspension below the leaf-spring assembly. This schematic of the assemblies was an exhibit at trial:
[82]*82[[Image here]]
Aerts testified that, during the manufacturing process, one of the legs on the u-bolt was made longer than the other leg. Aerts theorized that the unequal lengths of the legs on the u-bolt prevented the nuts from being adequately tightened. As a result, the u-bolt did not adequately clamp the axle housing. Aerts’s second theory was that the flattened portion of the u-bolt was off center. As a result, the round portion of the bolt could cause an indentation in the axle housing and eventually cause the axle housing to become loose. Under both theories, the loose axle housing shifted back and allowed the driveshaft to disengage and fall onto the street.
David Hall, an accident reconstructionist, testified based on photographs taken of the accident scene. Hall said that there were marks present in the street that were caused by the driveshaft striking the pavement. On the truck’s path, these marks were before the collision with the first parked car, the Firebird. Hall also testified that the impact between the truck and the Firebird occurred at a very low speed. Ford’s version
Ford’s theory is that Ledesma was driving inattentively and hit the parked cars with the truck’s protruding wheels, sending the truck out of control, careening into cars and over a curb onto a lawn. Ford contends that the impact with the curb caused the axle to shift back and the drive-shaft to disengage.
Edward Pylant, the owner of the Fire-bird that was hit, testified by deposition that he witnessed the accident.2 Pylant said that Ledesma was speeding at between 40 and 50 mph. Pylant testified that Ledesma was not paying attention and drove very close to the cars parked along the curb. As the truck passed the Firebird, the right wheel tub, the rear axle, and the outside right rear tire of the truck hit the Firebird. Pylant claims the impact caused the truck’s back end to [83]*83come off the ground and bounce away from the Firebird. The truck then hit a Honda Civic. The truck’s wheels hit the curb, went over the curb, and separated from the truck.
Based on Ledesma’s pretrial motion, the court excluded some testimony from Dan May, Ford’s expert witness. May is a mechanical engineer with thirty-one years of experience at Ford, including twenty-five years as a truck engineer. For seven years, May worked as a product design engineer and was involved in the manufacture and assembly of components. The trial court allowed May to testify about matters relating to design and manufacture but prevented him from testifying on matters relating to accident reconstruction because he was not an accident reconstruc-tionist. During trial, the court said it would let Ford question May about the cause of the accident but would then admit the report of Ralph Stern, Ford’s original accident reconstruction expert. In his report, Stern opined that the rear spring rear spring bolt broke, causing the rear wheels to slip far enough to pull the drive shaft out of the transmission. Ford elected not to have May testify about the cause of the accident.
The jury verdict
The trial court returned a verdict in favor of Ledesma. The jury concluded that a manufacturing defect was the cause of the accident and that Ledesma was not negligent. The jury awarded Ledesma $215,380 as damages for the repair of the truck, loss of the use of the truck, Ledes-ma’s lost profits, and damage to Ledesma’s credit reputation. In addition, the trial court awarded Ledesma $65,912 in prejudgment interest.
DISCUSSION
Ford raises four issues on appeal that relate to Ledesma’s failure to carry the burden of proof that should be imposed upon a plaintiff alleging a manufacturing defect. In Ford Motor Co. v. Ridgway, the Texas Supreme Court described the plaintiff’s burden as follows:
A manufacturing defect exists when a product deviates, in its construction or quality, from the specifications or planned output in a manner that renders it unreasonably dangerous. Torrington Co. v. Stutzman, 46 S.W.3d 829, 844 (Tex.2000); American Tobacco Co. v. Grinnell, 951 S.W.2d 420, 434 (Tex.1997). A plaintiff must prove that the product was defective when it left the hands of the manufacturer and that the defect was a producing cause of the plaintiff’s injuries. Torrington Co., 46 S.W.3d at 844.
135 S.W.3d 598, 600 (Tex.2004). Ford complains that Ledesma’s experts did not supply such evidence, that the court erroneously excluded some of its expert’s testimony, that the court’s instruction did not require Ledesma to prove all of the elements of his cause of action, and that the evidence was legally insufficient to support the judgment.
Expert witnesses
Expert witnesses are permitted if scientific, technical, or other specialized knowledge will assist the trier of fact to understand the evidence or to determine a fact issue. Tex.R. Evid. 702. Experts may be qualified by knowledge, skill, experience, training, or education. Id.
Upon objection by the opponent of the evidence, the proponent of expert testimony has the burden to prove that the evidence is admissible. E.I. du Pont de Nemours & Co. v. Robinson, 923 S.W.2d 549, 557 (Tex.1995). A two-part test governs whether expert testimony is admissible: (1) the expert must be qualified; and (2) the testimony must be relevant and be [84]*84based on a reliable foundation. Helena Chem. Co. v. Wilkins, 47 S.W.3d 486, 499 (Tex.2001). The proponent must show that the witness has special knowledge on the subject of his testimony. Gammill v. Jack Williams Chevrolet, Inc., 972 S.W.2d 713, 718 (Tex.1998). The testimony must be shown to be reliable before it is admitted. Id. at 726. The trial court is required to assess the reliability — not the truth or falsity — of the expert’s opinion. Robinson, 923 S.W.2d at 558. The criteria for assessing reliability vary depending on the type of expert and the nature of the evidence. Gammill, 972 S.W.2d at 726-27; see also Kumho Tire Co. v. Carmichael, 526 U.S. 137, 150, 119 S.Ct. 1167, 143 L.Ed.2d 238 (1999). In some cases, an expert’s experience alone may provide a sufficient basis for such testimony. Gammill, 972 S.W.2d at 726. But the court is not required to admit an expert’s opinion merely because the expert opines that his conclusion is valid. General Elec. Co. v. Joiner, 522 U.S. 136, 146, 118 S.Ct. 512, 139 L.Ed.2d 508 (1997); Merrell Dow Pharm., Inc. v. Havner, 953 S.W.2d 706, 712 (Tex.1997). There must not be too great an analytical gap between the data or observations and the expert’s conclusions. Joiner, 522 U.S. at 146, 118 S.Ct. 512; Volkswagen of Am., Inc. v. Ramirez, 159 S.W.3d 897, 904-905, 2004 Tex. LEXIS 1429, at *12 (Tex. 2004). The trial court has broad discretion in determining admissibility of an expert’s testimony. Robinson, 923 S.W.2d at 556-57.
We can reverse the district court’s determination regarding admission of the expert testimony only if the trial court abuses its discretion. Id. at 558. We gauge abuse of discretion by whether the trial court acted without reference to any guiding rules or principles. See id. We cannot find an abuse of discretion merely because we disagree with the decision of the trial court. See id.
Ford challenges the trial court’s admission of testimony by Ledesma’s experts, Geert Aerts and David Hall, and the exclusion of some of the testimony from Ford’s expert, Dan May.
Geert Aerts
, Ford does not challenge Aerts’s qualification to give opinions, but contends that the foundational data underlying his opinion is unreliable and that his methodology in arriving at his opinion was flawed.
Aerts is a metallurgical and mechanical engineer. His job is analyzing automotive components and accidents that happen to automobiles. His primary expertise concerns leaf-spring suspensions in trucks. In the five or six years preceding trial, he analyzed 150 failed leaf springs. He and his company began examining leaf springs at the request of a leaf-spring manufacturer that had a problem with its product. The manufacturer made 40-50% of all leaf springs in the United States, including the one on Ledesma’s truck.
The leaf spring is a stack of metal strips that parallel the length of the truck on each side of the truck. On top of each leaf spring is a spring plate, or tie plate, to which two u-bolts are bolted, one on each side of the leaf spring. The legs of the u-bolts extend through and above the tie plate, with the legs secured by a nut tightened onto each leg. There is also a pin that passes through the leaves up through the center of the tie plate. The axle housing rests in the base of the u-bolt, which is flattened to provide a stronger and more stable cradle for the housing.
Aerts opined that the axle displaced and the driveshaft dislodged because a u-bolt in the rear leaf-spring and axle assembly was defectively manufactured. He testified that the u-bolt was manufactured so [85]*85that the legs of the u-bolt were of different lengths and the flattened part of the base of the u-bolt was off center. Aerts found that the nuts on the u-bolts on the spring assemblies were tightened well below factory standards, but said some of the loosening might be attributed to either use or the force of the collision.
Uneven leg length
Aerts testified that the unequal leg lengths of a u-bolt caused the nuts to be tightened unevenly on that u-bolt — thirteen threads down on one leg, and only ten threads down on the other — diminishing its ability to clamp and allowing vibration in the tie-plate holding together the leaf spring and the axle housing. This movement, he opined, eventually broke the center pin of the tie plate and allowed the axle to shift back enough to allow the drive-shaft to separate, fall, and strike the ground, pushing the axle back and causing Ledesma to lose control of the vehicle. Aerts based his theory on measurements of u-bolts, Ford specifications, and his experience.
Aerts measured the u-bolt that failed, a u-bolt that did not fail, and an exemplar u-bolt. He also looked at Ford’s design specifications. He found that the nuts on the u-bolt were not at the specified tightness; he found that they were torqued between 30 and 130 foot-pounds — well short of the 185 foot-pounds specified. On cross-examination, he admitted that this was not conclusive evidence of the tightness at the assembly line because they might have loosened during use or because of the accident. He found that the legs of the defective u-bolt differed from each other by 5 mm; Ford’s design specifications permit only a ±2 mm variance between the legs. Aerts testified that, although the u-bolts twisted, these measurements indicated that they did not stretch during the accident or, if they did, that they each stretched the same amount. He testified that no forces during the accident would cause the nuts to move along the bolt without stripping the threads. Because he found no stripped threads, he concluded that the position of the nuts on the bolts did not change between the manufacture and the accident. He concluded that the u-bolt was outside of Ford’s manufacturing specifications, rendering the truck defective.
Aerts theorized that the defective u-bolt allowed vibration that caused the assembly to fail. He noted dents in the tie plate holding the spring assembly together, indicating to him that the center pin of the assembly was vibrating and knocking the nut on top of the center pin against the tie plate. Aerts testified that the indentation pattern showed that the vibration had been going on for “quite a while” and fatigued the center pin. He said that, when the assembly is properly secured, there is no such vibration and no contact between the nut and the tie plate. He testified that, although the center pin remained in place after the accident, the top of the pin and the nut sheared off in a way inconsistent with a fatigue fracture. He said that the force of the driveshaft falling to the ground and pronging in the street would provide enough force to shear the center pin. He said that the truck did not show damage indicating sufficient force from the collision that would have sheared the pin; he conceded, however, that he is not an accident reconstructionist.
Ford contends that Aerts’s opinion is subjective belief and unsupported speculation because his foundational data was either suspect or absent. Ford complains that Aerts never inspected the torquing equipment that tightened the nuts on the u-bolt, had no idea how deep the sockets on the equipment were, and did not know whether it could tighten the nuts [86]*86sufficiently despite the disparity in the length of the legs. Aerts conceded on cross-examination that Ford’s torquing equipment as described by counsel could accommodate any difference in the length of the legs of the u-bolt and tighten the nuts sufficiently to meet the specifications. He also conceded that the indentations on the tie plate were deeper on the front side, which Ford posits is consistent with the sudden impact with the Firebird. Ford’s expert, Dan May, testified that the sockets on the torquing equipment were deep enough (4 inches) to offset any inequality in the length of the u-bolt legs and that the wrench could tighten the nuts to the requisite torque. Ford complains that Aerts calculated the forces created by the accident without knowing how fast Ledesma was driving or how much damage was done to the Firebird. Ford also complains that Aerts reasoned backward from the conclusion that the u-bolt was defective, rather than forward from assembled facts.
Ford’s challenge to Aerts’s theory does not persuade us that the district court abused its discretion by admitting Aerts’s testimony. Ford’s attacks on Aerts’s theory of why the nuts were not tightened sufficiently do not undermine his observation that the nuts on the u-bolt were insufficiently tightened in a way that allowed the u-bolt to fail. To prove a manufacturing defect, Ledesma need not prove that the manufacturing process is flawed, only that it produced a flawed product. See Ridgway, 135 S.W.3d at 600; see also Parsons v. Ford Motor Co., 85 S.W.3d 323, 329-30 (Tex.App.-Austin 2002, pet. denied). The plaintiff can prove the condition of the product by direct evidence, circumstantial evidence, or inferences. See Ridgway, 135 S.W.3d at 601; Torrington, 46 S.W.3d at 844-45. Aerts measured the failed u-bolt and found that the legs were uneven and that the nuts were not tightened the same depth on those legs or as on other bolts. He found that, after the accident, the nuts were not at the levels of tightness set by Ford’s specifications; although he conceded that the nuts might possibly have loosened during use or the accident, he saw no signs (such as stripping of threads) that the nuts moved suddenly as a result of the accident. He also observed the indented tie plate and concluded that it showed that the rear assembly was loose and vibrating for. quite a while before the accident. He also noted that the center pin of the tie plate sheared off, indicating a sudden failure. Aerts theorized that the deviations from manufacturing specifications, the signs of unusual wear, and the nature of the detachment of the driveshaft all show that the u-bolt on this three-month-old truck was defectively manufactured in a way that caused the accident. Aerts does not appear to have tailored this accumulated data to support his opinion. Cross-examination of Aerts and May’s testimony cast some doubt on Aerts’s theory, but do not render it so unreliable as to make the district court’s admission of Aerts’s testimony an abuse of discretion.
Off-center flattened base of u-bolt
Aerts also testified that the manufacturing asymmetry of the u-bolt manifested in a flattened portion of the base of the u-bolt being off-center by three-quarters of an inch. He testified that flattening the part of the u-bolt surrounding the axle housing helps it mate better with the axle housing and better bear that load. He stated that the flat part being off center caused it to mate improperly with the axle housing. Aerts theorized that the round portion of the u-bolt could have indented the axle housing and relaxed the u-bolt nut torque; he stated alternatively that the u-bolt could have been deformed by the axle, leading to the same reduction in the u-bolt’s clamping power. He asserted that this, by itself or in conjunction with the [87]*87unequal leg length of the u-bolt, could have allowed the axle to move away from the spring assembly and allowing the drive-shaft to disengage.
Under cross-examination, Aerts admitted that he did not know whether the axle housing was actually indented. He conceded that the axle housing could have been hard enough to resist indentation, and that he did not know how hard the axle housing was. He never looked at Ledesma’s truck’s axle housing. Ford’s expert, Dan May, inspected the truck and testified that he did not see any indentation on the axle housing.
Ford argues that Aerts’s theory regarding the effect of the off-center flattening of the u-bolt lacks solid foundation. Ford argues that his failure to research the hardness of the axle housing or to examine the axle itself undermines his theory. Ford contends that the uncontroverted testimony that the axle was undented demonstrates the invalidity of Aerts’s theory.
We conclude that Ford has not shown an analytical gap sufficient to make the district court’s admission of this part of Aerts’s testimony error. Evidence casting doubt on whether Aerts’s theory was realized in this case may affect the credibility of his opinion, but it does not demonstrate that the district court abused its discretion by finding that his theory regarding the effect of an asymmetrically flattened u-bolt was rehable. See Robinson, 923 S.W.2d at 568.
David Hall
Hall testified as Ledesma’s accident reconstruction expert based largely on his review of photographs. He said that marks on the pavement and the damage to the vehicles supported Ledesma’s version of events. Hall testified that his testimony was based on his experience and training, and that his methods were acceptable in the field of accident reconstruction.
Hall said that an enlargement of a photo of the accident scene, taken the day after the accident by Ledesma’s father with a one-use camera, showed a mark on the pavement that Hall believed was consistent with “pronging” by the driveshaft. He conceded that he could not tell how far from the Firebird the marks were because he did not know the focal length of the camera that took the picture. Hall opined that the mark was in advance of the Fire-bird in the direction from which Ledesma’s truck was traveling, thus supporting Le-desma’s view that the driveshaft disengaged before the truck hit the Firebird. Hall did not visit the site until long after the accident and after the marks on the pavement shown in the picture had disappeared. He did not test to see how easy it was to make a mark on the road.
Hall said that the photo of the damage done to the Firebird indicated to him that the impact was at a low speed and that the fender on the side of the truck — not the tire or the wheel — hit the Firebird. He never personally examined the truck or the Firebird involved in the accident, but he did look at a Firebird with a similar body style. He noted that the truck’s fender was made of plastic and the exemplar Firebird was made of metal. Hall testified that the low speed of the collision supported his view that the driveshaft dislodged before the accident, because the impact forces were not sufficient to have caused the driveshaft to dislodge.
Ford contends that Hall’s testimony contains too many assumptions and logical leaps to be rehable. Ford complains that his conclusion about the prong mark was based on his view of a mark in a grainy photo. Ford attacks the reliability of Hall’s assertion that there was actually a mark on the pavement (rather than a flaw in the film) and his decision that this [88]*88mark was made by the truck’s driveshaft. Ford asserts that Hall’s testimony is further undermined by his failure to take measurements, prepare diagrams, or otherwise reconstruct the accident. Ford noted that a police officer who responded to the accident recalled no gouge marks in the pavement.
Ford also attacks Hall’s testimony about the impact on the Firebird. Ford complains that Hall’s failure to examine either vehicle renders speculative his calculations based on the depth of indentation of the Firebird. Ford also complains that Hall failed to consider damage to the Firebird not evident from the pictures, including the left wheel and axle housing being bent, the axle tube and seam being broken, as well as damage to the left rear control arm, the panhard bar, and the left rear shock absorber — all of which amounted to $2568 in repairs. Ford criticizes Hall’s reliance on the picture despite this repair evidence. Ford contends that this deliberate refusal to consider contrary evidence renders Hall’s opinion unreliable.
We do not find any error in the admission of Hall’s testimony. Ford’s criticisms go to the credibility, not the reliability of Hall’s theories. He was subject to vigorous cross-examination regarding the accuracy of his opinions. We find no abuse of discretion in its admission.
Dan May
May testified for Ford regarding its manufacturing process, but the court prohibited him from offering opinions related to accident reconstruction because he was not trained in accident reconstruction. The court also prevented him from giving his opinion that the damage to the u-bolt demonstrated that it broke because of a collision, not simple failure.
Ford argues that the law does not require special training in accident reconstruction and that it need only show that an expert possesses special knowledge on the matter on which he proposes to give an opinion. See Broders v. Heise, 924 S.W.2d 148, 152-53 (Tex.1996). Ford contends that May’s engineering training (bachelor’s and master’s degrees in mechanical engineering) and his experience in the design and manufacture of trucks (31 years at Ford, 25 of which were spent as a “truck engineer”) qualify him to give an expert opinion. Ford particularly touts his years planning the manufacture of leaf-spring and rear-axle assemblies in medium and heavy trucks similar to the parts at issue here and his more than six years as a design analysis engineer, which required him to investigate approximately 120 “events” to determine the root cause of any failure in the vehicles. Ford asserts that this was harmful because May’s testimony would have supported Ford’s theory that the u-bolt failed as a result of direct contact between the tires of the truck and the Firebird, and that the driveshaft did not disengage until after the collision.
We find no abuse of discretion in the exclusion of May’s testimony regarding the cause of the axle displacement. May undisputedly lacks training in accident reconstruction. He has some practical experience in his job with Ford, but his testimony contained sufficient uncertainty to support the court’s decision. May did not explain how the relatively minor external damage to both vehicles squared with his theory that the tire-to-tire collision caused the axle to displace. He repeatedly insisted that the u-bolt stretched during the accident sequence, even though he did not measure it, before eventually conceding that he had no way of knowing whether it stretched. May had no test data to know the likelihood that the collision could cause the axle to displace.
The standard of review defeats Ford’s complaint. The abuse-of-discretion stan[89]*89dard requires that we let court decisions stand so long as the court acted with reference to guiding rules or principles. Robinson, 923 S.W.2d at 558. We may not reverse for abuse of discretion merely because we disagree with the decision of the trial court. See id. The range of discretion means that the same record could support affirming opposite conclusions on the same expert. In this case, no abuse of discretion is shown.
The court’s charge
Ford contends that the district court gave legally incorrect definitions of manufacturing defect and producing cause in the jury charge. Although the court’s definitions track those provided in the pattern jury charge, Ford contends that the pattern jury charge is incorrect. Whether definitions used in the charge misstate the law is a legal question that we review de novo. St. Joseph Hosp. v. Wolff, 94 S.W.3d 513, 525 (Tex.2003). But we reverse for error in a civil jury charge only if, in light of the entire record, the error probably caused the rendition of an improper judgment. Reinhart v. Young, 906 S.W.2d 471, 473 (Tex.1995); see also Tex. R.App. P. 44.1. An incorrect instruction is more likely to cause reversible error with sharply contrasting evidence in a contested trial. Quantum Chem. Corp. v. Toennies, 47 S.W.3d 473, 480 (Tex.2001).
Manufacturing defect
The supreme court has written that “[a] manufacturing defect exists when a product deviates, in its construction or quality, from the specifications or planned output in a manner that renders it unreasonably dangerous.” Ridgway, 135 S.W.3d at 600. A plaintiff must prove that the product was defective when it left the hands of the manufacturer and that the defect was a producing cause of the plaintiffs injuries. Id.
Ford asserts that the court misstated the law because it omitted language that distinguishes manufacturing defects from other conditions rendering a product unreasonably dangerous. The court asked, ‘Was there a manufacturing defect in the 1999 Ford F-350 pickup truck at the time it left Ford’s possession that was a producing cause of the June 5, 1999 incident in question?” The district court then provided the following definition:
A “defect” means a condition of the product that renders it unreasonably dangerous. An “unreasonably dangerous” product is one that is dangerous to an extent beyond that which would be contemplated by the ordinary user of the product, with the ordinary knowledge common to the community as to the product’s characteristics.
Ford requested that manufacturing defect be defined as “a physical departure from the product’s intended design that renders it unreasonably dangerous.” Ford contends that the omission of language requiring proof of deviation from the manufacturer’s plans or specifications omits an essential element of Ledesma’s cause of action on which he bore the burden of proof. Without this element, Ford argues, Ledesma cannot recover from Ford for a manufacturing defect.
The issue presented, therefore, is whether the district court misstated the law by using the term “condition” in the definition instead of Ford’s more detailed requested language: “physical departure from the product’s intended design.” The El Paso court recently faced a strikingly similar challenge to an almost identical jury instruction for which the manufacturer proposed, but was denied, a similar alternative definition of manufacturing defect. Cooper Tire & Rubber Co. v. Mendez, 155 S.W.3d 382, 390-91 (Tex.App.-El Paso 2004, pet. filed). The El Paso court con-[90]*90eluded that the more specific definition requested by the manufacturer was not required by Texas law. Id. at 391. We find no case holding that this charge is erroneous.
Even if the definition were erroneous, we find no harm in this case. Although the evidence contrasts sharply, the clarity of the contrast and the nature of the instructions given and denied makes the error, if any, harmless. Manufacturing defect is the only theory that Ledesma presented. In fact, Ford directly asked Aerts, “The opinion that you’re offering here today is a manufacturing defect related to the U-bolt, is that right?” Aerts responded, ‘Tes.” The only evidence of defect presented to the jury is the deviation of the u-bolt from specifications, which is the essence of manufacturing defect under Ford’s definition. Ledesma’s theories of the accident rely on the manufacturing defect permitting the driveshaft to dislodge abruptly and cause the accident. The jury found Ledesma’s evidence of defect credible and, as a result, found a manufacturing defect. The omission of the words limiting the nature of the defect did riot affect the verdict because the only evidence of defect was that the truck as manufactured departed physically from the manufacturer’s specifications. Even if the district court erred by rejecting Ford’s requested definition, the error did not harm Ford.
Producing cause
A producing cause is “a substantial factor which brings about the injury and without which the injury would not have occurred.” Doe v. Boys Clubs of Greater Dallas, Inc., 907 S.W.2d 472, 481 (Tex.1995). The court defined producing cause in accord with the pattern jury charge as “an efficient, exciting, or contributing cause that, in a natural sequence, produces the incident in question.” Ford objected that this definition does not accurately state the law. Ford asserts that the charge given is wrong because it erroneously replaces “substantial” with the words “efficient,” “exciting,” and “contributing” to modify the term “cause.” Ford also complains that the definition given did not inform the jury that the cause must be one without which the event would not have occurred. In sum, Ford contends that the court should have told the jury that it must find causation-in-fact. See Prudential Ins. Co. v. Jefferson Assocs., Ltd., 896 S.W.2d 156, 161 (Tex.1995).
Ford ignores supreme court precedent using language almost identical to that used in the charge:
Producing cause requires that the acts be both a cause-in-fact and a “substantial factor” in causing the injuries. Union Pump Co. v. Allbritton, 898 S.W.2d 773, 775 (Tex.1995); Prudential Ins. Co. v. Jefferson Assocs. Ltd., 896 S.W.2d 156, 161 (Tex.1995). A producing cause is an efficient, exciting, or contributing cause, which in the natural sequence of events, produces injuries or damages. Haynes & Boone v. Bowser Bouldin, Ltd., 896 S.W.2d 179, 182 (Tex.1995).
Brown v. Bank of Galveston, Nat’l Ass’n, 963 S.W.2d 511, 514 (Tex.1998). Even in the cases cited by Ford, there is no indication that the supreme court has rejected the “efficient, exciting, or contributing” definition quoted in Brown. See generally, Trinity Universal Ins. Co. v. Bleeker, 966 S.W.2d 489, 491 (Tex.1998) (decided less than one month after Brown); Doe, 907 S.W.2d at 481. A case Ford uses to define producing cause as causation-in-fact uses the “efficient, exciting or contributing” definition of producing cause. See Union Pump, 898 S.W.2d at 775. Ford cites no authority rejecting the definition given.
[91]*91We find no error. The district court’s requirement that the cause be one that “produces the incident” is not discernibly different from Ford’s request that the cause be one “without which the event would not have occurred.” Although the terms “efficient, exciting, or contributing cause” do not necessarily conjure precisely the same meaning as the term “substantial factor,” they do not clearly differ significantly; more important, the supreme court has used both to define producing cause. See Trinity, 966 S.W.2d at 491; see also Brown, 963 S.W.2d at 514. Absent a supreme court declaration that “efficient, exciting, or contributing cause” is improper, the district court did not err by using that definition. See Carousel’s Creamery, L.L.C. v. Marble Slab Creamery, Inc., 134 S.W.3d 385, 399 (Tex.App.-Houston [1st Dist.] 2004, pet. filed) (using Brown definition); see also Compton & Adams Warehouse, Inc. v. Powers, No. 04-04-00099-CV, 2004 WL 2346293, *1-*2, 2004 Tex.App. LEXIS 9192, at *4 (Tex.App.-San Antonio Oct. 20, 2004, no pet.) (mem.op.).
Legal sufficiency of the evidence
Ford contends that the evidence is legally insufficient to support the verdict. A party attacking the legal sufficiency of an adverse finding of an issue on which it did not have the burden of proof must demonstrate on appeal that there is no evidence to support the adverse finding. See Croucher v. Croucher, 660 S.W.2d 55, 58 (Tex.1983). In reviewing a no-evidence issue, we are to consider only the evidence favoring the finding, disregarding all direct and circumstantial evidence to the contrary. Lenz v. Lenz, 79 S.W.3d 10, 13 (Tex.2002); Continental Coffee Prods. Co. v. Cazarez, 937 S.W.2d 444, 450 (Tex.1996). We must determine whether the proffered evidence as a whole rises to a level that would enable reasonable and fair-minded people to differ in their conclusions. See Transportation Ins. Co. v. Moriel, 879 S.W.2d 10, 25 (Tex.1994).
Ford’s central contention regarding the lack of evidence is that Ledesma’s experts’ testimony was unreliable and provided no evidence of a manufacturing defect. We have discussed these arguments above. We have concluded that the testimony was sufficiently reliable to support its admission. It is likewise some evidence to support the jury’s verdict.
Ledesma’s testimony provides some evidence to support his theory of manufacturing defect. Evidence of malfunction can provide circumstantial evidence of manufacturing defect. General Motors Corp. v. Hopkins, 548 S.W.2d 344, 349-50 (Tex.1977); see also Ford Motor Co. v. Gonzalez, 9 S.W.3d 195, 199 (Tex.App.-San Antonio 1999, no pet.); Sipes v. General Motors Corp., 946 S.W.2d 143, 155 (Tex.App.-Texarkana 1997, writ denied). The circumstantial evidence must provide more than mere suspicion. Ridgway, 135 S.W.3d at 601. In Ridgway, the supreme court found that circumstantial evidence of the source of a vehicle fire provided no evidence of a manufacturing defect. Id. But the court’s determination centered on facts not present here: the truck had been driven 54,000 miles by previous owners who had modified and repaired the truck, the plaintiff testified only that a fire occurred, and the expert testified that he suspected the electrical system was to blame and could not rule out the fuel system as a cause. Id. Here, Ledesma testified that he was the sole owner, that he had driven the truck 4100 miles in his three months of ownership, and that the truck had not been repaired or modified. In addition to evidence that the driveshaft disengaged, there was evidence that it happened suddenly and while the truck was driving down a clear street. Ledes-ma’s experts pinpointed the u-bolt that did [92]*92not meet manufacturing specifications as the cause, and the disengagement of the driveshaft under the conditions Ledesma described provide some evidence of a manufacturing defect. See Parsons v. Ford Motor Co., 85 S.W.3d 323, 329-30 (Tex.App.-Austin 2002, pet. denied).
Ford provided eyewitness testimony that contradicted Ledesma’s version of events and expert testimony contradicting his experts’ testimony. The jury rejected Ford’s version and believed Ledesma’s evidence. More importantly, the evidence contrary to the verdict is not within the scope of our review under the legal sufficiency standard. See Lenz, 79 S.W.3d at 13.
The record contains more than a scintilla of evidence that Ledesma’s truck differed, in its construction or quality, from the specifications in a manner that rendered it unreasonably dangerous. The record also contains evidence that this defect caused this accident and Ledesma’s damages.
CONCLUSION
The district court did not abuse its discretion by admitting the testimony of Le-desma’s experts and excluding some of the testimony of Ford’s expert regarding the accident. The instructions challenged were not erroneous and, even if they were, did not harm Ford. The record contains legally sufficient evidence to support the judgment. We affirm the judgment.
Concurring Opinion by Justice PEMBERTON.