Per Curiam.

Plaintiff Michael Muniga suffered *757 serious and permanent injuries in an automobile accident on February 7, 1972. His automobile was a 1966 Buick LeSabre manufactured by defendant General Motors Corporation (GMC) and sold to him new by defendant P. L. Grissom & Son Buick, Inc. (Grissom). Plaintiff alleged that while driving southbound on Telegraph Road in Dearborn at a constant speed of approximately 40 m.p.h., his car accelerated suddenly to 60 to 70 m.p.h. He applied the brakes but could not stop the car. In an effort to slow his car, he struck the curb. The car struck another curb and came to rest on a small guardrail on the median. The car was dangling over the northbound lane for 15 to 20 seconds when it was hit by a northbound car.

Plaintiff brought a products liability action against defendants on July 18, 1974. In Count I, plaintiff alleged that GMC negligently designed the automobile by failing to provide adequate engine mounts, failing to provide limiting equipment that would prevent the engine from moving and opening the throttle in the event of an engine mount separation, failing to inspect and test the engine mounts, failing to ascertain the consequences of engine mount separation and failing to warn of defective engine mounts. Count I also alleged negligence by Grissom for failing to inspect for defects and in failing to warn. Count II alleged breach of implied warranty by both defendants due to defects in the engine mounts and lack of equipment to prevent movement of the engine. Defendants denied the material allegations of liability and asserted as affirmative defenses that plaintiff was contributorily negligent and that he had abused or misused the product. A jury trial resulted in judgments of no cause of action in favor of both defendants.

*758 Plaintiff’s theory of the case was that the sudden acceleration was due to an engine mount failure that allowed the engine to rotate, opening the throttle. Defendants maintain that, at 40 m.p.h., insufficient torque was present so rotation of the engine would not occur if, in fact, the mounts had separated before impact through fatigue. Defendant also argued that the engine mounts fractured as a result of an impact in the accident — not through fatigue.

Plaintiff raises several issues on appeal. Plaintiff challenges the admissibility of evidence of certain tests performed by defendants’ expert witness.

The first test, which was performed for use in a different case, showed the effect of a frontal impact on engine mounts. For the test, a Buick 430 cubic inch engine from a Wildcat or Electra was placed into a chassis with no body, exhaust system, or fenders, and a 6,000-pound pendulum was swung to hit the front of the vehicle, resulting in a fracture of the engine mounts. High speed and regular speed cameras made films of this experiment. The engine mounts in this experimental vehicle were similar in configuration and were identical in mechanical properties to those in plaintiff’s 1966 LeSabre, but the mounts were not interchangeable because of differences in size and placement of the mounting hole. Plaintiff’s objection to admission of the film and testimony concerning the test was overruled. The witness compared photographs of stress marks on the rubber from the engine mounts from plaintiff’s car with those of the engine mounts fractured by the impact and with those separated by fatigue. The witness concluded that impact — not fatigue— caused separation of plaintiff’s engine mounts.

The admission of test results rests within the *759 wide discretion of a trial court. Moldovan v Allis Chalmers Manufacturing Co, 83 Mich App 373, 384; 268 NW2d 656 (1978), Pohlod v General Motors Corp, 40 Mich App 583, 585; 199 NW2d 277 (1972). Experimental data are admissible if the tests from which the data are drawn involved conditions sufficiently similar to those involved in the particular case to assist the trier of fact in reaching its conclusions. Smith v Grange Mutual Fire Ins Co of Michigan, 234 Mich 119; 208 NW 145 (1926), Thorp v Dayton Tire & Rubber Co, 51 Mich App 514; 215 NW2d 600 (1974).

The trial court was acting within its discretion in admitting evidence of this test. The test was designed to show only the effect of impact-caused fracturing on the rubber in an engine mount. Defendants laid a proper foundation by showing that the engine mounts in the experiment and in plaintiff’s car had identical properties. One would expect the striations on the rubber in the test vehicle and in plaintiff’s car to form similar patterns if separated from the metal by impact. The differences between the patterns formed by impact and the patterns formed by fatigue were not subtle. The test in no way concerns the amount of force necessary to cause an engine mount to fracture, so the differences between the test vehicle and plaintiff’s automobile were immaterial.

The other test challenged on appeal was designed to measure the torque generated by driving in various conditions. Defendants’ expert witness obtained a car similar to plaintiff’s in all respects except it had no air conditioner and it was well-tuned. The left engine mount was unbolted to simulate plaintiff’s car, assuming its left engine mount had in fact been separated as alleged. The test was performed in July. The expert testified *760 that poorer performance of the engine as a result of not being tuned or of cold weather would have generated less torque so that any differences between the road test and the actual case were biased in plaintiffs favor. The trial court admitted data and testimony regarding this test over plaintiffs objection.

As a result of this test, defendants’ expert concluded that 50 to 55 foot-pounds of torque resulted from driving the car at a constant rate of 40 m.p.h. The car was also driven with sudden acceleration to determine the torque necessary to lift the unbolted engine. The witness found this to be aproximately 450 foot-pounds and indicated that this amount of torque was generated when accelerating suddenly from a stop to 10 m.p.h. The expert witness indicated sufficient torque to lift the engine could not be generated when the car was traveling at a constant speed of 40 m.p.h.

The trial court did not abuse its discretion in admitting the evidence to show the impossibility of plaintiffs theory of the accident. The foundation was more than adequate to establish the validity of the test.

Plaintiff also alleges error in the trial court’s refusal to admit evidence of GMC’s recall of 1965 and 1967 Buick Electras and Wildcats. The purpose of this recall was to install restraining cables that would prevent engine rotation in the event of a motor mount separation. Defendants’ expert, who testified regarding impact tests performed on Electras and Wildcats, stated that mounts from the 1967 models had the same composition and exhibited identical wear characteristics to the mount used in plaintiffs car. The mounts were not, however, interchangeable due to differences in size and placement of holes. Moreover, the various *761 parts in the engine compartment of the test vehicles and plaintiffs car were arranged substantially differently.

After arguments on defendants’ motion to exclude evidence of the recall, the trial court found such evidence to be insufficiently relevant to outweigh the possible prejudice and confusion that could result from admitting evidence of a recall of a different model car.

As a general rule, a trial court has broad discretion in ruling on the relevancy of evidence submitted at trial, Wilson v W A Foote Memorial Hospital, 91 Mich App 90, 96-97; 284 NW2d 126 (1979), Orquist v Montgomery Ward, 37 Mich App 36, 41; 194 NW2d 392 (1971). This Court will not reverse a trial court’s decision absent an abuse of discretion. Wilson, supra, 97.

MRE 401 defines relevant evidence as that which has any tendency to make "the existence of any fact that is of consequence to the determination of the action more probable or less probable than it would be without the evidence”. Plaintiff sought to offer the recall evidence to show that the motor mounts on plaintiffs car, like the motor mounts recalled, could separate "spontaneously, go into full throttle” and create a hazard to the public.

Evidence of the recall would have no relevance as to whether a separated engine in plaintiffs car could rotate, thus opening the throttle. The potential for rotation is determined not only by the engine mounts, but also by the parts adjacent to and adjoining the engine itself. Where testimony established that the recalled models had substantially different engine compartment layouts than plaintiffs model, evidence of the recall would have no bearing on the ability of plaintiff’s engine to rotate and go into full throttle.

*762 Plaintiff also sought to prove, however, that the engine mounts on his car had the potential for separating through wear and fatigue, rather than only through impact. Evidence of the recall of similar engine mounts would be relevant in establishing this fact. This fact was not, however, disputed at trial, for defendant’s expert admitted that plaintiff’s motor mount could, in fact, separate through fatigue. The disputed evidence had no probative value with respect to the probability of preaccident separation of motor mounts. We therefore find no reversal error in excluding such evidence under MRE 403, which permits relevant evidence to be excluded on grounds of prejudice, confusion or waste of time.

Defendant also raises several objections to the instructions as given. No specific objections were made to the instructions on weight and exposure to known risks and no manifest injustice has been shown, so any error that might be alleged in these instructions is not properly before this Court. Javis v Ypsilanti Board of Education, 393 Mich 689; 227 NW2d 543 (1975), Reed v Stretton, 69 Mich App 519, 523; 245 NW2d 117 (1976). The instruction that there is no duty to warn of a general possibility of danger, when read in context with the other instructions on burden of proof, is not an inaccurate statement of the law. Fisher v Johnson Milk Co, Inc, 383 Mich 158, 161; 174 NW2d 752 (1970). Instructions on misuse and abuse of a product were supported by evidence that plaintiff himself had installed a rebuilt carburetor on his automobile, had experienced carburetor trouble just before the accident and was driving the car to have his carburetor replaced when the accident occurred. There was no reversible error in instructing on these defense theories.

*763 Plaintiff also argues that the trial court erred in instructing the jury on contributory negligence. Plaintiff made no specific objections to instructions on contributory negligence. As the conduct of the plaintiff, which defendant contended was negligent and proximately contributed to the accident, would have been the sole proximate cause of the accident, such an erroneous instruction would not have affected the jury’s verdict and must be regarded as harmless. Moore v Foster, 96 Mich App 317; 292 NW2d 535 (1980).

Affirmed. Cost to defendants.