This appeal involves a personal injury action grounded in strict liability.
On January 10, 1979, Michael Troja accidentally amputated his thumb while he was operating a radial arm saw manufactured by Black and Decker Manufacturing Company, Inc. He instituted suit in the Circuit Court for Anne Arundel County against Black and Decker in both negligence and strict liability. The negligence count was voluntarily withdrawn prior to trial. The strict liability counts were based on two theories:
1) The absence of a safeguarding system, which would prevent the saw from operating unless the guide fence was in place, was a design defect which caused the saw to be unreasonably dangerous.
2) The manufacturer’s failure to warn consumers of the risks of using the saw without the guide fence caused the saw to be unreasonably dangerous.
At the close of Troja’s case before Judge Raymond G. Thieme and a jury, the judge granted Black and Decker’s motion for a directed verdict on the defective design portion of Troja’s strict liability count. Judge Thieme ruled that Troja had failed to produce any legally sufficient evidence of the economic feasibility of a proposed alternative radial arm saw design, or of the existence of the technology necessary to produce such a product in 1976, the year the particular saw was manufactured.
The failure to warn issue was submitted to the jury on seven questions. The jury found that the absence of special warnings at the time the saw was marketed caused the saw to be defective and that the saw was unreasonably dangerous. Nevertheless, the jury concluded that the manufactur*106er did not know of the defect at the time it placed the saw on the market. Based on those findings, the judge entered a judgment in favor of Black and Decker.
Troja argues that the trial court: 1) abused its discretion in precluding an expert witness from stating an opinion in regard to the economic and technological feasibility of a proposed safeguard for the radial arm saw; 2) erred in directing a verdict on the issue of design defect based on the court’s findings that Troja failed to produce legally sufficient evidence to warrant submission of that issue to the jury; 3) abused its discretion in refusing to admit evidence of warnings the manufacturer placed on its saw subsequent to the marketing of the saw that caused Troja’s injury; 4) abused its discretion in not allowing cross-examination of the appellee’s expert as to warnings Black and Decker provided on models post-dating the saw in the matter at bar; and 5) erred in admitting testimony of Black and Decker’s compliance with federal and industry standards as they existed in 1976.
The alleged villain in this litigation is a twelve inch radial arm saw, DeWalt Model No. 780, manufactured by Black and Decker in 1976. Troja had borrowed the saw from Robert Krohn, who had hired Troja to build a bar. Troja and Krohn removed the saw from its metal base and stand in order that it could be carried from Krohn’s basement to the work site in the bar. The guide fence and metal base were left behind. Bereft of its base, the saw was placed directly on the floor. Troja rigged a makeshift guide fence by securing an aluminum level to the saw with two “C” clamps. At the time of his injury, Troja was using the saw to make a cross-cut. He had dispensed with his makeshift fence and guided the wood into the saw blade with his bare hand.
A “DeWalt Instruction & Maintenance Manual” accompanied the 1976 DeWalt Model No. 780. The manual contained instructions, illustrated with photographs, on the subjects of assembly, operation, and maintenance of the *107saw. The manual included instructions for performing various types of cuts. The proper procedure for executing a cross-cut was explained and illustrated.
The Design Defect
Troja contends that because the saw was designed so that the guide fence was easily removable, the absence of a safety feature that would prevent the saw from running when the fence was not in place rendered it unreasonably dangerous.
Maryland, in Phipps v. General Motors Corp., 278 Md. 337, 363 A.2d 955 (1976), embraced the theory of strict liability in tort actions. There the Court expressly adopted the elements contained in the Restatement (Second) of Torts, § 402A (1965). That section provides:
“(1) One who sells any product in a defective condition unreasonably dangerous to the user or consumer or to his property is subject to liability for physical harm thereby caused to the ultimate user or consumer, or to his property, if
(a) the seller is engaged in the business of selling such a product, and
(b) it is expected to and does reach the user or consumer without substantial change in the condition in which it is sold.
(2) The rule stated in Subsection (1) applies although
(a) the seller has exercised all possible care in the preparation and sale of his product, and
(b) the user or consumer has not brought the product from or entered into any contractual relation with the seller.”
Section 402A requires a court, in a design defect case, to weigh “the utility of risk inherent in the design against the magnitude of the risk.” Phipps v. General Motors Corp., 278 Md. at 345, 363 A.2d at 959. The court, Phipps tells us, ought to implement a balancing process to *108decide whether the product in question was unreasonably dangerous.
One helpful guide to the balancing process was recommended in Wade, On the Nature of Strict Tort Liability for Products, 44 Miss.L.J. 825, 837-38 (1973). Wade suggests seven factors that should be weighed in determining whether a given product is “reasonably safe.” Those factors are:
“(1) The usefulness and desirability of the product — its utility to the user and to the public as a whole.
(2) The safety aspects of the product — the likelihood that it will cause injury, and the probable seriousness of the injury.
(3) The availability of a substitute product which would meet the same need and not be as unsafe.
(4) The manufacturer’s ability to eliminate the unsafe character of the product without impairing its usefulness or making it too expensive to maintain its utility.
(5) The user’s ability to avoid danger by the exercise of care in the use of the product.
(6) The user’s anticipated awareness of the dangers inherent in the product and their avoidability, because of general public knowledge of the obvious condition of the product, or of the existence of suitable warnings or instructions.
(7) The feasibility, on the part of the manufacturer, of spreading the loss by setting the price of the product or carrying liability insurance.”
See also Phipps v. General Motors Corp. 278 Md. at 345, n. 4, 363 A.2d at 959, n. 4.
In some instances the risk is “inherently unreasonable,” and no balancing test is necessary. An example of an inherently unreasonable risk is where, as in Phipps, the gas pedal of a new automobile suddenly and without warning sticks, causing the vehicle to accelerate.
The failure of the manufacturer, in the case sub judice, to incorporate a safety system such as the one *109proposed by Troja is not an inherently unreasonable risk. Therefore, in order to create a jury issue on the liability of Black and Decker because of a defective design, Troja was required to produce evidence from which the jury could determine the former’s unreasonableness in manufacturing a saw, in 1976, without a safety system. In Singleton v. International Harvestor Co., 685 F.2d 112 (4th Cir.1981), the Fourth Circuit held that in order to carry a case to the jury, the evidence should show: the technological feasibility of manufacturing a product with the suggested safety device at the time the suspect product was manufactured; the availability of the materials required; the cost of production of the suggested device; price to the consumer, including that of the suggested device; and the chances of consumer acceptance of a model incorporating such features. Id. at 115-16. The design alternative proposed in Singleton was a roll-over protective structure for a tractor which had upended or toppled, trapping the operator beneath it. Because Singleton’s experts failed to provide the foundation information required, the court affirmed the trial court’s directed verdict on the issue of defective design. Id. at 116.
Troja contends that Judge Thieme erred when he refused to allow an expert in the field of “machine guarding safety systems” to testify that a radial saw design, incorporating a safety device described as an “interlock system,” could have been developed in 1976. Gerald Rennell, who was offered by Troja as an expert, testified that he had taken courses in machine guarding and industrial safety. He said that he had been employed as a “safety engineer” and a “loss-control inspector.” Although Rennell suggested that Black and Decker could have incorporated a “safety interlock feature,” which would have prevented the saw from running when the guide fence was not in place, he acknowledged that he had no experience in radial arm saw design. Rennell was unable to furnish a design demonstrating the actual placement of such a system, or to explain how it could be integrated in the saw without interfering with the *110functions for which the fence would normally not be employed. The expert’s bald statement that a safety interlock device could be implemented without great cost to the manufacturer was not supported by any data regarding the cost of the materials necessary to include such a feature. Moreover, Rennell did not conduct any tests to determine the feature’s actual utility. See Singleton at 115-116. Despite the trial court’s pointed comments that it would exclude Rennell’s testimony as to the feasibility of incorporating the proposed safety device, unless the expert established a foundation, Troja opted not to question his witness with respect to those areas. Inasmuch as there was an absence of a proper foundation for the expert’s statement relative to a safety interlock device, the trial court excluded all testimony regarding the economic/technological feasibility in 1976 of producing a saw with the proposed safety system. We agree with Judge Thieme. The record simply does not support Troja’s assertion that the trial court incorrectly excluded Rennell’s testimony.
The decision to admit or exclude “expert” testimony is within the broad discretion of the trial court and that decision will be sustained on appeal unless it is shown to be manifestly erroneous. I.W. Berman Properties v. Porter Brothers, Inc., 276 Md. 1, 344 A.2d 65 (1975). To qualify as an expert, a witness must have such skill, knowledge or experience in the field in question that his opinion will aid the trier of fact. Pennsylvania Threshermen & Farmers’ Mutual Casualty Insurance Co. v. Messenger, 181 Md. 295, 298-99, 29 A.2d 653, 655 (1943). The expert must have knowledge of the facts necessary to support his opinion. Fink v. Steele, 166 Md. 354, 363, 171 A. 49, 53 (1934); Uhlik v. Kopec, 20 Md.App. 216, 223, 314 A.2d 732, 737, cert. denied, 271 Md. 739 (1974). The total absence of any information caused Rennell’s testimony regarding the alleged feasibility of an alternative design to be a mere proposal, unsupported by evidence. Judge Thieme did not abuse his discretion in declining to admit Rennell’s testimony.
*111When we view the evidence and all the inferences which may reasonably be drawn therefrom in a light most favorable to Troja, we are left with the inescapable conclusion that there simply was no legally sufficient evidence of design defect for the jury to consider. Dix v. Spampinato, 278 Md. 34, 358 A.2d 237 (1976). The trial court properly granted the manufacturer’s motion for a directed verdict on the issue of design defect.
The Failure to Warn
Troja’s sole remaining ground for recovery, in the wake of the directed verdict against him on the issue of design defect, was that the warnings supplied with the DeWalt 1976 model were inadequate. He maintains that because of the manufacturer’s knowledge in 1976, the year in which the saw was manufactured, the warnings included in the manual with the saw were insufficient. Specifically, Troja contends that Black and Decker’s failure to affix a label cautioning that it could be hazardous to use the saw without the guide fence made the saw unreasonably dangerous.
To bolster his position, Troja produced the expert testimony of Dr. Robert Cunitz, who testified extensively on the factors generally used to design warning systems and the criteria he used in evaluating existing warning system devices. Based on his examination of the 1976 model owner’s manual that was provided with the unit, Dr. Cunitz deduced that then existing warnings were inadequate.1 He *112said that the absence of a specific warning to persons using the saw without the guide fence could cause serious injuries, and that lack made the saw unreasonably dangerous. Cunitz admitted that he had not performed any surveys of the type of warnings used by other manufacturers’ of radial arm saws marketed in 1976. He agreed, however, with Rennell’s testimony that the warnings provided by Black and Decker conformed with 1976 federal and industrial standards. Nevertheless, Cunitz opined that those standards were insufficient because they failed to address the particular hazard that was present in the case at bar. Although Cunitz stated that accident and injury information “should have” existed to corroborate his opinion, he conceded that Black and Decker may have been unaware of the data. The precise nature of that evidence was never clarified by Cunitz or any other witness. Notwithstanding a direct admonition by the court that it was concerned with the status of the technology as it existed in 1976, not 1984, Troja did not produce that evidence. See Singleton v. International Harvester Co., 685 F.2d at 115. Instead, Troja proffered that Black, and Decker’s 1984 model contained a stronger warning than the 1976 model. Specifically, the 1984 warning cautioned the user “never to saw freehand.” Troja argues that he should have been allowed to introduce evidence of remedial warnings existing on subsequent models in order to prove that the manufacturer *113had the knowledge of the need to issue a stronger warning to consumers in 1976.
The admissibility of evidence of subsequent remedial measures in design or warnings in product liability cases appears to be one of first impression in this State. A review of the considerable case law existent in federal reports and those of other States leads us to observe that the bulk of decisions on the subject can be divided roughly into two schools: 1) those which apply Federal Rule of Evidence 4072 (barring the admission of evidence of subsequent remedial measures to prove culpable conduct in strict liability actions), and 2) those holding that Federal Rule of Evidence 407 does not apply in strict liability actions. See Ault v. International Harvestor Co., 13 Cal.3d 113, 117 Cal.Rptr. 812, 528 P.2d 1148 (1974).
Understandably, Troja urges that we adopt the Ault rationale and hold that the trial court’s decision forbidding him to introduce evidence of subsequent warnings was erroneous. We, however, opt to follow the Fourth Circuit’s decision in Werner v. Upjohn, 628 F.2d 848 (4th Cir.1980), because we think its reasoning sounder than that of Ault.
We rely particularly on the language used by the Werner court in rejecting Ault and its progeny:
“The rationale behind Rule 407 is that people in general would be less likely to take subsequent remedial measures if their repairs or improvements would be used against them in a lawsuit arising out of a prior accident. *114By excluding this evidence defendants are encouraged to make such improvements.
The [Ault ] court assumes that the product is defective, and thus overlooks the situation where the product is not defective but could be made better. The manufacturer who undertakes precautionary measures in this setting will face the risk of liability for an injury caused by the earlier nondefective version of the product based on evidence of his subsequent act which made the product safer but in no way supports an inference that the initial version of the product was defective.”
Id. at 857. (Epiphasis added.)
We stand beside Werner and hold that in a strict liability case evidence of subsequent remedial measures is not admissible to prove culpable conduct.
Aside from the policy reasons already discussed, another more traditional ground exists for excluding Troja’s proffered evidence. The standard usually applied in dealing with the admissibility of evidence is whether the probativeness of the evidence outweighs its potential for prejudice. Evidence is relevant if it has a tendency to make the existence of a fact more or less probable than it would be without that evidence. In theory, evidence of any subsequent modification may be relevant, because the inference could be drawn that the product was defective before the manufacturer implemented the remedial measures. Yet the mere fact that evidence may be “relevant” does not assure its admission. When, as here, the “modification” sought to be introduced was not made until approximately seven years after the allegedly defective saw was manufactured, the disparity in time is too great to permit a party to interject the manufacturer’s changes. The introduction of such evidence under those circumstances would have the potential of confusing the jury by diverting its attention from the question of whether the product or warning was defective at the time it was manufactured. See Herndon v. *115Seven Bar Flying Service, Inc., 716 F.2d 1322, 1328-30 (10th Cir.1983), cert. denied, Piper Aircraft Corp. v. Seven Bar Flying Service, Inc., — U.S.-, 104 S.Ct. 2170, 80 L.Ed.2d 553 (1984); Grenada Steel Industries v. Alabama Oxygen Co., 695 F.2d 883, 887-89 (5th Cir.1983). The trial court correctly refused to permit evidence of subsequent modifications.
In embracing Werner v. Upjohn, 628 F.2d 848 (4th Cir. 1980), we explicitly adopt the exception contained in Fed.R. Evid. 407.3 That exception permits evidence of subsequent remedial measures to be introduced “for another purpose, such as proving ownership, control, or feasibility of precautionary measures, if controverted, or impeachment.”
Troja next argues that Black and Decker raised the question of feasibility of remedial measures through the testimony of its own expert, Howard Hershock, and, therefore, Troja should have been permitted to cross-examine Hershock on the issue. This contention is similar to that submitted by the plaintiff in Werner. We reject Troja’s argument and adopt as our reason the language of the Werner court in turning back a similar assertion:
“[Defendant] does not argue that it could not have written a stronger warning [at the time of manufacture], it argues that the [earlier] warning was adequate given the knowledge it had at that time. This defense simply does not raise an issue of feasibility.”
Id. at 855.
Troja next avers that the trial court erred in permitting Black and Decker’s expert to testify that the saw complied with all federal and industry practices and standards in effect on the date of manufacture. Troja contends that the evidence was prejudicial because the jury could infer that if the industry did not require an instruction with respect to free hand cuts, then the power tool industry had no knowledge of the danger involved in making those cuts. He *116submits that the jury might also speculate that Black and Decker, as part of that industry, had no such knowledge.
Troja seemingly overlooks the testimony of his own expert. Rennell told the jury that in 1976 the Black and Decker radial arm saw construction and warnings were in compliance with the federal regulations propounded by the Office of Safety and Health Administration, and the industry standards dictated by the American National Standards Institute (ANSI) and Underwriter’s Laboratory (UL). Furthermore, Dr. Cunitz repeated Rennell’s statements. By introducing that evidence, Troja rendered harmless any error committed by the trial court in overruling the objections to similar testimony by Black and Decker’s expert.
JUDGMENT AFFIRMED.
COSTS TO BE PAID BY APPELLANT.