Donald J. and Catherine A. Duchess appeal from the judgment entered after a jury determined that appellee, Langston Corporation, was not liable in this products liability action. We conclude that the trial court made two erroneous evidentiary rulings; we are constrained to reverse and remand for a new trial.
Appellants instituted this action on August 21, 1991, to recover damages they sustained after Mr. Duchess was injured on May 12, 1990. Mr. Duchess was employed by 4M Manufacturing in Eighty Four, Pennsylvania. On the day in question, he was working on a Langston Saturn III Flexo-Folder Gluer (“Saturn III”) which fabricates corrugated board into corrugated boxes. Specifically, he had removed the ink shield and was cleaning dry ink from the printer section of that machine. He had depressed a red button designed to stop the machine. However, unknown to Mr. Duchess, the wiper roller continued to run after the button was depressed.
While adjusting the spray nozzle, which was located above the wiper roller on the drive side of the machine, Mr. Duchess’s gloved hand touched the rotating wiper roller, and his hand was drawn into the nip point of the machine. Mr. Duchess, who testified that he did not see the roller running, lost his little finger, ring finger, and part of his middle finger. He was not able to return to work for one year.
The case proceeded to jury trial from July 11 through 14, 1995, based solely on a products liability theory. Appellants contended that the Saturn III was defective because it did not have an interlock on the ink shield, which covered the print rollers. Appellants argued that appellee should have installed an interlock to shutdown automatically the running wiper roller when the ink shield was removed. Appellee countered that the machine was not defective and that Mr. Duchess assumed the risk by placing his hand near the running wiper roller.
The case was submitted to the jury on special interrogatories. The jury determined that the lack of an interlock device on the guard or ink shield of the Saturn III did not constitute a design defect, which was the first question submitted. This appeal followed denial of appellants’ post-trial motions.
Appellants contend they are entitled to a new trial on a number of grounds. We believe two of those grounds possess merit.
Appellants first argue that the trial court erred in precluding them from introducing into the evidence the operator’s manual for the Saturn III. They sought to introduce the manual because it incorrectly indicates that when the stop button is depressed, the entire machine stops. The trial court initially ruled that the manual was not admissible since Mr. Duchess admitted that he never read the manual. Thus, the court reasoned, *412the description in the manual was not relevant.
However, appellants again moved for admission of the manual after Mr. Duchess was cross-examined, arguing that appellee had opened the door to the manual’s contents. We agree. Mr. Duchess was cross-examined about warnings that appear on the Saturn III. In particular, he was shown a picture of a warning and read it to the jury:
WARNING: This equipment was shipped with a detañed Operator’s Manual and a Maintenance Manual. These manuals contain vital information for safe use of this equipment. Read these manuals carefully before operating or performing maintenance on this equipment. Fañure to foñow the safety practices in these manuals could result in serious bodñy injury or property damage. Replacement manuals can be obtained by calling or writing the Langston Corporation, 111 Woodcrest Road, Cherry Hill, New Jersey, the zip and the telephone number.
Reproduced Record (“R.R.”) at 222a.
Mr. Duchess then was asked whether he was concerned with the fact that his employer did not give him the manual to read, and he responded in the negative. Next, he was questioned as to whether he paid attention to warnings. After this cross-examination, appellants moved again to introduce the manual into evidence to show that if Mr. Duchess had the manual, it would not have prevented the accident since the manual indicated incorrectly that the entire machine was shutdown after the stop button was depressed. Appellants argued that by introducing the warning and implying that the machine was not defective because it contained warnings to read the manual, appellee had opened the door to introduction of the manual.
We agree. There is no question that this cross-examination was a direct attempt to suggest that if Mr. Duchess had read the manual, the accident would not have occurred. It created the implication that the warning to operators of the machine to read the manual rendered the machine safe. However, the manual contains no warning that the rollers continue to rotate when the stop button is depressed and in fact, indicates the contrary.
We disagree with appehee’s suggestion that introduction of the manual would have raised the issue of the negligence of Mr. Duchess’s employer in not distributing the manual. Appellee argues that since the employer was not a party, its negligence could not be raised. However, it is clear that appellee’s questioning of Mr. Duchess raised the issue of Mr. Duchess’s own negligence and not that of his employer. Appellee suggested that Mr. Duchess was negligent in not heeding the warning by either asking his employer for a copy of the manual or obtaining a copy of the manual himself by calling or writing to appellee. Appellee opened the door to the introduction of this manual by its line of questioning.
Since Mr. Duchess had not read the manual, the manual was not relevant to this accident. By suggesting that the manual would have prevented these injuries, appellee opened the door to the evidence, which was not otherwise admissible, that the manual would not have prevented the accident. Jamison v. Ardes, 408 Pa. 188, 182 A.2d 497 (1962); Gigliotti v. Machuca, 409 Pa.Super. 50, 597 A.2d 655 (1991).
We also believe that the trial court erred in faffing to allow appeUants to introduce evidence that in 1991, appellee installed an interlock device on the ink shield for the Saturn III. We address this issue since, unlike the other issues raised by appellants, it will reoccur during the re-trial of this matter. In interrogatories, appellee admitted that it incorporated an interlock on the ink shield of Saturn III in July, 1991, approximately one year after this accident.
In Pennsylvania, the subsequent repair rule, forbidding admission of evidence of subsequent repair in a negligence action, is inapplicable in a strict liability ease. Matsko v. Harley Davidson Motor Co., Inc., 325 Pa.Super. 452, 454-162, 473 A.2d 155, 156-159 (1984); Wilkes-Barre Iron v. Pargas of Wilkes-Barre, 348 Pa.Super. 285, 293 n. 2, 502 A.2d 210, 214 n. 2 (1985); Gottfried v. The American Can Co., 339 *413Pa.Super. 403, 403, 489 A.2d 222, 222 (1985).
Leaphart v. Whiting Corp., 387 Pa.Super. 253, 564 A.2d 165, 173 (1989).
This rule first was announced in Matsko v. Harley Davidson Motor Co., 325 Pa.Super. 452, 473 A.2d 155 (1984). Therein, we examined the subsequent repair rule, which forbids admission of evidence relating to safety repairs performed following an accident. We determined that the rationales for the subsequent repair rule were “inapplicable to a products liability case.” Id. 325 Pa.Super. at 455, 473 A.2d at 156 (footnote omitted). We examined the two rationales underlying that rule. First, subsequent repairs are irrelevant to the issue in a negligence action, that issue being the reasonableness of the defendant’s actions under the applicable duty of care at the time of the accident. Subsequent repairs are not relevant because actions taken with hindsight are not probative of actions taken without such hindsight. The second justification for the rule is one of public policy: defendants would be unwilling to take safety measures following an accident if they knew that such measures would be introduced into evidence against them.
We ruled in Matsko that neither rationale applied in the strict liability setting. We found that evidence of subsequent repairs becomes relevant in products liability cases since the issue in such cases is the character of the product rather than the defendant’s conduct. Further, we noted that the public policy justification did not apply. We observed that manufacturers would not be deterred from making safety repairs to then-products if the subsequent repair rule did not apply:
The contemporary corporate mass producer of goods, the normal products liability defendant, manufactures tens of thousands of units of goods; it is manifestly unrealistic to suggest that such a producer will forego making improvements in its product, and risk innumerable additional lawsuits and the attendant adverse effect upon its public image, simply because evidence of adoption of such improvement may be admitted in an action founded on strict liability for recovery on an injury that preceded the improvement. In the products liability area, the [subsequent repair rule] ... does not affect the primary conduct of the mass producer of goods, but serves merely as a shield against potential liability. In short, the purpose for the [subsequent repair rule] is not applicable to a strict liability case and hence its exclusionary rule should not be gratuitously extended to that field.
Id. 325 Pa.Super. at 457, 473 A.2d at 157, quoting Ault v. International Harvester Co., 13 Cal.3d 113, 117 Cal.Rptr. 812, 815-16, 528 P.2d 1148, 1151-52 (1974). We also noted that insurers would not allow manufacturers to refuse to implement safety devices simply because evidence of such implementation could be used in a lawsuit.
Herein, the cases upon which the trial court relied, Connelly v. Roper Corp., 404 Pa.Super. 67, 590 A.2d 11 (1991), and Gott-fried v. American Can Co., 339 Pa.Super. 403, 489 A.2d 222 (1985), did modify this rule somewhat. In Gottfried, the plaintiff was injured opening a can. “In order to show that the ‘state of the art’ was such that safety features were available” to reduce the risk of injury suffered by the plaintiff, the plaintiff’s expert produced cans placed on the market by other manufacturers in 1977, the year in which the can that injured plaintiff was manufactured. Id. 339 Pa.Super. at 409, 489 A.2d at 225. This evidence was allowed. However, the trial court refused to allow evidence that in 1981, the defendant manufactured a can containing some of the same features. We upheld this ruling, concluding that the 1981 can was not relevant.
We noted that Gottfried did not change the rule regarding relevancy of evidence, stating:
In cases in which it has been alleged that a product was defectively designed, the jury is required to determine whether the product lacked any safeguard necessary to make it safe for normal use. In such cases it is relevant to show the “state of the art.” However, it is only the state of the art at the time of design and/or manufacture that is relevant. Design improvements made four years after an injury has occurred are not relevant to show the state of the art at the time that the injury occurred.
*414Id., 389 Pa.Super. at 410, 489 A.2d at 226 (emphasis added). In Connelly, the reasoning of Gottfried was applied.
The Gottfried case must be confined to its holding that subsequent design changes cannot be introduced to show the state of the art at the time the product was manufactured. However, in the instant case, appellants were not attempting to prove state of the art at the time of the accident by introducing subsequent design changes. The testimony of appellants’ expert witness, mechanical engineer John Frank, establishes that the interlock was available at the time of manufacture. Appellants asked Mr. Frank to explain why he determined that the Saturn III was defective, and he responded:
[T]he defect is simply that the guard that was provided over the dangerous machinery, [which was] the nip point between the wiper roll and the anilox roll, was not an adequate guard because it was hinged and was easily opened and exposing the danger area. So, the guard is not effective in protecting workers from coming in contact with that dangerous machinery. Workers have to go into that area on a very frequent basis and open the machine and change the setup and change the print plates. The manufacturer recommends going in and cleaning, in fact, on a daily basis. So, there is an expectation that this is a type of a guard that would be frequently opened. A guard of that type that would be frequently opened is, to be proper, it has to include interlocks that workers will not be exposed to the dangerous machinery; that any time that guard is opened and the danger is exposed, the machinery is stopped so that it is not in a dangerous condition.
Q. What is an interlock?
A. An interlock is simply a device that senses a condition of a piece of machinery and prevents other functions of the machine from operating while it is in that condition. There are interlocks of all various types around. The type that we are talking about here is a safety interlock that when a dangerous condition exists, what it does, it puts another part of the equipment or machinery in a condition so that it is safe while the danger is presented.
R.R. at 311a.
Mr. Frank suggested that the . ink shield should have been fitted with an interlock which shutdown the rollers when lifted. He further explained that interlocks are and were commonly used on any number of products:
One example that is very common is the residential washer and dryer. I am sure you are all familiar with your washing machine whenever it goes into spin cycle, and you open the door, that machine will stop, or your dryer if you open the door while it is running, it will stop and you have to close the door. Then, you start the machine. It is exactly the same principle that is applied to the door of your washer and dryer [which] acts as a guard because it prevents the user from coming in contact with the dangerous moving machinery inside.
Id. at 311a-12a. Finally, Mr. Frank noted that interlock principles frequently are applied, stating that millions are contained in washers and dryers and that many thousands of other interlocks are used in industrial machinery. With those interlocks, the same principle is applied: “Any time you have a guard that has to be opened with some frequency thereby exposing the dangerous machinery, an interlock must be provided so to make that machinery be in a safe condition while the danger is exposed.” Id. at 312a.
At the time of this Saturn Ill’s manufacture, interlocks were used daily in all types of machines. Thus, the evidence showed that appellee chose to produce an item which was not as safe as it could have been considering the state of the art at the time of production. The fact that appellee later incorporated the interlock was a subsequent repair that should have been admitted in this products liability action under the general rule announced in Matsko.
Furthermore, we agree with appellants’ contention that when appellee’s expert questioned the practicability of installing the interlock device, appellants should have been *415permitted to rebut this evidence with evidence regarding appellee’s installation of the interlock device. There is no doubt that defense expert, Peter James Sehwalje, questioned the practicability of placing an interlock on the shield:
Q. Are there any downsides to using an interlock device?
A. Well, the downsides are that if you have an interlock in this particular application, you preclude the ability of the people setting up the machine to set the printing unit properly. There is also the inherent downsides associated with any form of an interlock device; the unit becomes a reliance mechanism to shut down the machine instead of shutting down through switches in this environment that they can accumulate ink inside and the internals will not operate. There are normal downsides associated with any control device.
R.R. at 527a. By posing issues relating to the practicality of installing the interlock, appeUee clearly opened the door to evidence that the interlock was practical. The fact that it was practical was demonstrated by appellee’s incorporation of the interlock the following year into machines it manufactured.
Appellee protests that it never questioned the feasibility of the interlock device and thus, did not open the door to this evidence. It points to in-chambers conferences where it told the judge it would not contest the interlock’s feasibility and points to its expert witness’s admissions that the interlock was feasible to install. Feasibility, which means that something is capable of being done, is not the same as practicality, which questions whether it is better, considering various factors, to do something. Semantics aside, however, Mr. Sehwalje clearly testified that while it was feasible to install the interlock, i.e., it could be done, installation of the interlock caused other problems that made it impracticable. This testimony is rebutted directly by the fact that appellee actually installed the interlock in 1991. When appellee presented evidence that the interlock was not practical by listing the “downsides” of the interlock, appellee rendered evidence as to the practicality of the interlock relevant. Appellants then should have been permitted to introduce evidence that tended to show that the interlock was not impractical. The fact that ap-pellee’s subsequently incorporated the interlock made it more likely than not that the interlock was not impractical. Thus, we agree with appellant’s contention that once appellee’s expert questioned the practicality of the interlock, appellants should have been permitted to show that the interlock had been incorporated on the machine.
Judgment reversed. Case remanded for a new trial. Jurisdiction relinquished.
DEL SOLE, J., Files a Dissenting Opinion.