J. B. Williams Co. v. United States

Per Curiam: This case was referred to Trial Commissioner William E. Day with, directions to prepare and file his opinion on the issues of the cross-motions for summary judgment filed by the parties under the order of reference and Rule 166 (c). The commissioner has done so in an opinion and report filed on August 12, 1971, wherein such facts as are necessary to the opinion are set forth. On October 6,1971, defendant filed a motion requesting that the commissioner’s recommended conclusion of law be adopted 'by the court as the basis for its judgment pursuant to Rule 54(b) (3) (iii), no request for review of the commissioner’s opinion and recommended conclusion of law being before the court. Since the court agrees with the opinion and recommended conclusion of the trial commissioner, it hereby grants defendant’s motion and adopts the same, as hereinafter set forth, as the basis for its judgment in this case. Therefore, plaintiff’s motion for summary judgment is granted and defendant’s cross-motion is denied. Judgment is entered for plaintiff with further proceedings stayed pursuant to Buie 167 for 90 days to afford the parties an opportunity to obtain an administrative resolution by the agency of the adjustment in the contract price commensurate to any additional amounts expended by plaintiff in adhering to the change order. OPINION OP COMMISSIONER Dat, Commissioner: This case involves a Wunderlich Act (68 Stat. 81, 41 U.S.C. §§ 321-22 (1964 ed.)) review of a decision that was rendered by the General Services Administration Board of Contract Appeals under the disputes provision of contract No. GS-00S-37509. Cross-motions for summary judgment have been filed by the parties. On April 16, 1962, the General Services Administration (GSA) contracted with Landers, Frary & Clark (hereinafter Landers)1 for the production of certain radiological survey meters, Model CD-V-715, designed for use by civil defense personnel to detect and measure high range fallout radia — 495 — tion in the event of nuclear disaster. In accordance with, the contract, and in order to begin full scale production, plaintiff submitted for the contracting officer’s approval 300 initial production units for performance testing and evaluation under the specification requirements. The contracting officer found these meters unable to meet the requirements in three vital areas — namely, spectral dependence and directionality, shock resistance, and voltage saturation of the ionization chamber — and thus withheld approval to proceed. Plaintiff subsequently redesigned its meters to conform to the specifications, thereafter claiming entitlement under the standard changes clause to an equitable adjustment in the contract price reflecting the additional costs incurred in meeting these requirements. The contracting officer denied this claim and plaintiff took timely appeal to the GSA Board of Contract Appeals (hereinafter Board). In a detailed opinion filed on September 25, 1988,2 the Board upheld the decision of the contracting officer, one member dissenting. Stated briefly, plaintiff now argues that while there is no dispute as to the facts as found by the Board, proper weight was not accorded certain facts while other relevant facts were overlooked, which, taken together, would have led to a contrary decision. Furthermore, plaintiff attacks the Board’s interpretation of Landers’ contractual obligation with defendant. Thus plaintiff claims the Board committed reversible errors in its findings of fact and conclusions of law and that its decision is therefore not entitled to finality by authority of sections 1 and 2 of the Wunderlich Act.3 The pertinent facts of the case are as set forth below. In 1960 the Federal Supply Service (a branch of GSA) issued an invitation for bids to furnish it with certain radiological survey meters “* * * in accordance with Office of — 496 — Civil Defense Mobilization Standard Item Specification CD-V-715 dated July 1, 1960 * * Included among tbe bidders were Anton Electronic Laboratories, Inc. (soon after bought out by Lionel Electronics Laboratories, and hereinafter Lionel), and plaintiff. As part of the bid requirements, preaward samples were submitted for testing for such things as accuracy, stability, temperature, atmospheric pressure, immersion and light sensitivity, and spectral dependence. While it is not clear to whom the contract was formally awarded (Anton or Lionel), the latter, whose samples were found to be in compliance with the above-mentioned performance characteristics, actually manufactured the original Government contracted CD-V-715 survey meters referred to as Lionel Model No. 1. The aforesaid specification called for (among other things) testing of three of the initial 300 production units taken at random against certain performance requirements set forth in 21 separate specification paragraphs. In addition, another paragraph required the contractor to certify and, if requested, furnish test data indicating that certain components of the CD-V-715 had been tested for and found to be in conformance with designated specification paragraphs, most notably the one relating to the 95 percent voltage saturation requirement for the heart of the instrument, the ionization chamber. While the record does not indicate whether Lionel tested these units with respect to voltage saturation or whether the mandatory certificate was in fact issued, it appears as though the resulting test data confirmed Lionel’s adherence to the specification requirements because Lionel subsequently produced and the Government thereafter accepted 11,729 Model No. 1 units, deliveries being completed on March 28,1962. Prior to the award of the Landers contract, but subsequent to the issuance of the invitation for bids for that contract, the Government entered into a second agreement with Lionel to produce approximately 14,000 additional CD-V-715 survey meters. These units were known as Lionel Model No. 1A and were to be “* * * equivalent to OCDM Model 1 as manufactured by Lionel Electronic Laboratories and in ac — 497 — cordance with. Office of Civil Defense Mobilization Standard Item Specification CD-V-715 dated July 1, 1960 * * 4 On March 2,1962, invitations for bids were dispatched by the Federal Supply Service for the production of another 63,185 CD-V-715 survey meters, as before, to be “equivalent to OCDM Model 1 as manufactured by Lionel Electronic Laboratories and in accordance with Office of Civil Defense Mobilization Standard Item Specification CD-V-715 dated July 1, 1960 * * Attached to said invitation was a notice to prospective bidders pointing out 15 significant changes from the initial CD-V-715 invitations, included among which were the availability of sample instruments for bidders in the present invitation, deletion of the preproduction model requirement, and the availability of model instruments for initial testing and inspection. Paragraph 25 of the Terms and Conditions of the said invitation, making sample instruments available to contract bidders, also contained the following language: Samples of instruments specified in this Invitation for Bids, will be made available to prospective bidders upon request * * * if* # i|i % # In the event there is a discrepancy between the sample and the specifications cited herein, bidders are advised that the specifications will govern. While the requirement for preproduction models was omitted, the contractor was nevertheless required as a prerequisite to full scale production, to submit within 45 days of award a random sampling of the initial 300 production units to be tested and evaluated against the 21 performance specifications. In addition, the requisite for a certificate evidencing at least 95 percent voltage saturation remained. On March 30, 1962, Landers responded to this invitation by bid stating in pertinent part: We are attaching to this letter our quotations on Kadi498 ological Surrey Meters CD-V-700 and CD-V-715 and are offering an “equal” product in each, instance. * * * * sjr * * * * * we have had extensive design and manufacturing experience in making our own chambers and feel this poses no significant problem. Models of the CD-V-715 were purchased in accordance with the terms of the invitation for bid and were tested for accuracy using the 100 Curie Cobalt 60 source located at our plant. Performance was found to be satisfactory and we would therefore, due to the urgency of delivery requirements, follow closely the design of the previous manufacturer. Our ability to produce instruments quite similar to those you are presently purchasing, has been demonstrated and we believe there should be no question of our ability to produce the Landers, Frary & Clark Model Numbers 700 CD-V and/or 715 CD-V, provided we are the low bidder under this invitation. This letter m no way talces conception to any of the provisions of the in/oitation for bid. [Emphasis added.] On April 6, 1962, Messrs. Carroll and Congdon (representatives of the contracting officer administering the aforesaid contract) and Mr. Brownell (the contracting officer’s technical advisor in connection with this contract) visited the Landers’ plant on a preaward survey designed to evaluate Landers’ production capability. They were accompanied by representatives of Landers (among whom were Jerome Strauss, vice president and Gordon Baymond, chief engineer) . After a tour of the facilities, a meeting was held in which each paragraph of the specifications was reviewed and discussed in detail with Mr. Baymond. Having had experience with the previous Lionel CD-V-715 contract, Mr. Brownell pointed out several possible problems that might arise during manufacture, most notably involving the 95 percent voltage saturation. Also, it was repeatedly emphasized that certification of the 95 percent saturation level would be demanded, along with adherence to the specifications. There was no mention made of any actual defects present in the Lionel model, however. — 499 — It is significant to note that prior to bidding Landers purchased a number of Lionel Model No. 1 meters for testing and bid preparation purposes. These instruments were not, however, put through the gamut of tests that would have indicated whether or not they complied with all of the performance requirements, but rather were tested only for accuracy. The instruments so tested displayed a slight downdrift in the read-out (a characteristic purportedly undesirable from the point of view of the operator of the instruments) but were considered to be within calibration accuracy or close enough thereto to be properly readjusted. At the Board hearing, plaintiff attempted to explain away this allegedly inadequate testing procedure by pointing out that the time period between the date of receipt of the instruments and the bid due date (only 3 to 4 weeks) was insufficient to conduct what would have amounted to 6 to 8 weeks of testing. Moreover, Mr. Raymond stated in his testimony that the accuracy test was in any case the most significant and revealing test. A contrary view was taken by Mr. Carl Siebentritt, Jr. (a radiological expert from the Office of Civil Defense — OCD—) who testified that Landers could have, at the very least, run tests on the three most crucial areas of the 21 mentioned in the specifications (voltage saturation, spectral dependence and shock) in no more than 10 working days. The results, he stated, would have given a “pretty good fix” on the instruments’ performance. On April 16, 1962, Landers was awarded the contract to produce 63,135 of the CD-V-Y15 survey meters at a total contract price of $1,066,350.15.5 Thereafter, Landers purchased 10 additional Lionel Model No. 1 survey meters from OCD; the purpose being, according to Mr. Raymond, to dissect, sketch, test and analyze the various component parts and materials in order to duplicate as closely as possible the given model. It was believed both by Mr. Raymond and Mr. Strauss that by closely following the Lionel design, plaintiff would be offering defendant an instrument in full compliance with the performance requirements of the specifi — 500 — cations.6 The factors leading to this conclusion were, as stated in their testimony, the short term delivery schedule (they had approximately 45 days in which to submit the initial 300 production units), the elimination of the preaward testing procedures, and the fact that the defendant previously contracted under the same OCD specification with Lionel for survey meters (now issued as models by defendant) and accepted full delivery thereof. Consequently, Landers viewed this more or less as a production contract, i.e. an agreement to manufacture an already tried, tested, and approved product.7 In furtherance of producing such a product, Landers checked various detailed parts in its first 12 production models for performance capability and also substituted interchangeable parts into the Lionel models, evaluating to see if the same performance standards were met. Prior to producing the 300 initial units, however, Landers did not go — 501 — so far as to test these models against all of the 21 performance requirements. These initial 800 meters were completed by Banders on or about July 17, 1962, and were (according to Mr. Raymond), with but a few exceptions, basically identical (both physically and pictorially) to the original Lionel model. Mr. Raymond did admit to certain functional and physical differences in the units, but attributed these to the fact that new suppliers were at times used8 and also that certain changes were purposely initiated to increase production efficiency and accuracy.9 The differences cited included use of a thinner circuit board with a slightly different layout; smaller conductors; induction rather than hand soldered parts; an aluminum, rather than zinc, cover; thinner paper in the ionization chamber; different rotary switches, indicating meters, electrometer tubes, transformer winding, handles; and a switch cover positioned slightly off from where it was located on the Lionel unit. In July 1962, Landers commenced testing on a random sample taken from these 300 units for compliance with the 21 performance characteristics. These tests were completed on August 21, 1962, and revealed satisfactory results in all but two categories — the humidity and hammer shock tests. Mr. Carroll recommended to the contracting officer that Landers be allowed to deviate from the specifications on these two points and conduct less stringent tests thereon. This suggestion was later denied. It should be noted that a saturation test of the ionization chamber was not conducted — 502 — at this time so that Landers was as yet Tinware of any such deficiency. The record contains two reports from the National Bureau of Standards on tests performed during the summer of 1962 on the Lionel models. The first (dated June 20) reveals that upon testing three Lionel Model CD-V-715 survey meters (neither report specified whether the models tested were No. 1 or No. 1A, although OCD assumed them to be the latter) against seven of the 21 specification characteristics including spectral dependence, the instruments were found to be overly sensitive but probably within calibration of the operational requirements. The second report (dated August 1) reflects the results of voltage saturation testing initiated on May 16. This was the first instance of voltage saturation testing on these models. The finding was that all three instruments tested lacked the necessary voltage saturation level by 10 to 15 percent. In September 1962, representatives of Lionel and OCD held a meeting at the Pentagon occasioned by the above-mentioned test results and also unfavorable field reports coming in on the Lionel Model 1A. In reply to assertions by defendant that Lionel Models 1 and 1A were not 95 percent saturated,10 the Lionel representative claimed that this lack of saturation in the Model 1 had been known by the Government and that they had not required Lionel to submit the certificate warranting such saturation achievement. As a result of these discussions and the lack of saturation information on the Landers model, Edward Williams (a member of the Instrument Evaluation Committee of OCD) was dispatched to the Landers plant to ascertain whether or not Landers’ ionization chamber was properly saturated. Mr. Williams arrived at the Landers plant shortly thereafter and met with (among others) Messrs. Strauss, Bay-mond, Congdon and Carroll. He related what had transpired at the Pentagon meeting, produced a copy of the August 1 National Bureau of Standards Beport for their inspection, — 503 — and requested that the Landers unit 'be subjected to humidity and voltage saturation testing in his presence. These tests were conducted the following day on one Lionel No. 1 and three Landers instruments (this was the first time a saturation test had been run on the Landers unit). The Lionel unit operated at 20 volts and was found to be approximately 60 percent saturated, while the Landers model indicated approximately Y0 percent saturation at the chamber operating voltage of 22 volts.11 It was also observed that during these tests Landers failed to connect the ionization chamber guard ring, an omission which would tend to lower the saturation reading. Evaluation of the data obtained from this test led Landers’ engineers to conclude that an increase in power supply to 55 volts would be necessary to achieve the required 95 percent saturation level. By letter dated November 1, 1962, the contracting officer, at the request of OGD, advised Landers that approval to proceed with production was not forthcoming until certain problems were ironed out, namely, achieving 95 percent saturation in the ionization chamber, meeting the specified hammer shock test, and maintaining the required spectral dependence parameters at intermediate test points as well as at the extreme ranges. (The modified test which Mr. Brownell allegedly orally granted and which the Landers units passed required testing only at the extreme energy ranges.) On December 29, 1964, Landers wrote the contracting officer asking for an equitable adjustment in the contract price corresponding to the extra cost incurred in redesigning the instrument to fulfill the above-mentioned requirements. The contracting officer denied this request and on appeal taken therefrom stated that Landers did no more than that already required by the contract and hence no change within the meaning of the changes clause took place. This decision was subsequently appealed to the GSA Board which, as previously mentioned, denied the claim. The Board first held that the principle enunciated in Sea-view Electric Co., ASBCA No. 6966, 61-2 BCA ¶ 3151, the — 504 — primary case relied upon by plaintiff, was not applicable under the factual circumstances of the present case. In iSeavieto, the Armed Services Board concluded that where a contract specifically called for a product built in accordance with a Government furnished model and specifications, the model became part of those specifications and the Government thereby impliedly warranted that a product manufactured in accordance with that model necessarily complied with the performance requirements of the specifications. This even though the contract contained a discrepancy clause similar to that employed in the Landers contract. The GSA Board, however, found several distinguishing factors which led it to reject a similarly based argument made by plaintiff herein, to wit, representations were made to Seaview’s president to the effect that the model met the performance requirements of the specifications whereas no such representations were made to Landers. The Board, in fact, points out that Landers was warned of possible problem areas. No opportunity was afforded Seaview to test and disassemble models prior to the contract award, as was afforded Landers. The contract with Seaview actually called for parts physically, mechanically and electrically interchangeable with the model, thus prescribing the design, while Landers had total control over the design of its unit. Finally, the Seaview contract imposed a more stringent burden to reproduce the model by calling for an instrument “in accordance with” rather than “equivalent to” the designated model. The Board further stated that the mere fact that Landers bid to closely follow the Lionel design (to which the Government took no exception), did not in itself make the model a part of the contract specifications. Patterning after a model design, the Board reasoned, does not equate to reproducing that model. Moreover, the Board viewed the concluding sentence of Landers’ bid [“This letter in no way takes exception to any of the provisions of the invitation for bid.”] as indicating an intent not only to closely follow the Lionel design but also to maintain adherence with the performance requirements of the specifications, using the model more as a guide for production and design. — 505 — The Board went on to state that, even assuming arguendo that the Government did prescribe the design of the meters and that the furnished model should be considered part of the specifications, nevertheless, the Landers finished product was not “equivalent to” the Lionel model. In so doing, the Board accepted Mr. Siebentritt’s definition of equivalency between two survey meters, i.e. when one “is shown to detect and measure radiation over the spectrum of energies for which it is designed to measure it in exactly the same fashion” as the other. It was Mr. Siebentritt’s opinion that the initial 300 units produced by Landers were not equivalent to the Lionel model because of various material and design alterations. In weighing this testimony against that of Mr. Baymond (who took the contrary view, that these preproduction units did detect and measure radiation in the same manner as the Lionel model), the Board chose to accept the Siebentritt conclusion, feeling he possessed greater expertise in this area. The Board bolstered its conclusion by analyzing and comparing statistical reports on both the Landers and Lionel models with reference to the three crucial performance requisites now at issue — voltage saturation, spectral dependence and directionality, and shock resistance. In all but the first category, the Board found the Landers models to be deficient, while the Lionel models appeared to satisfy the contract requirements. Therefore, the Board stated, even assuming the interpretation given the contract by Landers was reasonable in light of the surrounding circumstances (an assumption the Board later rejected), plaintiff’s case still fails because it neglected to manufacture a meter that was “equivalent to” the furnished model. Following the reasoning of the contracting officer, the Board further decided that no actual change took place within the meaning of the changes clause. It felt the contracting officer’s letter of November 1, 1962, calling upon Landers to “take all necessary steps to demonstrate conformance of your instruments to Standard Item Specification CDV-715” demanded no more than that already required under their agreement with defendant. The Board, therefore, concluded that Landers itself opted for redesign of the in — 506 — strument in order to meet the specifications, and not the contracting officer. Landers’ final argument before the Board sought to place the risk upon the Government for assuming that Landers would follow the Lionel model. In support of this argument plaintiff asserted that the defendant had knowledge of Lionel’s failure to issue a voltage saturation certificate and was thus obligated.to inform plaintiff thereof; that the defendant was put on notice by the terms of Landers’ bid that it took the invitation for bids to mean the Lionel model was tested, approved, and met the specification requirements; and finally, that due to the short term delivery schedule, the contract could have been nothing but a production contract and not a design contract. The Board found that (a) the time elapsed from the date of award to final delivery (which with an extension of time encompassed the greater part of a year), did not amount to a short term delivery period, even though the preproduction models had to be available within 45 days of award and (b) the apparent interpretation Landers gave the phrase “equivalent to OCDM Model No. 1 as manufactured by Lionel Electronics Laboratories” to mean that reproducing the model would satisfy the contractual obligation was not a reasonable one because as the Board put it: [I]t is clear that Landers knew that it could not make true reproductions of the Lionel model and this is borne out by Landers’ bid and the fact that Landers used some materials and parts in its said pre-production instruments which were different from those used in the Lionel model and made design changes in its said instruments from the design of the Lionel model to accommodate these materials and parts. Accordingly, the Board denied Landers’ claim for an equitable adjustment in the contract price. Plaintiff now urges that the Board erred as a matter of law in interpreting Landers’ contractual obligation with defendant to require strict adherence to the specifications regardless of whether the Lionel unit was capable of complying with the same specifications. Furthermore, plaintiff argues that the Board did not use the proper legal test for determining equivalency and that had it, the Board could have concluded — 507 — only that the Landers units were “equivalent to” the Lionel models. While plaintiff does not take exception to the facts as found by the Board, it does challenge the Board’s reluctance in finding other pertinent facts and according to them proper weight. Finally, plaintiff contends that the contracting officer’s letter of November 1 did in fact amount to a change order within the meaning of the changes clause and that it is entitled to the adjustment it seeks. After an extensive review of the record and supporting briefs, we conclude for reasons set forth below that plaintiff is indeed entitled to recover an amount commensurate to the additional expenditures incurred in redesigning the meters to conform with the specifications. The central issue in this case is whether or not Landers was obliged by the contract to provide an instrument satisfying the performance requirements of the specifications irrespective of whether the Lionel model was capable of meeting these specifications. Since this is a matter of contractual interpretation — a question of law — we are not constrained by the interpretation given the contract by the Board, but rather are free to give independent judicial consideration to the language therein. 41 U.S.C. §§ 321-22 (1964). Defendant maintains that a strict construction of the phrase “equivalent to the OCDM Model 1 as manufactured by Lionel * * * and in accordance with * * * Standard Item Specification CDV-715” leaves but one clear and unambiguous interpretation — that calling for an equivalent product adhering as well to the specification requirements. If viewed in confinement from the circumstances surrounding the procurement and other contract language, defendant’s position might well be controlling. However, it is a well known principle of contract construction that an instiument must be viewed in its entirety and the language contained therein given that meaning which would be derived by a reasonably intelligent contractor acquainted with the contemporaneous circumstances. Hol-Gar Mfg. Corp. v. United States, 169 Ct. Cl. 384, 388, 395, 351 F. 2d 972, 975, 979 (1965), and authorities cited. Taken in this light, we find Landers’ conclusion that producing an equivalent product necessarily implied conformance with — 508 — the specifications to be justified and reasonable under the circumstances. Landers knew, when preparing its bid, that defendant had previously contracted for and accepted full delivery of a large number of CD-V-715 survey meters manufactured by Lionel under exactly the same specification requirements. There was never any indication that these Lionel meters were in any way deficient or not acceptable to the defendant. Landers also was aware of the inspection and testing procedures outlined in this earlier contract as well as the certification requirement contained therein. From this, Landers drew the logical conclusion that the Lionel product fully satisfied defendant (enough, at least to offer it as a model for subsequent contracts for the same meters) and met all of the performance requirements. Landers chose to harmonize the two requisites of the phrase “equivalent to the OCDM Model 1 as manufactured by Lionel * * * and in accordance with * * * Standard Item Specification CD-V-715” rather than find them in conflict. [Emphasis supplied.] The language employed in the invitation for bids served further to augment this belief. Paragraph 25, which made the Lionel units available for prospective bidders, identified these units as the “instruments specified in this Invitation for Bids.” Understandably, a contract bidder reading this would likely conclude that the Lionel meter being offered as a model conformed to the specifications. In addition, the instant invitation expressly deleted the requirement calling for pre-award performance testing and inspection, leading a bidder to believe that in reproducing the Lionel model, the need for such testing was obviated by the fact that the Lionel unit had already passed these tests. Underscoring Landers’ interpretation of said invitation accepting the model as an approved product meeting all the performance standards was its responsive bid. Plaintiff therein promised to supply an “equal” product which “due to the urgency of delivery requirements” would “follow closely the design of the previous manufacturer.” The bid stated finally: This letter in no way takes exception to any of the provisions of the invitation for bid. — 509 — This last sentence is extremely significant in understanding Landers’ thinking when making its bid. Unfortunately, the Board missed the point when it stated: * * * Certainly, in view of this statement, the Contracting Officer must have concluded, and rightly so, that Landers intended to produce a CD-V-715 that followed closely the design of the Lionel model and yet met the performance requirements of the specifications. * * * This notion, that Landers differentiated between the Lionel design and that which the specifications required, is patently incorrect and without evidentiary support. The one logical meaning derived from this sentence is that Landers felt it was being responsive to defendant’s demands as expressed in the invitation for bids by offering a product equal to the Lionel product, this being accomplished by closely following the Lionel model. Landers’ bid also clearly indicates that plaintiff construed the delivery schedule to be short term in nature, thus leaving Landers the impression that little or no alteration was necessary in the design of the model but that plaintiff had only to duplicate or “follow closely” the Lionel design. The Board found otherwise, stating that the time differential between the date of award and date of completed delivery (a period of almost 1 year) did not amount to a short term delivery ¡slchedule. This conclusion however, totally overlooks the realities of the situation facing Landers at the time. Under the provisions of the contract, the initial 300 production units had to be completed and available for inspection and testing within 45 days of award, a time span which the Board concedes is “undoubtedly a short period.” During this time Landers had to test the Lionel models and prepare sketches of parte therefrom, job out production of certain parts, construct all necessary tooling, assemble the production line, and manufacture the initial lot of 300. This time period can be the only true measure of whether the delivery schedule was long or short, and certainly under the circumstances described above, it must be considered an abbreviated schedule. Coupled with the Board’s finding that the specification was “both a performance and design specification, primarily the former” (a fact which, due to prior Government contract experience, — 510 — Landers recognized) and that such things as voltage saturation, spectral dependence and shock resistance are all design characteristics, it is easy to appreciate why plaintiff contemplated only incidental design changes, decided that what defendant wanted was a duplicate of the furnished model, and reasoned that the Lionel model must have adhered to the specification requirements. As has been earlier noted, defendant did not take exception to the terms of this bid when awarding Landers the contract. The Board placed great emphasis on the discrepancy clause contained in paragraph 25 of the contract [“In the event there is a discrepancy between the sample and the specifications cited herein, bidders are advised that the specifications will govern.”] stating that, taken together with the warnings given Landers at the preaward meeting concerning adhering to the specifications, plaintiff was put on notice that the specifications had to be met regardless of the condition of the Lionel model. However, the mere existence of such a clause does not amount to a warning that discrepancies actually do exist, nor does it shift responsibility onto the contractor to ascertain at his peril whether or not such conditions exist. When confronted with this boilerplate provision and in view of the circumstances of this procurement, Landers chose to discount its meaning and effect and rely instead on the integrity of the Government in holding out the Lionel model as the “instruments specified in this Invitation for Bids.” As the testimony repeatedly indicated, Landers would not have bid on this contract had any major design alterations been contemplated for it knew these could not be accomplished in the 45 days allotted between date of award and submission of initial production units. The Board also placed too much weight on the fact that Landers was warned of certain problems concerning voltage saturation at the preaward meeting. What Brownell in fact indicated, was that certain problems had cropped up in the manufacturing process with regard to voltage saturation, but no mention was made of any possible problem in achieving the desired saturation level, as the Board seemed to imply. In actuality, there was never as much as a hint that the Lionel unit did not measure up to Government standards. — 511 — What we are holding, in effect, is that the Government cannot always absolve itself of all responsibility, particularly when making certain representations (either express or implied) , by merely inserting a clause into a contract to that effect.12 Having established plaintiff’s interpretation of the contract to be reasonable, we must still determine whether or not plaintiff in fact carried out its promise to provide a survey meter “equivalent to” Lionel Model No. 1. Plaintiff argued before the Board the equivalency should be measured functionally, i.e., whether the meters detect and measure radiation in the same way and perform similarly. The Board chose instead to adopt the definition given the term by Mr. Siebentritt [“When it is shown to detect and measure radiation over the spectrum of energies for which it is designed to measure it in exactly the same fashion.” (Emphasis added.) ] Since this is again a question of law — a matter of contractual interpretation — we are not bound by the meaning ascribed by the Board. Borrowing from the area of patent law in which the equivalency doctrine often arises to settle infringement disputes, this court has defined equivalency between two items as follows: “if it performs substantially the same function in substantially the same way and for substantially the same purpose * * *”13 [Emphasis supplied]. This then, will be the measure of whether or not plaintiff lived up to its contractual obligation. In finding the Landers model not equivalent to that of Lionel, the Board relied on the expert testimony of Mr. Siebentritt over that given by Mr. Baymond. A review of Mr. Siebentritt’s testimony, however, reveals nothing but educated speculation lacking any supportive proof as to — 512 — Landers’ equivalency. Mr. Sibentritt goes down a list of alterations made by plaintiff, stating for example, that the difference in outer cover materials could tend to alter the readings or that machine soldering could tend to cause more absorption or that the different screws used to fasten the chassis to the top of the case could affect the mechanical characteristics of the instrument. As Siebentritt said: “You can’t say for sure that it wouldn’t.” Similarly, however, you can’t say for sure that it would. Theoretically speaking, these various changes made to accommodate new suppliers, speed up production, and put out a more uniform product could have resulted in an instrument functionally different from the Lionel model, but the true measure of equivalency must come from actual comparative testing and analysis of the instruments. In order for plaintiff’s claim to be supportable, then, the Lionel and Landers meters must prove to be deficient in substantially the same way with respect to voltage saturation, spectral dependence and directionality, and shock resistance. We will analyze each characteristic separately. Voltage Saturation As far as this characteristic is concerned, there is little doubt that both the Lionel and Landers units were nearly equally deficient. In fact, defendant’s brief virtually concedes this point on page 39: The Government admits that both meters, the Lionel model and the Landers model, were almost equally deficient in voltage saturation. The Board is a bit more evasive on the point, stating only that * * * the record indicates that the Lionel model was not 95% voltage saturated * * * but not actually making a comparative study of the two instruments. Had the Board done so, the inescapable conclusion would have been that the units were almost identically deficient, with the Landers unit actually exhibiting a higher reading and thereby outperforming the Lionel model.14 — 513 — Spectral Dependence and Directionality Spectral dependence is an indicia of the constancy and evenness with which, an instrument measures radiation over the energy spectrum, while directionality points out whether the instrument measures this radiation with uniformity when placed in different positions relative to the energy source. The specifications called for measurement within certain tolerances over an energy range from 80 KeV [thousand electron volts] to 1.2 MeV [million electron volts]. Both the Lionel and Landers units met the above-mentioned tolerances when measured at the extreme ends of the energy spectrum. However, in his letter of November 1, 1962, the contracting officer required Landers to meet these tolerances at intermediate energy levels as well, these points being at 120 and 170 KeV. This plaintiff was unable to do. The question now becomes whether or not the Lionel unit supplied Landers was likewise unable to meet the specification requirements at these or similar intermediate points. The Board chose to answer this question by referring to tests conducted on the preproduction and bid models submitted by Anton Electric, wherein the spectral dependence was found to be within the aforementioned tolerances. This, however, does not mean to say that the production units eventually supplied plaintiff also met these requirements. To answer such a query, attention should be focused, if possible, on the most recent data compiled on the Lionel production units. This is what plaintiff attempted by introducing a National Bureau of Standards Keport dated June 20, 1962, which stated in reference to spectral dependence of the Lionel model tested that All instruments are too sensitive, by more than the allowed 15%, reaching a peak at about 120 KeV. As they are all over sensitive to Co-60 and Cs-137 [gamma] radiation, an adjustment of calibration will probably bring them within specification for spectral dependence while maintaining accuracy. Concerning directionality, the same report stated: “At 120 KeV and lower energies, directional dependence is excessive.” These observations are reinforced by the testimony of Mr. Edward Williams (an expert witness called by defendant) — 514 — wlio conceded that the Lionel model was too sensitive in both areas. It should be further noted that not only did the Board rely solely on tests conducted on prebid and preproduction models, but also the December 20,1960 test results came from testing of but one instrument, and then only from one direction and not all directions as called for in the specifications. The evaluation of relative deficiency should apply comparable standards if there is to be any logical correlation. Thus, the action of the Board appears to have been in error. Furthermore, the Board’s conclusion that the June 20 report relates to Model 1A testing is without any factual basis. The defendant could not say for sure which instrument it tested at that time. Taking into account the almost identical nature of Model 1 and Model 1A (the only discernible difference being the increased voltage supply), we believe that it would be more in line with the facts as presented to treat the instrument tested as the Model 1, or at least to conclude that it really makes no significant difference. Thus, we conclude the Board erred in holding the units were not equivalent with respect to spectral dependence and directionality. Shook Resistance The Board relies solely on a memorandum dated January 21, 1968, signed by engineers of the Naval Research Laboratory and a memo relating to this report signed by Mr. Williams of OCD, in finding that the Lionel Model No. 1 withstood the shock testing and was therefore not deficient in this category. Again, we must measure the two models against the same criteria and under the same conditions if we are to draw any relevant conclusions. The Board stated that these exhibits “show clearly that three Lionel Model No. 1 instruments were tested for ability to withstand shock * * * and that two of the instruments passed the required shock test.” What it failed to consider, however, was the final conclusion registered by Mr. Williams in reviewing the results with respect to both Model 1 and Model 1A. The results of this testing indicated that only 50 per cent of the instruments successfully passed the shock — 515 — test and that the primary reason for instrument failure was due to poor construction and quality control. These results indicate that it is probable that a large percentage of the total procurement of CD V-Ylfi’s manufactured by Lionel Electronics, Inc., are incapable of _ meeting the shock test requirements of [the specifications]. [App. Ex. 44, p. 56.] Thus, while two of the three No. 1 units passed this test, only one of three No. lA’s passed. More significant, however, is the fact that the units were mounted by means different than that used in the shock test performed on the Landers units, from which the Board concludes the Landers deficiency. In the Lionel test “[t]he meters were clamped to a plywood mounting board 'bolted directly to the mounting shelf of a lightweight high-impact shock machine * * *”15 whereas in the Landers testing procedure the instruments were bolted directly to a steel shock plate mounted onto the shock machine. The nature of the mounting is, of course, all important when testing for shock resistance. The fact that the Landers units did not pass the test, while the Lionel models did, is, therefore, of no real comparative significance. What is notable is that in testing four Landers units and two Lionel units under the latter type conditions, all six units failed to withstand the impact of the 2-foot drop test. As the resultant report signed by Mr. Williams states: * * * The principal reasons for the instrument being unable to meet the requirements of this test were due to meter and printed circuit board failure. Three of the LFC [Landers., Frary & Clark] instruments contained a 3/64" thick circuit board, and the remaining instrument contained a thick circuit board. Both the Lionel instruments were manufactured with a thick printed circuit board. The thickness of the circuit board seemed to have little effect on the capability of the instruments to pass these tests. The reason for the circuit board failure was that the transformer is mounted on the rear edge of the circuit board, which is not supported on the rear edges. Under impact the transformer acts as a lever on the circuit board, causing it to break at the rear mounting pins. Meter failures were due primarily to the inability of the spring coil to withstand the shock. — 516 — It is thus apparent that both brands of survey meters were unable to pass the shock testing standards imposed on plaintiff due to identical faults in the design and placement of the circuit board relative to the transformer. This leaves as yet unanswered only the question of whether, as argued by plaintiff, the contracting officer’s letter of November 1,1962, called for a change in the contract requirements or, as the Board concluded, made no change in the specifications but merely called upon plaintiff to conform to the contract as originally stated. This court was recently confronted with a similar situation in Ets-Hokin Corp. v. United States.16 There plaintiff contracted with the Army Corps of Engineers to construct the Integrated Mission Control Center at the NASA site in Harris County, Texas. Included among the contract requisites was installation of an automatic fire detection and alarm system. In bidding on the project, plaintiff based its final figure in part on projected usage of an alarm system identical to one it had previously installed in other buildings at the Ctenter. The contracting officer, however, disapproved of this system, stating it did not meet contract requirements, and demanded that an ionization type system be installed, even though the contract did not specifically call for such a system. Plaintiff proceeded to install this system and, treating this directive as a constructive change, thereafter sued for an equitable adjustment under the changes article. While the Corps of Engineers Board of Contract Appeals denied plaintiff’s appeal, it did not however hold that the system originally contemplated by plaintiff was incapable of meeting the contract requirements. The Board merely held that plaintiff did not supply defendant with the system it wanted, even though (as the Board conceded) plaintiff did not know defendant’s intention nor could it reasonably have been expected to know such. This court reversed the Board’s decision, concluding that under the experienced and intelligent general contractor standard, plaintiff did act reasonably. The court held that plaintiff was entitled to an equitable adjustment “by reason of the constructive change ordered by the' contract — 517 — ing officer with respect to the fire detection equipment.”17 In applying the same standard in the present case, we conclude that plaintiff’s reading of the contractual provisions and its actions subsequent thereto were reasonable under all the prevailing circumstances. Therefore, a 'directive from the contracting officer calling upon plaintiff to proceed beyond that which was reasonably interpreted to be the contractual obligation must be viewed as a constructive change order affording plaintiff recovery as a result. CONCLUSION OF LAW Plaintiff’s motion for summary judgment is granted and defendant’s cross-motion is denied. Judgment is entered for plaintiff with further proceedings stayed pursuant to Pule 167 for a period of ninety (90) days to afford the parties an opportunity to obtain an administrative resolution by the agency of the adjustment in the contract price commensurate to any additional amounts expended by plaintiff in adhering to the change order. Footnotes [1] In May 1966 Landers, Frary & Clark merged with The J. B. Williams Company, thereby assuming the parent’s name. At the close of the Board hearing, The J. B. Williams Company was substituted as appellant in the administrative proceeding. [2] GSABCA No. 1460, 68-2 BCA ¶ 7276. [3] Plaintiff alternatively claims entitlement to damages for breach of contract on the theory that (a) the defendant misrepresented facts and failed to disclose vital Information to plaintiff concerning the lack of voltage saturation of the Lionel model, and (b) a more stringent and demanding testing procedure was required of Landers for spectral dependence and shock resistance than that contemplated by the contract. Since we find the argument under the changes clause persuasive as to plaintiff’s claim, we feel it unnecessary to resort to any analysis under this separate count of the petition. [4] This Is the Identical specification as that under which the Lionel Model No. 1 had been produced. [5] Soon thereafter the Government exercised an option to increase deliveries to 82,063, thereby boosting the total contract price to $1,440,000. [6] The transcript contains several passages which emphasize this point. For example: “Q. * * * In making this bid did you prepare it in the belief you were offering to duplicate the design of a product that had been fully developed and approved by the Government? “A. [Mr. Strauss] I just assumed that this thing was letter perfect. I knew it was being made in production; I knew it was a critical instrument and I knew that Brownell, the same man that was coming to see us was inspecting the thing down at Lionel. * * *” ***** “Q. Mr. Strauss, was there anything in the IFB for this procurement or the circumstances surrounding the procurement which indicated to you at the time you prepared the bid for this contract that the Lionel Model 1 was not entirely acceptable to the Government? “A. No.” ***** “Q. Did the invitation and circumstances of this procurement otherwise indicate to you when you prepared the bid for this contract that you were bidding for a regular production contract? “A. Yes, regular production contract. “Q. And did you arrive at that conclusion from the circumstances of the procurement? “A. Certainly. The models, the need for preproduction models with the bid and after the bid had been deleted and the exceptions to the specifications had been listed and I assumed that was it and I also believed in the integrity of the engineering design that the Government furnished as made by a previous manufacturer on a manufacturing basis and accepted by the Government because we got the things from the depot. “Q. Did the delivery schedule have anything to do with your conclusion? “A. Naturally. There was no time to do anything except production engineering.” [7] Both Strauss and Raymond testified that Landers would not have bid on this contract had anything more than only incidental engineering changes been contemplated. [8] “Q. Are electronic components identical when manufactured by two vendors? “A. No. “Q. Are you speaking physically or electrically? “A. Principally physically. If they are purchased to the same procurement documents, drawings, specifications, then they must be equivalent electrically. They do not necessarily appear the same physically or have the same physical dimensions.” [9] For example, on pp. 82-3, Mr. Raymond testified: “We made minor changes in the artwork of the printed circuit board. These were done to accommodate the components of different suppliers or more than one supplier. * * * “We also noticed the Lionel unit in some regards the printed wiring was in very close proximity which we thought would lead to production problems of bridging in the way of soldering marks causing short circuits, so we made what we thought was an improved layout in the circuit board. “These were the changes as opposed to the Lionel for getting a better yield and making instruments more producible.” [10] While OCD assumed the June 20 and August 1 test reports dealt with the Lionel Model 1A, they inferred, due to the similarity of the instruments and the increased voltage in the Model 1A (a factor which would tend to produce a better saturation level), that the Model No. 1 was likewise deficient as to voltage saturation. [11] It should be noted that these Landers units were not taken from the Initial lot of 300 but rather were units to which Landers had made a further design change, namely, centering the collector electrode to provide a more ideal geometry and accurate output. [12] See. e.g., Aerodex, Inc. v. United States, 189 Ct. Cl. 344, 350 n. 7, 417 P. 2d 1361, 1367 n. 7 (1969) ; North American Philips Co. v. United States, 175 Ct. Cl. 71, 80 n. 4, 81, 358 F. 2d 980, 985 n. 4, 986 (1966) ; Seaview Electric Co., ASBCA 6966, 61-2 BCA ¶ 3151, at pp. 16,371-3 where general discrepancy clauses similar to the one employed In the Instant Invitation were held not sufficient to provide the specific exception required to place responsibility on the contractor for defects in the drawings or models. Por cases holding that this court will not give literal effect to broad exculpatory clauses if the result is to negate other contract provisions and thereby relieve the Government of liability, see, e.g., Poorvu v. United States, 190 Ct. Cl. 640, 652, 420 F. 2d 993, 1000 (1970) and cases cited therein. [13] Autogiro Company of America v. United States, 181 Ct. Cl. 55, 67, 384 P. 2d 391, 400 (1967). [14] See Appeal File Dock. 80 and App. Exh. 38. After careful analysis, engineers concluded that the chamber voltage was insufficient in both instruments to attain the required 95 percent level. [15] App. Ex. 41, p. 1. [16] 190 Ct. Cl. 668, 420 F. 2d 716 (1970). [17] 190 Ct. Cl. at 681-2, 420 F. 2d at 724.