BALDWIN, Circuit Judge.

Western Marine Electronics, Inc. (Wes-mar) appeals from an October 4, 1983 judgment of the United States District Court for the Western District of Washington holding claims 1, 14, 16, and 17 of U.S. Patent No. 3,553,638 (the Sublett patent) invalid under 35 U.S.C. § 102(b) and not infringed by certain sonar devices manufactured by Furuno Electric Co., Ltd. (Fu-runo). Furuno cross-appeals from the district court’s denial of an award of attorney fees. We affirm those parts of the district court’s judgment holding the Sublett claims invalid and denying attorney fees. In view of our decision on invalidity, we do not reach the district court’s non-infringement finding.

Background

The invention of the Sublett patent relates to gravity stabilized sonar devices used on fishing boats. A sonar device includes a drum-shaped acoustical transducer, which sends out short-duration and highly directional sound waves (analogous to a beam of light) which travel through the water and are reflected back from objects in the water, such as fish or other submerged objects. By knowing the direction that the transducer is facing when it detects this reflected sound or “echo”, one can determine the location of the detected object. The direction of the trans *842 ducer’s beam is defined by a vertical component which measures the angle of the beam with respect to a horizontal plane such as the surface of the water, referred to as the “tilt” angle, and by a horizontal component which measures the beam angle about a vertical axis such as the angle between 0 and 360 degrees around the vessel, referred to as the “scan” angle.

One of the objects of the Sublett invention was to stabilize the transducer against the pitching and rolling motions of a ship. The sonar system is gravity stabilized by means of a pendulum suspension mechanism. The mechanism includes a rotatable frame with support arms extending from the frame which carry the transducer like a pendulum. The pendulous suspension allows the transducer to remain relatively stable despite the pitch and roll of a ship. In addition, the transducer may be rotated and tilted to alter the orientation of the transducer independently of the stabilizing suspension mechanism.

The inventor, Kenneth Sublett, who was chief engineer of Wesmar, conceived of the stabilized sonar in late 1967. The Sublett patent application was filed on June 19, 1969, and the patent issued on January 3, 1971. Claim 14 of the patent is reproduced in the Appendix. Claims 1, 16, and 17 are reexamination claims.

Sublett’s conception was reached under intensive pressure from Wesmar’s president, Bruce Blakey, who perceived a substantial need in the fishing industry for a sonar device which would be capable of the tilt and scan functions, but be unaffected by the pitch and roll of a vessel. Although the prior art disclosed equipment capable of stabilizing the transducer against pitch and roll, the equipment used motion sensing devices such as gyroscopes which were expensive and bulky. For this reason, the fishing industry had opted for unstabilized sonar devices. Two such devices were Wesmar’s SS-100 and SS-200 sonar devices.

Wesmar’s SS-300 sonar model embodies the Sublett invention. The SS-300 also includes a dual frequency feature which permits operation at low frequency (long range) and high frequency (short range), whereas the earlier SS-200 model operated at a high frequency and thus had a short range.

District Court Decision

In a judgment dated October 4, 1983, rendered after trial without a jury, and based upon a separately filed opinion and findings of fact and conclusions of law, the district court held the claims in suit invalid under 35 U.S.C. § 102(b) 1 on the ground that the Sublett invention was placed on sale more than one year prior to application for patent.

The court determined that prior to June 19, 1968 (the critical date, i.e., one year before the application filing date), Wesmar had placed the SS-300 sonar device on sale. Early in its discussion of the on sale issue, the district court noted that Wesmar had not been forthright in producing certain documents, and in light of other key documents produced by Furuno, Wesmar’s non-production was suspicious. In addition, the court found that Wesmar’s interrogatory answers relating to the on sale issue were “at best, slipshod and, at worst, deliberately misleading” and that certain interrogatories, answered falsely by Wes-mar, were corrected “two years after the answers were served and only after [Furu-no] obtained proof of their falsity.”

Key documents relied on by the district court included:

1. production drawings, dated in February, March, and April 1968, showing parts critical to the stabilization function;

2. invoices and delivery receipts, dated January through May of 1968 and found in the possession of a machinist and a found *843 ry, showing the manufacture and delivery to Wesmar of parts necessary for Wesmar to construct the stabilized sonar devices;

3. a June 5, 1968 letter from a Wesmar manager, Fenwick Macintosh, to a customer stating in part:

Incidentally, the SS300 is now in production. Among other new features, this unit is equipped with stabilized sound dome and a 360 degree sweep. It is a dual-frequency unit providing long and short range; for search and then pinpointing the exact location of a school of fish. This unit will be in the price range of $9,000.00.

and,

4. a June 7, 1968 letter from Wesmar’s president, Bruce Blakey, to Peter G. Schmidt, president of Marine Construction and Design Company (Marco), a distributor of Wesmar products in Peru, which provides in relevant part:

We have completed our pricing exercise on the SS300 and have established the f.o.b. Seattle selling price of $7,300. Marco’s cost on this item is $4,830, and Marco’s margin is therefore $2,470.

To accomplish introducing the unit .in Peru, it is recommended that Marco switch the customer who has ordered (2) Model SS200’s for installation prior to November 1, to the new Model SS300 unit. Delivery can be accomplished, we believe, within two weeks.

In conclusion, it is my opinion that from studying the John O’Neal reports, the Mel Souter report, attending the debriefing session with Marco and from our subsequent discussions with John that the trip was worthwhile. The technical points raised can be corrected easily, with the exception of range, of course. The fact that we now have the SS300 with its low frequency capability enhances our position in this regard. In my opinion, with aggressive action, we can now move to capture a fair share of the market.

The letter contains a handwritten notation “John — OK to buy one SS300 at SS200 price per PGS 11 June.”

The court found that the Sublett invention, “insofar as the stabilization mechanism is concerned,” was completed well prior to the critical date. From the documentary evidence, the court determined that the stabilization mechanism was made and in production by early June of 1968. An engineering model was built by mid-April of that year, parts for at least three additional stabilization mechanisms were ordered in May and delivered in June; and on June 12, Wesmar ordered six additional “yokes” (a critical part of the mechanism) from a foundry “Due ASAP.” The court also found that Wesmar had an SS-300 permanently mounted on a test boat in Seattle for demonstrating the sonar to potential customers in the “summer” of 1968.

As to testing, the court found that Wes-mar had ample time to field test the SS-300 prior to the critical date, and that the relative simplicity of the mechanism did not require in-depth testing to determine utility or functionality.

Although Wesmar contended that the June 7 letter proposes that the device be “evaluated” in Peru and that shipping documents and internal documents between Marco and Wesmar refer to the shipment of the SS-300 as being for experimental purposes, the court concluded that the documentary evidence and the testimony from the witnesses established that the transaction with Marco was not for experimental purposes. For example, the court found it significant that a Marco manager, Chuck Hart, who was the Marco employee most familiar with the transaction, testified that the Marco purchase of the SS-300 was not connected with any experimental purpose. From the testimony and the exhibits, the court found that the shipping documents referred to “experimentation” in order to avoid substantial Peruvian duty payments. The court also credited the deposition testimony of Wesmar’s Mr. Macintosh that he remembered seeing stabilized sonar devices being constructed at Wesmar *844 in April 1968 and had specifically installed stabilized devices in vessels in Canada during April, May, and June of 1968. As for the June 7 letter, the court found that it imposed no conditions of reporting, testing or any other condition that might restrict use or resale of the SS-300, and did no more than suggest market testing. Finally, the court found that any evaluation contemplated in Peru was of the SS-300’s dual frequency innovation rather than Sub-lett’s claimed stabilization invention.

The district court denied Furuno’s request for attorney fees finding that the case was not “exceptional” for purposes of 35 U.S.C. § 285. The court noted that Fu-runo had been compensated to some extent for Wesmar’s discovery abuse by court-imposed monetary sanctions.

The thrust of Wesmar’s appeal concerning the on sale determination is directed to the issue of reduction to practice. Wesmar contends that the district court applied an improper legal standard in determining when the Sublett invention had been reduced to practice; erroneously determined that the invention was reduced to practice prior to the critical date; and therefore erroneously held the claimed invention to be on sale prior to the critical date. In particular, Wesmar argues that there is no evidence that a stabilization mechanism embodying the claimed invention was installed on any vessel prior to the critical date. And, it argues, in the absence of on-board testing of the invention in the environment for which it was intended, there can be no on sale bar here. Wesmar also disputes numerous credibility determinations made by the trial court.

OPINION

It is rare that an appellate court will have before it an “easy” section 102(b) case. One court has observed that this area of the law “encompasses an infinite variety of factual situations which, when viewed in terms of the policies underlying § 102(b), present an infinite variety of legal problems wholly unsuited to mechanieally-applied, technical rules.” 2 Rigid standards are especially unsuited to the on sale provision where the policies underlying the bar, in effect, define it. See TP Laboratories, Inc. v. Professional Positioners, Inc., 724 F.2d 965, 973, 220 USPQ 577, 583 (Fed.Cir. 1984) (public use).

As a result, this court has been careful to avoid erecting rigid standards for section 102(b). In Barmag Barmer Maschinenfabrik AG v. Murata Machinery, Ltd., 731 F.2d 831, 221 USPQ 561 (Fed.Cir.1984), the court reviewed a lower court’s application of the three-part test of Timely Products Corp. v. Arron, 523 F.2d 288, 302, 187 USPQ 257, 267-68 (2d Cir.1975), for determining when an offer of sale of equipment to be later manufactured started the running of the statutory time period for filing a U.S. patent application. While approving of the Timely Products criticism of the “on hand” doctrine because the doctrine often produces results contrary to the basic policies underlying the on sale bar, this court nevertheless declined to adopt the Timely Products test for all cases:

The Timely Products standard, while requiring less for “on sale” activity than the “on hand” test, is, nevertheless, restrictive in that an offer to sell, without the existence of a physical embodiment of what is offered, does not start the running of the time period. It is not difficult to conceive of a situation where, because commercial benefits outside the allowed time have been great, the technical requisite of Timely Products for a physical embodiment, particularly for a simple product, would defeat the statutory policy and we, therefore, do not adopt the Timely Products test as the answer in all cases.

Barmag Barmer, 731 F.2d at 837, 221 USPQ at 565.

Recently, in Shatterproof Glass Corp. v. Libbey-Owens Ford Co., 758 F.2d 613, 225 USPQ 634 (Fed.Cir.1985), the court emphasized that a bare offer to sell cannot be viewed in a vacuum but must be considered *845 in connection with the surrounding circumstances. Although Shatterproof had begun soliciting, prior to the critical date, orders for products which would be made by the claimed apparatus and method, the court determined that a reasonable jury could have found that the apparatus and method of the claims were not functional as of the critical date. Based on this lack of functionality and on the surrounding circumstances, the court held that a reasonable jury could have determined that the on sale bar was not satisfied.

The authorities agree that the facts of each case must be weighed in view of public policy. Public policy favors prompt and widespread disclosure of inventions to the public, while giving the inventor a reasonable amount of time (1 year, by statute) to determine whether a patent is worthwhile, but precluding attempts by the inventor or his assignee from commercially exploiting the invention more than a year before the application for patent is filed. See, e.g., In re Caveney, 761 F.2d 671 at 676 (Fed.Cir., 1985); Barmag Barmer, 731 F.2d at 836, 221 USPQ at 565; D.L. Auld Co. v. Chroma Graphics Corp., 714 F.2d 1144, 1147, 219 USPQ 13, 16 (Fed.Cir.1983); General Electric Co. v. United States, 654 F.2d 55, 61, 211 USPQ 867, 873 (Ct.Cl.1981).

A determination of whether an invention has been placed on sale within the meaning of section 102(b) must be decided on the facts of each case. To give effect to the underlying policies, the court will want to consider the totality of circumstances relating to the character and extent of commercial activities, the type of invention and its stage of development as evidenced by engineering models, prototypes, and production models, along with the character and extent of bona fide experimentation.

After considering the record and the arguments presented on appeal, we hold that the district court did not err in concluding that the Sublett invention as covered by claims 1, 14, 16, and 17 was on sale prior to the critical date, and that the claims were therefore invalid. 3

In answer to Wesmar’s reduction to practice argument, we reply that the district court properly looked to the totality of circumstances in determining that the claimed invention, as embodied in a structure meeting all the elements of the claim, was on sale prior to the critical date. The district court found the sonar system “complete” well before the critical date. And, although the evidence on testing is less than clear, 4 we see no clear error in the district court’s determination that it would be inconsistent with the surrounding circumstances to find that Wesmar required further testing to determine the invention’s utility or functionality.

The starting point must be the June 7, 1968 letter from Mr. Blakey to Marco’s Mr. Schmidt offering to sell the SS-300 device. The letter establishes the selling price f.o.b. Seattle, Marco’s cost and profit margin, and estimates delivery within two weeks. “To accomplish introducing the unit into Peru,” Mr. Blakey recommends that Marco switch a customer who has ordered two SS-200 models to the new SS-300 unit. The letter concludes by saying that “with aggressive action, we [Wesmar and Marco] can now move to capture a fair share of the market.” The handwritten notation “OK to buy” on the exhibit evidences Marco’s understanding that Wesmar was communicating an offer to sell which Marco could (and did) accept.

*846 The letter must also be considered in light of Mr. Macintosh’s June 5 letter announcing to a sales agent that the SS-300 “is now in production,” and correspondence in early March from Marco to Wesmar indicating that a competitor had introduced into Peru a new sonar with five times the range of Wesmar’s SS-200 sonar. The latter might reasonably supply Wesmar a motivation to quickly commercialize its SS-300 sonar in Peru.

We see no clear error in the district court’s finding that stabilization mechanism was complete prior to the critical date. Documents show that certain parts drawings had been made and machined parts had been received by Wesmar from a machinist as early as January 1968. Mr. Sub-lett admitted that an engineering prototype existed at least by mid-April. Three additional SS-300 devices were constructed from parts ordered by Wesmar in May and delivered in June. Significantly, certain parts were made out of aluminum. Yet at least one Wesmar employee testified that Wesmar had built stabilized units using brass parts before switching to aluminum which suggests that there existed other units before the three units constructed with aluminum parts.

Moreover, two employees testified that SS-200 housings had been reworked to receive stabilized mechanisms. Mr. Macintosh, testifying by deposition, said that he had personally installed stabilized SS-200 devices in 16 vessels in Canada prior to the critical date.

Wesmar argues that Mr. Macintosh’s testimony is entitled to no weight because it was given fifteen years after the fact and is inconsistent with the invoices produced by Furuno during discovery showing that only enough parts were available to construct one device by April and three others in June. Wesmar faults the district court for noting that “Wesmar used several different suppliers” and “the absence of documents showing the manufacture of other stabilization mechanisms does not indicate that such mechanisms were not made or offered for sale prior to June 1968.” Wesmar says this amounts to imposing upon Wesmar the burden of proving a negative. We think not. The district court could properly draw adverse inferences from Furuno’s production and Wesmar’s non-production, and instead choose to credit the testimony of Mr. Macintosh. Macintosh reconstructed the events based on trip reports he had kept. He specifically identified by name the vessels in which he had installed stabilized SS-200 devices. Wes-mar did not controvert his testimony at trial.

Wesmar challenges the trial court’s findings by pointing to testimony that the SS-300, not the SS-200, was the first stabilized unit. Furuno, in support of the findings, points to a machinist’s invoice dated in January 1968 which refers to a “reworked” SS-200 housing, a gear drawing dated in April 1968 referring to the SS-200, and testimony that Wesmar did at times rework SS-200 housings to accommodate the stabilization mechanism.

We are not persuaded that the trial judge clearly erred in making these findings. Making credibility determinations and weighing the evidence are the unique province of the trial court.

Wesmar also challenges the district court’s findings that Wesmar’s activities were not primarily experimental. Wesmar characterizes the June 7 letter as expressing a purpose to “evaluate” the SS-300 device in Peru. According to Wesmar, shipping documents, Wesmar control over the device, and “test” reports all show an experimental purpose. However, we agree with the district court that the evidence points to market testing rather than experimentation.

A plain reading of the June 7 letter supports the finding that Wesmar made an unequivocal offer to sell the SS-300 sonar device. As noted by the court below, the letter imposes no conditions of reporting, testing, or control or restriction over use or resale. Mr. Hart, the Marco employee most familiar with the transaction, testified that there was nothing experimental about the transaction. Indeed, the court deter *847 mined that the testimony of Mr. Hart and the documents show that the real reason the shipping papers refer to experimental purposes was to avoid substantial Peruvian import duties. Wesmar’s argument as to its “control of the testing” also fails. Although the importation documents for the Peruvian transaction indicated that the SS-300 unit would be returned to Seattle in two weeks, in fact the unit was not returned but was later resold and used on a different vessel. Finally, Wesmar points to “test” reports prepared by Wesmar’s Mr. Brinkerhoff and a Marco employee as showing experimental purpose. The trip report prepared by Mr. Brinkerhoff never mentions the stabilization function at all while a later trip report prepared by the Marco employee (a copy of which was sent to Wesmar) merely states that the stabilized transducer was working very effectively. Brinkerhoff testified that the main thing he was interested in was the sonar’s dual frequency characteristic and that he “really [didn’t] remember how the stabilization operated.” The trial court properly recognized that testing or experimentation performed with respect to non-claimed features of the device does not show that the invention was the subject of experimentation. In re Theis, 610 F.2d 786, 793, 204 USPQ 188, 193-94 (CCPA 1979). The record shows, as corroborated by Mr. Hart of Marco, that the purpose of the Brinker-hoff trip was to help Marco “install” and “set up” the SS-300 sonar on a vessel, as well as to check out reported problems with a number of SS-200 devices which were already in use on other vessels.

In sum, we must agree that Wesmar’s June 7 offer letter together with the surrounding circumstances evidence a primarily commercial rather than experimental purpose. See, e.g., Pennwalt Corp. v. Ak-zona, Inc., 740 F.2d 1573, 1580-81, 222 USPQ 833, 838-39 (Fed.Cir.1984).

Attorney Fees

The district court responded to Furuno’s claim for fees as follows:

While Wesmar’s conduct in this litigation has been less than exemplary, the Court cannot find that the case is so exceptional as to fall within the category of cases embraced by the above statute. To some extent, defendant has been compensated for Wesmar’s discovery abuse by monetary sanctions imposed by the Court. Beyond that, the parties must be left to pay their own attorney fees.

Furuno’s main argument on appeal is that this case is exceptional because Wes-mar conducted the litigation in bad faith especially during discovery by intentionally concealing its “on sale” activities and by answering interrogatories falsely. Furuno contends that the lower court mistakenly believed that an award based on an exceptional case under 35 U.S.C. § 285 is mutually exclusive from an award for abuse of discovery under rule 37 of the Federal Rules of Civil Procedure, i.e., relief from discovery abuse is available only under rule 37.

While this court agrees that bad faith conduct of the litigation can make a case “exceptional” for purposes of 35 U.S.C. § 285, see Hughes v. Novi American, Inc., 724 F.2d 122, 125, 220 USPQ 707, 710 (Fed.Cir.1984), it also recognizes that judgments of the trial court concerning good and bad faith conduct are not easily overturned. We discern no error in the trial court’s determination of the extent to which the present discovery misconduct merited sanctions and whether it rendered the case exceptional for purposes of 35 U.S.C. § 285.

Conclusion

Having considered all of the arguments, we affirm those parts of the judgment holding claims 1, 14, 16, and 17 of the Sublett patent invalid under 35 U.S.C. § 102(b), and denying attorney fees.

AFFIRMED.

APPENDIX

A system for scanning and detecting schools of fish in a body of water comprising

a marine vessel;

*848 APPENDIX — Continued

a sonar system mounted in said vessel, said sonar system comprising

(1) a frame,

(2) a acoustic transducer capable of generating a sonar scanning beam and detecting sound waves reflected from objects such as fish,

said transducer pendulously suspended from said frame so that said scanning beam is substantially stabilized against the roll and pitch of said vessel, the face of the oscillatory driving element of said transducer mounted for rotation about an axis perpendicular to the vertical axis of said frame,

(3) transducer drive means for varying the angle of said transducer and scanning beam with respect to said vertical axis independently of the suspension of said transducer from said frame,

(4) scan drive means for rotating the frame and transducer about said vertical axis,

(5) a water tight housing enclosing said frame and transducer, and

(6) a high dielectric constant liquid within said housing surrounding said transducer; and

means for advancing said transducer to a scanning position below the keel of said vessel and for retracting it to a rest position when not in use.

Claim 14 of the Sublett Patent

Notes

. 35 U.S.C. § 102 provides in relevant part:

A person shall be entitled to a patent unless—

(b) the invention was patented or described in a printed publication in this or a foreign country or in public use or on sale in this country, more than one year prior to the date of the application for patent in the United States____

. Philco Corp. v. Admiral Corp., 199 F.Supp. 797, 815, 131 USPQ 413, 429 (D.Del.1961).

. Claims 1, 16, and 17 are directed to the sonar system. Claim 14 includes a "marine vessel.” That claim would technically be invalid under a combination of § 102(b) and § 103. In re Kas-low, 707 F.2d 1366, 217 USPQ 1089 (Fed.Cir. 1983). Wesmar does not assert that the claim is valid even if the sonar system is held to be on sale.

. There was testimony that Wesmar had "at that time” a boat for testing equipment in Puget Sound or in "one of the lakes,” and during the "summer" of 1968 had an SS-300 device permanently mounted on a test boat in Seattle for demonstrating the device to potential customers. Wesmar argues that the first stabilization mechanism to be tested "on the high seas” was the SS-300 shipped to Marco in July 1968 as a result of the June 7 letter.