Andrew Corp. v. Gabriel Electronics, Inc.

Citations: 785 F. Supp. 1041; 23 U.S.P.Q. 2d (BNA) 1019; 1992 U.S. Dist. LEXIS 2701; 1992 WL 43441Docket: Civ. 83-0372 P

Court: District Court, D. Maine; March 2, 1992; Federal District Court

SummaryOriginalCytator Case TreatmentBackward CitatorAnalysis

EnglishEspañolSimplified EnglishEspañol Fácil

Andrew Corporation sued Gabriel Electronics, Inc. for infringing on its Knop patent (U.S. Patent No. 4,410,892) for a horn reflector antenna. The court previously established Gabriel's liability for infringement, and the damages phase was held from July 8 to July 16, 1991. The infringement period spans from the patent's issuance on October 18, 1983, to 1989. Under 35 U.S.C. § 284, the patent owner can recover adequate damages, specifically lost profits due to the infringement.

To successfully claim lost profits, the patent owner must demonstrate causation—showing that sales would have occurred but for the infringement—and provide adequate evidence for calculating the lost profits. Causation can be established in two ways: showing that the patent owner was one of two suppliers in the market, or by satisfying a four-part test from Panduit Corp. v. Stahlin Bros. Fibre Works, Inc., which requires proof of demand for the patented product, absence of acceptable noninfringing substitutes, capability to meet that demand, and the amount of profit the patent owner would have earned.

The evidence presented indicates that when Andrew Corporation entered the market in 1980, two competitors, Gabriel and Antennas for Communication, Inc. (AFC), were selling similar antennas. Gabriel introduced an infringing TH-10 antenna in late 1983, while AFC's antennas were non-infringing and remained on the market until at least 1988. The court determined that there was a third supplier in the horn reflector antenna market during the infringement period, impacting Andrew's claim for lost profits.

The Court of Appeals for the Federal Circuit established that a market can still be classified as a two-supplier market even if a third supplier exists, provided that the third supplier's products are not acceptable substitutes for the patented product or their sales are negligible compared to the main suppliers. In fiscal years 1984 and 1985, AFC held market shares of 16.4% and 9.8%, respectively, indicating a significant presence in the horn reflector antenna market. However, by fiscal 1986, AFC's market share declined to 6.8%, with only 30 antennas sold out of 442 total. Testimony from AFC's Vice-President of Sales and Marketing indicated that AFC's sales were not significant after 1985, noting a market dominated by the two main competitors. Sales further decreased in fiscal years 1987 and 1988, with AFC selling only 10 and 15 antennas, respectively. The court underscored that the mere existence of a competing device does not render it an acceptable substitute, particularly if purchasers prefer specific features exclusive to the patented product. The patented horn reflector antenna, designed to reduce interference and improve radiation patterns, met a long-felt need in the market, as evidenced by a significant increase in sales of antennas with these patented features following their introduction, while sales of alternative models declined sharply. This data supports the conclusion that the features of the patented antenna were not replicable by competing products, reinforcing its unique market position.

Eric Brooker, Vice-President of Engineering at Andrew, qualified as an expert in terrestrial microwave systems, testified that the AFC antennas proposed by the Defendant are inferior to the patented SHX10A antenna based on radiation pattern analysis. He stated that the AFC antennas do not serve as acceptable substitutes. Although Brooker's potential bias was noted, his conclusions were supported by evidence. Joe Echols, a former AT&T employee, corroborated the importance of radiation patterns in antenna selection, especially under route congestion conditions, despite asserting that patterns were not the sole consideration. Echols acknowledged that the infringing TH-10 antenna had superior patterns compared to the AFC model and indicated that other antennas could perform adequately, yet emphasized that buyers prioritized the better patterns of the patented antenna.

The Court determined that the AFC antenna is not a viable alternative, establishing that Plaintiff and Defendant operated within a two-supplier market for horn reflector antennas. Consequently, Plaintiff is entitled to a presumption of lost profits due to Defendant's infringement under the Panduit test. The first inquiry under this test confirms demand for the patented product, supported by evidence of substantial sales during the infringement period, particularly from AT&T. The second inquiry regarding acceptable noninfringing substitutes reaffirmed the rejection of the AFC antennas, and the Court also dismissed the Defendant's proposal of UHR antennas as satisfactory substitutes.

Defendant's witnesses acknowledged that Andrew's patented antenna demonstrated significantly superior radiation patterns compared to the UHR antennas produced by the Defendant. Engineer Dan Allen characterized the difference as akin to comparing a "Cadillac to a Ford." An internal memorandum from Gabriel highlighted serious deficiencies in the UHR antennas, which were deemed a competitive disadvantage. Notably, Gabriel ceased production of the UHR series shortly after launching the infringing TH-10 antennas, indicating a lack of viable alternatives.

Defendant posits that the TH-10C and TH-10X antennas are noninfringing alternatives. However, these models violate an existing court injunction, with the TH-10C exceeding the absorber lining requirement and the E- to H-plane ratio stipulated in the injunction. Similarly, the TH-10X also contravenes the injunction specifications. The Court has determined that any antenna violating the injunction is considered infringing, thus ruling the TH-10C and TH-10X as unacceptable noninfringing alternatives.

Additionally, Defendant claims that the ATH-10 antenna is a suitable noninfringing alternative. The ATH-10 features a corrugated feed horn, which does not meet the patent's requirement for a smooth-walled conical feed horn, thus falling outside the injunction's coverage. The Court concluded that the ATH-10 does not infringe the patent, either literally or under the doctrine of equivalents. Furthermore, the ATH-10 was not market-ready until late 1988, and precedents from cases like Panduit and Kaufman emphasize that the evaluation of noninfringing alternatives must focus on the period of infringement. Therefore, the ATH-10, not being fully developed until the end of the infringement period, cannot be considered a viable noninfringing substitute, despite Defendant's claims about potential earlier availability.

The Court determined that the ATH-10 antenna was not a viable noninfringing substitute from 1983 to the end of 1988, but was acceptable in 1989. To satisfy the third Panduit requirement for lost profits, the plaintiff demonstrated robust manufacturing capabilities through credible testimony from its manufacturing manager, indicating the ability to produce at least one antenna per shift and operate up to three shifts daily. The combined production of both the plaintiff and defendant did not exceed 569 horns annually, while the defendant sold approximately 1,200 infringing antennas during the infringement period. The plaintiff had sufficient marketing and sales capacity to address demand if the defendant had not infringed, as evidenced by its large sales force and the nature of the market for horn reflector antennas.

In calculating lost profits, the Court noted the Federal Circuit's endorsement of the incremental income method, which excludes fixed costs from profit calculations. The plaintiff claimed lost profits on horn antennas totaling $9,673,892 and further losses from price erosion exceeding $4 million. However, the Court found the plaintiff's accounting expert, Ms. Kone, to be less persuasive than the defendant's expert, Mr. Hoffman, who provided a broader and more compelling analysis. The Court largely rejected Mr. Hoffman's critique of Ms. Kone's assumption that the plaintiff would have captured all sales made by the defendant during the infringement period.

In Kaufman Co. v. Lantech, Inc., the Court established that a patentee who demonstrates a reasonable inference that infringing sales caused lost profits meets the burden of proof for entitlement to those profits. The Plaintiff has shown entitlement to lost profits for infringing sales through 1988 due to existing demand, lack of substitutes, and capability to make sales. The Defendant failed to rebut this claim regarding the Panduit factors for that period. However, in 1989, the presence of a noninfringing product, the ATH-10, led the Defendant to successfully argue that inferring causation for lost profits from infringing sales was unreasonable. Testimony from Ms. Kone indicated that 1,216 infringing antennas were sold by Gabriel from fiscal 1983 to 1989, but excluding sales from 1989 results in 1,188 antennas for which lost profits could potentially be claimed. 

Ms. Kone's methodology for calculating expected profits involved subtracting production costs from sale prices, considering both direct and overhead costs. She assessed fixed and variable costs by analyzing production processes at the Andrew Orland Park facility and engaging with department managers to understand cost activities. This analysis determined certain costs as fixed while others varied by 25-50% with production levels. The Court noted criticisms of Ms. Kone's cost analysis approach from Polaroid Corp. v. Eastman Kodak Co., which highlighted its imprecision and potential bias, supported by Mr. Hoffman's trial testimony. Additionally, Ms. Kone lacked documentation for her interviews that informed her cost assessments, making it difficult for the Court to evaluate their thoroughness or bias. Mr. Hoffman pointed out that her analysis is typically used for internal cost accounting rather than for litigation purposes.

Mr. Hoffman and the court in Polaroid emphasized a cost accounting method that evaluates fixed and variable costs by analyzing empirical data over a 4 to 5 year period. Significant changes in costs indicate they are not fixed, regardless of how they are labeled. Hoffman analyzed the Plaintiff's costs and found substantial changes over time, challenging the classification of many costs as fixed. His findings demonstrated that both macro-level selling and administrative costs, and micro-level costs related to specific departments, varied with sales. For instance, costs associated with the tool crib significantly changed during the damage claim period, contradicting fixed cost assumptions.

Hoffman's analysis also revealed that depreciation, contrary to Ms. Kone's conclusions, aligned closely with sales, indicating it is not a fixed cost. He noted a consistent pattern of rising depreciation expenses relative to revenue across various businesses. Additionally, as Gabriel's sales grew, so did its general and administrative overhead costs, which was anticipated in Plaintiff's own business plan regarding horn antennas. This evidence casts doubt on Kone’s lost profits analysis, which categorized many costs as fixed despite rising sales.

Further, Hoffman pointed out that Plaintiff's analysis overlooked costs from departments like accounting and personnel, which would typically increase with sales volume. Consequently, the court found Kone's cost analysis unreliable, discrediting her lost profits calculations. Additionally, the Plaintiff seeks recovery for profits lost on accessory items sold alongside the horn reflector antenna, known as convoyed sales. If the Plaintiff can reasonably demonstrate that it would have achieved similar sales as the infringer, the expected net profits from those sales will measure the loss, according to the precedent set in Paper Converting Machine Co. v. Magna-Graphics Corp.

Plaintiff claimed a loss of $2,891,792 due to lost convoyed sales, supported by Ms. Kone's testimony and documentation. The Court acknowledged that the items listed as accessories in the documentation would typically be sold with the antennas, which the Plaintiff would have sold but for the infringement. However, the Court determined it was not reasonably probable that Plaintiff would have made all sales of the 281 waveguide sold by the Defendant during the infringement period, as these were not consistently sold together with the antennas. Consequently, the Plaintiff failed to demonstrate entitlement to lost profits from convoyed sales.

Ms. Kone's methodology for calculating the cost of goods sold for convoyed sales was deemed flawed, undermining her lost profits figures. She also testified to experiencing a 15% price erosion on horn antennas due to competition with Gabriel, with a corresponding 13.46% erosion on Gabriel's products, leading to a claim of $4,088,835 for price erosion. The Court rejected this price erosion figure, noting it did not consider the established economic principle that price increases typically result in reduced sales volume. The Plaintiff's business plan indicated that a price increase would significantly impact sales, further questioning the reliability of Ms. Kone's calculations.

Ultimately, the Court concluded that the Plaintiff did not provide proper evidence for calculating lost profits and, therefore, was not entitled to a damages award for lost profits. In instances where lost profits cannot be established, the patent owner is entitled to a reasonable royalty under 35 U.S.C. 284. The Court emphasized that determining a reasonable royalty should not be viewed as equivalent to standard royalty negotiations, as this would ignore the reality of the infringement.

Determining a "reasonable royalty" following infringement relies on a legal fiction, as there is typically no actual willingness on either side to negotiate a license. The reasonable royalty is assessed based on specific facts of each case, including the nature of the plaintiff's property, the extent of the defendant's infringement, the property's usefulness and commercial value, and the broader commercial context. The plaintiff possessed rights to a significantly improved horn antenna, which was commercially successful by the start of the infringement period. The plaintiff also had a policy against licensing its patents to competitors to protect its investment in research and development.

The defendant presented evidence of historical licensing rates around 4% for similar microwave technologies, which could be reduced in certain instances. Prior to the infringement, the plaintiff predicted a profit margin of 4.8% on the antenna's sale. The absence of viable non-infringing alternatives indicated that any licensing would negatively impact the plaintiff's own sales. The defendant recognized the need for an antenna comparable to the plaintiff's successful SHX10A model, as its own products had not met customer expectations.

Taking into account the plaintiff's effective product and the defendant's competitive pressures, the court concluded that a royalty rate of 10% would be reasonable, acknowledging that this would limit the defendant's profit potential. The court rejected the defendant's suggestion of a 2.8% royalty, stating that it did not reflect the realities of the marketplace and would not represent a rate that a willing patent holder and a willing license seeker would negotiate.

The Court rejects the Plaintiff's proposed reasonable royalty rate of 34.36%, which was based on the average of its lost incremental profit from the patented product. The disparity between the Plaintiff's profitability projections for the SHX10A and the Defendant's actual profits from the TH10 led the Court to conclude that no hypothetical willing parties would agree to such an unreasonable figure. According to the Federal Circuit's interpretation of the entire market value rule, damages may be based on the value of an entire apparatus when the patented feature drives customer demand. Therefore, sales of unpatented items that hypothetically would benefit from the patented invention should be included in the royalty base. The Plaintiff successfully demonstrated that the Defendant frequently sold accessories with its horn reflector antennas, as evidenced in the trial transcripts. However, while the Defendant likely sold some waveguides with the antennas, it was not consistent. The Court found insufficient evidence to establish what portion of waveguide sales constituted lost convoyed sales due to the infringement, thus excluding these figures from the royalty computation.

Regarding prejudgment interest, the Court states that it should generally be awarded on reasonable royalty patent damage awards, as supported by General Motors Corp. v. Devex Corp. The Court acknowledges that interest may be denied in cases of undue delay caused by the patentee; however, it does not attribute the lengthy proceedings solely to the Plaintiff or the vagueness of the Knop patent. The Court concludes that the Plaintiff is entitled to prejudgment interest calculated at the prime rate, compounded monthly, to fairly compensate for the loss of use of funds during the litigation. Schedule IV of PX 291 indicates the average interest rate during the infringement period, but its calculations are deemed inaccurate due to an incorrect base.

Finally, the Court notes that under 35 U.S.C. § 284, it may enhance patent damage awards if the infringement is found to be willful.

The Court initially believed that the Defendant had likely copied the Plaintiff's antenna design for its TH-10 model. However, after reviewing new evidence and trial records, the Court concluded that the Defendant did not copy the Plaintiff’s design in the early 1980s. Testimony from Mr. Whiting, the Defendant's Vice-President of Engineering, was found to be sincere and convincing regarding the development of the antenna, supported by equally persuasive testimony from Mr. Gemme. The Court determined that the Defendant had not copied the horn antenna design and acted appropriately to avoid patent infringement while remaining competitive. The Court faced challenges in determining the validity of the patent and the specifics of the alleged infringement. The Defendant sought legal advice during the proceedings and justifiably relied on what appeared to be sound counsel, leading the Court to find no basis for willful infringement. It also concluded that this case was not exceptional enough to warrant attorney fees under 35 U.S.C. § 285. Consequently, the Defendant is required to pay the Plaintiff a 10% royalty on sales of infringing horn reflector antennas, excluding the ATH-10, from October 18, 1983, to 1989, with interest accruing at the prime rate, compounded monthly. The parties must attempt to agree on the award amount and submit a proposed judgment within 15 days; if they cannot agree, both parties must submit their proposals along with supporting arguments within the same timeframe for the Court to resolve.

The Court prioritizes PX 291 and its detailed testimony over DX 1A due to its superior clarity. Ms. Kone's testimony indicates that AFC's sales for fiscal 1986 were 27 units, suggesting a reduced market share. In fiscal 1987, PX 291 reports 48 antennas sold by AFC, but after testimony, the Court finds that 38 were sold to a U.S. company for overseas installation, rendering them irrelevant for domestic market analysis. 

The proposed noninfringing substitute for the AFC antenna features a fiberglass horn, which, according to Mr. Crisman's testimony, had fallen out of favor by 1986 due to reliability concerns compared to metal. The absence of fiberglass in competing horns is deemed a desirable trait, further disqualifying AFC's antenna as an acceptable substitute. 

The Defendant argues against evaluating the TH-10C's infringement status without considering the injunction. The Court is unconvinced by the Defendant's claims that the TH-10C does not infringe due to differences in absorber type and function. It applies the doctrine of equivalents, which assesses whether the TH-10C performs the same function in a similar manner to achieve the same result as the patented invention. The Court previously ruled that placing the absorber 54 inches deep below the shelf is akin to the patented method, and an E-to-H-plane ratio of 1.3 achieves similar results. 

Despite the Defendant's claims regarding differences in absorber type and function, the Court finds no compelling evidence that the pyramidal absorber in the TH-10C serves a different purpose than that in the patented device, affirming the testimony from the prior trial that highlighted essential functional distinctions.

The absorber serves a specific purpose by establishing a frequency-independent boundary condition that supports the HE11 mode, distinct from merely absorbing signals. The Court notes that the pyramidal absorber's deep placement in the TH-10C cone, rather than just in the shield, indicates its role in shaping the main beam by reducing the E-plane pattern. Consequently, since the TH-10C operates similarly to the patented antenna, it infringes under the doctrine of equivalents.

The Plaintiff sought to exclude Mr. Hoffman's testimony, claiming insufficient discovery of his opinions before trial. However, the Court admitted his testimony de bene esse, noting that the details of his critique of Ms. Kone's assumptions were disclosed in his deposition. The Court found no grounds for exclusion, as Plaintiff's counsel effectively cross-examined Mr. Hoffman, and Ms. Kone's presence enabled adequate preparation for this cross-examination. Additionally, rebuttal testimony was provided by Mr. Brooker, reinforcing the absence of surprise or prejudice to the Plaintiff.

Regarding the relevance of Mr. Hoffman's regression analyses, the Plaintiff argued they were improperly used to establish a cause-and-effect relationship between sales volume and costs. While Mr. Hoffman acknowledged that regression analysis alone cannot prove causation, he clarified its utility in demonstrating correlations between variables. The Court found his regression analyses valuable, in conjunction with other evidence, in supporting the Defendant's argument that production costs increase with sales volume.

Rule 401 of the Federal Rules of Evidence defines "relevant evidence" as evidence that influences the probability of a consequential fact in a legal action. The regression analyses presented at trial satisfy this relevance standard. Citing Polaroid Corp. v. Eastman Kodak Co., the Court emphasizes that regression analyses are more accurate than the interview method used by the Plaintiff. The Court finds the business plan relevant, not for its specific projections but for the economic principles it illustrates, particularly regarding price increases and volume reductions. 

While the Plaintiff claims a conservative 15% price erosion figure, the Court cannot determine if this adequately offsets potential sales declines due to price increases. Despite the Plaintiff's increased antenna sales in 1989, the Court is hesitant to generalize this trend over the entire infringement period, noting that external factors, including competition and technological developments, influence market behavior. 

The Defendant achieved a 7.2% profit on TH-10 antenna sales and argues that the royalty awarded must allow for profit, which the Court clarifies is not a legal requirement. The Court considers profit as a factor in determining a reasonable royalty, particularly since horn reflector antenna sales are a significant part of the Defendant's business and the Plaintiff sought legal advice to modify its products to avoid patent violations post-decision.

The Court rejects the Defendant's argument that the reasonable royalty rate should be calculated as half the difference between the production costs of the ATH-10 and the TH-10, citing the known disadvantages of the corrugated horn. It highlights the significant risk the Defendant faced in not marketing the successful patented device, which could have led to a competitive disadvantage in a challenging market. The Court believes it is unlikely the Defendant would risk its competitive position by relying on a new product with known drawbacks when it could license and market the successful technology while developing its own. Additionally, the Court recognizes that the period during which the Plaintiff was deprived of funds coincided with a favorable investment climate, considering the prime rate a conservative measure for compensating the Plaintiff. Evidence from a related case, Andrew Corp. v. Gabriel Electronics, is acknowledged as providing context on the development of horn reflector antennas by the Defendant.