Patents

Pfaff v. Wells Electronics: When an Idea Becomes 'Ready for Patenting'

The Supreme Court replaced the Federal Circuit's vague 'substantially complete' standard with a two-part on-sale-bar test that can start the clock before a prototype ever exists.

February 10, 2025
A close-up of a semiconductor chip socket on a circuit board
Wayne Pfaff sold a custom chip-carrier socket to Texas Instruments from detailed drawings, months before he built one. Shutterstock
Educational content, not legal advice. This article explains general legal concepts. It does not create an attorney–client relationship. For your specific situation, consult a licensed attorney.

Pfaff v. Wells Electronics, Inc., 525 U.S. 55, No. 97-1130 (U.S. Nov. 10, 1998), is the decision that gave the on-sale bar its modern shape. Writing for a unanimous Court, Justice Stevens discarded the Federal Circuit’s malleable “substantially complete” inquiry and announced a crisp, two-part test: an invention is “on sale” within the meaning of 35 U.S.C. § 102(b) once (1) it is the subject of a commercial offer for sale and (2) it is “ready for patenting.” The second condition could be satisfied without a working prototype — drawings detailed enough to let a skilled artisan build the invention would do. On the facts, Wayne Pfaff had sold his socket to Texas Instruments from designs alone, more than a year before he filed, and so his patent claims were invalid.

At a glance

  • Case: Pfaff v. Wells Electronics, Inc., 525 U.S. 55, No. 97-1130 (U.S. Nov. 10, 1998)
  • Author: Justice Stevens, for a unanimous Court
  • Below: Federal Circuit reversed the district court in part, holding the patent invalid under a “substantially complete” rationale
  • Patent: U.S. Patent No. 4,491,377, a socket for mounting and removing leadless semiconductor chip carriers
  • Statute: Pre-AIA on-sale bar, 35 U.S.C. § 102(b)
  • Holding: The on-sale bar applies when, before the critical date, the invention is (1) the subject of a commercial offer for sale and (2) “ready for patenting,” shown either by reduction to practice or by drawings/descriptions enabling a skilled artisan to practice it
  • Disposition: Affirmed; claims held invalid

The facts: a sale from a sketch

The timeline did most of the work. In November 1980, Texas Instruments asked Wayne Pfaff to design a new device for mounting and removing leadless semiconductor chip carriers. Pfaff prepared detailed engineering drawings. By March 17, 1981, he had shown a sketch of his concept to TI representatives. On April 8, 1981, TI sent written confirmation of a previously placed oral purchase order for 30,100 of the new sockets, a transaction worth $91,155. Pfaff did not fill the order — or build a working prototype — until the summer of 1981.

He filed his patent application on April 19, 1982. That date set the critical date one year earlier, April 19, 1981. The commercial offer for sale, embodied in TI’s purchase order, plainly fell before the critical date. The only doubt concerned the second half of the eventual test: at the moment of the sale, the invention existed only on paper. Under the Federal Circuit’s then-prevailing approach, the question was whether the invention was “substantially complete,” a totality-of-the-circumstances judgment that had produced inconsistent results across cases. The Supreme Court granted certiorari to resolve that uncertainty.

The two-part test and the rejection of “substantially complete”

The Court’s central move was to read § 102(b) as written and to anchor the bar in something more administrable than a court’s after-the-fact impression of how far along an invention was. The statute denies a patent if the invention was “on sale” more than one year before the application. Justice Stevens reasoned that the phrase carries a definite meaning only if the invention has reached a point at which it could, in principle, be patented. An inventor cannot be penalized for selling something that does not yet qualify as an invention; but once it does qualify, the one-year grace period begins running on any commercial offer.

Hence the two conditions. First, there must be a commercial offer for sale — not merely experimental activity, internal development, or a tentative inquiry, but a definite offer that the other party could accept to form a binding contract. Second, the invention must be “ready for patenting.” The Court identified two ways to prove that readiness: by reduction to practice (a working embodiment), or “by proof that prior to the critical date the inventor had prepared drawings or other descriptions of the invention that were sufficiently specific to enable a person skilled in the art to practice the invention.” That second pathway is the doctrinally significant one. It severs the on-sale bar from the existence of a physical prototype and ties it to the enabling content of the inventor’s disclosure.

Applied to Pfaff, both conditions were met before April 19, 1981. TI’s purchase order was a commercial offer accepted in the ordinary course of business. And Pfaff’s drawings were detailed enough that the manufacturer built the sockets to his specifications without alteration — strong evidence that, as of the sale, the invention was ready for patenting. The Court therefore affirmed the invalidity of the asserted claims.

Why the “ready for patenting” prong matters

The genius and the hazard of Pfaff both lie in the second prong. By holding that an invention can be “ready for patenting” without a prototype, the Court closed a loophole and opened a trap. The loophole: an inventor could otherwise sell a fully specified design, reap commercial rewards, and delay reduction to practice in order to extend the de facto monopoly. The trap: inventors who think of patenting as something that happens only once a device physically works may not realize that mailing a complete drawing package to a customer can start an unforgiving statutory clock.

The decision also rationalized the doctrine’s purpose. The on-sale bar exists to discourage inventors from commercially exploiting an invention long before filing, to encourage prompt disclosure, and to give the public a settled expectation that ideas in commerce for more than a year are free to use absent a timely patent. A bright-line “ready for patenting” standard serves all three aims better than a fuzzy completeness inquiry, because parties can identify, at the time of a transaction, whether the clock has started.

Open questions

Pfaff answered the threshold question but left a generation of follow-on issues to the Federal Circuit. What exactly counts as a “commercial offer for sale” — does it track the Uniform Commercial Code, or some federal common-law gloss? Later cases such as Group One, Ltd. v. Hallmark Cards, Inc. leaned toward traditional contract principles, but the boundary between an offer and a mere negotiation remains fact-intensive. How does the experimental-use exception interact with a sale that is partly commercial and partly for testing? And how “specific” must drawings be to render an invention ready for patenting when the claimed advance lies in a parameter the drawings do not fully pin down? These questions continue to generate litigation, because the two-part test supplies the frame but not every answer.

Implications

  • Calendar the critical date early. Any commercial offer — a quote, a purchase order, a binding price — more than one year before filing can be disqualifying. Date every transaction and file before the anniversary.
  • A prototype is not a safe harbor. Detailed, enabling drawings can make an invention “ready for patenting” before anything is built. Do not assume the clock waits for a working unit.
  • Distinguish offers from negotiations. Preliminary discussions and non-binding inquiries are not “commercial offers for sale.” Structure early customer contacts with that line in mind.
  • Document experimental purpose. Where pre-filing activity is genuinely for testing, contemporaneous records of experimental intent can preserve patentability — but the burden is real.
  • The principle survives the AIA. Although Pfaff construed pre-AIA § 102(b), the “ready for patenting” framework continues to guide on-sale-bar analysis, and the Supreme Court reaffirmed the durability of “on sale” in Helsinn v. Teva (2019).

Frequently asked questions

Does the on-sale bar require that the inventor actually deliver a product? No. The bar turns on a commercial offer for sale, not on delivery or even acceptance in every instance. Pfaff had not delivered a single socket — or built one — when the clock started; TI’s purchase order for a fully specified design was enough.

What is the difference between “reduction to practice” and “ready for patenting”? Reduction to practice means a working embodiment of the invention exists (or a constructive reduction via a filed application). “Ready for patenting” is broader: it is satisfied either by reduction to practice or by enabling drawings or descriptions. Pfaff’s innovation was recognizing the second route.

Did the America Invents Act overrule Pfaff? No. The AIA rewrote § 102 and folded the on-sale bar into § 102(a)(1), but courts continue to apply Pfaff’s analytical structure. In Helsinn Healthcare S.A. v. Teva Pharmaceuticals USA, Inc. (2019), the Supreme Court held that “on sale” retained its established meaning under the AIA.

Authorities and sources