Federal Circuit Clarifies Design Doctrine of Equivalents Limits

Patent law has many moods, but one of its favorites is this: “Nice try, but words still matter.” That is the big lesson from the Federal Circuit’s decision in Smartrend Manufacturing Group v. Opti-Luxx, a case that dealt with both a design patent and a utility patent for illuminated school bus signs.[1] The ruling is worth attention not just because it reversed one infringement judgment, vacated another, and wiped out an injunction, but because it showed how quickly a loose infringement theory can run into a brick wall when the patent record is more specific than the patentee would like.[1][7]

The headline phrase “design doctrine of equivalents” makes for catchy legal copy, but let’s keep the doctrine neat and the coffee unstirred: the Federal Circuit did not create some new free-floating design-patent version of the doctrine of equivalents. Instead, the opinion clarifies a set of limits that matter in design cases and in mixed design-and-utility disputes. On the design side, the court reinforced that design patents live and die by the claimed visual presentation and any narrowing words that accompany it. On the utility side, it reminded everyone that the doctrine of equivalents cannot be used like a magic cape to cover an accused product that lacks the specific functions the patent itself treats as important.[1][2][3][5]

What Happened in the Case?

Smartrend asserted two patents against Opti-Luxx: a design patent covering the ornamental design of an LED light panel, and a utility patent covering an illuminated school bus sign with a frame structure.[1] At trial, Smartrend won on both. The jury found infringement, the district court denied judgment as a matter of law, and a permanent injunction followed.[1][7] Then the Federal Circuit arrived and, in classic appellate fashion, turned the table over with surprising elegance.

For the design patent, the appellate court held that the district court had misconstrued the term “transparency.” The district court told the jury that “transparency” could mean both transparent and translucent. The Federal Circuit said no: transparent is not the same as translucent, and the description in the patent narrowed the claim accordingly.[1][7] For the utility patent, the court agreed that the claimed “frame” had to be a separate and distinct component, and then held that no reasonable jury could find infringement under the doctrine of equivalents because the accused integrated frame did not perform the functions the patent assigned to the claimed frame.[1][5][7]

In other words, the court did not merely tweak a jury instruction. It reinforced an old patent truth with new force: if your patent claims and specification talk like they mean something specific, courts may take you at your word. Dangerous, I know.

The Design Patent Side: Why “As Shown and Described” Matters

Design patent law has its own architecture. The governing infringement standard is the ordinary-observer test: would an ordinary observer, giving the attention a purchaser usually gives, think the accused design is substantially the same as the patented one?[2][6] Since Egyptian Goddess, that is the sole test for design patent infringement; the older “point of novelty” test is no longer a separate hurdle.[2]

But “sole test” does not mean “anything visually close goes to the jury.” The claim still has to be construed, and design patents can be limited by their drawings and by the words used to describe them.[1][3][4] That is where Smartrend gets interesting. The design patent claimed the design “as shown and described,” and the description stated that oblique shading lines in certain views “denote transparency.”[1] The district court treated that as broad enough to capture translucency too, relying in part on the MPEP’s explanation that oblique line shading can be used for transparent, translucent, or reflective surfaces.[1][4]

The Federal Circuit rejected that move. The MPEP explains what oblique shading may depict in drawings generally, but the patent’s own description narrowed what those lines meant in this particular patent.[1][4] Once the patentee chose the word “transparency,” the court refused to let later litigation quietly stretch it into “transparent or translucent.” Think of it this way: if a restaurant menu says “lemon pie,” the kitchen does not get to defend a key lime pie by saying, “Well, both are citrus-adjacent.” Patent law is even less forgiving than dessert law.

This part of the decision matters because it confirms that words in a design patent are not decorative filler. Under Curver Luxembourg, the design claim can be limited by the article and by the language that accompanies the drawings.[3] The USPTO’s own guidance also reflects that while drawings are usually the best description, narrative text can supplement and narrow scope, and applicants cannot use vague “variations” or “equivalents” language to float beyond what the drawings actually disclose.[4] Put differently, if you claim a design “as shown and described,” courts may read the “described” part like it actually belongs there. Wild concept.

The Utility Patent Side: Doctrine of Equivalents Is Not a Rescue Raft for Missing Functions

The utility-patent half of Smartrend is where the doctrine of equivalents took the real hit.[1] The doctrine remains good law, of course. The Supreme Court reaffirmed it in Warner-Jenkinson, while also emphasizing that equivalence must be assessed on an element-by-element basis, not by waving broadly at the invention as a whole.[5] Festo then reminded everyone that prosecution history estoppel can cut the doctrine back when claims were narrowed during prosecution.[5]

In Smartrend, the relevant claim required a frame that surrounded the perimeter of the translucent panel and formed a perimeter of the sign for mounting it to the school bus.[1] The district court construed “frame” to mean a separate and distinct component from the rest of the sign. Because Opti-Luxx used an integrated, one-piece design, the district court had already granted summary judgment of no literal infringement.[1][7]

So Smartrend leaned on the doctrine of equivalents. The theory was simple: yes, the accused product lacks a separate frame literally, but its integrated structure does substantially the same thing. The Federal Circuit was not convinced. Looking to the claim language and the written description, the court concluded that the claimed separate frame did more than merely sit around the sign and help mount it.[1] According to the patent, the separate frame also let the sign be removably received, serviced, replaced, swapped out for customized signage, and reused without damage from direct mounting.[1] Those were not side quests. They were core functions.

And here is where the case becomes a cautionary tale for expert testimony. Smartrend’s expert tried to define the frame’s function narrowly as surrounding the sign and forming a perimeter for mounting.[1] But the court held that expert testimony cannot contradict the patent’s own description of the claimed function. Because the accused integrated frame could not perform the patent’s identified removable-receipt and customization functions, no reasonable jury could find equivalence.[1][7] The result was reversal of the infringement judgment under the doctrine of equivalents.[1]

That holding fits squarely with broader doctrine-of-equivalents law. Courts have long worried that the doctrine can become too mushy, too unpredictable, and too eager to punish products that are merely similar in a broad practical sense.[5][7] Smartrend narrows that mush by insisting that if the intrinsic record defines the function of a claim element, the patentee cannot ignore those disclosed functions and present a thinner version to the jury. The doctrine protects against trivial changes, not against patents that were drafted more specifically than their owner later prefers.

What the Federal Circuit Really Clarified

1. Design patents stay narrow when the patent text narrows them

The court reiterated that design patents are given narrow scope and that the claimed design may be limited by the accompanying description.[1][3] In practice, that means careful word choice in a design patent can help precision during prosecution but may also box the patent in during litigation.

2. The ordinary-observer test remains central, but claim construction still matters

Egyptian Goddess remains the foundation, yet the court in Smartrend showed that the ordinary observer only compares the design that was actually claimed, not the one the patentee wishes had been claimed.[1][2][6]

3. Expert testimony can help, but only within limits

The Federal Circuit suggested that expert testimony about the ordinary observer may be permissible if the expert is familiar with that perspective, although Opti-Luxx forfeited its challenge in this case.[1][7] On remand, and in future cases, lawyers should expect tighter fights over whether the expert really understands the relevant purchaser’s viewpoint and whether objective support exists for any claimed industry-specific meaning.[1][7]

4. The doctrine of equivalents is anchored to the patent’s intrinsic evidence

If the claim language and specification identify the real function of a claim element, that function controls the equivalence analysis.[1][5][7] A patentee does not get to sand off inconvenient details just because they complicate the jury story.

5. Mixed design-and-utility cases require strategic consistency

One of the most practical lessons from Smartrend is that plaintiffs cannot afford internal contradiction. If the design patent depends on narrow descriptive language and the utility patent depends on structural separateness, the infringement case must respect both realities from the beginning. Trying to run a broad visual story on one patent and a trimmed-down functional story on the other can create appellate trouble fast.[1][7]

Why This Decision Matters for Patent Owners, Defendants, and Product Teams

For patent owners, the message is not “never be specific.” Specificity can be powerful. It can distinguish prior art, sharpen a design, and make the invention feel concrete. But specificity comes with a receipt. If your patent repeatedly celebrates a separate frame because it enables replacement, servicing, and customization, a court may later insist those are part of the frame’s function.[1] If your design description says shading denotes transparency, you may not get to sue a translucent product and call it close enough.[1][4]

For defendants, Smartrend offers a roadmap. Read the drawings, yes, but also mine the written description for narrowing language, functional commitments, and repeated advantages the patentee emphasized. In the right case, those statements can shape claim scope, defeat literal infringement, and collapse a doctrine-of-equivalents theory before it reaches the finish line.[1][3][7]

For product designers and in-house counsel, this decision is also a drafting and freedom-to-operate lesson. The safest design-around is still a meaningful difference that changes the overall visual impression in design-patent space and changes the claimed element’s actual function in utility-patent space.[1][2][5] A product team should not merely ask, “Can we make it look a little different?” It should also ask, “Does our version actually work differently in a way the patent record would recognize?”

And that is where the opinion connects to broader 2026 design-patent developments. Later commentary around Range of Motion reflects a Federal Circuit increasingly comfortable with narrowing design patent scope through functionality analysis and, in some cases, resolving noninfringement without a jury when designs are deemed plainly dissimilar.[8] Smartrend fits that climate: courts are signaling that patent owners do not get a generous halo simply because design and utility issues can sound fuzzy when summarized at cocktail hour.

Practical Experiences From the Real World Around This Issue

Anyone who has worked around design-patent and product-shape disputes has probably seen a version of this movie before. The first screening usually happens in the conference room, not the courtroom. A product manager puts the accused product next to the patent drawings and says, “Come on, that is basically the same thing.” A litigator nods halfway. Then someone opens the patent file, reads the claim language and specification out loud, and the room gets a little quieter. That is the Smartrend lesson in everyday form: infringement theories often feel strongest before anyone reads the patent carefully.

Another common experience shows up in expert preparation. Teams sometimes assume a polished expert can smooth out awkward claim language by reframing what a feature “really” does. But seasoned patent litigators know that this strategy gets dangerous fast. If the patent repeatedly praises a feature for several concrete functions, an expert who trims those functions down to one litigation-friendly sentence may look less like a guide and more like a screenwriter doing late rewrites.[1][7] Judges tend to notice. Appellate judges, as it turns out, notice with even more enthusiasm.

In-house counsel also feel this issue long before a complaint is filed. During product clearance, the hard question is rarely “Is there a patent?” The hard question is “What, exactly, did the patent owner choose to claim, and how tightly did they describe it?” Engineers often think in gradients: close, close-ish, pretty different, definitely different. Patent law sometimes thinks in sharper categories. Separate can mean separate. Transparent can mean transparent. A removable component can be very different from a molded-in one, even if both accomplish some broad commercial objective.[1][3][4]

There is also a drafting experience buried in cases like this. Patent prosecutors are often under pressure to make an application clear, persuasive, and distinct over prior art. That pressure can lead to useful language explaining why a feature matters. The catch is that those explanations may later become litigation handcuffs. Prosecutors and litigators therefore need to act like members of the same band, not rival garage groups. If one side writes poetry about the irreplaceable benefits of a separate frame, the other side should not be surprised when a court treats those benefits as part of the feature’s legal identity.

Finally, there is the business-side experience. Companies that successfully design around patents usually do not win by making cosmetic tweaks at the last minute. They win by understanding the patent’s structure early and building meaningful differences into the product from the start. The smartest teams ask not only how a rival product looks, but how the patent defines what matters about the look or the structure. That habit saves money, reduces panic, and lowers the chance that your Friday afternoon ends with outside counsel saying, “Well, there is good news and there is appellate news.” In that sense, Smartrend is not just a case about bus signs. It is a practical reminder that precision in patent documents eventually becomes precision in litigation outcomes.

Conclusion

The Federal Circuit’s decision is important because it clears away two kinds of overreach at once. First, on the design side, it confirms that descriptive words attached to drawings can meaningfully limit claim scope. Second, on the utility side, it confirms that the doctrine of equivalents cannot erase functions the patent itself treats as central. Patent owners still have strong tools. Defendants still need to be careful. But after Smartrend, one thing is clearer than ever: when the patent record draws a line, litigants ignore it at their peril.[1][2][3][5]