Skip to Content


The Cervelo P3 Patent

Written by: Douglas Robinson
Date: Mon Jun 20 2011

Cervelo has a U.S. patent on the P3 bicycle frame, but do you know what aspect of the frame was the basis for the patent? Since patents are foreign to many people, a little background is in order.

For the most part U.S. utility patents are valid for 20 years from the date a patent application is filed. Therefore, a patent that takes three years to issue from the date it was filed will have a term of 17 years, assuming no overly long delays by the U.S. Patent and Trademark Office, after which it falls into the public domain and anyone can practice the invention. Besides a utility patent the other type of U.S. patent is a design patent. These patents are strictly for designs that have no utility. They last only 14 years and the drawings define the protected elements.

This is a basic premise of the patent system. Assuming the invention is novel and non-obvious, in exchange for providing a description that is detailed enough so that a person of ordinary skill in the art can practice the invention, the patentee is granted a patent for a limited number of years. In this way the general knowledge of the Country is advanced and others can build on the idea.

One aspect of patents that many people do not understand is the rights a patent gives the owner. A patent provides the patentee the right to EXCLUDE others from making, using or selling the invention. What does this mean? Well, most people think a patent give one the right to practice their invention. This may not be correct. Let's assume I receive a patent for a center pull brake, and that such a brake is new and non-obvious. If however, someone before me received a patent to a basic bicycle brake that did not specify center pull or side pull, a device made according to my patent would infringe the earlier patent. Therefore, while I can prevent others from making, using or selling my patented center pull brake, the owner of the first general brake patent can prevent me from practicing my invention.

The next area of difficult is the part of the patent called the "claims." A U.S. patent must have at least one claim. The claims define the scope of the rights the patent is intended to cover. Generally, claims are hard to read. Each claim is a single run-on sentence and they can be quite long. Certain words in claims are called "terms of art" and have defined meanings and unless you deal with patents on a regular basis you would have no way of knowing these terms. As an example in a claim that recites "a widget comprising:…A, B and C" the word "comprising" means that a widget having elements A, B and C would infringe the claim regardless of whether it had more elements, such as D, E, and F. On the other hand a claims that recites "a widget consisting of:…A, B and C" is limited to a device that has only A, B, and C and nothing more.

With that brief background, let's look at U.S. patent 6,889,992 entitled: Aerodynamic bicycle frame, which Cervelo obtained on May 10, 2005. Looking at the figure, you may recognize this frame as the Cervelo P3. The patent issued with 15 claims of which claim 1 was the only independent claim. (Claims 2-15 all added details to claim 1.)

Here is claim 1:

1. A bicycle frame comprising:
a first tube;
a second tube extending transversely from an upper region of said first tube, in a direction;
a third tube extending transversely and downwardly from a lower region of said first tube, and towards a same direction as said second tube; and
a two-part tube joining respective ends of said second and third tubes, said two-part tube having a straight upper part and a curved lower part, said straight upper part being substantially vertical and said curved lower part being curved longitudinally towards said first tube such as to conform to a curvature of a bicycle wheel.

The language does take some getting used to. The first tube is the head tube 20, the second tube is the top tube 30, and the third tube is the down tube 40. The two-part tube is the seat tube 50 (number not shown in Fig.5) which includes the straight upper part 52 and the curved lower part 56.
The first, second and third tubes can be found in bicycles going back decades. It is the last paragraph and the two-part tube that is essential to this patent.
Claim 1 of the patent was initially rejected as being anticipated by Busby (U.S. Patent 6,036,213). Figure 1 of Busby is shown here.

The Examiner argued that Busby disclosed a bicycle frame comprising a first (head) tube 14, a second (top) tube 12, and a third (down) tube 20 extending transversely and downwardly from the lower region of the first tube 14, and a seat tube 18 joining the other ends of the top and down tubes. The Examiner remarked that the elongated seat tube is a two-part tube and that the lower part 18 has an arcuately contoured or bowed configuration, and a straight, nearly vertical portion 22.

In arguing against the rejection, the Applicant argued that the arcuately contoured or bowed configuration of the lower part of Busby does not conform to the curvature of the bicycle wheel. Applicant goes on to emphasize that having the curved tube match the curvature of the rear wheel offers aerodynamic benefits.

Applicant further argued that the upper part 22 of Busby is outside the triangle constituted by the frame. In contrast, the frame according to claim 1 recites the upper and lower parts of the seat tube are parts of the triangle constituted by the frame, since the second tube joins the seat tube at an upper region of the upper parts of the seat tube. It is interesting that while the figure of the '992 patent show this two part structure as the Applicant describes, those details are not claimed. The only feature that is claimed is the conformance of the curved portion to the curvature of the wheel.

The Examiner allowed the claims in response to the amendment containing the foregoing arguments filed by the Applicant. So, a bicycle frame having a head tube, top tube, down tube and a two-part seat tube, where the seat tube has a straight portion and a curved portion that curves in conformity with the curvature of the rear wheel is covered by claim 1 of the '992 patent.
Interesting as well are figures 1(a) and 1(b) of the patent, which are acknowledged prior art. These figures are described in the patent as follows:

"Trimble in U.S. Pat. Nos. Re. 33,295; 4,941,674; and 4,982,975 discloses a bicycle frame made of composite material. The frame comprises a seat tube having a curved section which is cut into a one-piece straight seat tube. Such seat tube is also illustrated in FIGS. 1(a) and 1(b). The resulted design of the bicycle frame according to Trimble is that of a traditional "double diamond" frame, which has two open triangles with a common side. The curved section of Trimble's seat tube does not allow, however for a full coverage of the upper portion of the rear wheel and thus an optimal aerodynamic shielding is not achieved."

Thus a frame with a curved section which cuts into a straight tube is distinguished from the claimed frame as it does not allow for "full coverage of the upper portion of the rear wheel."

The '992 patent will expire on filed Sep. 20, 2020, which is twenty years from the date of the filing of the first patent application for a Canadian patent. It appears that the P4 carries this feature as well.
How about some of Cervelo's competitors frames? Specialized has the Transition which definitely has a curved portion that conforms to the rear wheel. The real question for this frame is does the seat tube have a straight portion? Not necessarily an easy question to answer as the top tube and rear stays all seem to meet at the upper end of the curved portion. The "straight portion" almost extends above the diamond.
Pinarello makes a few frames that appear similar. In particular the Graal, where the seat tube has a curved lower portion that conforms to the shape of the rear wheel and a straight upper portion. The Pinarello FT3 is a different matter; the lower portion curves but is less conforming than other lower tubes. The entire seat tube appears more "S-shaped" and therefore this is a grayer area. This frame appears closer to the prior art that was distinguished during prosecution of the Cervelo patent. Pinarello also makes the FM1. Like the Graal there appears to be a distinct curved lower portion conforming to the rear wheel and a straight upper portion inside the diamond.
Finally there is the Orbea Ordu. Similarly to the Specialized Transition, the Ordu, while having a curved lower portion, does not have a clear straight upper portion. Interestingly, Orbea obtained U.S. Patent 7,918,471 on April 5, 2011, entitled: Bicycle Frame in this area. The patent only has two claims with claim 1 being the independent one:

1. A bicycle frame comprising:
(a) a front quadrilateral of tubular elements (C), the front quadrilateral (C) having
(i) a horizontal upper tube (4) joined at one end to a head tube (5),
(ii) a slanted lower tube (6) joined at one end to the head tube (5) and having a pedal attachment point (P) at the other end of the slanted lower tube (6), and
(iii) a sealing element (2) joined at one end to the other end of the slanted lower tube (6), the sealing element curved for covering a rear wheel of a bicycle;
(b) a rear triangle (t) of tubular elements, the rear triangle (t) having
(i) a lower brace (7) joined at one end to the other end of the slanted lower tube (6),
(ii) an upper brace (3) joined at one end to the other end of the lower brace(7), and
(iii) the sealing element joined (2) at the other end to the other end of the upper brace (3);
(c) a coupling element (A) joined to the other end of the horizontal upper tube (4) and to other end of the upper brace (3) to form a joint between the horizontal upper tube (4) and the upper brace (3), an axis (e2) of the joint between the coupling element (A) and the upper brace (3) approximately coplanar with an axis of the upper brace (3);
(d) a seat tube (1) projecting outward from the joint between the coupling element (A) and the upper tube (4), the seat tube (1) having an axis (e1) which intersects the center of the pedal attachment point (P);
(e) a comfort angle (α) between an axis (e3) of the horizontal upper tube (4) and the axis (e2) of the upper brace (3) is greater than or equal to 120; and
(f) the frame is a single piece manufactured of carbon-compound material.

(reference numbers have been added to the claim).
A couple of caveats, some of the frame builders discussed herein may have licenses to the Cervelo '992 patent, in which case these builder's frames would not infringe. Second, the foregoing is only a cursory review of the patent and the frames. An in depth review of the file history, and prior art has not been undertaken. Finally, as was discussed at the beginning of this article, the '992 patent was cited during the prosecution of the Orbea '471 patent, showing how two patents can be issued in the same area as long as the ideas are non-obvious and novel.

[Publisher's note: Douglas Robinson is a triathlete, a Slowtwitcher, a patent attorney, and is or new IP editor. He'll be writing from time to time about the IP landscape as it pertains to the events in which you might compete, the equipment you might use, the products you might ingest. If you have topics or questions you might like covered, you can contact Douglas at DougRob@Slowtwitch.com]

  

  

  

Articles related to this one
WTC and the term Ironman
Slowtwitch's new IP editor Douglas Robinson explores the legal issues and ramifications regarding the Ironman trademark. 3.20.11

Comments

Prior Art 4 out of 5 stars

Reviewed by: Tom Galvani, Jun 24 2011 8:14PM

Great article, Doug. I can't believe a) that there wasn't more prior art cited in this case and b) that the arguments were persuasive that Busby did not read on the seat tube formed of two parts as claimed, as you note. I've added a link to the article on my own blog - www.galvanilegal.com/blog.

Defending patent 5 out of 5 stars

Reviewed by: Bart Turner, Jun 22 2011 4:42AM

I have also worked with patents around the world in my business but with software. My understanding is that Cervelo needs to enforce the patent or they risk losing the the opportunity to enforce. We don't know if companies are licensing the IPR from Cervelo. If Cervelo is not licensing the IPR and letting, IMO, infringing to take place then the patent is worthless. You don't get to selectively enforce. This is at least how it works with Software in the US. Other countries are different on patent law but Canada is very similar to the US.

On the comment above on the Transition. I do not believe the article was inferring that the Transition does not infringe. I believe it was questioning if it did or not. I actually read it as it was more suggesting that it does but maybe Specialized licensed the IPR from Cervelo.

Good stuff! 5 out of 5 stars

Reviewed by: Tom Anhalt, Jun 21 2011 7:14AM

Although this sort of thing might cause most folks eyes to glaze over...this mechanical engineer (who's had to deal with patents at times) REALLY enjoyed this piece, especially the part about how/why other seemingly similar frames (i.e. the Transition) are not in violation of the '992 patent.