Appeal 2007-1650
Application 11/111,799
The Examiner argues that separating the clock pulses is the very
essence of the Walter reference, and thus it is a valid reference (apparently
the little overlap at the bottom of Fig. 3(b) notwithstanding). (Answer 4).
Secondly, he argues that independent claims do not require that the
“overlapping between the output clock signals be eliminated.” (Answer 5,
top). The Appellants strenuously object to this interpretation of the claims.
(Reply Br. 2).
Considering the limitations of Claim 20, the key phrase is “wherein
the coupling of delay elements along the first feedback path controls the
amount of time that clock edges associated with the first and second clock
outputs are non-overlapping”. Both this limitation, and its counterpart in
method claim 27, call for the delay elements being able to control the
amount of time that clock edges are non-overlapping. This limitation,
however, does not absolutely require that the pulses ever be non-
overlapping, but rather that the controls merely be in place to establish the
amount of non-overlapping.
We agree with the Examiner’s first argument that the term “clock
edges” can be broadly but reasonably read on the top corners of the clock
pulses of Walter, as thereafter the pulses decline rapidly below a threshold of
effectiveness. Furthermore, a second point to consider is that even Figure
3(b) indicates a time in which the clock edges are non-overlapping -- namely
from the top right hand corner of pulse 36a to the point where line 38a starts.
Since 38a does not exist in that region, the lines are non-overlapping. In
addition, there is another span of time in this cycle in which the outputs are
non-overlapping, namely from the point when 36a ceases to exist, onward to
a later point of 38a. As can be seen by the various Figures 3(a), (b), and (c),
9
Page: Previous 1 2 3 4 5 6 7 8 9 10 11 12 Next
Last modified: September 9, 2013