Browse Prior Art Database

Method for automatically determining signal pulse width tests via waveform definitions Disclosure Number: IPCOM000029486D
Original Publication Date: 2004-Jun-30
Included in the Prior Art Database: 2004-Jun-30
Document File: 1 page(s) / 40K

Publishing Venue



In a static timing tool, automatic calculation of appropriate minimum acceptable pulse width values for non-50% duty cycle can be accomplished using the multiplier and offset technique and taking into account the potentially different values of active and inactive interval specified by the user when he defines the clock shape for the timer.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 86% of the total text.

Page 1 of 1

Method for automatically determining signal pulse width tests via waveform definitions

When clock signals are defined, there is no check in our timing tool to check that the active or inactive pulse width is appropriate for an unbalanced - not 50% duty cycle - signal. As a signal travels through a chip, its waveform can become distorted to the point that a failure can occur. To prevent this, a procedure based on the clock frequency was developed. However, the procedure did not address non-symmetric waveforms. The solution is to use the waveform definition to determine the appropriate active/inactive pulse width values. This way a waveform of any shape can be accurately modeled. The existing code measures how far away a signal is from a 50-50 duty cycle pulse, while the new code would measure how far away a signal is from its ideal waveform shape, which includes the 50-50 case.

     This invention uses the signal definition as it enters the design along with some parameter's entered by the designer to determine the appropriate pulse width. This allows non-50% duty cycle (distorted) waveforms to get the appropriate checks as well as 50% duty cycle waveforms.

     This procedure works as an enhancement to the code that calculates pulse width values based on the cycle time. Instead of using the cycle time, it uses the values given for the ideal rise and fall times of a waveform. From those numbers, the widths of ideal active and inactive intervals are calculated independently....