Eldo and Pyxis ensure coverage of expected spread and achievement of parametric yield targets: Superior design rule checking; Using Calibre’s foundry gold standard design rules removes uncertainty and shaves one.
“Where Mentor really comes into its own is in the back-end. Once you’ve done the layout, the Calibre design rule checking is far superior to anything else that’s on the market. It gives you confidence that when you’re ready to release a design to the foundry, it will go through the foundry checks with minimum problems.”
The Problem: “We need tools that can provide the appropriate Monte Carlo capabilities, and so on, to ensure that during simulation we cover the expected spread and can achieve our set of parametric yield targets,” Dr. Churcher explains. “If the accuracy of the analog system is dependent on the master band-gap reference and that drifts out because it’s CUSTOMER PROFILE Pioneers in intelligent battery management AMS SoC technology.
- Design Challenge
- Meet foundry requirements on first tape out
- Fast and accurate verification
- Eliminate risks and respins when using design rules not supported by foundries
- Support for smaller technology nodes Solution
- Eldo and Pyxis ensure coverage of expected spread and achievement of parametric yield targets
- Superior design rule checking
- Using Calibre’s foundry gold standard design rules removes uncertainty and shaves one or two weeks at the back-end
- Foundries support only Calibre at more aggressive process nodes
- Common cockpits and shared data within the AMS/Calibre flow makes it easier to move from one design phase to the next and reduces errors, reducing time to tape out
- AMS/Calibre flow adopted by Dukosi for future pioneering battery management designs
The Solution: Dukosi adopted the fully-integrated suite of Mentor AMS tools. By using common cockpits and shared data, the AMS/Calibre flow makes it easier to move from one design phase to the next and reduces errors, speeding the road to tape out. First, the design is captured in schematic form using Pyxis Schematic. Then from the integrated simulation control interface in Pyxis Schematic, they invoke Eldo Classic or Eldo Premier to verify that the design functions correctly and meets performance specifications. Verification of performance coverage over the range of process corners, temperature, voltage, and target application variation is managed using the ICAnalyst cockpit environment built around Eldo Classic or Eldo Premier. Typically, they focus on the process extremes first, and then as the circuit design is refined, they perform Monte Carlo and other types of analysis. Once they achieve the performance they want, physical design creation is performed using Pyxis Layout, starting with schematic driven layout (SDL), which provides connectivity driven layout data behind the layout data. Final verification of the physical design is achieved using Calibre nmDRC and nmLVS products before the design is released to the foundry for production.
The Results: Because of their success using Mentor’s AMS/Calibre flow on this exploratory technology, Dukosi plans to stick with this flow as they continue to break new ground in battery management.
“Calibre helps close the loop at the back-end of a design far more easily and quickly. If we look at past designs that we’ve done with the other tool, we would have saved one or two weeks on each design at the back-end if we had been using Calibre.”