Stats

  • Locked Locked
  • Replies 3
  • Subscribers 125
  • Views 13326
  • Members are here 0
This discussion has been locked.
You can no longer post new replies to this discussion. If you have a question you can start a new discussion

Running ADE L several times in a day

ArmanHR

Dear All,


I am designing robust SRAM memory cells for space application. I have to calculate the minimum critical charge that each memory node can tolerate until data flips.
I run ADE L several times (let’s say around 300-400 times) to find the minimum critical charge and optimum transistors sizes. Each time, I opened up the waveform viewer to check the waveforms. When I found the optimum sizes, I wrote down them and turned off my laptop. The day after that, when I again run ADE L, my saved results were not the optimum! Again, I had to run the simulator several times to find the optimum transistor sizes; they were close to the previous data, but different numbers! I am sure I did not make any mistakes. What can be the problem? Is it a bug?


Bests,

  • Andrew Beckett

    I've not heard of this happening before (and I'd be very surprised if the tool is being used correctly - it can't invent results...). Are you doing Results->Select and clicking on the plot button to re-evaluate the results each time you restart ADE to look at previous results? Otherwise the scalar values shown in the ADE L outputs pane just show the results as they were at the time the state was last saved, which may not be what  you are expecting?

    Regards,

    Andrew.

  • ArmanHR
    in reply to Andrew Beckett

    Thank you for the answer.

    Maybe it is better I explain what I am doing now:

    • I inject a current noise to each SRAM node (my memory has two back-to-back NMOS and PMOS pairs and two positive feedbacks that connect the internal nodes to each other.)
    • I run ADE L and open the waveform viewer to check the signals.
    • If it still operates as a normal memory, I increase the current noise value and again run the ADE L.
    • I repeat this procedure until the memory does not operate correctly. It means the injected noise can destroy the stored data.
    • I repeat it many times a day (by increasing each transistor sizes, by changing the injected noise) to find out the optimum area, power, access time and noise reliability.

     

    I only click on the “plot output” after finishing the simulations. So do you think it is a problem?

    As a solution, I wanted to increase or reduce the number of sampling during simulation!!!

    Bests,

    Arman

  • Andrew Beckett
    in reply to ArmanHR

    Given that the netlist is presumably changing each time (if you're changing the transistor sizes by editing the schematic - or have you parameterised them?), I cannot see how running the same simulations the following day can give different results.

    I suggest you contact customer support.

    Andrew.

Community Guidelines

The Cadence Design Communities support Cadence users and technologists interacting to exchange ideas, news, technical information, and best practices to solve problems and get the most from Cadence technology. The community is open to everyone, and to provide the most value, we require participants to follow our Community Guidelines that facilitate a quality exchange of ideas and information. By accessing, contributing, using or downloading any materials from the site, you agree to be bound by the full Community Guidelines.