Home > Community > Forums > Custom IC Design > High capacity IR analysis problems

Email

* Required Fields

Recipients email * (separate multiple addresses with commas)

Your name *

Your email *

Message *

Contact Us

* Required Fields
First Name *

Last Name *

Email *

Company / Institution *

Comments: *

 High capacity IR analysis problems 

Last post Tue, Nov 20 2012 7:49 AM by oojah. 2 replies.
Started by oojah 26 Oct 2012 02:37 AM. Topic has 2 replies and 953 views
Page 1 of 1 (3 items)
Sort Posts:
  • Fri, Oct 26 2012 2:37 AM

    • oojah
    • Top 500 Contributor
    • Joined on Thu, Feb 26 2009
    • Posts 17
    • Points 210
    High capacity IR analysis problems Reply

     Hi,

    I'm using IC6.15-64b.500.6 on openSUSE 11.4.

    I'm trying to carry out a high capacity analogue IR drop analysis but am running into problems. I've successfully done "regular" IR drop analyses in the past using an extracted view and a post layout simulation but haven't used the high capacity version before.

    I've been following the flow described in the Virtuoso Power System manual to carry out static rail analysis. I have a layout that has successfully passed LVS and I have run QRC extraction on it to generate a PGDB. As I'm only interested in the VDD net for the moment, I did a Full-chip selected nets extraction and only extracted that net. The PGDB was created successfully as
    far as I can tell.

    The next step is the CurrentAnalysis. I've chosen Peak, have the top cell name set to the layout cell, net names set to VDD, the spice netlist set to the .spice file in the PGDB directory, the power and ground pins set as VDD and GND respectively, and the voltage set. I've loaded the manufacturer model files and have set the all to typical mean corner only. I've set an output directory that exists. I've left all of the advanced options blank.

    When I click run, the log looks like the simulation files are being loaded, but then everything stops very quickly.

    The Thunder.log file looks like this:

    Including /local_storage/ral/vps_simulation_2/thunder.ckt.
    Including /eee/vlsi/DesignKits/AMS/v4.00/spectre/c35/cmos53.scs.
    Including /eee/vlsi/DesignKits/AMS/v4.00/spectre/c35/bip.scs.
    Including /eee/vlsi/DesignKits/AMS/v4.00/spectre/c35/cap.scs.
    Including /eee/vlsi/DesignKits/AMS/v4.00/spectre/c35/ind.scs.
    Including /eee/vlsi/DesignKits/AMS/v4.00/spectre/c35/res.scs.
    Including /home/ral/vlsi/Cadence/pgdb2/attoX_Core.spice.
    Reading /home/ral/vlsi/Cadence/pgdb2/attoX_Core.spice at line 32754.
    ...
    INFO (THUNDER-5361): Found spectre at location:
    /eee/vlsi/Cadence/2011/MMSIM/tools/bin/spectre
    INFO (THUNDER-5382): Processing device models.


    The HCP window log ends with:

    Process exited with status 0


    Thunder.errors only contains its header, no actual errors.

    The only files in the output directory are:

    thunder.ckt
    thunderCktFile.ckt 
    thunderCmdFile.cmd 
    Thunder.cmdlog
    thunderDevModel0.sp
    Thunder.errors
    Thunder.log
    Thunder.mcap
    Thunder.pinfo
    vps_probe_file


    So it looks like something is happening somewhere to stop the simulation, or I've missed something out. Any hints on what I need to do?

    Cheers,

    Roger
     

    • Post Points: 20
  • Tue, Nov 20 2012 5:53 AM

    • Quek
    • Top 10 Contributor
    • Joined on Wed, Oct 14 2009
    • Singapore, 00-SG
    • Posts 1,051
    • Points 15,990
    Re: High capacity IR analysis problems Reply

    Hi Roger

    It might be good to file a service request with a testcase to your local Cadence suppor. More detail investigation is most probably needed.

    Best regards
    Quek

    • Post Points: 20
  • Tue, Nov 20 2012 7:49 AM

    • oojah
    • Top 500 Contributor
    • Joined on Thu, Feb 26 2009
    • Posts 17
    • Points 210
    Re: High capacity IR analysis problems Reply

     Hi Quek,

     Thanks for the reply. Unfortunately I don't have that option as we get our licences via Europractice and don't have a support contract. I think we're due an updated version soon so I'll give that a try when it appears.

    Cheers,

    Roger

    • Post Points: 5
Page 1 of 1 (3 items)
Sort Posts:
Started by oojah at 26 Oct 2012 02:37 AM. Topic has 2 replies.