The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Simulators >> Circuit Simulators >> Need Spectre Help on initial values
https://designers-guide.org/forum/YaBB.pl?num=1172066635

Message started by kiran123 on Feb 21^st, 2007, 6:03am

Title: Need Spectre Help on initial values
Post by kiran123 on Feb 21^st, 2007, 6:03am

Hi Experts,
Can anyone of you tell us on how to give an initial condition for vsource/isource or any passive components in spectre ? because i wanted in one senario where i initially wnated to force some voltage and current value at some nodes and then do a DC analysis with the existing DC sources which i declared in body of simulation deck.

by the way is there any possibility of giving stimulus of pulse type with specified format say "1001110" where i can define pulsewith of pulse and tr and tf

i can do this with pwl but the definiing will be very lengthy when it comes to long 1's and 0's format

Thanks for your support
Best Regards
Kiran Kumar

Title: Re: Need Spectre Help on initial values
Post by Andrew Beckett on Feb 22^nd, 2007, 11:49am

An initial condition for a vsource or isource is rather meaningless - it's implicitly defined by the source itself.

For capacitors and inductors you can specify an initial voltage or or current (see "spectre -h capacitor" or "spectre -h inductor" and look at the ic parameter). Alternatively you can use spectre's nodeset or ic statement (spectre -h nodeset or spectre -h ic) to do this kind of thing on particular nodes (say).

As for the bit type stimulus, you can do that:

Code:

//

vsource (1 0) vsource type=bit val0=0 val1=1 rise=1n fall=1n period=10n data="101100101"
r1 (1 0) resistor r=1k

tran tran stop=150n

Best Regards,

Andrew.

Title: Re: Need Spectre Help on initial values
Post by achim.graupner on Feb 25^th, 2007, 11:39pm

Dear Kiran,

as regards your 2nd question "0101..." the best thing is to use a verilogA-Model. If you have a discrete timing you can use the
always @ (timer())-Statement to write a variable and the
transistion-Statement to drive the output. If this description is to short for you. please let me know, I will look for an example for you.

Another possibility is to (automatically) create a VerilogA-Modell from the .vcd-output of a digital simulator. This is helpful if the required waveforms are complicated and do not change between different analog simulations. I have a script available which does this conversion. I can provode it if you like.

Achim

Title: Re: Need Spectre Help on initial values
Post by kiran123 on Feb 27^th, 2007, 6:58am

Achim,

if there is another way declaraion, Yes, i would like more explanation on defining that
So could please expalin me on how to define with an example how Mr.Andrew suggested

Best Regards
Kiran Kumar

Title: Re: Need Spectre Help on initial values
Post by kiran123 on Mar 13^th, 2007, 6:29am

Andrew,

Thanks for your reply but when used the same statement i am getting error message as shown below
could you please let me know what needs to be done . One thing is i remember this is SPICE format of declaration so it would be helpful if you tell me what modification should i do to suppress this error

Notice from spectre during hierarchy flattening.
vsource: Invalid enumeration `bit' used as value of parameter `type'. Ignored.
Error found by spectre during hierarchy flattening.
vsource: Waveform type must be specified if any waveform parameters are given.

spectre terminated prematurely due to fatal error.

Thanks & Best Regards
Kiran Kumar

Title: Re: Need Spectre Help on initial values
Post by Geoffrey_Coram on Mar 13^th, 2007, 7:34am

Kiran -
I think you need a newer version of the simulator. I see "bit" is not available in IC5.1, but is available in MMSIM6.1. Andrew can probably tell you which exact version it was added in.

Title: Re: Need Spectre Help on initial values
Post by Andrew Beckett on Mar 15^th, 2007, 6:34am

It was added in MMSIM6.1 as Geoffrey suggested.

Note, it's not currently a choice in the analogLib vsource/isource/port components - PCR (Product Change Request) 952356 has been filed to get that implemented.

Regards,

Andrew.