Home > Community > Forums > RF Design > trouble with getting correct ro through sp analysis in spectre

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: *

 trouble with getting correct ro through sp analysis in spectre 

Last post Thu, Mar 29 2012 12:52 PM by snaildr. 2 replies.
Started by snaildr 24 Mar 2012 08:15 PM. Topic has 2 replies and 2791 views
Page 1 of 1 (3 items)
Sort Posts:
  • Sat, Mar 24 2012 8:15 PM

    • snaildr
    • Not Ranked
    • Joined on Sat, Mar 24 2012
    • Seattle, WA
    • Posts 5
    • Points 55
    trouble with getting correct ro through sp analysis in spectre Reply

    Hi there,

    I'm pretty new to Spectre, and was simulating some basic stuff. Here's a question about getting ro using sp analysis that puzzls me.

    I'm under the impression that for the drain-source impdance, the large signal value is 'ron' which is Vds/Id, and the small-signal value is ro==1/gds, which is dVds/dId. So, for small-signal gain, I expect to use ro to match my hand calculation to simulation, and it did work.

    I also want to do a sp analysis at the output port of a single-transistor common-source amplifier. I set the load to be an RF choke so that the output impedance should be the drain-source impedance. In the MOSEFT and bias condition I used, ron=40 Ohm, ro=1/gds=220 Ohm, so there's a big difference between the two. However, when I look at the Z11 (input impedance looking into the output port, i.e. drain-source port), the real(Z11) is close to ron (40Ohm) ! I varied bias voltage, MOSFET dimensions etc, this issue remains. It's supprising to me that as a small-signal simulation, sp analysis picked the large-signal impedance!

    As a control experiment, I replaced the 'port' element with a 50 Ohm resistor and a 'vsin' element in series and watch the voltage before and after the 50 Ohm resistor, it indeed appears that the voltage ratio is about 0.8, so the impedance should indeed be ~250Ohm.

     

    ro is a key parameter in circuit design and being able to get the correct value through sp analysis is very important in RF designs. I'd really appreciate your take on this!

     

    Ran 

    • Post Points: 20
  • Thu, Mar 29 2012 12:12 PM

    • Tawna
    • Top 25 Contributor
    • Joined on Fri, Jul 11 2008
    • Snohomish, WA
    • Posts 216
    • Points 5,720
    Re: trouble with getting correct ro through sp analysis in spectre Reply

     Hi Ran,

    Without seeing your netlist/model files, it's difficult to say what is going on. 

    I suspect it's a setup issue - or a misunderstanding of how the port operates.

    sp is a small signal analysis, like ac.   

     

    I recommend contacting Cadence Customer support ( http://support.cadence.com ) with your netlist/model files (You cannot post model files here...they belong to whomever developed them) so an AE can look at this and assist you.

     

    best regards,

     

    Tawna

    Best regards, Tawna Wilsey Staff Support AE, Global Customer Support Cadence Design Systems, Inc.
    • Post Points: 20
  • Thu, Mar 29 2012 12:52 PM

    • snaildr
    • Not Ranked
    • Joined on Sat, Mar 24 2012
    • Seattle, WA
    • Posts 5
    • Points 55
    Re: trouble with getting correct ro through sp analysis in spectre Reply

    Issue resolved, I was looking at Z22 to get the port2 input impedance, but Z-parameter is really Z-parameter not (s11+1)/(1-s11). So, to get the port input impedance one should look at ZM parameter or equivalently the Z-smith of S-parameter.

     

    Thanks Tawna for your kind reply.

     

    thanks,

    Ran 

    • Post Points: 5
Page 1 of 1 (3 items)
Sort Posts:
Started by snaildr at 24 Mar 2012 08:15 PM. Topic has 2 replies.