The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl Simulators >> Circuit Simulators >> Size Dependent Mismatch Variation's Description don't work in Agilent RFDE2006U3 https://designers-guide.org/forum/YaBB.pl?num=1276330403 Message started by pancho_hideboo on Jun 12th, 2010, 1:13am |
Title: Size Dependent Mismatch Variation's Description don't work in Agilent RFDE2006U3 Post by pancho_hideboo on Jun 12th, 2010, 1:13am I use following mismatch variation model parameters which have size dependent mismatch variation's descriptions in Cadence Spectre, Agilent RFDE(ADSsim) and Agilent GoldenGate. Here standard deviation of mismatch variation is proportional to "geo_fac=1.0/sqrt(w*l*m)". But this size dependent mismatch variation does not work correctly in Agilent RFDE2006U3(ADSsim2006U3). On the other hand, this works correctly after Agilent RFDE2008U2(ADSsim2008U2). Is there anyone who have same experience ? There is no official support for Agilent RFDE any more, although newest ADSsim engine still can understand Cadence Spectre's syntax netlist directly. Is there anyone who knows method to describe size dependent mismatch variation valid even for RFDE2006U3(ADSsim2006U3) ? Quote:
|
Title: Re: Size Dependent Mismatch Variation's Description don't work in Agilent RFDE2006U3 Post by pancho_hideboo on Jun 12th, 2010, 1:19am The attached figure is a statistics of Vth of various size PMOS. P0 is "m=1". P1 is "m=1". P2 is "m=2". P3 is "m=4". P4 is "m=8". As you can see, variation of "Vth" is small for larger "m" in RFDE2008U2(ADSsim2008U2). But this is not true for RFDE2006U3(ADSsim2006U3). Here Vth of each PMOS with different sizes behave as mismatch variation with same standard deviation. |
Title: Re: Size Dependent Mismatch Variation's Description don't work in Agilent RFDE2006U3 Post by pancho_hideboo on Jun 15th, 2010, 5:03am Self follow up. :D I know "m" is a reserved keyword for multiplier in Netlist of Cadence Spectre's Syntax. For example, Cadence Spectre shows following "Warning". Quote:
But "m" works properly as both instance multiplier and pass parameter in Cadence Spectre, Agilent GoldenGate and Agilent ADSsim2008U2. On the other hand, "m" don't work properly in Agilent ADSsim2006U3. If I change model file like following, I can make it valid even for Agilent ADSsim2006U3. Quote:
|
The Designer's Guide Community Forum » Powered by YaBB 2.2.2! YaBB © 2000-2008. All Rights Reserved. |