Forum
May 25th, 2020, 5:29pm
 Pages: 1
 Simulation for Ferroelectric Varactor (Read 354 times)
 aguntuk Junior Member Offline Posts: 12 Dresden, Germany Simulation for Ferroelectric Varactor Nov 29th, 2018, 2:24am   Hi,I have created a Ferroelectric Varactor model using VerilogA. Now, I wan to simulate the C-V curve. I have added a voltage source with the varactor. Can anyone help to setup in ADE to simulate C-V curve?I am using Cadence Virtuoso with spectre simulator.Version: ICADV12.3-64b.500.21I have attached my ADE and testbench screenshot. Back to top IP Logged
 Ken Kundert Global Moderator Offline Posts: 2273 Silicon Valley Re: Simulation for Ferroelectric Varactor Reply #1 - Dec 2nd, 2018, 5:07am   Run an AC analysis.Configure the AC analysis so that it sweeps DC voltage rather than frequency.Plot the current through the capacitor.You are left with two free variables: AC magnitude of the source and the analysis frequency. Set one to 1, set the other so that the output current is scaled so that you are plotting in Farads directly.-Ken Back to top IP Logged
 Ken Kundert Global Moderator Offline Posts: 2273 Silicon Valley Re: Simulation for Ferroelectric Varactor Reply #2 - Dec 2nd, 2018, 5:10am   That procedure is described in Modeling Varactors.There is a trick to modeling varactors. You should read and follow the guidance of that paper carefully.-Ken Back to top IP Logged
 aguntuk Junior Member Offline Posts: 12 Dresden, Germany Re: Simulation for Ferroelectric Varactor Reply #3 - Dec 14th, 2018, 2:22am   Ken Kundert wrote on Dec 2nd, 2018, 5:07am:Run an AC analysis.Configure the AC analysis so that it sweeps DC voltage rather than frequency.Plot the current through the capacitor.You are left with two free variables: AC magnitude of the source and the analysis frequency. Set one to 1, set the other so that the output current is scaled so that you are plotting in Farads directly.-Ken thanks for your answer. I like this manual idea. I am confused about the analysis frequency. I put the ac magnitude to 1 Now in ac analysis, what should be the analysis frequency so that I can make it to farad? I didn't get the conversion here. Can you help me with this? Back to top IP Logged
 Geoffrey_Coram Senior Fellow Offline Posts: 1985 Massachusetts, USA Re: Simulation for Ferroelectric Varactor Reply #4 - Dec 14th, 2018, 5:55am   If I analyze this circuitv1 1 0 1 ac 1c1 1 0 1pat 1 Hz, then the ac current in the voltage source is 6.28e-12, that is, 2*pi*the capacitance in Farads. So, you could change either the ac magnitude or the frequency to scale out that 2*pi. Back to top If at first you do succeed, STOP, raise your standards, and stop wasting your time.   IP Logged
 aguntuk Junior Member Offline Posts: 12 Dresden, Germany Re: Simulation for Ferroelectric Varactor Reply #5 - Dec 14th, 2018, 7:42am   V=I*Z=I/(jwc) , so |C|=|I/(w*V)| if V=1,w=1 then C=I. w=2×Π×f, if I wanna have w=1 I have to setup f=0.159 HzNow I put this frequency in the analysis... but the current which gives me actually the impedance is not the capacitance that should be for a varactor. My verilogA model contains leakage current also which also counts. This curve is ok for the current curve but with the frequency and ac magnitude of 1V the conversion to capacitance is opposite. Back to top IP Logged
 Ken Kundert Global Moderator Offline Posts: 2273 Silicon Valley Re: Simulation for Ferroelectric Varactor Reply #6 - Dec 14th, 2018, 12:05pm   If the drive voltage is 1, then plotting the current is the same as plotting admittance (not impedance). And if the admittance is purely imaginary and the frequency is 1/2π, then plotting the admittance is the same as plotting the capacitance in Farads.How are you passing an amp through your capacitor. Is C0 on the order of a Farad?-Ken Back to top IP Logged
 aguntuk Junior Member Offline Posts: 12 Dresden, Germany Re: Simulation for Ferroelectric Varactor Reply #7 - Dec 17th, 2018, 2:37am   I have applied equation of charge and current from a paper where they implemented the ferroelectric capacitor using the charge and considering the leakage current:The equations have been provided in the screenshot. I used their equation and the curve that they have generated is not matching maybe my conversion is not right.I have used the below code before ending module as in the paper described:                I(in,out) <+ itemp + ddt(qq);            V(in,vr) <+ ddt(L*1e-12*I(in,out));            V(vr,vc) <+ R*I(in,out); Back to top IP Logged
 Ken Kundert Global Moderator Offline Posts: 2273 Silicon Valley Re: Simulation for Ferroelectric Varactor Reply #8 - Dec 17th, 2018, 12:13pm   Please give the whole model. It is hard to make sense of what little you have provided.-Ken Back to top IP Logged
 aguntuk Junior Member Offline Posts: 12 Dresden, Germany Re: Simulation for Ferroelectric Varactor Reply #9 - Dec 20th, 2018, 12:33am   There was a mistake in conversion of units in the model. Now it works Thanks a lot. Back to top « Last Edit: Dec 20th, 2018, 2:16am by aguntuk »     IP Logged
 Pages: 1