Simulating MOS Varactor

Dear wgtkan,

I’ve had a lot of experience simulating varactor impedances over both frequency and voltage bias and should be able to help you a little.

A couple of questions if I may...

1. In your test bench, are you also intending to vary the DC voltage of your bias voltage “vtune”?

2. What frequency range are you intending to cover? In your test bench, with the 100 H inductor in series with your device under test, my concern is its impedance is too large.

3. I have provided a test bench in this forum that does what, I believe, you want if you are interested. Basically, it allows you to determine the C-V characteristic of the varactor at a given frequency. Let me know if you are interested.

Shawn 

Hello Shawn,

Thank you so much for your response.

My intention is the following:

1. To plot the Capacitance versus the bias voltage which is vtune at 2.45GHz

2. To plot the Capacitance versus frequency over the range of  1Hz to 3GHz.

I have looked at your setup where you used Voltage source in series with an inductor making the two (the series connection of the voltage source and inductor) in parallel with an AC current source and then using an Iprobe that is connected to one port of the  Varactor and the other port of the Varactor grounded.

I did AC sweep where my X axis is frequency and then went to direct plot > main form and plotted the voltage but weren’t able to get the real or imaginary showing up in the direct form window.
I even tried the XVal function of the calculator by multiplying it with 2*pi and still nothing.

What am I doing wrong?

Dear wgtkan,

wgtkan said:

My intention is the following:

Perfect - I understand totally and should be able to help a little anyway.

wgtkan said:

I did AC sweep where my X axis is frequency and then went to direct plot > main form and plotted the voltage but weren’t able to get the real or imaginary showing up in the direct form window.

I think you need to open the Results browser, select the node voltage of interest and send it to the calculator. From there, you may use the real() and imag() functions to determine the real and imaginary components. You may then apply the value() function to get the value of the real or imaginary part at one or more frequencies to compute the effective capacitance or real part of the impedance (after dividing by the current through the device under test). I've assembled a zip file that contains a Portable Document Format file entitled "example_spectre_small_signal_impedance_test_bench_test_cap_smlogan_v1p1_060420.pdf" with an example test bench, ADE-L setup and analysis panel, and example C-V and Re{Zin}-V characteristics. I've also include the state file as a directory with the outputs and the exported outputs with the latter in the comma separated file "test_cap_smlogan_v1p1_outputs_060420.csv". From that file, you may view the ADE-L/ADE-XL/Assembler expressions to compute the impedances at several frequencies - or even import the file to your own test bench. Obviously, the state file directory I included will not load the models I am using in your environment. I only include it so you have a copy of the variables, analysis, and outputs. Finally, if you use ocean scripts, there is a sample ocean script file I use to compute the real and capacitive parts of a device under test and place them in a comma-separated variable file called "z.ocn". The only function it calls is getWave() which is just a version of getData() with some error checking. I included it as well.

The zip file is attached.. Hence, download the file, extract its contents and take a look. Let me know if you have any questions wgtkan!

Shawn

test_cap_smlogan_ade_L_sample_test_state_and_example_060420.zip

Hello Shawn,

Thank you so much indeed. I will give it a try today and will let you know.