I am a fresh college graduate and have recently started working as analog design engineer. I am working on 22nm FDSOI technology, And characterizing this device has been trouble for me. The device is way too sensitive for every parameter. I have done this before with 180nm technologies and there, hand calculations and simulator results used to be pretty close. Please suggest how should I characterize the device and what current equation should I use as the regular one isn't helping.

  • 2
    \$\begingroup\$ Characterizing such transistors is not an easy thing to do, particularly if your end goal is to create some kind of SPICE model. The subject may be too broad for this site. In any event, you should tell us all you can about what you have done, what you expected, and what you actually got. \$\endgroup\$ – Elliot Alderson Oct 16 at 17:08
  • 1
    \$\begingroup\$ I think "Characterization" is enough of an overloaded term in this space, that you need to be a bit more explicit and detailed with what you're talking about \$\endgroup\$ – W5VO Oct 16 at 17:44
  • \$\begingroup\$ with the gain-bandwidth of nanometer FETs, and some inductance in the shared RETURN path, you may simply have an oscillator. Does the use of a Keithley prevent oscillation? \$\endgroup\$ – analogsystemsrf Oct 17 at 3:22
  • \$\begingroup\$ Characterization means, that you are measuring the transistors on chip, deembed the pads, and transmission lines, plus the transistor wiring, and you try to fit the measurement results to a transistor model, like BSIM-IMG in this case. Is it really, what you are doing? If you just want to get an easy model for hand calculation, than I have to disappoint you, there is no good easy to use model for hand calculation which even just approximate the precision of such models at 180nm. Just take a look at the DC characteristic! It is in velocity saturation right after the threshold voltage. \$\endgroup\$ – Horror Vacui Oct 18 at 8:44
  • \$\begingroup\$ @ElliotAlderson, Sir my objective is to design a two stage miller compensated operational amplifier. For that I need to know the transistor parameters, I am taking square law current equation, so I tried calculating Kn, vth, lambda, but I observed that all these parameters are sensitive with width and length. I mean if I keep W = 500n, L = 250n and if I keep W = 1000n and L = 250n, the current is not simply doubling, I don't understand how it should be done. From all this comments I think the square law equation is not useful. So can you please suggest how I should do my hand calculations? \$\endgroup\$ – sumita sahu Oct 19 at 15:38

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.