Stats

  • Locked Locked
  • Replies 3
  • Subscribers 63
  • Views 15014
  • Members are here 0
This discussion has been locked.
You can no longer post new replies to this discussion. If you have a question you can start a new discussion

Huge performance difference between using the ideal inductors and the non-ideal ones in RF circuit. Please help~

Alex Liao

Hi,

In my VCO circuit, I want to do a layout. Anything comes from analog lib does not have layout view, like 'ind' 'cap'. When I replace the device of 'ind' by the 'spiral_std" (while keeping the inductance the same), the fundamental frequency gives similar values but the AC transient output waveform is way different. The output using the ideal device 'ind' has a large output swing which is good while using the later device, the output swing is almost 0, say varying from 1.1959V to 1.1961V, almost un-oscillating.

It seems there are large difference between the ideal inductance and the 'spiral_std" in terms of the modelling.

I also did similar experiment on ideal 'cap' VS the 'mim_cap" which has the size information like 'spiral_std' does.

With same capacitance value, I did not notice anything different. It means the modelling for those two are similar or at least the circuit is insensitive to the difference between them.

This is not the case for the inductor, which is a very large device in terms of geometry. I could think of many effects would be involved like parasitic resistance and capacitance.

In this circumstance, what should I do? Due to layout needs, how to use any non-ideal device like 'spiral_std'  given the inductance value is already solved/known? Any extra compensation action I can take or I need to re-size the design or even re-design?  

Kind Regards,

Alex

  • Dear Alex,

    I am not sure I fully understand your specific observation or question, but I think you are asking if the large difference you observe in the amplitude of your VCO when using an ideal inductor and using a non-ideal inductor model is reasonable. You also note that the magnitude of the VCO stead-state frequency when using the two different inductor models is similar.

    If that is your question, the amplitude of oscillation can be a very strong function of the inductor real impedance if the magnitude of the negative resistance of the sustaining amplifier is not much larger than the inductor real impedance. The frequency of oscillation is far less dependent on the real part of the inductance. From your observation of the magnitude of the oscillation using the "spiral_std" model whose real part is non-zero, I would guess that the magnitude of the sustaining amplifier's negative resistance at the desired VCO frequency is nearly the same as the the real part of "spiral_std". You must include a realistic model for the inductor you are planning to use if you are studying the actual VCO performance - which will always include a non-zero real part that is, in general, a function of frequency.

    Shawn

  • Dear Shawn,

    smlogan said:
    You also note that the magnitude of the VCO stead-state frequency when using the two different inductor models is similar.

    For this point, what I said something similar is about the results resulting from using ideal Capacitor (cap) and the non-ideal Capacitor( mim_cap). But the effect of the capacitance by using different models seems not a big deal.

    smlogan said:
    The frequency of oscillation is far less dependent on the real part of the inductance.

    smlogan said:
    which will always include a non-zero real part that is, in general, a function of frequency.

     

    I am wondering if you wrote a typo here. "Far less dependent on the real part?" Should it be "the oscillation frequency is very dependent on the real part?" according to the context. Or the term "far less dependent on" means "almost not dependent on" which looks strange according to whole idea you want to express.

    Regards,

    Alex

  • Dear Alex,

    >I am wondering if you wrote a typo here. "Far less dependent on the real part?" Should it be "the oscillation frequency is very dependent on the real part?" 

    Unfortunately, no. The oscillation frequency of your VCO will be far more dependent on the imaginary part of your inductor than its real part. The real part impacts the Q of the inductor and has a second order effect on oscillation frequency. I know you have used the formula 1/root (LC) to estimate the frequency in your previous posts...note that this formula does not include the real impedance of the inductor or capacitor.

    Shawn

Community Guidelines

The Cadence Design Communities support Cadence users and technologists interacting to exchange ideas, news, technical information, and best practices to solve problems and get the most from Cadence technology. The community is open to everyone, and to provide the most value, we require participants to follow our Community Guidelines that facilitate a quality exchange of ideas and information. By accessing, contributing, using or downloading any materials from the site, you agree to be bound by the full Community Guidelines.