• State TI Thinks Resolved
  • Locked Locked
  • Replies 1 reply
  • Answers 1 answer
  • Subscribers 38 subscribers
  • Views 606 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

INA233 and other Spice models

Paul Buckman
Prodigy 100 points

Other Parts Discussed in Thread: INA233, TINA-TI

G'day everyone, hope all are well.

Ok, I am using Tina 12 Industrial as my circuit capture and simulation program, a continual issue is that the chips that I wish to use are manufactured by TI. I find it extremely difficult to find spice models for these (for example INA233). I have downloaded the IBIS model. It seems to me that spice models are for use with analogue simulation whereas IBIS are used in digital simulation.

My design interest is in that mixed mode arena that is microprocessor/ I2C/ analogue interface which is precisely where more and more of TI's and other manufacturers products sit. Why is it so difficult to find these mixed mode libraries? I would have thought that it would have been in TI's explicit interest to provide these to make it extremely easy to study/ experiment/ design and produce marketable product which in turn would drive TI's production.

Can any anybody please advise whether I am looking in the wrong place or where I am going wrong.

Thanks and regards

Paul Buckman

  • 0
    TI__Mastermind 32395 points

    Hello Paul,

    Thank for posting to the forum.

    You are right that SPICE models are mostly meant for analog simulations, while IBIS models are meant for digital/signal integrity simulations. A SPICE model is different from an IBIS model and simulating both at the same time requires powerful simulators like HSPICE or so it seems, TINA Industrial (I see that TINA Industrial does support IBIS analysis on their website).

    There are plenty of TINA-TI models for TI ADCs, but these are not complete mixed-signal models. They only simulate the analog performance of the device such as signal settling time or supply current. As far I know, there is no one ADC model that simulates this analog performance, translates it into digital codes, and then drives these codes on a digital communication bus that has its pins modeled to the IBIS model. 

    There are some ways to convert an IBIS model into a SPICE model. See this link: www.tina.com/.../

    The standard practice is to separate analog and digital integrity into two simulation domains and analyses. Ensuring the ADC (or a power monitor like INA233) will successfully capture the input signal requires following the datasheet guidelines and then possibly simulating in SPICE as well. Ensuring the digital communication lines have matched impedances is where the IBIS model comes in. The IC stuff that is actually happening in between the signal capture and the corresponding output codes is never going to be fully modeled as this could be proprietary.

    Now unfortunately, the INA233 does not have a TINA-TI model. The main reason is that use-case of this device is pretty much always limited to measuring a small voltage drop across a high-side or low-side shunt resistor and as long as the device is used this way, the accuracy (aka, the signal settling time) will easily meet the datasheet specifications. This is further affirmed by the fact that the INA233 and most other digital power monitors use delta-sigma dominated architectures, where settling times are much less important compared to SAR ADCs.

    For stand-alone ADCs from TI, these devices can be used in many more applications where the signal is provided by a sensor or buffer that may not have the BW or output impedance needed to satisfy a proper settling time, thus SPICE simulation is used to determine the optimal charge bucket filter, which is usually never necessary when measuring a shunt resistor on a power rail that won't be affected by INA233's dynamic input impedance.

    Overall, if the INA233 is being used in its typical application, then proper analog performance can be verified by looking at the datasheet and performing standard error analysis equations, which we cover in our training videos (see below). Once this is verified, the digital signal integrity can be verified with the IBIS model.

    Sincerely,

    Peter