• State
  • Locked Locked
  • Replies 3 replies
  • Subscribers 37 subscribers
  • Views 2263 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

Tina (designsoft): small signal model, bode plot....

Bishnu Ban
Intellectual 500 points
Other Parts Discussed in Thread: TPS54218, TINA-TI, LM5009

John recommend me to use TPS54218 as a substitute simulation for TPs57112Q1.

This design is just to step down 5V to 3.3V/500mA.

problem No.1: if I use the battery as  input voltage 5V start to calculate DC nodal voltage, it results a warning,  operating point not be found, related mistake on switching controller (Component: U1.X_U8_U296.D_D11). I could go around this problem by replacing battery with voltage generator and defining a voltage piecewise linear.

refer the schmatics (attached file)

3730.TPS54218.TSC

Problem2 No. 2 (main problem): I like to do small signal analysis. I have to add a voltage generator (sinusoidal ) at feed back loop  to calculate the ac transfer function.

response from Tina: two signals source can not be used (because there is one already on input side & second in feed back loop).

As suggested there is a slope compensation therefore simple/ideal condition is not enough for loop compensation.

Perhaps I have to buy Tina-Ti instead of free version from designsoft.





  • 0
    TI__Mastermind 41310 points

    Bishnu,

    I believe that you have several issues that are giving you the first problem. First, you are applying a transient signal to the load before the device has begun to operate. This is not a very good practice unless you are worried about how startup will be effected by your transient load. Using a battery will effect the DC operating point, however, if you choose one of the other options (Zero initial conditions or Use initial conditions), the simulation begins properly I would prefer that you use a PWL source, but the choice is yours. The startup waveform is ugly, but the device does function. Next issue is that your simulation is not long enough for the device to reach 3.3V output. You will need at least 800us of simulation time to reach 3.3V and your transient waveform did not change until after 1ms of simulation. Please see the revised circuit for the proper operation if you were looking for a load pull simulation.

    8054.TPS54218_up.TSC

    The second problem is that your are running a small signal analysis on a transient model (AC simulation on a time based model). This will not work and any information you obtain will be incorrect. In order to do a small signal analysis, you will need an Average (or AC) model for the TPS54218, which is available in the product folder. Please note that the average model will look very different from the Transient model since ONLY the loop information is contained. Please follow the directions in the .TSC file in order to simulate your custom circuit.

    You may purchase TINA from DesignSoft, however, this will not help the issues you are seeing.

    Britt

  • 0 in reply to Britt Brooks - Online Design Tools
    Intellectual 500 points

    Hi Britt!

    your information are very helpful.

    I have one problem more.

    the simulation of attached model (last post) takes around half hour.

    I have attached here another one from some community member, this takes only some seconds.

    I have changed the analysis parameters same to that of LM5009, but still takes half hours.

    6567.LM5009.TSC

    I am sure you will find the causes very fast.

    thanx

    Bishnu

  • 0 in reply to Bishnu Ban
    TI__Mastermind 41310 points

    Bishnu,

    The two product models are very different, hence, the simulation times will vary. The analysis parameters only control how the simulator interacts with the models and the circuit components. they do not establish a simulation speed. The TINA models do run faster than the models would run in PSpice, however, the models themselves are not the same.

    I do not know exactly how different the two models are, however, it appears that from a simulation standpoint, they are very different. I do not know how to make the TPS54218 simulate at the same speed as the LM5009. I doubt that anyone would know how to do that unless they made both models.

    Britt