• State Resolved
  • Locked Locked
  • Replies 8 replies
  • Subscribers 38 subscribers
  • Views 1396 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.

Is the INA333 TINA model working properly?

Andrea Fanelli
Prodigy 60 points
Other Parts Discussed in Thread: TINA-TI, INA333, INA118

Hi,

I'm trying to simulate a circuit using TINA-TI.

I'm using the the INA333 component in the simulation, but when I run the simulation, I get this error:

 

"Component: U3.D17"
I think there is a problem with the INA333 model.
Indeed if I replace the INA333 model with another instrumentational amplifier, like the INA118, everything works properly. 
Is there anyone who can help me in solving this problem? Can someone check and fix the INA333 model?
Thanks in advance for your replies.

 

  • 0
    TI__Expert 5515 points

    Andrea,

    I just tested the model using the INA333.TSC circuit found in the Examples->TI Test Circuits folder (select Open Examples... from TINA-TI's File menu). It ran fine for me, and I'm pretty confident the model is OK. Perhaps you can run the same test circuit and verify that it works for you.

    If it does, then the problem may be something in your particular circuit. If you can post the TSC file, we can take a look and maybe help you find the problem.

  • 0 in reply to Rick Downs - WEBENCH Design Center
    Prodigy 60 points

    Dear Rick,

    thanks for your answer. I finally found out what I was doing wrong with my circuit. I was using a sinus wave generator as positive input of the ina333 and I grounded the negative input, using single supply operation.

    I've just found out that the input should be over 0.1 V both for the positive and negative input while using single supply!

    Thanks for your help,
    Andrea 

  • 0 in reply to Andrea Fanelli
    Prodigy 60 points

    Dear Rick,

    after some trial, I still have convergence problems.

    If I visualize the output using an Oscilloscope, it is correctly displayed. But if I try to run Analysis -> AC analysis -> AC transfer characteristic or calculate nodal voltages, it gives me an error.

    Can you please help me to find out which is the problem in my circuit? 

    You can find the schematic of the circuit attached. It's weird that, if I replace the capacitance of the integrator with a shortcut and if I remove the first order passive low pass filter, the AC analysis converges.

    Thanks in advance for your help.

    Cheers,
    Andrea

    0447.Analog with MFB.rar

  • 0 in reply to Andrea Fanelli
    Prodigy 20 points

    We could not get the TINA model to behave, nor the SPICE model. They are useless. However, the hardware behaves as promised.

    Kevin Buchs

  • 0 in reply to Andrea Fanelli
    TI__Expert 5515 points

    Andrea,

    SPICE often has problems converging in circuits with lots of feedback loops - you sometimes have to play some tricks to get it to find a point to work from. In your case, simply adding a 1M resistor from the output of the servo amp to ground (only for simulation purposes) made it work for me. Sometimes adding these resistors, or node set components, can solve these types of problems.

     

  • 0 in reply to Kevin Buchs
    TI__Expert 5515 points

    Kevin,

    See my answer to Andrea - it may not be the model, per se - some circuits give SPICE simulators fits as they try to find operating points. I'm glad you found the hardware suitable!

  • 0 in reply to Rick Downs - WEBENCH Design Center
    Prodigy 60 points

    Thank you all for your support and for the great answers!

    It is working now :) :)

    You've made my day!


    Best,

    Andrea

  • 0 in reply to Rick Downs - WEBENCH Design Center
    Prodigy 20 points

    Rick,

    I appreciate your point, having had a lot of experience in getting SPICE to converge. In our case (ECG circuit) we were unable even with shunt resistors. With super-heroic efforts, with shut resistors and initial conditions, we could obtain stabilization, but then the tiniest change in inputs would sent the circuit into saturation. So, the results never were correct.

    My guess would be that the model is created with (mathematically ideal) behavioral elements (V controlled V sources, etc.) which then hurt the stability. With real transistor models, you wouldn't have that extreme because they have real resistance and capacitance. Obviously the real physical circuit is stable so I am pretty sure the modeling can be done more effectively.

    - Kevin