• State
  • Locked Locked
  • Replies 4 replies
  • Subscribers 46 subscribers
  • Views 270 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.

ADS8920B: ADS8920B convergence issues with TINA.

Matt Moody
Prodigy 40 points
Part Number: ADS8920B
Other Parts Discussed in Thread: ADS8900B

I've been trying to simulate a circuit taking low and higher level photodiode currents and digitizing them with the ADS8920B.  Attached is the schematic I am using.  It works fine when the ADC is not in the circuit, then it has convergence problems at the 3mA input when the ADC is inserted back into the simulation.  Any idea why I can't get a good simulation with the ADC inline with the input at 3mA?

  • 0 in reply to Matt Moody
    TI__Mastermind 18037 points

    The ADS8900B model released to the web is a very comprehensive model that covers many of the data sheet specifications. Unfortunately, this means that the model is complex, and sometimes has convergence issues. This is especially true if the support driver circuit is complex. In your case I am not surprised that it doesn’t converge, based on the complexity. In these cases I recommend using a simplified model. I can help you with that. What this in mind, I have a few questions / comments that I would like to ask before I begin.

    1. The main goal of SAR ADC simulation is generally to assure that the driver amplifier settles by the end of the acquisition period. This type of simulation is documented in detail in Driving the SAR Video. If I create a simplified model, this is all that the model would do. Do you want this type of model?What sampling rate will you use.
    2. For other types of analysis (i.e. Noise, Bandwidth, Stability). I suggest using your circuit without the ADC. The total noise from your circuit can be calculated to include the effects of the voltage reference and ADC noise. See ADC+Amp+Reference Noise calculation video.

    Let me know if you want this kind of model and if so I will send you it later next week. I will confirm that it converges.

     

    I hope this helps.

     

  • 0 in reply to Art Kay
    Prodigy 40 points

    Thank you for the background and helpful tips.  I do typically run the simulations without the ADC, more so for speed of the analysis.  I do however like to integrate the ADC into the full simulation for a complete end-to-end view of what is going on (even if it's not as comprehensive), so if you can provide the simplified model as you called out in #1, that would be great.  I'm running at the 1MSPS.  Thanks for the help.

  • 0 in reply to Matt Moody
    TI__Mastermind 18037 points

    Matt,

    I took your final stage and used it to drive the simplified ADC model.  I noticed that you used the EVM circuit to drive the ADC (nice choice as this is a proven circuit).  I did not include all of your circuitry as it did not converge for me (without the ADC).  Can you double check the rest of the circuit and see if you can get it to simulate?  I can help here, but I as you to look at it, as you mentioned that it converged without the ADC.  In any case, the driver circuit settles well.  I suggest increasing the filter resistors from 2.21 ohms to 3 ohms as this looks more stable.  Below is a link to the circuit.  Run "Analysis>Transient Analysis" and look at the "Verror" signal.  It should be less than 1 LSB which is 76.3uV.  Note that the driver circuit was designed for the ADS8900B (20 BIT Version).  So, you shouldn't have trouble settling to 16 bits.

    ADS8920b.TSC