• State TI Thinks Resolved
  • Locked Locked
  • Replies 9 replies
  • Answers 3 answers
  • Subscribers 62 subscribers
  • Views 644 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.

TPS43060: impedance looking into the FB feedback pin

Nitish Agrawal
Expert 1595 points
Part Number: TPS43060


Hello,

I would like to know the impedance looking into the FB feedback pin.

I am trying to measure loop gain of my boost circuit using TPS43060 and would like to make sure that voltage perturbation I am injecting in the Spice simulation circuit is at an appropriate node.

The (compensated) loop gain I am measuring using my simulation is quite different from that predicted by the TPS4306x_BOOST_CALCTOOL_REVB.xls (total phase plot).

Following is what the two methods are predicting:

Frequency    EXCEL TOOL              SIMULATION (using TPS43060 transient Spice model)

1kHz        16.7dB / -78deg.              14.7dB / -113.5deg.

2kHz        12.5dB / -79deg.              6.5dB / -114deg

It is my understanding that the Excel tool shows the total phase plus 180deg. (resulting in a positive phase margin).  In actuality, the Excel tool was showing the following (from which I subtracted 180deg. to get the above reported phases):

Frequency    EXCEL TOOL

1kHz        16.7dB / 102deg.

2kHz        12.5dB / 101deg.

Please correct me if I am wrong.

I would appreciate your help in trying to figure out the discrepancy.

Regards,

Nitish Agrawal

  • 0
    TI__Guru* 78675 points
    Hi Nitish,

    Thank you for considering the TPS43060 in your project. When you use the Pspice, I believe the Pspice has a goal function that can directly display the gain and phase without requiring the user to adjust by 180 deg. Since you got -114deg, I am afraid your sim plot may refer to a wrong goal function.

    The excel tool should be correct and it does need to adjust by adding the 180deg to get the actual phase margin. However, there will be small difference between the excel tool and the sim, because the excel tool is based on small signal model ignored nonlinearty, while the sim model may be more close to reality owing to some higher order effects being included. Either the excel tool or the sim should be used as guidelines to have a good start point of a design, and after all the experiment is always needed to verify and fine tune the circuit for optimal performance.

    Thanks,
    Youhao
  • 0 in reply to Youhao Xi
    Expert 1595 points
    Hi Youhao,

    Thank you for your response. TPS43060 appears to be a good part.

    What is the input impedance of the FB pin in TPS43060?
    The reason for this question is because I would like to make sure that I am injecting the voltage perturbation at an appropriate node (to calculate the loop gain).
    I am surprised there is so much difference between the Excel tool and the simulation....I just want to make sure I am doing everything correct.

    Kind regards,
    Nitish Agrawal
  • 0 in reply to Nitish Agrawal
    TI__Guru* 78675 points
    Hi Nitish,

    The info is given in the datasheet EC table. The FB pin input bias current is 20nA typical, so you can assume its in the scale of multi Mega Ohm, and the effects are negligible.

    Thanks,
    Youhao
  • 0 in reply to Youhao Xi
    Expert 1595 points
    Hi Youhao,

    Thank you for your response. A couple of follow up question:

    a) Is there a way for me to see the small signal gain/phase characteristics for frequencies below 10Hz shown in the TPS4306x_BOOST_CALCTOOL_REVB.xls tool (say, down to 0.01Hz)?
    b) Does the gain and phase calculation in the WEBENCH tool use the exact same as in the Excel tool?

    Kind regards,

    Nitish Agrawal
  • 0 in reply to Nitish Agrawal
    Expert 1595 points
    Hi Youhao,

    I had an additional question regarding the phase of Gvc shown in the TPS4306x_BOOST_CALCTOOL_REVB.xls tool:
    Would you please confirm if the phase of Gvc is shown as "actual phase + 180deg".

    Can you also confirm if the above is true for Gea and total phase?

    Kind regards,
    Nitish Agrawal
  • 0 in reply to Youhao Xi
    Expert 1595 points
    Hi Youhao,

    I had an additional question regarding the phase of Gvc shown in the TPS4306x_BOOST_CALCTOOL_REVB.xls tool:
    Would you please confirm if the phase of Gvc is shown as "actual phase + 180deg".

    Can you also confirm if the above is true for Gea and total phase?

    Kind regards,
    Nitish Agrawal
  • 0 in reply to Nitish Agrawal
    TI__Guru* 78675 points
    Hi Nitish,

    Yes it is already adjusted, so the phase curve reading is the phase margin.

    Thanks,
    Youhao
  • 0 in reply to Youhao Xi
    Expert 1595 points
    Hi Youhao,

    Thank you for your continued support. A few more questions:

    a) My min duty cycle is 0.4153 and Dmax = 0.5363. Based on these values, the TPS43060 datasheet section 9.2.1.2.1 (selection of switching frequency), the max fsw = 415kHz.
    However, WEBENCH tool recommends fsw = 426kHz for 96% efficient design. Is that desirable?

    Also, the Excel design tool doesn't flag any error when a 500kHz switching frequency is used.

    b) Is there a way for me to see the small signal gain/phase characteristics for frequencies below 10Hz shown in the TPS4306x_BOOST_CALCTOOL_REVB.xls tool (say, down to 0.01Hz)?

    c) Does the gain and phase calculation in the WEBENCH tool use the exact same as in the Excel tool?
    I would appreciate your comments.

    Kind regards,

    Nitish Agrawal
  • 0 in reply to Youhao Xi
    Expert 1595 points
    Hi Youhao,

    I realized I made an error in calculation in fsw_max when posting my previous question (a), repeated below (so please disregard it):

    a) My min duty cycle is 0.4153 and Dmax = 0.5363. Based on these values, the TPS43060 datasheet section 9.2.1.2.1 (selection of switching frequency), the max fsw = 415kHz.

    CORRECTION: The correct fsw_max = 4.15MHz when calculated based upon the minimum off-time of TPS43060.

    Please just focus on my questions (b), (c).

    Regards,

    Nitish Agrawal