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.

LM5175-Q1: Short Circuit Protection; Half Switching Frequency; Constant Current Loop Gain Test Method

Part Number: LM5175-Q1
Other Parts Discussed in Thread: LM5175

I have three questions about the LM5175-Q1:

1.Is the calculation of the inductor current during a short circuit an algorithm such as the overcurrent protection threshold voltage (buck:76mV, boost:160mV) / sampling resistor, so that when the calculated threshold is biased to the maximum, the inductor current is relatively large during a short circuit. Is there any other short circuit protection mechanism?

2.My customer found that after entering the working condition of buck-boost, the switching frequency will be halved, which is the data measured in the actual application. There is no corresponding description in the specification and datasheet. Is this a normal phenomenon? Is there any parameters and explanations?

3.My customer wants to use LM5175-Q1 to charge the battery, and the current-limit mode (constant current loop) is set. The customer wants to know how to test the open-loop gain of the current loop, since the test result of them seems abnormal (oscillation). The method they used is the inject the small signal to the Rsense(R448 in the figure) of PIN ISNS+ and ISNS-.

/resized-image/__size/3200x2400/__key/communityserver-discussions-components-files/196/LM5175_2D00_Q1.gif

  • Hello, 

    Thanks for reaching out. The device has in addition to cycle by cycle peak current limit also the hiccup protection. Do you have a screenshot to share of the switch nodes with halved switching frequency? We're looking forward hearing from you. 

    Kind regards,

    EM

  • Hi EM,

    Thanks for your reply. But I have other questions:

    1. LM5175 has two kinds of current protection: 1. cycle by cycle peak current link & 2. input/output average current limit. And the limitation boundaries are set by the Rsense (R448 and R436) of both protection circuit. Is this right? 

    2.The switching frequency is halved, maybe my customer wrongly observed the single switching node when the device works in a buck-boost mode. It seems that the buck node and boost node switches alternately. So that the switching frequency seems halved. But it may be right when both switch nodes are observed. It this right?

    3.My customer want to measure the open-loop gain of the current inner loop of LM5175, what should they do? According to the circuit given below, where is the small signal should be injected and where should the observation probe be set? You can mark it in the circuit.

    It will be appreciated for your reply~

  • Hi Matt,

    sorry for the delayed reply.

    Please see my answers to your question:

    1. correct

    2. yes, this is also correct - just could not be seen easily in the datasheet - see Figure 13 and 14

    3. stability measurements are done adding a small resistor above the high side FB resistor divider. At that point the error is injected and measured.

    Best regards,

     Stefan

  • Hi Stefan,

    The stability measurement is expected on the inner current loop. Is still the small signal injection at the FB resistor (which is at the voltage loop)?

  • Hi Matt,

    maybe I misunderstood, what do you mean with the inner current loop?

    Best regards,

     Stefan

  • Hi Stefan,

    Yes, I mean the inner current loop.

    Best Regards,

    Matt

  • HI Matt,

    can you sketch a picture or explain in more detail what you understand as the inner current loop.

    Thanks,

     Stefan

  • Hi Stefan,

    Since LM5175 use peak current control/ valley current control in the voltage regulation.

    According to the PCM control diagram, there is a peak current control inner loop of the whole control system.

    My customer wants to know how to measure the inner loop open-loop gain, but they don't know the methods.

    Thank you~

  • HI Matt,

    copying the info from the privat chat here again:


    I think it should work like this but the injection signal needs to be very small (just a few mV). this on the other hand make the noise ratio critical.

    Best regards,

     Stefan

  • Hi Stefan,

    I have one more question, my customer wants to confirm:

    If they want to test the inner current loop response during the average current limitation mode, they give an injection method to the resistor connected to the ISNS+ and ISNS- (instead of the CS and CSG connection given by you), whether this kind of measurement valid?

    Thank you very much!

  • Hi Matt,

    it depends on the filter for the ISNS input. The information is only valid till (lower) the corner frequency of this filter.

    (As disturbances from the sensor would be filter.)

    Best regards,

     Stefan