LT3972 failure

To whom it may concern,

   LT3972 switching regulator is widely used at some of my designs on RF equipement as main input regulator.

  We have experienced diferent problems with mentioned device and its related LT3971.

In some ocasions regulator fails and sets its output at the same value of its input voltage (+24V) therefore circuitry supplied is widely damaged. 

In some other intetgrations regulator does not works properly due to a, what we think, poor conexion to ground through the bottom pad. In these cases we are forced to resolder manually. 

Both cases are being repeated in time, at a low ratio, but repeated... and are looking for some advice to avoid any occurrence of these issues.

You will surely need further information regarding PCB design and auxiliary circuitry used, etc to provide some support... 

I'm really lookig forward to receiving your feedback.

thanks in advance. 

  • 0
    •  Analog Employees 
    on Jan 28, 2021 3:17 AM

    Hi MaR,

    You mentioned an issue related to poor soldering and what you do to solve it is to repeat it manually.

    Please consider the guidelines regarding soldering and PCB layout as prescribed in the LTC371 Datasheet.
    Lead Temperature (Soldering, 10 sec) is 300degC.
    See page 19 for the PCB Layout.


  • Thank you for answering John.

    As I wrote there are two different issues. The issue coming for a poor soldering does not lead to a high output voltage... in that case regulator simply doesn't work. Soldering recomendations are being taking into account so probably some other aspect take part.

    But what we are really worried about is on the other kind of issue when input voltage comes through regulator and is set at its output causing serious damages. Layout recomendations, as well as auxiliry circuitry recomendations, was studied and applied as far as intagrations constraints permit. 

    Have you been reported about similar failures before?

    Is there any kind of application note related to this regulator advicing how to isolate regulation steps to minimize possible damages?... Any recomendation in this sense?

    Thank you again for your valuable answers.

    Best Regards

  • 0
    •  Analog Employees 
    on Jan 29, 2021 6:47 AM in reply to MaR

    Hi MaR,

    From your description of the problem you are encountering, its seems that the fault occurs during start-up of the circuit.

    I recommend two things:

    1. Please review your design according to the recommendations in the LT3972 datasheet regarding the "Hot Plugging Safely" (see page 17)

    The LT3972 has an overvoltage protection feature which disables switching action when the VIN goes above 35V typical (33V minimum). When switching is disabled, the LT3972 can safely sustain input voltages up to 62V.

    2. You may evaluate your design, using our LTspice application model for LT3972, which you can modify to represent the particular circuit you are working on.
    This will help you debug the problem in your circuit without encountering actual damage in your circuit.

    see LTspice Overview:

    How to get the LT3972 LTspice model simulation:
    1. Go to
    2. See section Tools & Simulations

    When LTspice is already installed in your computer, you can access the LT3972 using the following steps:
    1. In the LTspice window, create new schematic
    2. Add a component, type LT3972
    3. Click "Open this macromodel's test fixture" option that will provide you a ready to use sample application model of LT3972

  • Thank you very much for your support. I will review all suggested information before going on with the thread.