Post Go back to editing

ADM13307 - Can the Sense Pins Be Driven When Unpowered?

    Good afternoon.  I have a customer using the ADM13307-4ARZ in a supervisory function.  The maximum voltage that the SENSE pins can see are as follows:

          SENSE1_MAX_INPUT = 1.1V + 5% = 1.155V (not sure if SENSE1 and SENSE3 are backwards and have asked for verification)

          SENSE2_MAX_INPUT = 1.8V + 5% = 1.89V

          SENSE3_MAX_INPUT = 2.5V + 5% = 2.625V

 

     where VDD is 3.3V +/- 5%.  Now, with a reference voltage of 0.6V (Figure 2) and looking at the "Absolute Maximum" ratings for the "Sense" pins on page 6 of the data sheet, the equation given yields the following as the absolute maximum voltage allowed on each sense pin where VIT1 = 2.25V, VIT2 = 0.6V and VIT3 = 0.6V according to the ordering guide on page 11.

 

                        MAX_SENSE - [(VDD + 0.3V) * VIT] / VREF → from page 6

 

                        MAX_SENSE1 = [(VDD + 0.3V) * VIT1]/VREF1 = [(3.3V + 0.3V) * 2.25V] / 0.6V = 15V  so good here

 

                        MAX_SENSE2 = [(VDD + 0.3V) * VIT2]/VREF2 = [(3.3V + 0.3V) * 0.6V] / 0.6V = 3.6V so good here

 

                        MAX_SENSE3 = [(VDD + 0.3V) * VIT3]/VREF3 = [(3.3V + 0.3V) * 0.6V] / 0.6V = 3.6V so good here

 

      The ADM13307 is used in an application where the main power goes away but the SENSE voltages they are monitoring stay on because of a SuperCap for 5-10ms.

     QUESTION:  Is it safe to have voltages present on SENSE1, SENSE2 and SENSE3 when no voltage is applied to the VDD pin 8?

  • Hi there,

    by breaking down and rewrite that equation it can be changed to:

    SENSEmax=(VDD + 0.3V) * [VIT1/VREF1]

    where Vit1/Vref1 is the divider ratio.

    VDD + 0.3V is a indication of ESD diode clamping to VDD.

    What this means is the node after the divider is clamped to VDD pin via a ESD diode.

    If it is a 2.5V threshold input (SENSE1), the divider is internal and adjustable ones has external divider.

    Based on the equation you can not have SENSE2 at 0.6V while VDD is at 0V.

    In reality, what happens is the when VDD goes down first, current will leak from the sense pins through the internal ESD diode to VDD pin. and because the current are limited by the dividers, it is safe enough for a short (or even long) period of time.

    In such event the current into the pin will be limited by the top resistor in the divider, so user needs to make sure it is reasonably big for the external ones. For Sense1 with internal divider, the spec table says the max input current is 8uA at 5.5V, that's mostly due to leakage through the internal divider, with 2.25 threshold, the divider ratio is about 1:1. So leakage with only top resistor should be around 16uA. so you get 4.8uW across the ESD diode during the 5-10ms power down, which I'm sure the diode can handle.

    Ren

  • Hi Richard,

    i think the value you've chosen are reasonably safe.

    the leakage on these pins are low enough (50nA and 25nA) so even double the resistance wouldn't cause much accuracy shift due to leakage.

    Ren

  •     Good morning Ren.  My customer is asking for verification of his current circuit as shown below for each SENSE input.  I have included my comments in parentheses based on my understanding of your previous response.  Please confirm.

         Thank you.

     

         1)  SENSE1 is used for 2.5V.  We do not have an external resistor on SENSE1 (Rich - since the divider is internal, there should be no need for an external current limiting resistor).  Do we need one to limit current further or is the internal resistor R1 good enough?

     

         2)  SENSE2 is used for 1.8V and we have 16.9K as top resistor so current is 1.8V/16.9K = 106.5uA (Rich - therefore the power would be 185.9uW which seems high to me given your example for SENSE1). Is this good enough or do we need a higher resistor?

         3)  SENSE3 is used for 1.1V and we have a 6.49K as top resistor so current is 1.1V/6.4K = 169.5uA (Rich - therefore the power would be 181.7uW which seems high to me given your example for SENSE1).  Is this good enough or do we need a higher resistor?

     

  • This question has been assumed as answered either offline via email or with a multi-part answer. This question has now been closed out. If you have an inquiry related to this topic please post a new question in the applicable product forum.

    Thank you,
    EZ Admin