TLV809 vs TPS3808 accuracy in monitoring a 12 volt line

Hi Marc,

Below is an application note that discusses the design considerations for the resistor divider. Basically, the current through the divider should be at least be 100x the leakage current into the device. This will yield the optimum accuracy for the least amount of current through the divider. This app note also discusses the accuracy of the SVS when the tolerance of the resistors is included.

http://www.ti.com/lit/an/slva450/slva450.pdf

For the pull-up resistor at the RESET output of the SVS, if there is no design requirements in the datasheet, the RESET voltage can be pulled up as high as the abs max rating of the RESET pin. However, it should be pulled up less than this so as to prevent damage to the device.

Please let me know if this helps.

Regards,

Darwin

Darwin,

Thank you for the help. I wanted to add some of the offline help you provided me to benefit others.

Sources of inaccuracy

The team identified three sources of inaccuracy for circuit there are as  follows:

• Voltage ranges on which the device can be trigger, Vit in accuracy
• External Resistor inaccuracy used to create the external voltage divider.
• Internal Resistor Inaccuracy which can affect the leakage current and the effective R2 of the external voltage divider.

Proposed circuit and accuracy

 Using the app note highlighted above as a guide we formulated a voltage divider that should draw 100 times the leakage current of the device to optimize the power vs. accuracy curve. Assuming a requirement to use E96 1% accurate resistors, below are the recommended resistor values to use:

Using the above resistors the team calculated the minimum voltage a reset would happen, maximum voltage, and nominal voltage that would trigger a reset considering all three sources of inaccuracy above. These are highlighted in the table below.