AN 896: Multi-Rail Power Sequencer and Monitor Reference Design

ID 683778
Date 3/15/2024
Public
Document Table of Contents

2.4. Output Voltage Data Formats and Related Parameters

The reference design uses the DIRECT format—defined in the PMBus* Specification—to store all data for the input or output voltages for current status, warning levels, or error levels.

The power coefficients are:

  • Determined by you
  • Specific to each voltage rail for every page
  • Based on the voltage scaling resistors in the design

The ADC in dual supply MAX® 10 devices can measure from 0 V to 2.5 V with 12-bits resolution. In single supply MAX® 10 devices, the ADC can measure up to 3.0 V or 3.3 V depending on your power supply voltage. To provide a sufficiently large scale that retains enough resolution for accurate measurements, select appropriate values for the voltage divider.

For example, on a 3.3 V input, if you set your OV_Fail at 115%, you would need to be able to measure a range from 0 V to 3.8 V. This example assumes that you use a 2.5 V external reference voltage (ADC_VREF) and you apply a voltage divider, as shown in the following figure, to the monitored voltage rail.

Figure 12. Voltage Divider Example on a 3.3 V Rail


In the Platform Designer, the parameter editors automatically calculates the values for you. You just need to ensure that the output of the voltage divider does not exceed ADC_VREF for an overvoltage condition. The calculations show you how the settings and reported values relate to the PMBus* specification.

Calculations Related to PMBus* Specifications

Given the DIRECT format definition of :

Where:

  • X is the calculated "real world" value in the appropriate units such as A, V, and °C
  • m is the slope coefficient—a two-byte, two's complement integer
  • Y is a two-byte, two's complement integer received from the PMBus* device
  • b is the offset—a two-byte, two's complement integer
  • R is the exponent—a one-byte, two's complement integer

You can determine the coefficients knowing that:

Using the 16-bits resolution available for m, you get these constants:

Therefore, if you read back a value of 3549 after sending the READ_VOUT command, you can apply the constants to the formula and solve: