SDAA315 March   2026 AM2432 , AM625 , AM62A7 , AM62D-Q1 , AM62L , AM62P , AM6442

 

  1.   1
  2.   Abstract
  3.   Trademarks
  4. 1 Introduction
  5. 2 Terminology
  6. 3 PHY Tuning Algorithm
  7. 4 Key Tuning Parameters
    1. 4.1 Parameter Configurations for Non-DQS PHY Tuning Algorithm
    2. 4.2 Parameter Configurations for DQS PHY Tuning Algorithm
  8. 5 Prerequisites for PHY Tuning Algorithm
    1. 5.1 Hardware Requirements
      1. 5.1.1 Flash Device Preparation
      2. 5.1.2 PHY Configuration
    2. 5.2 Attack Vector
    3. 5.3 Passing vs Failing Region
    4. 5.4 Master vs Bypass Mode
      1. 5.4.1 Bypass Mode
      2. 5.4.2 Master Mode
  9. 6 Need for a Newer Tuning Algorithm
    1. 6.1 Temperature Variations
  10. 7 Algorithm Implementation
    1. 7.1 DQS PHY Tuning Algorithm
      1. 7.1.1 Diagonal Selection
      2. 7.1.2 Valid Read Delays Selection
      3. 7.1.3 Corner Points Identification
        1. 7.1.3.1 Corner Point Selection for Only One Read Delay Value
        2. 7.1.3.2 Corner Point Selection for Two Different Read Delay Values
      4. 7.1.4 Tuning Point Selection
    2. 7.2 Non - DQS PHY Tuning Algorithm
      1. 7.2.1 Fix Tx DLL Value
      2. 7.2.2 Find Rx Window 1
      3. 7.2.3 Find Rx Window 2
      4. 7.2.4 Choose Larger Rx Window
      5. 7.2.5 Calculate the OTP
      6. 7.2.6 Temperature Consideration
  11. 8 Tuning Enhancements
    1. 8.1 Tuning Time Optimization – Skip Tuning Feature
    2. 8.2 Runtime Validation – Validate OTP
  12. 9 Summary
  13. 10References

Abstract

This Application Note provides an overview of the OSPI DQS PHY Tuning algorithm, a new feature introduced in MCU+ SDK 11.02, designed to optimize high-speed serial NOR and NAND flash access on Texas Instruments' Sitara™ processor family, including AM243x, AM62x, AM62Ax, AM62Dx, AM62Px, AM64x, and AM275x devices. The PHY Tuning process automatically calibrates three timing parameters which include Tx DLL, Rx DLL, and read delay. These parameters compensate for board-level propagation delays, process, voltage, and temperature variations. The target audience includes embedded software engineers, system architects, and hardware designers working with OSPI flash memory on Sitara processors.