TI Friends and Family,
Our customer is experiencing about a 15% rate of AM5718 devices that do not boot-up on first boot-up. They are failing the TI recommended I/O calibration.
The team is not having any success resolving this issue quite yet so I hoping maybe to get further assistance.
We believe they have tried an automatic reboot but this does not always work so I don't think this is a smoking gun. One first thought is that there is a problem in the schematic or in the layout with the impedance matching which is why the AM5718 is getting mad on boot-up. I can forward both separately.
In an attempt to help get to a root cause, I have pasted our initial comments below:
----
First off, from what I digest on the TI E2E forums, rather than a simple reboot, it looks like the suggested steps below are what was recommended (and still would be) by TI for a proper IO recalibration sequence:
- IODELAY recalibration sequence:
- - Complete AVS voltage change on VDD_CORE_L
- - Unlock IODLAY config registers.
- - Perform IO delay calibration with predefined values.
- - Isolate all the IOs
- - Update the delay mechanism for each IO with new calibrated values.
- - Configure PAD configuration registers
- - De-isolate all the IOs.
- Relock IODELAY config registers.
We have asked if this is being followed?
Secondly, if not using AVS (Adaptive Voltage Scaling) then you can actually create the iodelay files by using the TI pinmux tool for the processor. You can get the tool from the below link: http://www.ti.com/tool/PINMUXTOOL
Lastly, if you are using AVS then you need to recalibrate the iodelay. Please refer to the TRM Technical Reference Manual, as it has all the details to help do this.
***One other thing to keep in mind is that when using AVS you have to program a register that changes the IOdelay for each peripheral. Please again review the latest TRM and take a look at section 18.4.6.1.8 IO Delay Recalibration. This should not affect what you are seeing but in the final product you need to do this to make sure you meet the timing for each IO.***
-----
As always I appreciate anyone's support. Any additional comments or guidance appreciated.
TY,
CY