MCU-PLUS-SDK-AM263PX: SDK 10.x Drivers_open() Fails Assert Check

Kier

Part Number: MCU-PLUS-SDK-AM263PX
Other Parts Discussed in Thread: TMDSCNCD263P, SYSCONFIG

Tool/software:

Hello,

The following line fails the assert check in Drivers_open() in Core2 (No RTOS). Note: I don't get the same problem for ADCs in FreeRTOS Core 0.

[r5f1-0] 0.000981s : .c:Drivers_open:67: SOC_getAdcReferenceStatus(0) == true failed !!!

However, if I increase the sleep time from 1us to 1000us on line 64, the check passes.

<Question>

How did you decide on 1us sleep time on line 64 please? Why is this even needed? The code comment "Wait for ClockP_sleep" doesn't say anything useful, it just states the obvious.

11 months ago

0 Brennan Hartigan 11 months ago

TI__Genius 13629 points

Hi Kier,

Are you using one of the TI AM263Px EVMs? You may have an ADC VREF switch in the wrong position.

Regards,

Brennan

0 Kier 11 months ago in reply to Brennan Hartigan

Mastermind 9870 points

Thanks Brennan.

Yes, I'm using the TMDSCNCD263P. How would I determine "wrong"? Please elaborate.

In any case, it is unlikely that it is "wrong". I've been using SDK 9.x just fine for my application. SDK 10.x introduces the code in question above (for the same configuration) and breaks my application.

I think an answer to my question in the opening topic is appropriate here.

0 Brennan Hartigan 11 months ago in reply to Kier

TI__Genius 13629 points

Kier,

Please confirm that the ADC VREF switches are set to the following positions on the TMDSCNCD263P:

SW9.1: Pin 1-2 (DOWN)

SW9.2: Pin 4-5 (DOWN)

These switch positions allow for the ADC VREF to be the AM263Px on-die LDO, which is required for the AM263Px MCU+ SDK ADC Examples to function properly.

Regards,

Brennan

0 Kier 11 months ago in reply to Brennan Hartigan

Mastermind 9870 points

My switches are both up because we use the output of SW7 (PMIC LDO1V8) for reference in our application.

I suppose the next question would be why does the code in Drivers_open() enable the internal reference in SDK 11.0 but not in SDK 10?

0 Brennan Hartigan 11 months ago in reply to Kier

TI__Genius 13629 points

Hi Kier,

While I wait on a response from the ADC driver author, can you confirm that the example runs correctly if you change the switches to reference the internal VREF?

Regards,

Brennan

0 Kier 11 months ago in reply to Brennan Hartigan

Mastermind 9870 points

Hi Brennan,

Brennan Hartigan said:
can you confirm that the example runs correctly if you change the switches to reference the internal VREF?

I'm using our application not an example but yes, changing SW9 to both "down" passes the assert with sleep set to 1us.

So to recap, when I updated the SDK from 9.x to 10.x, SYSCONFIG added the mandatory extra SOC code in Drivers_open(). This is the subject of this topic/complaint:

MCU-PLUS-SDK-AM263PX: Drivers_Open() Now Mandatory To Call - Arm-based microcontrollers forum - Arm-based microcontrollers - TI E2E support forums

It seems that in my haste to keep my application going having been forced to use Drivers_open() with the added SOC code, I did not notice that the internal reference is enabled when my application has been using external reference with SW9 up.

I could not find a way to affect the choice of reference and that is subject of this topic:

MCU-PLUS-SDK-AM263PX: How To Select External ADC Reference in SYSCONFIG - Arm-based microcontrollers forum - Arm-based microcontrollers - TI E2E support forums

0 Brennan Hartigan 11 months ago in reply to Kier

TI__Genius 13629 points

Kier,

Understood, the appropriate experts will follow up with you on the new threads you created.

Regards,

Brennan