STM32L476 USB inconsistencies.

Sorry I have to head out and Support requested an overnight run with what they sent.

It works too but just skips PLL's and runs sysclock off HSI (LIKE DFU)   Yeah that gets 48Mhz too.

I'll fix the HSI / MSI cal line 204 tomorrow too!

THANK you Jan!

And THANK you ST !

It ran all night fne 48.007Mhz.

Time for some cruelty testing.

Skip PLL's = Skip the OTHER PLL's generating the 80Mhz.

Offered by Jan on the forum. Quite possibly the REAL cause all along. I'd vote with Jan . :)

And it still warrants more attention from ST.

Overnight test finished. Stable at 48 MHZ after tortured a bit.

Original demo code loaded back in. Initially came up at 48Mhz(?)  Flakey....

Unplug - Replug = 57Mhz.    Okay that's what I thought..

Changing line 204 makes it happy.

BAD WAS -  RCC_OscInitStruct.

HSICalibrationValue

= RCC_MSICALIBRATION_DEFAULT;

GOOD IS -  RCC_OscInitStruct.

MSICalibrationValue

= RCC_MSICALIBRATION_DEFAULT;

In the BAD configuration the

MSICalibrationValue

Unplug - replug 10 times. No change = HAPPY   = 48Mhz

My DELAY LINES fix, while just a crude TEST, were probably getting the stack/ram RANDOMLY loaded with accidental values that left the MSI trim values in a random HAPPY (48Mhz) state instead of NO delay lines leaving it in a random UNHAPPY (57Mhz) state. That's just my guess.

Many thanks to Jan and ST for the helpful exchanges!

That is great to know!!! (No other update than replacing

HSICalibrationValue by

MSICalibrationValue?)

Last request please (as you have the failing HW): could you please make a test using following clock configuration (if possible to run it for a long time in a failing board)?

It is generated based on CubeMX and using MSI to generate system clock (@ 80MHz) and USB clock with LSE trimming.

/** System Clock Configuration
*/
void SystemClock_Config(void)
{
 RCC_OscInitTypeDef RCC_OscInitStruct;
 RCC_ClkInitTypeDef RCC_ClkInitStruct;
 RCC_PeriphCLKInitTypeDef PeriphClkInit;
 /**Initializes the CPU, AHB and APB busses clocks 
 */
 RCC_OscInitStruct.OscillatorType = RCC_OSCILLATORTYPE_LSE|RCC_OSCILLATORTYPE_MSI;
 RCC_OscInitStruct.LSEState = RCC_LSE_ON;
 RCC_OscInitStruct.MSIState = RCC_MSI_ON;
 RCC_OscInitStruct.MSICalibrationValue = 0;
 RCC_OscInitStruct.MSIClockRange = RCC_MSIRANGE_11;
 RCC_OscInitStruct.PLL.PLLState = RCC_PLL_ON;
 RCC_OscInitStruct.PLL.PLLSource = RCC_PLLSOURCE_MSI;
 RCC_OscInitStruct.PLL.PLLM = 3;
 RCC_OscInitStruct.PLL.PLLN = 10;
 RCC_OscInitStruct.PLL.PLLP = RCC_PLLP_DIV7;
 RCC_OscInitStruct.PLL.PLLQ = RCC_PLLQ_DIV2;
 RCC_OscInitStruct.PLL.PLLR = RCC_PLLR_DIV2;
 if (HAL_RCC_OscConfig(&RCC_OscInitStruct) != HAL_OK)
 {
 Error_Handler();
 }
 /**Initializes the CPU, AHB and APB busses clocks 
 */
 RCC_ClkInitStruct.ClockType = RCC_CLOCKTYPE_HCLK|RCC_CLOCKTYPE_SYSCLK
 |RCC_CLOCKTYPE_PCLK1|RCC_CLOCKTYPE_PCLK2;
 RCC_ClkInitStruct.SYSCLKSource = RCC_SYSCLKSOURCE_PLLCLK;
 RCC_ClkInitStruct.AHBCLKDivider = RCC_SYSCLK_DIV1;
 RCC_ClkInitStruct.APB1CLKDivider = RCC_HCLK_DIV1;
 RCC_ClkInitStruct.APB2CLKDivider = RCC_HCLK_DIV1;
 if (HAL_RCC_ClockConfig(&RCC_ClkInitStruct, FLASH_LATENCY_4) != HAL_OK)
 {
 Error_Handler();
 }
 PeriphClkInit.PeriphClockSelection = RCC_PERIPHCLK_USB;
 PeriphClkInit.UsbClockSelection = RCC_USBCLKSOURCE_MSI;
 if (HAL_RCCEx_PeriphCLKConfig(&PeriphClkInit) != HAL_OK)
 {
 Error_Handler();
 }
 /**Configure LSE Drive Capability 
 */
 __HAL_RCC_LSEDRIVE_CONFIG(RCC_LSEDRIVE_LOW);
 /**Configure the main internal regulator output voltage 
 */
 if (HAL_PWREx_ControlVoltageScaling(PWR_REGULATOR_VOLTAGE_SCALE1) != HAL_OK)
 {
 Error_Handler();
 }
 /**Configure the Systick interrupt time 
 */
 HAL_SYSTICK_Config(HAL_RCC_GetHCLKFreq()/1000);
 /**Configure the Systick 
 */
 HAL_SYSTICK_CLKSourceConfig(SYSTICK_CLKSOURCE_HCLK);
 /**Enable MSI Auto calibration 
 */
 HAL_RCCEx_EnableMSIPLLMode();
 /* SysTick_IRQn interrupt configuration */
 HAL_NVIC_SetPriority(SysTick_IRQn, 0, 0);
}�?�?�?�?�?�?�?�?�?�?�?�?�?�?�?�?�?�?�?�?�?�?�?�?�?�?�?�?�?�?�?�?�?�?�?�?�?�?�?�?�?�?�?�?�?�?�?�?�?�?�?�?�?�?�?�?�?�?�?�?�?�?�?�?�?�?�?�?�?�?�?�?�?�?�?

-Amel

Yep. Test results hopefully BEFORE lunch and if not - shorty after...I'm gathering up the 3 DUDS to see if it fixes them as well..

Amel -  Still good. 48.0069Mhz

And about the 3 DUD production PCB's that we had......

They Enumerate now! Same Load  ----   'JAN.DFU'

Now I have to find out where that was that I can give Jan points. :\

Thanks everyone

I saw another question I didn't answer. Summary:

The EVAL board CDC example code in STM32CubeL4 (1.6 ) can have USB clocking issues on the direct connected USB connections PA11 and PA12 when using MSI trimmed with LSE clock configuration. This can cause the device to be either ignored completely or cause DEVICE NOT RECOGNIZED responses from a USB host. This is because the 48Mhz MSI can be randomly TRIMMED way -WAY out of whack. The offending line in main.c is where it says :

RCC_OscInitStruct.HSICalibrationValue = RCC_MSICALIBRATION_DEFAULT;

rather than

RCC_OscInitStruct.MSICalibrationValue = RCC_MSICALIBRATION_DEFAULT;

So what does that bother? The first line leaves the *MSI* calibration UNTOUCHED, at some undefined value.

Why?  Well aside from it being TYPEO 'H' instead of 'M' causing the MSI value to not be set as desired here,

this struct is not initialized. Making the occurrence particularly, annoyingly RANDOM....

This is the only mod needed for this issue.

(Amel - No other update than replacing 

HSICalibrationValue by 

MSICalibrationValue!)

Thanks to Jan for spotting this and ST for helping track it down.

Thank everyone for the help so far. I am able to talk to the LCD, using The BSP example. Too bad there is no easy way to copy the example without messing up the original example. John