cancel
Showing results for 
Search instead for 
Did you mean: 

STM32H753 HSE Not Generating Clock with NX2016SA crystal

SRAravind2410
Associate II

Hi Team,

     I’m currently working on a project using the NX2016SA 25 MHz crystal resonator from NDK
(Datasheet: https://www.ndk.com/images/products/catalog/c_NX2016SA_e.pdf).

All configurations have been done through STM32CubeIDE , and the setup is as follows:

Step 1: IOC Configuration

  • Enabled High-Speed External (HSE) clock.
  • Selected Crystal/Ceramic Resonator mode.
  • Enabled PH0 (RCC_OSC_IN) and PH1 (RCC_OSC_OUT) in the pinout.

Step 2: Clock Configuration

  • In the Clock Configuration tab:
  • Set HSE as the PLL input.
  • Used the “Resolve Clock Issues” option to fix initial PLL setup errors.

Step 3: Code Generation and Flash

  • Code was successfully generated and flashed to the board.
  • No compilation or linking errors.
  • The microcontroller powers on as expected.

Issue Observed

  • When probing PH0 and PH1 using a DSO (Digital Storage Oscilloscope), no clock waveform is observed.
  • The crystal does not appear to be oscillating.
  • There is no sign of activity or ringing — both lines are flat.

Step 4: Hardware (Schematic)

  • Crystal: NX2016SA-25M connected to PH0 and PH1.
  • Two 10 pF load capacitors connected from each crystal pin to GND.
  • Pads 2 and 4 of the crystal are connected to GND (as per datasheet recommendation).

Additional Notes

  • The project setup was done directly using STM32CubeIDE.
  • Target MCU: STM32H753VIT6 (100-pin LQFP).
  • VCAPs, VDD, and other power lines are confirmed stable.

 

Request

Could you please help identify potential reasons why the crystal oscillator is not functioning?

Possible areas for guidance:

  • Is any additional RCC or PWR configuration needed manually in code for STM32H7 series?
  • Does HSE require any delay or stabilization time in startup code?
  • Could the load capacitors (10 pF) be too low or too high for this crystal?
  • Should the internal bypass resistor or drive strength be configured?

Please let me know if any additional details such as full schematic, startup code, project are needed.

 

Screenshot (303).png

 

 

 

Thanks & Regards,

Aravind.SR

 

4 REPLIES 4
Mike_ST
ST Employee

Hello,

>> The microcontroller powers on as expected.

What actuallly happens? Can you debug ?

 

>> Enabled High-Speed External (HSE) clock.

Does the RCC HSE READY bit turn on ?

 

>> When probing PH0 and PH1 using a DSO (Digital Storage Oscilloscope), no clock waveform is observed.

It might be not a good idea to probe the PH0-PH1 directly as probes introduce a load on the xtal's inputs and can prevent it to to oscillate, please use the MCO outputs to check the frequency.

On the nucleo 144 STM32H753, our BOM foresee to use 5,6 pF capacitors in case the xtal is fit.

MHoll.2
Senior III

In adition to the points from Mike_ST:

Check that You PCB componente has the pin out as requested (tipical error is that the case was designed for bottom view, while the data sheet shows top view).

Hi Mike,

Thank you for your valuable inputs.

  • Regarding your question on debugging: Yes, the MCU boots and runs code correctly when the HSI clock is used. But when HSE is selected as the source, the code stalls during clock initialization (likely waiting for HSE to stabilize).

  • I’ll check the RCC_CR.HSERDY bit status during debug and confirm whether it ever sets.

  • Also noted about the DSO probe – I agree, probing PH0 and PH1 directly might load the crystal. I will try using MCO1 (PA8) to output the system clock for a safer observation.

  • Thanks for sharing that the Nucleo board uses 5.6 pF capacitors for this crystal range. I’ll try replacing the 10 pF load capacitors with 5.6–6.8 pF and check for startup.

 

Best regards,
Aravind.SR

Hi MHoll.2,

Thank you for your valuable inputs.

  • Excellent point regarding the crystal footprint. I’ve double-checked and confirmed that the crystal orientation matches the top view layout from the datasheet (pins 2 and 4 to GND, 1 and 3 connected to PH0 and PH1).

  • However, I’ll reverify the PCB footprint again just to be sure, in case the footprint was flipped in the layout unintentionally.

 

Thanks again for the guidance!

Best regards,
Aravind.SR