STM32F303RET Clock Issue

[bradbas]

I have a design that I have used in the past but a latest build I am having trouble with the system clock. I am using 8mhz crystal, IOC file HSE is setup to use the crystal. I cannot get the processor to run above 16 Mhz. I have older projects that use the same processor and no issue with running at max 72Mhz. I have this problem with 1.12 and 1.19 so I don't believe it's CubeIDE. Is there any known issues with these chip? 

[AScha.3]

>Is there any known issues with these chip? 

No, afaik.

+

show your x.ioc , Clocktree in Cube ;

like this:

 

[bradbas]

 

[AScha.3]

And if you set it to its typ. 72 M , as in my clocktree , then ..what ?

+

>Keep in mind this identical configuration in the past year has worked at 72Mhz 

On same board ? or just similar design?

(because from your schematic, no decoupling caps etc. visible .)

[TDK]

What symptoms are you seeing? Show your schematic.

[bradbas]

Here is the processor side of the project. Keep in mind this identical configuration in the past year has worked at 72Mhz no problem.

[bradbas]

Update,

Figured it out. My circuit was the same as past but not the power supply. Turns out my 3V3 supply was actually running at 2.0 and once the processor loading got up due to increasing clock speed it dropped below 2.0 which would then make the core unstable. Moral of the story always check your power rails. Thanks everyone.

 

[TDK]

Target voltage of 2.0 V should have been a smoking gun in STM32CubeProgrammer.

[bradbas]

That's where it get very interesting. My power supply had a feedback problem on 3V3 and when 3.3V rail was not loaded much it was running 3.1 or so but when it got loaded it dropped. When I used the IDE Cube Programmer it did indicate 3.1V which I knew should work and it programmed the device no issues. Just when it started running loading down the 3V3 rail did it become unstable. I found it by monitoring it with a scope and it became very clear what the issue was.