What is the reason why ESP-32 can still work after the capacitor of the crystal oscillator is shorted by the ESP-32 hard

[li_jun]

In normal operation, the software control chip outputs a high and low level signal with a frequency of 100HZ at GPIO5. When we short-circuit the load capacitance of the 40MHz crystal oscillator, the output signal of GPIO5 changes to a signal of 10HZ-20HZ, and the CPU seems to be working at reduced frequency. What is the reason? Why can the CPU still work?

[ESP_Sprite]

Yeah, don't do that, it runs the ESP out of specs and we can't guarantee anything wrt how it works in that case. For an explanation, I'm just guessing here, but the CPU runs off the internal PLL. While that PLL it uses the Xtal as a reference, it has its own VCO that gets tuned to run in sync with the xtal frequency. Could be that shorting the xtal makes the VCO still run, but not very stably and at the lowest frequency it happens to be able to reach, and that's what you see.

[li_jun]

Thank you very much for your reply
I still have some questions to ask you：
1.What is the relationship between PLL_CLK and XTAL_CLK? No description of this part can be found in the SYSTEM_CLOCK section of the manua.

[ESP_Sprite]

PLL_CLOCK is the clock that is generated by the PLL, as described above. It's 320 or 480MHz, depending on, well, mostly what CPU clock speed you select (80/160 or 240MHz)