9S12GC16 EMI Problem

取消
显示结果 
显示  仅  | 搜索替代 
您的意思是: 

9S12GC16 EMI Problem

1,839 次查看
ttian
Contributor I

Hi,

I design a dashboard and do EMI test.

The EMI biggest noise points are 49.152MHz73.728MHz98.304MHz.

The bus clock of MCU is 24.567MHz. It's 2, 3, 4 times of bus clock.

 

the capacitor C19(0.1uF) is one capcitor between VDDPLL and VSSPLL.

When I removed C19, the cluster passed the EMI test.

But I found the system is not stable without C19.

 

The question is C19 could be removed during production? Is there any risk if removing C19?

I will change Colpitts oscillator to Pierce type and PE7 is connected to VDD to see if this has influcen for EMI.

Do you have any suggestions how to reduce EMI?

标签 (1)
0 项奖励
回复
2 回复数

991 次查看
Alban
Senior Contributor II
Ni Hao,
 
Decoupling is necessary to ensure the PLL has enough power when needed.
 
Improving thr oscillator layout is a better option.
What are the value of these noise peaks ?
 
The peaks are normal because are harmonics of the bus clock. But their values can be controlled with layout and other considerations (slew rate, no 90° angles on PCB...)
 
祝好,
Alban.
 
0 项奖励
回复

991 次查看
ttian
Contributor I
xiexie!(Thank you!) the problem has been solved by changing PCB layout.
0 项奖励
回复