• State
  • Locked Locked
  • Replies 1 reply
  • Subscribers 46 subscribers
  • Views 395 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

Noise in PCB

Santosh Kumar Singh
Prodigy 180 points

Hi,

I designed a multi-layer mixed signal PCB. I separated analog and digital section planes(i.e. power supplies and grounds). Then I combined all the grounds at the power supply like as star topology. Now when I tested the PCB I got more noise than earlier ones having a common ground plane. What may be the problem???

  • 0
    Genius 3295 points

    Hello Santosh,

    Your question is hard to answer without seeing the actual layout. I will try to make a few observations based on the small amount of information provided;

    PCB related noise can be caused by many phenomena other than circulating ground currents -which is what star grounding prevents. The power trace also needs to run parallel to the "return" or ground.

    The power traces should also be separated. Optimally you have two sources -one dedicated to digital and one dedicated to analog. Even if you only have one source you need to star the power traces and keep the analog power trace separate from the digital. You also need to make sure the digital power trace runs parallel to the digital ground as close as it possible. This will couple the noise from the digital power trace to the digital ground and form a distributed PCB capacitor. The analog power trace needs to be routed the same way.

    If you run an analog power trace over a digital ground or the other way around, you will get noise coupling between the two circuits.

    You can also create noise by placing high frequency signal traces too close together. Another major source of noise is long, un-terminated traces carrying high freqency signals. If the trace inductance and capacitance is resonant to the signal frequency you can get noise gain from the RLC peaking. 

    Pulse signals that have fast rise and fall times can cause L di/dt spikes if they are sent down long, high inductance traces -this is true for even low frequency pulse signals.

    There really are no hard and fast rules for PCB design. Most of what you do is learn from past mistakes. It is a very painful experience I know. 

    If you don't have to worry about proprietary issues, you might want to provide a drawing with more details about the layout. This way it is easy to spot a problem area.

    Good luck with your design,

    Chuck