Author Topic: radiated emissions inside power supply effecting microcontroller?  (Read 2394 times)

0 Members and 1 Guest are viewing this topic.

Offline SArepairmanTopic starter

  • Frequent Contributor
  • **
  • Posts: 885
  • Country: 00
  • wannabee bit hunter
radiated emissions inside power supply effecting microcontroller?
« on: September 27, 2014, 05:20:11 am »
Can I put a MCU on the same board as a switchmode power supply? It is galvanically/optically isolated on its own ground plane. The MCU is 3 inches away from any mains connections, which is switching 5A@110V@10KHz. The gap between the two ground planes is ~0.3 inches.

Should I put the option for a shield can around the mains SMPSU? There is no analog stuff that is exposed to anything, just a MCU/RS232 interface.

Could a shield like this possibly be effective?

*star washer under screw
« Last Edit: September 27, 2014, 05:34:45 am by SArepairman »
 

Offline Rerouter

  • Super Contributor
  • ***
  • Posts: 4694
  • Country: au
  • Question Everything... Except This Statement
Re: radiated emissions inside power supply effecting microcontroller?
« Reply #1 on: September 27, 2014, 06:47:13 am »
there should only be 2 cases where a micro can be upset inside a noisy enviroment, the first is noise coupling through long or high impedance traces, the second is its supply rails / reset line, follow good decoupling practices on the supply rails, and add a pull up or pull down resistance on every pin that is not being driven hard, to ensure that both noise cant cause a glitch and secondly positive or negative going spikes have somewhere to go if coupled that strongly,
 

Offline SArepairmanTopic starter

  • Frequent Contributor
  • **
  • Posts: 885
  • Country: 00
  • wannabee bit hunter
Re: radiated emissions inside power supply effecting microcontroller?
« Reply #2 on: September 27, 2014, 06:55:02 pm »
If there is a long trace between two devices (such as a RS232 chip) should I distribute the pull down resistor between the two chips b/c of the trace inductance (5 inches )?

Would the shield that I described get rid of a need for the extra pulldown?
 

Offline T3sl4co1l

  • Super Contributor
  • ***
  • Posts: 21681
  • Country: us
  • Expert, Analog Electronics, PCB Layout, EMC
    • Seven Transistor Labs
Re: radiated emissions inside power supply effecting microcontroller?
« Reply #3 on: September 27, 2014, 11:21:30 pm »
I don't see how there'll be enough noise to cause problems at pretty much any distance... it's just logic...

Tim
Seven Transistor Labs, LLC
Electronic design, from concept to prototype.
Bringing a project to life?  Send me a message!
 

Offline T3sl4co1l

  • Super Contributor
  • ***
  • Posts: 21681
  • Country: us
  • Expert, Analog Electronics, PCB Layout, EMC
    • Seven Transistor Labs
Re: radiated emissions inside power supply effecting microcontroller?
« Reply #4 on: September 27, 2014, 11:22:31 pm »
If there is a long trace between two devices (such as a RS232 chip) should I distribute the pull down resistor between the two chips b/c of the trace inductance (5 inches )?

Would the shield that I described get rid of a need for the extra pulldown?

Pulldowns have very little effect on signal quality.  You need resistors on the order of Z_0 to do that.  Typical trace impedances are around 100 ohms.

Tim
Seven Transistor Labs, LLC
Electronic design, from concept to prototype.
Bringing a project to life?  Send me a message!
 

Offline poorchava

  • Super Contributor
  • ***
  • Posts: 1672
  • Country: pl
  • Troll Cave Electronics!
Re: radiated emissions inside power supply effecting microcontroller?
« Reply #5 on: September 29, 2014, 07:36:39 pm »
10kHz seems way low for a mains psu. It's gonna emit audible noise. Why don't you move the frequency to somewhere around 30kHz?
I love the smell of FR4 in the morning!
 

Offline SArepairmanTopic starter

  • Frequent Contributor
  • **
  • Posts: 885
  • Country: 00
  • wannabee bit hunter
Re: radiated emissions inside power supply effecting microcontroller?
« Reply #6 on: September 30, 2014, 02:41:40 am »
10kHz seems way low for a mains psu. It's gonna emit audible noise. Why don't you move the frequency to somewhere around 30kHz?

as the pic programs it I can put a time delay on it, so it fools the customer into thinking that it is quiet and then wham bam 5KHz switching :-DD

i wonder if its possible to play music with a switcher, or even speech...
« Last Edit: September 30, 2014, 02:44:09 am by SArepairman »
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf