Transient noise on AC lines

Discussion in 'Embedded Systems and Microcontrollers' started by signalflow, May 16, 2014.

  1. signalflow

    Thread Starter Member

    Mar 12, 2014
    50
    0
    Hi, I am trying to filter transient noise that is on the AC lines.

    I have a 120Vac to 24Vac transformer that goes to an LM2596 5V switching regulator which then powers my microcontroller and LCD, etc.

    I plug the transformer into the same power strip as a fan or other AC motor.

    When I switch ON or OFF the AC fan or motor, I get lots of transient noise into the system.

    I am mainly using SPI to communicate between ICs.

    I have tried TVS diodes across the 24Vac lines as well as across the 5V lines. I have tried zener diodes between 5V and GND. I have tried ferrites, inductors, LC filters. I have put 100-ohm resistors in series with some of the SPI lines. I have tried external EMI line filters. I have 220uF caps across 5V to Gnd to help smooth out any transients. These things may be helping but nothing seems to filter out the noise. I have even tried all of these things combined.

    Nothing works.

    If I put a quencharc (RC snubber) across the fan or AC motor, then that helps. But I am testing without the quencharcs because I want the system to be robust and not be susceptible to noise even without the quencharcs.

    Any advice is appreciated. I'll try anything at this point.
     
  2. MrChips

    Moderator

    Oct 2, 2009
    12,449
    3,365
    You are going in the wrong direction.

    220μF capacitors are no good for reducing HF transients.

    You need 0.1μF ceramic capacitors placed liberally around your circuit.
     
  3. signalflow

    Thread Starter Member

    Mar 12, 2014
    50
    0
    Yes, I have 0.1uF caps across every IC power pins.
     
  4. nsaspook

    AAC Fanatic!

    Aug 27, 2009
    2,912
    2,178
    First you need to isolate the source of the upset on your circuit and find out what really happens. Are all unused pins set to outputs or grounded, do you have a shielded enclosure with a input power filter?
     
  5. THE_RB

    AAC Fanatic!

    Feb 11, 2008
    5,435
    1,305
    TVS diodes won't do anything at all for negative noise spikes, they only clamp positive noise spikes.

    Are you switching relays on your 24v line?

    On your +24v DC line going to the SMPS, put a resistor that drops about 1v, followed by a big cap 2200uF 35v etc. That will totally elimimate any HF noise going into the input of the SMPS.
     
  6. signalflow

    Thread Starter Member

    Mar 12, 2014
    50
    0
    All unused pins are set to outputs and set to 0.

    I'm not using relays at this moment. I disconnected them and I am switching the AC motors/fans on and off with a manual switch so that the only wires going into my board are the 2 power lines (24Vac).

    I'm wondering if the motor transients are radiating (RF) and then coupling onto the board since no type of filtering on the 24Vac lines has worked.
     
  7. signalflow

    Thread Starter Member

    Mar 12, 2014
    50
    0
    Just figured something out. I had 3-foot long IDC cables going to my LCD and 7-segment...I put a 6-inch cable on and that dramatically improved the noise susceptibility. I still get some noise glitches, but now I know a lot of it was coming through RF.
     
  8. MrChips

    Moderator

    Oct 2, 2009
    12,449
    3,365
    3-foot long interface cable is asking for trouble.
    Glad you figured that out on your own.
     
Loading...