Definitive answer to AC interference

Discussion in 'General Electronics Chat' started by clintonb, May 7, 2015.

  1. clintonb

    Thread Starter Active Member

    Mar 6, 2011
    42
    0
    I upgraded one of my projects with a stronger motor this week. Its a 220V AC motor(first AC I've done) and the idea was to turn on and off depending on the where its needs to be.

    When I ran the program, I noticed that the display which shows its status displays random junk when the motor turns on and off.

    The mcu controls the motor by controlling a mosfet which in turns opens and closes a relay. All works great except for the junk output. I read of all the possibilities and sat on the relay for a while.

    If I remove the motor and just have the controller control the relay there is no interference. If I add the motor then interference is back. If I remove the motor and use an AC filament bulb then same issue.

    I read a few articles about AC line interference but nobody ever gets to a "its working" stage. Is there a way this can be solved or should I place the controller in a steel box to shield it.
     
  2. cmartinez

    AAC Fanatic!

    Jan 17, 2007
    3,573
    2,539
    What sort of sensors are you using? Using shielded cable is key to this kind of application.
     
  3. MaxHeadRoom

    Expert

    Jul 18, 2013
    10,548
    2,373
    cmartinez likes this.
  4. AnalogKid

    Distinguished Member

    Aug 1, 2013
    4,539
    1,251
    Since both a motor and a light bulb are inductive loads, relay contact bounce turns the inductance into a high voltage generator, and excellent source of both radiated and conducted noise. An appropriately rated capacitor across the relay contacts might squelch the noise at the source. Plus all of that grounding stuff.

    ak
     
    MaxHeadRoom and cmartinez like this.
  5. cmartinez

    AAC Fanatic!

    Jan 17, 2007
    3,573
    2,539
    Would you recommend a capacitor in series with a resistor? or just the cap? Plus, wouldn't the cap need a resistor in parallel for it to be able to discharge between cycles?
     
  6. crutschow

    Expert

    Mar 14, 2008
    13,014
    3,234
    A solid-state relay in place of the mechanical relay would likely eliminate the problem.
     
  7. cmartinez

    AAC Fanatic!

    Jan 17, 2007
    3,573
    2,539
    Yeap... that's probably the simplest solution... how much current does the motor draw?
    In fact... using an SSR would also allow you to eliminate the mosfet.
     
  8. MaxHeadRoom

    Expert

    Jul 18, 2013
    10,548
    2,373
    If the motor is not grounded even using a SSR may not cure it.
    Max.
     
  9. MikeML

    AAC Fanatic!

    Oct 2, 2009
    5,450
    1,066
    Is there now galvanic (optical, relay) isolation between the MCU and the motor?
     
  10. cmartinez

    AAC Fanatic!

    Jan 17, 2007
    3,573
    2,539
    That's a very good point... complete isolation of the MCU from the rest of the system (including the relay) would further improve things
     
  11. clintonb

    Thread Starter Active Member

    Mar 6, 2011
    42
    0
    The sensors are nothing fancy, they are just pots on the adc pins. The wires I'm using on them aren't shielded. Would an unshielded sensor effect the lcd screen? I would expect iratic reading at worst.
     
  12. crutschow

    Expert

    Mar 14, 2008
    13,014
    3,234
    He mentioned "random junk when the motor turns on and off" which would imply that it's due to relay contact bounce, and that would be eliminated by a solid-state relay.
     
  13. clintonb

    Thread Starter Active Member

    Mar 6, 2011
    42
    0
    I don't think this is the issue. If I remove the relay and just connect the motor to the power source just using a plain switch it gives the same result. The relay currently has a resistor in series and a capacitor across the pins together with a diode.
     
  14. clintonb

    Thread Starter Active Member

    Mar 6, 2011
    42
    0
    I'll give the solid state relay a shot. I have a solid state from an old machine.
     
  15. AnalogKid

    Distinguished Member

    Aug 1, 2013
    4,539
    1,251
    The cap is across the contacts where the AC load is , not across the coil where the DC drive is. There is no "discharg(ing) between cycles." Also, no series resistor; an X or Y rated cap is designed to sit across the power line continuously.

    ak
     
  16. AnalogKid

    Distinguished Member

    Aug 1, 2013
    4,539
    1,251
    That resistor, capacitor, and diode are around the coil, not the contacts. Also, mechanical switch contacts bounce just like mechanical relay contacts. Another possibility is the start capacitor centrifugal switch, if you motor has one.

    A solid state relay probably is the best solution. They create their own kind of interference, but most have circuits and components to reduce and/or snub this.

    ak
     
  17. cmartinez

    AAC Fanatic!

    Jan 17, 2007
    3,573
    2,539
    Sorry about my "cycles" quote.... I meant to say between motor "turn on - turn off" work cycle... not electric ac-cycles
     
  18. clintonb

    Thread Starter Active Member

    Mar 6, 2011
    42
    0
    I see it now, sorry I'm still a bit green. With the inductive load you get something like a boost converter. When you rapidly open and close the switch you get a higher voltage than the input. I switched out the button to a more bouncible button and removed the relay. Its not difficult to replicate the result by tapping the button.
     
  19. Lestraveled

    Well-Known Member

    May 19, 2014
    1,957
    1,215
    An filament bulb is non-inductive, yet you still get interference with it when it turns on and off.

    I don't think inductive kick back is the source of your problem. In your troubleshooting you have removed both inductive sources, your relay and motor, yet you still get interference. I would look more in the direction of AC mains. Perhaps a voltage drop or a ground loop when the motor turns on. Connect your motor or instrumentation to a different AC outlet/circuit. Separate them AC wise.
     
    cmartinez likes this.
Loading...