DC motor EMI problem

Discussion in 'The Projects Forum' started by ShaunManners, Jan 25, 2009.

  1. ShaunManners

    Thread Starter Active Member

    Feb 16, 2008
    Hi All,

    I have a problem I was hoping to discuss with you...

    I have a Velleman USB interface board which has a PWM output - I am using it to send a signal to a MOSFET/IGBT driver (optically isolated) which supplies the 15v to the IGBT gate...

    It all works perfectly... the only problem is that when the motor is running, the analogue inputs from my USB interface appear to be affected... they are represented on my computer by a range of values from 0-255 and they are randomly jumping around...

    I have tried a pull down resistor to the analogue input... but that makes no difference.
    I have tried the whole circuit with a dummy load instead of the motor... and there is no interference. so it has to be the motor.

    I have a diode across the motor... the motor is running on a 24v supply, and the gate of the IGBT is driven from a 15v regulator from the same 24v supply. the other side of the optically isolated driver runs off a 12v battery... the grounds are not connected.

    I'm thinking I need some sort of suppression on the motor.. but have no idea how to go about this...

    Do you have any suggestions?

    Thanks in advance.
  2. Skeebopstop

    Active Member

    Jan 9, 2009
    DC motors are particularly noisy given the brushing if you are using a simpled brushed DC motor.

    How far away is the USB circuit from the motor/switching circuit? Give them some distance. Why would the USB circuit run off a battery? Why not just use the 500mA of the PC and make sure the pc and motor drive circuits both reference the same ground.

    You may want to tie the grounds together at a common point with some filtering in between (i.e. as alberto mentioned) if the USB power option via PC isn't an option.

    May I ask if your analog input actually has a driving signal? I know you said you had a pull down resistor on there but generally they are intended to be quite weak, much weaker than the output impedance of a driving circuit delivering your analog voltage. Stick some chokes and filters on your analog signals to keep high frequency switching harmonics out (if possible), but never assume your analog inputs are accurate unless they are being driven.

    Do you have any resistance on your IGBTs to keep them from switching too fast and also potentially ringing? This can also be a great source of electrical noise.