Author Topic: How do I suppress EMI from fan?  (Read 1306 times)

0 Members and 1 Guest are viewing this topic.

Offline The Magic RabbitTopic starter

  • Regular Contributor
  • *
  • Posts: 108
How do I suppress EMI from fan?
« on: October 06, 2015, 04:22:34 pm »
I've got a PIC at 48MHz processing data and using the PWM module to drive two fans via individual IRL520 FETs. The PWM module is running around 44kHz, with duty cycles between 20% and 100% I have a signal diode across the fan to protect the FET.

When the duty cycle is below 100% the data on the I2C lines seems sporadic, so I'm convinced there must be some noise from the fan switching leaking back into the system.

I have 0.33uF caps between Vcc and Gnd on the PIC, and also between +5V and Gnd on the I2C lines that feed the LCD and sensors. I have 470uF cap on the 12V input and also between 5V and Gnd after the 5V regulator.

I'm sure I need to do something across the fans or around the FET to block electrical noise from the fans, but I don't know what. Sadly I don't have access to a scope to look at the signals.

Can anyone advise?

Sent from my XT1039 using Tapatalk

 

Offline marshallh

  • Supporter
  • ****
  • Posts: 1462
  • Country: us
    • retroactive
Re: How do I suppress EMI from fan?
« Reply #1 on: October 06, 2015, 04:32:50 pm »
Sounds like something similar I had where a buck regulator was creating noise upstream. The solution was to put a 100uh coil in series with the supply of the smps (in your case, the motor that gets PWMd) as well as a diode to suppress back-emf.
Verilog tips
BGA soldering intro

11:37 <@ktemkin> c4757p: marshall has transcended communications media
11:37 <@ktemkin> He speaks protocols directly.
 

Offline tggzzz

  • Super Contributor
  • ***
  • Posts: 19468
  • Country: gb
  • Numbers, not adjectives
    • Having fun doing more, with less
Re: How do I suppress EMI from fan?
« Reply #2 on: October 06, 2015, 05:04:40 pm »
You will need to understand where the currents flow, and that cannot be done by looking at the schematic: you have to look at the actual construction.

Remember there is no such thing as a wire, but there are long thin inductors and resistors which cause voltage drops when current flows through them. (Rule of thumb: 1nH/mm, and realise V = L di/dt and V = I R)

Maybe threading the leads multiple times through a ferrite toroid might help - but it depends on the construction.
« Last Edit: October 06, 2015, 05:08:41 pm by tggzzz »
There are lies, damned lies, statistics - and ADC/DAC specs.
Glider pilot's aphorism: "there is no substitute for span". Retort: "There is a substitute: skill+imagination. But you can buy span".
Having fun doing more, with less
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf