Which Microcontroller is the best immunity to EMI or Noise

[pak]

Hi friends,
 
   I'm planning a microcontroller project which is operating in high EMI or noisily enviroment. Anyone has experience in that area please suggest which microcontroller is the best choice?

thanks,
pak

   Happy microcontroller programming   

[zorx]

My suggestion is to use microcontrollers with integrated oscillator and as many of peripherals integrated into mcu's package (this depends on your project requirements). Also, wich mcu's have you used (8051, pic, arm, ...)?. For 8051 family my suggestion is Silicon Laboratory (www.silabs.com). Look on their site, wide choice, lots of code examples (especially if you download Silabs IDE).

Regards

[kevin8]

I used AVR ATmega16 and some MSP430 ,absolutely AVR is better

[pak]

thanks for suggestion,
     Actually, I'm looking for the microcontroller which is high immunity to AC line noise.

     I've done one project with Silab toolstick C8051F330. the project is simple ir controlled fan. The fan operates in 240VAC and driven by relay from main circuit.

     During the time of testing, I haven't connected fan with main circuit board. After finishing the hardware and firmware, I connected fan to main circuit, the big problem was come out. Microcontroller is keeping reset itself when I triggered ir signal. I've tried to change the layout of board and put AC line filter, unfortunately it did not work. 

thanks,
pak

  Happy microcontroller programming 

[bogdantk]

Hello
Most rugged is 80C31 and the worse is PIC18F family.
But it is also important to have a very good power supply
(decoupling, input filter) an a good PCB layout. I have done
3 or 4 projects for a very noise and EM contaminated environment
and I found out that the PCB is the critical Part (Need to have
large power and GND traces and a as large as possible Ground Plane
- if you can afford use 4 layer PCB with 1 GND layer). The box
is also part of the system - use metal !
But  EMC is most of all art and very little science !
Hope will help.

[otter]

Make sure that you have a pull-up on the reset line; it must not be left floating. You could also try putting a small cap (100nf) from reset to 0V. If this improves it, then that is were the noise is getting it. You may not be able to reprogramme the device with the cap on the reset (reset might be used by ISP programmer).
If the noise is coming along the DC power line, then try a  filter ( small inductor in series with the supply, and a 10uf right across the processor power pins.
There is some science in EMC, but first you have to find out where the problem is coming from. A good lump of luck is also useful.

[pelctronics]

if you use 7805 regulator put 100nf capacitor in input and 100nf on output pins to ground,that prevents oscillating of 7805,also on power pin,  much  possible closer to ic package .  That is very important to  do     
 

[edi14_10]

instead put a 100nF, it is also important to put a 220uF elco to the output of 7805 to prevent the voltage drop.

[eoasap]

The material of the capacitance matters too, Try using an OSCON capacitor instead of electrolytic.
what frequency is the clock? the bypass capacitors should be placed as close as possible to the VCC/GND
pins, and the value of the capacitor is dependent on the clock.


1) Add a ferrite bead or choke to the power supply source to help filter out the AC
2) You can't rely on the internal RC oscillator if you need a precision clock. if you must use
an external clock/crystal, use series terminating resistors to reduce harmonics on the oscillator
3) Run the traces close together as the crystal signal is like a differential pair, since the output
is 180 degrees out of phase with the input of the crystal, they have cancelling magnetic fields
making it immune to generating or receiving EMI noise.

4) Study books such as 'electromagnetics explained' (on this forum), howard johnson, eric bogatin, Lee ritchey, etc..
are all good authors to help deal with these subjects.

5) Read articles such as the following by Douglass Brooks
http://www.ultracad.com/article_bypass.htm