How much RF emission ?

Discussion in 'General Electronics Chat' started by fanie, Aug 2, 2009.

  1. fanie

    Thread Starter Active Member

    Jan 20, 2007
    63
    0
    Can someone please indicate if this circuit would emit any interference RF from it's switching, if any ?

    Thanks.
     
  2. bertus

    Administrator

    Apr 5, 2008
    15,649
    2,348
  3. t06afre

    AAC Fanatic!

    May 11, 2009
    5,939
    1,222
    Since you ask do you have any problems with noise from this circuit
     
  4. fanie

    Thread Starter Active Member

    Jan 20, 2007
    63
    0
    None what so ever. I'm trying to figger out if I should have any :D

    What will generate the harmonics ?
     
  5. bearing01

    Member

    Jul 29, 2009
    14
    0
    Mathematically, a square wave is generated by a summation of an infinite number of sine waves, all with odd multiples of the fundamental frequency and with a particular scaling factor. The first harmonic, which the 3rd harmonic, is a tone that is 9dB lower than the fundamental.

    ie: if you have a 1MHz square wave, on a frequency spectrum analyzer, you will see energy at 3MHz, 5MHz, 7MHz, 9MHz, ... and the 3MHz tone will be 9dB lower than the 1MHz tone. The same can be observed by doing a DFT or FFT on a digital oscilloscope.

    Your real culprit of interference is the instantaneous current drawn by the gates when changing states from 0->1 or 1->0. You get a spike of current drawn from the battery during the transition to charge any capacitance on any node from 0V to 1V. Capacitance comes from the gate capacitance looking into any gate and also at the output of any gate. This spike of current, drawn from the battery, will have infinite harmonics all of equal magnitude. Therefore, you will see 1MHz (fundamental) plus 2MHz, 3MHz, 4MHz, ..., harmonic tones in this current and all with equal strength to the fundamental. The problem here is that if there are other circuits connected to the same battery then those circuits will see voltage ripple on the battery due to the current spike drawn from it, and also due to any inductance or resistance in the wires or connections leading from the circuit back to the battery. To alleviate this interference between blocks you want to isolate both circuits by RC filtering the power supply / battery. The objective is to have any instantaneous switching current to be drawn from a nearby capacitor connected between supply and ground. Then between this capacitor positive note to the battery you put in a resistor, maybe a 100 Ohms or so, depending on the current drawn by the circuit and the tolerable IxR drop across that resistor. The resistor forces instantaneous switching current from being drawn from the capacitor and not the battery. Between logic transitions the capacitor gets recharged slowly through that resistor by the battery. This keeps the noisy switching current of your circuit from bouncing the battery's voltage and this coupling into other circuits using the same battery. Every switching logic circuit should have their own power supply capacitor with resistor. Then the resistor of each block all connect back at the battery terminal. Circuits don't share wires leading back to the battery. This is referred to "star connecting" the power supply.

    Also, to emit RF interference into the atmosphere you need a good antenna to radiate it. Switching currents being drawn through long wires will emit radio interference. Keeping circuits small and compact while keeping wires short will help reduce this. Also try to keep instantaneous switching current out of power supply wires or lines (going back to the battery) by putting RC filtering on the power supply to keep instananeous switching current out of the long wire - to prevent radiation of it.

    When doing the supply filtering you want to use a high quality capacitor with good RF performance. Short leads to keep self inductance, of the capacitor, low as well as low resistance. An electrolytic capacitor isn't the best choice.
     
    Last edited: Aug 2, 2009
Loading...