lower noise AC regulation circuit variants

[enterio]

Hi electrotechnicians, new here and new quite new to ele too..
..so bear with me, please 

bottom line: I want and advice to which type of circuit to pick to be able to pass EMC testing more easily...than the other alternative.
circuit purpose: to control resistive load power. load nominal operational voltage is 220VAC and it's 200W thing. A heater.

the two alternatives I have (or have come up with) are:
a)
diac controlled triac. Opto-diac is being controlled by microcontroler (arduino) which is sensing the AC phase, so it basically knows when will triac trurn off and it is turning the diac on to turn the triac on based on program/preference of such phase shift.
schema I found online:


b)
PWM controlled MOSFET on the low side that open-closes rapidly and controls the effective voltage (rectified) on the load that way:


please, assume some similar conditions for both circuits: the same load, the same mains, 2 layered PCB and some wires to the load...some reasonable conditions that are real enough and we can work this thought experiment out with.

hope it's enough for the starter,
thanks

[CosteC]

Group control will give you lowest noise IMHO. 200 W heater shall not notice your "long" on/off cycles.

[Roehrenonkel]

Hi,

Group control will give you lowest noise IMHO. 200 W heater shall not notice your "long" on/off cycles.

....never heard of that therm "Group-control". But i guess it's similar to the german "Impulspaket-Steuerung".
Combined with a SolidStateRelay with zero-crossing-switching one should be good for the emc-test.

Good luck

[Benta]

English: burst control.
Basically PWM at a much lower frequency than 50/60 Hz. With zero-crossing SSRs it's practrically noiseless (EMI-wise).

[CosteC]

Yes. You switch only at zero crossings -> low noise, easy for implementation. Some people will tell you resolution is poor, which is nonsense for 200 W heater - for such heater 10 ms "packets" of heat are very small.

[Marco]

Basically PWM at a much lower frequency than 50/60 Hz. With zero-crossing SSRs it's practrically noiseless (EMI-wise).

Ideally more like pulse skipping in a DC DC converter. You can approach it like PWM, but that's how you end up with all those shit induction cookers.