. > > Correct, it's the dv/dt that causes RMI. No matter if it's > positiv or negative. Also no matter of it's an traic, mosfet, IGBT > or whatever. That would be true if there is no snubber cap across the Fet or IGBT. That cap will allow a low dV/dT without causing losses in the fet. Like this the switch-off losses are zero, and as you stitch on at zero volts, the turn - on losses are also zero. Fets can switch much faster than triacs or SCRs. I will see if I can find my old circuit. > > The practical difference > > being that the triac "turns itself off" at the next z-c point, which > > keeps switching and s/w simpler > > Yes, you just have to "count" from the last z-c point ant "trig" > the triac, possibly with just a short puls. The shutoff is handled > by the triac itself at the next z-c. Note that for inductive loads > this gets a little more complicated... > > > From what I can gather, series inductors, shielding and good layout > > are fairly effective at minimising RFI > > I also thinks that you can get pretty low RMI by using suitable > shokes on the line (mains) terminals. The large part of RMI is > transitted from the mains cords attached to the switching device. > > > To keep it simple, it seems that > > something other than a triac is needed for PWM if you wish to use > > the leading z-c and turn the juice off mid-wave > > Yes, that's why people are looking at MOSFETs, IGBT's and other > devices. But note, the reason these gives less RMI are that they are > switching off slower (at mid cycle) then the triac turns on > (also at mid cycle). Slower switching also gives higher loss (heat) > in the swithing device. If you are switching the MOSFET hard (as > you normaly want to do) up to the same speed as the Triac, you stll > get the same level of RMI. There's no free lunch, right ? :-) > > Jan-Erik. > > -- > http://www.piclist.com hint: The list server can filter out subtopics > (like ads or off topics) for you. See http://www.piclist.com/#topics > > -- > http://www.piclist.com hint: The list server can filter out subtopics > (like ads or off topics) for you. See http://www.piclist.com/#topics > -- http://www.piclist.com hint: The list server can filter out subtopics (like ads or off topics) for you. See http://www.piclist.com/#topics