Originally Posted by A9X-308
At switch ON, both V and I are zero, so you want to switch at V zero cross. See the video I linked earlier.
In low current devices like a DVDP, there is so little current draw, there is never going to be an arc. In a high power amplifier, at idle the draw is still modest. For example my P7000S is 120W which is where most people are going to turn it off most of the time, so again, little risk of arcing with a suitable relay. Even then, an occasional arc is not going to destroy a relay, that takes may thousands of operations where arcing occurs.
Somewhere between 1 and 0.7 is what you are going to see in the network in the distribution phase. I am very dubious about a normal poweramp being around 0.3; I'd need to see that for myself.
Capacitors are pretty ineffective for arc suppression which is why they're not used very much in high power or high reliability switching.
You still have to switch the device OFF....and even with 1mA of current, an instantaneous change in current still approaches infinity volts. That's just how the math works out. In practice is definitely different, but my point about the math is the truth here about sparking - which is the only issue I was addressing. You won't get any spark switching with 120V and no current flowing because 120V won't conduct through the air. Sparking is entirely related to switching current - not voltage. EMI can be generated from interrupting either current or voltage. You're just choosing between a magnetic or electric field generation for the EMI wave generated.
An inductor has a power factor of zero. When an amplifier is in idle, you are driving the primary inductor of the input transformer....but since some power is being used and you're connected to a capacitor tank circuit on the output you won't see the power factor drop all the way to zero. You can doubt it all you want, but I've measured this personally - in fact, it's even part of my job to quantify the power factor of the audio equipment I design, which gets verified by compliance organizations all over the world. The power factor moves all over the place depending on the output power the system is delivering. Without PFC, 0.3 at idle is very reasonable. How many measurements have you seen personally? Keep in mind that I'm saying the PF will improve as output power is increased, and you generally don't get above 0.7 for non PFC designs.
The reason capacitors aren't used in AC systems is because the capacitor value to have an appreciable effect on the voltage spike will end up leaking AC current when the switch is off, plus the required voltage rating makes it physically very large, etc... I brought it up as a conceptual understanding of what the current is trying to do (which is keep conducting).
If you've got a bouncing relay drive, then you only need a few dozen switches before you can measure increased impedance across the relays switched nodes due to the arcing that occurs. You can even get them to latch up because they end up welding together.
To be honest, I'm not sure what problem you're trying to solve with zero-voltage switching on an AC relay. You're arguing the sparking doesn't matter.