DC on clip?

[joetama]

Ok, I did a search so hopefully I won’t get flamed for posting this up there...:takecover: Anyway here we go.

So, a few boys on another forum have been discussing if an Amp in clip throws DC or not. I know for a fact that certain old school pro amps will when they are under a MEGA load and being driven way too hard, at least this is how I have learned it. But the question is an amp under a relatively normal (if you can call it 'normal') clipping instance will throw DC or if you can consider what it outputs to be DC?

By the way, I am a strong believer in overhead so I am not a regular clipper....

-Joe

[Kevin P]

Well, I don't know if you want to call it "DC" or not, but when an amp clips, the peaks of the waveform gets lopped off, which may cause a momentary instance of DC. By no means is it continuous however. It's not so much the flat parts of the waveform that cause the damage but the sharp edges (transients) of the waveform are what fry tweeters. The flat (DC) parts could cause woofer coils to heat up but they're pretty resiliant for the most part.

[JonMarsh]

It depends on the feedback loop design. Many amplifiers nominally have unity gain at DC, by virtue of using a large series cap (typically electrolytic) in series with the shunt resistor. Audio signals are often asymmetric as regards which polarity has the highest peak voltage level, even when the average net plus and minus is balanced. When an amp which has a cap in the shunt leg of the feedback network clips on one polarity only, it "walks" the voltage across that cap, causing a charge on the cap which propagates through the loop and manifests as a DC output. Very common when overdriving so called pro amps.

Some amps are "DC coupled" in the feedback looop (for example, all the ones I've deisgned). this configuration will not develop DC at the output from clipping, and clip recovery is generally cleaner. Other factors affect clip recovery, too, including the amount of negative feedback used, and whether there are anti-saturation clamps in the base drive of the output (if it's a bipolar amps). There's a lot of ways for a designer to screw up, a lot of little details that need attention for a robust design. An amp that oversaturates the drivers will have sticking at clip- you can see it readily if you ease it into clipping on a 10 kHz sine wave. Many amps in the old days would blow up if you drove them into clipping with 20 kHz sine waves. Not mine....


~Jon