John Meyer wrote a paper on amplifier ratings called "Making Sense of Amplifier Power Ratings". He said that since companies like Meyer are producing self powered speakers the issue of how to compare them with amplifier ratings arises.
Peak power is a term that I have heard about. This paper clarified it for me. Meyer talks about the need to measure power ratings as the ability of the amplifier to reproduce signal at the speaker in a duration, with a particular amount of distortion. He appears to recommend that the distortion and voltage sag at peak durations be negligible. The distortion should not be significantly coloring the signal.
The way amplifiers are tested is by loading them with resistance like a speaker or another load. They then put a sine wave into the amplifier which drives the load. I gather that when the amount of tolerable distortion is reached that information is recorded. The sine wave is unique in that it is a signal with a crest factor of three decibels. This means that the heating power or average power compared to the peak power produced by a sine wave has a two to one ratio (three decibels). Meyer goes on to say that amplifiers need to make short bursts of square waves sometimes. The amount of time they need to be able to do this is a concern.
In the example John Meyer gives about their 18" subwoofer system, he says that the peak power was tested with a drum note for 40 milliseconds as a peak. When this happened the voltage sagged in the amplifier. This is not just true for amplifiers, it is true for electrical systems when demand goes up in a peak like way. This sag produces an unintended compression effect of this subwoofer system at this testing level. At lower testing levels there is not the same distortion. This leads me to the idea of conservative operation of equipment to gain desirable results. In my opinion certain amplifiers and speakers could be underrated in order to better their distortion specifications and their power ratings.
To put these ideas of how long amplifiers should sustain output into a usable rule, Meyer found a solution. The goal is to not compress the signal in the amplifier. They say that the sine wave should be able to reach its peak without significantly sagging the peak of the line voltage over half a second. This makes sense to me in that we should strive for a rating that does not include a significant sag during half a second. Hopefully the peaks that are demanded are no longer than half a second. If they were, that signal (or song) might not be listenable anyways.
Here is my way of understanding the rating concern. At the 40 millisecond rating, the amplifier says, "Yeah, I can handle that peak". When the same amplifier is asked to peak for 500ms it might say, "Okay, but I will distort, compress and sag in voltage". That is why this amplifier should be derated, so that it more accurately reflects its peak capabilities.
The power rating should reflect real use with little distortion if that is desirable.
No comments:
Post a Comment