Good grief... the underpowering myth again?
There is so much bad information perpetuating itself that it's almost embarrassing for our industry.
1. Clipping of a signal generally does no harm to a low frequency driver provided it can handle the thermal and mechanical power.
2. A severely clipped signal will be close to 2x the power of the same signal as it begins to clip. Therefore, it is easier than you think to unintentionally OVERPOWER a speaker not realizing this fact.
3. A clipped signal damages high frequency drivers very easily because the power density to the high frequency driver shifts upwards in relation to the increased harmonics present in a clipped signal. If you have a 1kHz "square wave", there will be very large componenets at 2, 3, 4... kHz which will damage the relatively fragile HF driver.
4. Speakers for bass gtr. fail due to exceeding mechanical limitations probably 90% of the time. For HF drivers, it's aboout 90% thermal failure.
5. Power handling ratings of a speaker are grossly simplified down to a single "marketable" number. In fact, it's a series of curves that don't even come close to a single number. At very low frequencies, this real number can be 1/2 of the marketing number or less.
6. By far and away the biggest failure mode of low frequency drivers is exceeding the mechanical power handling. Just because a driver doesn't fail immediately doesn't mean you aren't doing cumulative damage to the suspension components when you exceed X-max.
7. Speaker failures can indeed damage speakers. Voice coils can short out when the insulation system becomes damaged by excessive heat or mechanical injury that results in shorted turns or a completely shorted VC.
8. A shorted HF driver presents the same shorted load to an amplifier regardless of the crossover being 1st or 2nd order. A short circuit is "zero" ohms and anythingin parallel with zero ohms is still zero ohms.
9. JBL had a tech brief for OEMs back in the 60's regardinguse of LF drivers for musical instrument amps, and for various duty cycles they had a derating curve
down from the rated power. For compressed or clipped signals, derating by 1/2 was suggested.
10. The headroom argument that's commonly presented as the reason for powering at 2x RMS or greater is pure bunk if the system is actually driven to the point of clipping because at that point it's no longer headroom because it's being used.
11. Some manufacturers of power amps are guilty of recommending WAY more power than a speaker can handle except under ideal circumstances. Why? Because selling bigger amps is more profitable, not because it's in the customer's best interests. If it were in the customer's best intrests, then why do the same manufacturers power their powered versions of the same spaker at between 1.0-1.5x RMS, even with the perfect operating conditions (adaptive limiting, high pass filtering, optimum crossover, etc.)? Because it's a classic case of do as the do, not as they say. Also, because the engineers hold a lot more power over the marketing folks when they control what's inside the box.
12. Turning down the input sensitivity controls on amps does NOT reduce the ratedpower of an amplifier. The only thing it does is to increase the drive signal (input) level required to reach rated power. The average drive signal has 20dB of headroom above what's required to reach rated power on a typical amp.
13. Gain structure should not be set so that everything clips at the same time. This is one of the current myths that is so technically wrong as to be embarrassing to the industry. If a power amp (or drive processor) has a limiter (virtually all do these days), then you MUST have drive signal headroom in order for the limiter to limit down from. Otherwise as you cross the threshold into limiting, if the drive signal clips then for all intensive purposes the limiter doesn't matter one bit. The signal is still clipped. And at a high level too. I have seen more HF driver damage due to this myth leading the uninformed user to think they were "protected". For live sound, I like to see 10-12dB of headroom between the point that rated power is reached and the drive electronics clip. I believe this myth started because some of the DSP processors were pretty noisy and that was the manufacture's "BS" workaround. It should also be noted that the nominal drive levels of these processors are around +20dB rather than the typical pro level of +26dB which makes things even worse.
Hope this helps clear up some myths.
[end of rant]
