Re: Selling the Ampeg. What will I get now?


"RichL" <rpleavitt@xxxxxxxxx> wrote in message
news:1s6dnXyOMso5nD_UnZ2dnUVZ_hqWnZ2d@xxxxxxxxxxxxxxxx
JimmyM <mmm@xxxxxxx> wrote:
On Mon, 23 Feb 2009 00:19:57 GMT, Brian Running <brunning@xxxxxxx>
wrote:

JimmyM wrote:

Sorry dude, but Mike's right.

Speakers blow because of overpowering. Period.

He's not right, because he said that clipping is not dangerous to LF
drivers. That's wrong.

You are right when you say that the only thing that causes a voice
coil wire to melt and the speaker to melt is more current than it
can handle. That's true. But that begs the question, "How much is
too much?"

Speaker designers rate their speakers' power-handling assuming that
the input signal will be clean and unclipped. You find me any
manufacturer that will give a power-handling rating for a clipped,
square-wave input. Good luck. They always rate them assuming a
clean, unclipped signal.

Any wire has a limit to the amount of current it can handle before
melting. If the current is DC, then as current flows through the
wire, it will heat up. The lower the current level, the longer it
takes for the wire to heat up. There's a certain level of current
that a wire can handle all day long without heating. But take that
same wire and increase the current, it will heat up. Increase the
current high enough, and it melts. Now assume that the current is
coming in pulses. Wire takes time to heat up, it's not
instantaneous. If the current is pulsed, then there's a gap between
doses of current. That means there's time before the wire heats up,
and time for it to cool back down. That means that the pulses can
go to higher current levels without melting the wire than
steady-state current. Eventually, though, you'd reach a current
level so high that even a super-fast pulse would melt the wire.

Music signals cruise along at low current levels for the most part,
with intermittent peaks of high current. Those intermittent peaks
won't melt the voice coil because there's not enough time for the
coil to heat much, and there's time in between peaks for the wire to
cool down. Speaker designers assume that music is going to be run
through the speakers, not direct current, that's why they use RMS
ratings. "Continuous RMS" means power handling with continuous AC,
not continuous DC. If speaker manufacturers rated their speakers
for DC current, the rating would be way lower than the "continuous
RMS" rating, because there's no time lag for heating the voice coil
wire, or time breaks for cooling.

Square waves' RMS power is equal to their peak power. A square wave
is at constant peak power, just like DC. With a true square wave,
there's no pause in the current flow. So, if you're feeding a
square-wave signal into a speaker, the continuous RMS rating is
meaningless, the speaker's capability is way lower than with a music
signal.

So, yes, you're right, the only way to burn a voice coil is
overcurrent. But you're not right that square waves are not more
dangerous to speakers than music signals. Speakers are not designed
to handle square waves, or constant current. In effect, what
feeding square waves into your speaker cab does is drastically lower
their power-handling capacity. So, when Rieves says that square
waves or a clipped signal are not dangerous to LF drivers, he's
wrong.

I'm a little surprised, Jimmy, that you haven't run across this in
your years of experience. Melting voice coils with clipped signals
is not all that uncommon.

Dude, I'll be honest with you...my knowledge of such things could fit
in a thimble, and it's only been in the last year that I've taken any
interest in it. So being that I'm engineering-challenged and you're
not, I have to dumb down my argument. So in that case, I would use
the synth argument. I've yet to see a synth playing a square wave
blow a cab simply because it was a square wave. It still has to reach
a volume that causes a cab to blow, and as long as synths have a
volume control, I don't see that it's a problem in and of itself.

The problem really isn't that it's a square wave, the problem is that
you're overdriving amp circuits and producing a square wave as a result.

Synth outputs are well controlled; bass outputs aren't. As you pluck
harder and harder, the output signal sent to an amp gets larger and
larger until it begins to clip. At this point you've got a sine wave
(with harmonics added, of course) with its tops clipped off. You'd
think the average power would be saturated at this point but it isn't.
As you continue to drive the amp harder, you clip more of the top off
and the waveform shape begins to approach a square wave. And the
average power has changed by a factor of two in going from the case
where you're barely clipping to that in which you have a square wave.
It's the most extreme signal possible under those circumstances.

Also, I don't know if this is what you were saying in this
explanation, but if you're saying that a square wave will cause your
amp to send DC to the voice coils, I would argue that your amp is
broken and needs to be taken to a repair shop immediately because an
amp should never send DC to your speakers under any circumstances.

What he's saying is that the *magnitude* of the voltage is constant
throughout the duration of the square wave. It's still changing sign
every half period, so it's not DC.

Again, as I am very challenged in this area (but learning), I can only
take the word of the various speaker/amp designers I have access to,
and not a single one will say that a square wave will blow a cab
unless said square wave puts out too much power for the cab to handle,
and DC is never put out to the speaker under any circumstances unless
the amp is broken. But a clip that is too much for a woofer to handle
will definitely blow it, no doubt about it. So if Mike said that
clipping won't blow woofers under any circumstances, he's definitely
wrong.

There's also the issue, discussed in other posts, that a large-amplitude
square-wave signal will attempt to force the voice coil to undergo
abrupt excursions from "in" to "out" whereas an unclipped signal will do
this more gradually and gently over the course of a single period of the
signal.

As I said before, one needs to be aware that an amp driven into heavy
clipping can produce twice its RMS rated power. As long as the user is aware
of that and takes it into account, there's no problem. One can take a 100
Wrms amp and drive it into heavy continuous clipping into a speaker that is
rated at 300 watts continuous/600 watts peak (as many rock and metal
guitarists routinely do), and the speaker isn't going to be damaged, because
the clipped signal is still going to be well below the speaker's rated
power. If one is using a 300 Wrms amp and driving it into heavy continuous
clipping, there is a very real chance that the speaker will be damaged
because the speaker will be getting up around 600 watts continuous, well
beyond its design rating. However, from a practical standpoint, the
distortion produced is going to be so intense and so ugly that no one but an
idiot is likely to run it that way for long enough to damage the speaker.
The same speaker driven with a 300 watt amp that is occasionally driven
into clipping will work fine because the 600 watt peak rating of the speaker
isn't being exceeded so no mechanical damage is likely to occur, and the
peaks aren't of sufficient duration to cause overheating. In other words, no
one with any common sense is likely to damage a speaker by using an amp
rated at the same power as the speakers continuous power rating.




.

Relevant Pages

  • Re: Selling the Ampeg. What will I get now?
    ... Any wire has a limit to the amount of current it can handle before ... Music signals cruise along at low current levels for the most part, ... because there's no time lag for heating the voice coil ... With a true square wave, ...
    (alt.guitar.bass)
  • Re: Hang-Time??? Ping: Rich Koerner
    ... Rich is exactly right when he says that heavily overdriving an amp ... Now, as long the the switch is left in that position, the speaker cone, ... what instrument produces a 2 Hz square wave? ... The bass, in the hands of Bootsy Collins. ...
    (alt.guitar.bass)
  • Re: New Compact Monoblocks
    ... Built on a Hammond steel 10x6x2 chassis. ... the output signal with a 5kHz square wave at input ... not make the amp oscillate at HF ... even with an "8 ohm" speaker. ...
    (rec.audio.tubes)
  • Re: New Compact Monoblocks
    ... the output signal with a 5kHz square wave at input ... with an R load. ... not make the amp oscillate at HF ...
    (rec.audio.tubes)
  • Re: Kikusui ORC-27A tubed oscilator mods.
    ... 12AT7 for square wave, able to be switched in, and a final 6AQ5 CF ... NFB loop. ... it has 8Vrms across it and by calculating voltage and currents in the ... In music, bass signals ...
    (rec.audio.tubes)