"spamtrap1888"
"Phil Allison"
this short article, by yours truly, sets out to debunk the many absurd
myths
surrounding sudden failures of loudspeakers - particularly woofers used in
live music and disco sound systems.
http://sound.westhost.com/articles/speaker-failure.html
The article is rather tightly written, so you may need to read it a bit at
a
time and cogitate.
Take a careful look at the links provided at the end of the article too.
--------------------------------------------------------------------------
Pretty good! The write up does raise a few issues I would like to
address:
1. Back in the day we found that the pole piece vent is useless for
cooling. Without the vent, when the woofer compresses the air in the
enclosure, the dustcap (assume solid) pushes air between the voice
coil former and the pole piece. This causes air to flow out of the
space between the magnet and the pole piece, over the wires of the
voice coil and out the gap. On the expansion stroke, air flows back
over the voice coil and then between the VC former and the pole piece.
--------------------------------------------------------------------------------
** With most speakers, there is a sealed volume behind the magnetic gap so
very little air flow goes on.
Hot air adjacent to the voice coil circulates with cooler air and this
serves to heat the magnet structure, along with conduction and radiation of
heat from the voice coil.
The main reason for having a hollow pole piece is to relieve pressure on the
cone during large excursions, but it also allows moving air to cool the pole
piece too.
---------------------------------------------------------------------------
With the vented pole piece, most of the air simply goes in and out the
vent, and not past the voice coil. The forced air movement past the
voice coil caused by the unvented pole piece should cool it better
than simple conduction (plus some convection) to the pole and to the
top plate when the pole piece is vented.
----------------------------------------------------
** JBL designed a way to force cool air over at least part of the voice coil
in operation - provided the cone is moving significantly at low frequencies
of course. Their idea, called " Vented Gap Cooling " takes advantage of the
AES 50Hz to 500Hz testing method to get around double the previous published
power ratings.
http://www.jblpro.com/catalog/suppor...pe=3&docid=297
See the final two paras on the first page.
----------------------------------------------------------------------------
2. While the point is well taken that attempting to eliminate clipping
by using higher powered amps merely increases power supplied to the
speaker, for a given amplifier, tweeters are significantly more likely
to be destroyed when the amplifier clips, because amplifier clipping
sharply increases the amount of high frequency signal, and tweeters'
small excursions produce less cooling.
--------------------------------------------------------------------------------
** Tweeters burn out for the exact same reason woofers do when the ampler
clips - cos the *average power level* has gone up !!
With normal unclipped programme, a tweeter may receive 5% of the applied
power or 5 watts out of 100. With 6 dB of clipping, that same tweeter will
now receive 20 watts, a direct result of turning up the gain by 6dB.
Others have done tests to show that the increase in high frequency energy
due to peak clipping music programme is small in comparison to the above.
------------------------------------------------------------------------------
3. Saying that the minimum impedance is purely resistive grates a bit,
because the voice coil always has some inductive reactance, however
small, and there is some intrawinding capacitance as well.
-----------------------------------------------------------------------------
** Real speakers (woofers) test purely resistive at some frequency in the
200Hz to 500Hz range and this condition corresponds with the impedance
minimum.
If the impedance minimum for a particular driver is at 250Hz, then below
that frequency the impedance is capacitive and above that frequency it is
inductive. ( There is usually about an octave range where the impedance
varies by only 10%. )
If you sweep test a driver using a dual trace scope, one channel for voltage
and one showing current, you can observe the two traces coinciding in phase
at any impedance minimum or maximum.
The minimum
impedance is just the spot where the high electrical impedance of the
mechanical resonance rolls off and before the inductive reactance
takes hold.
** Correct.
If you were to jam the voice coil tight in the cap, then all you have is the
R of the coil plus some ( lossy) inductance.
..... Phil