Sign in with
Sign up | Sign in
Your question

Tweeter failure in speaker

Last response: in Home Audio
Share
Anonymous
September 19, 2004 12:52:42 AM

Archived from groups: rec.audio.pro,nz.tech (More info?)

We have a wedge speaker (stage foldback) which has twice had the tweeter
fail. It is a two way speaker with passive crossover.

The failure has been detected as distortion from the tweeter.

Since it happened only in that speaker out of four identical ones coupled
in two parallel pairs to the same amp, is there any possible crossover
fault that could cause such a failure to occur? It has been connected in
parallel with another speaker of the same type on the same channel of the
amp as was the case previously, but there have been no problems with the
other speaker.
Anonymous
September 19, 2004 12:52:43 AM

Archived from groups: rec.audio.pro,nz.tech (More info?)

"Patrick Dunford" <patrickdunford@nomail.invalid> wrote in message
news:MPG.1bb6be468fe5a83298a475@news.paradise.net.nz

> We have a wedge speaker (stage foldback) which has twice had the
> tweeter fail. It is a two way speaker with passive crossover.

> The failure has been detected as distortion from the tweeter.

> Since it happened only in that speaker out of four identical ones
> coupled in two parallel pairs to the same amp, is there any possible
> crossover fault that could cause such a failure to occur?

Possible.

> It has been
> connected in parallel with another speaker of the same type on the
> same channel of the amp as was the case previously, but there have
> been no problems with the other speaker.

Could be a shorted cap or an open parallel coil, or it could just be a
series cap that is at the high end of the possible +80% tolerance range.
Anonymous
September 19, 2004 12:52:43 AM

Archived from groups: rec.audio.pro,nz.tech (More info?)

Patrick Dunford <patrickdunford@nomail.invalid> wrote:
>We have a wedge speaker (stage foldback) which has twice had the tweeter
>fail. It is a two way speaker with passive crossover.
>
>The failure has been detected as distortion from the tweeter.
>
>Since it happened only in that speaker out of four identical ones coupled
>in two parallel pairs to the same amp, is there any possible crossover
>fault that could cause such a failure to occur? It has been connected in
>parallel with another speaker of the same type on the same channel of the
>amp as was the case previously, but there have been no problems with the
>other speaker.

It's possible. If the crossover is sending lower frequency stuff to the
tweeter, it will damage it. The most common crossover failures, though,
are capacitor failures that increase the lowest frequency going to the
tweeter rather than reduce it.

Try running a 500 Hz signal into both speakers, and measure the voltage
across the tweeter. You won't even need a true RMS meter since all you
want to do is compare the two. Are they close, or is one a whole lot
higher than the other? (You do need to have two good tweeters to do this.)

I think it's more likely that you're clipping your amp and that only chance
has kept the other tweeter from blowing, but it might be a bad crossover
and it wouldn't hurt to check.
--scott

--
"C'est un Nagra. C'est suisse, et tres, tres precis."
Related resources
Can't find your answer ? Ask !
Anonymous
September 19, 2004 12:52:43 AM

Archived from groups: rec.audio.pro,nz.tech (More info?)

"Patrick Dunford" <patrickdunford@nomail.invalid> wrote in message
news:MPG.1bb6be468fe5a83298a475@news.paradise.net.nz...
> We have a wedge speaker (stage foldback) which has twice had the tweeter
> fail. It is a two way speaker with passive crossover.
>
> The failure has been detected as distortion from the tweeter.
>
> Since it happened only in that speaker out of four identical ones coupled
> in two parallel pairs to the same amp, is there any possible crossover
> fault that could cause such a failure to occur? It has been connected in
> parallel with another speaker of the same type on the same channel of the
> amp as was the case previously, but there have been no problems with the
> other speaker.
>

Patrick,
Hook an audio oscillator up to the speaker terminals, and put an AC
voltmeter or scope across the tweeter. Do this for both speakers, and
compare the voltages.
Anonymous
September 19, 2004 12:52:44 AM

Archived from groups: rec.audio.pro,nz.tech (More info?)

Another possibility: does the crossover include some sort of device to
protect the tweeter from excessive levels? Such devices include relays and
light bulbs. If so, the one in the speaker that's blown may not be
functioning anymore, so the excess signal comes through and ZAP. (Possible
in the case of a relay, almost impossible with a light bulb.)

If there isn't such a device in the crossover, though, I'd look for a
shorted or out-of-spec cap.

Peace,
Paul
Anonymous
September 19, 2004 4:47:55 PM

Archived from groups: rec.audio.pro,nz.tech (More info?)

"Patrick Dunford"
>
> We have a wedge speaker (stage foldback) which has twice had the tweeter
> fail. It is a two way speaker with passive crossover.
>
> The failure has been detected as distortion from the tweeter.
>
> Since it happened only in that speaker out of four identical ones coupled
> in two parallel pairs to the same amp, is there any possible crossover
> fault that could cause such a failure to occur? It has been connected in
> parallel with another speaker of the same type on the same channel of the
> amp as was the case previously, but there have been no problems with the
> other speaker.
>

** Seen this sort of thing a few times - ie repeat, mysterious failures of
tweeters or horn driver diaphragms.

Case 1. EV horn in 2 way PA box fails - owner takes it to local EV
agent to get fixed. They replace the burnt diaphragm and charge a hefty fee.
The new diaphragm fails on the very next use. The EV agent wants to charge
again - claims abuse is the cause. Owner, knowing this is not true, brings
the PA box to me. The passive X-over looks fine, all caps and inductors
test OK but something is weird about the measured response - it has a 10
dB peak in the output at 1.4 kHz ?? I notice that the wood surface next
to a 33 ohm, 10 watt resistor is scorched - this resistor is in parallel
with an inductor so has to be disconnected to do a test - it tests OPEN !!
So I replace the resistor and the response is now as expected. But how did
it ever burn out like that ??

Running a 1.4 kHz sine wave into the x-over with NO load connected provides
the answer - when driven at even moderate power levels at or near 1.5 kHz,
the 33 ohm resistor starts to SMOKE !! With no load the x-over becomes a
series resonant circuit that *magnifies* the drive voltage by several times
at 1.5 kHz so that 33 ohm resistor cops about 10 times more power than it
ever would normally.

The only way to prevent a recurrence is to up rate that resistor to one
with a 100 watt power rating. This upgrade is done to the second box as
well after I informed the owner.


Case 2. Emilar horn diaphragm fails after a decade of use - the owner
fits a new one and it fails on the very next use. Examination shoes no
visible damage to the diaphragm. The same thing happens with a second
replacement diaphragm - again not the slightest visible sign of damage but
the coil is open circuit. Accusations and wild theories abound as to the
cause. The local dealer, after contacting Emilar, is eventually forced
admit he has been supplied a batch of faulty diaphragms.



.............. Phil
Anonymous
September 19, 2004 7:54:48 PM

Archived from groups: rec.audio.pro,nz.tech (More info?)

In article <cih7g9$k5q$1@panix2.panix.com> in nz.tech on 18 Sep 2004
07:51:37 -0400, Scott Dorsey <kludge@panix.com> says...
> Patrick Dunford <patrickdunford@nomail.invalid> wrote:
> >We have a wedge speaker (stage foldback) which has twice had the tweeter
> >fail. It is a two way speaker with passive crossover.
> >
> >The failure has been detected as distortion from the tweeter.
> >
> >Since it happened only in that speaker out of four identical ones coupled
> >in two parallel pairs to the same amp, is there any possible crossover
> >fault that could cause such a failure to occur? It has been connected in
> >parallel with another speaker of the same type on the same channel of the
> >amp as was the case previously, but there have been no problems with the
> >other speaker.
>
> It's possible. If the crossover is sending lower frequency stuff to the
> tweeter, it will damage it. The most common crossover failures, though,
> are capacitor failures that increase the lowest frequency going to the
> tweeter rather than reduce it.
>
> Try running a 500 Hz signal into both speakers, and measure the voltage
> across the tweeter. You won't even need a true RMS meter since all you
> want to do is compare the two. Are they close, or is one a whole lot
> higher than the other? (You do need to have two good tweeters to do this.)
>
> I think it's more likely that you're clipping your amp and that only chance
> has kept the other tweeter from blowing, but it might be a bad crossover
> and it wouldn't hurt to check.

Of course, now that I think about it, the tweeter would have a lower
power rating than the woofer. How does the crossover reduce the power
level in the tweeter?
Anonymous
September 19, 2004 7:54:49 PM

Archived from groups: rec.audio.pro,nz.tech (More info?)

"Patrick Dunford"

> Of course, now that I think about it, the tweeter would have a lower
> power rating than the woofer. How does the crossover reduce the power
> level in the tweeter?


** Errrr - by filtering out all the low frequencies ?




............ Phil
Anonymous
September 19, 2004 7:54:50 PM

Archived from groups: rec.audio.pro,nz.tech (More info?)

Phil Allison wrote:

> "Patrick Dunford"
>
> > Of course, now that I think about it, the tweeter would have a lower
> > power rating than the woofer. How does the crossover reduce the power
> > level in the tweeter?
>
> ** Errrr - by filtering out all the low frequencies ?

Also, the HF unit is typically more sensitive ( dB/W ) than the LF driver,
so attenuation of the input is required too, which lowers the input power
to the HF unit in addition to the filter.


Graham
Anonymous
September 19, 2004 7:54:51 PM

Archived from groups: rec.audio.pro,nz.tech (More info?)

Pooh Bear wrote:
>
> Phil Allison wrote:

>> ** Errrr - by filtering out all the low frequencies ?
>
>
> Also, the HF unit is typically more sensitive ( dB/W ) than the LF driver,
> so attenuation of the input is required too, which lowers the input power
> to the HF unit in addition to the filter.

Also, if the tweeter were fed the same power level (assuming it could
handle it), wouldn't the tweeters be blowing your ears out?
Anonymous
September 19, 2004 8:28:39 PM

Archived from groups: rec.audio.pro (More info?)

In article <2r4aaqF163s7cU1@uni-berlin.de> philallison@tpg.com.au writes:

> [the crossover network] - it has a 10
> dB peak in the output at 1.4 kHz ?? I notice that the wood surface next
> to a 33 ohm, 10 watt resistor is scorched - this resistor is in parallel
> with an inductor so has to be disconnected to do a test - it tests OPEN !!
> So I replace the resistor and the response is now as expected. But how did
> it ever burn out like that ??
>
> Running a 1.4 kHz sine wave into the x-over with NO load connected provides
> the answer - when driven at even moderate power levels at or near 1.5 kHz,
> the 33 ohm resistor starts to SMOKE !! With no load the x-over becomes a
> series resonant circuit that *magnifies* the drive voltage by several times
> at 1.5 kHz so that 33 ohm resistor cops about 10 times more power than it
> ever would normally.

It would help to have a schematic for reference. The first question
that came to mind was "how can you have a circuit with the load
disconnected, but then I thought about variations in crossovers. I've
seen crossover designs which have a series L-C circuit with the
tweeter in parallel with the inductor to provide a sharper cutoff for
the high-pass end. Is this how it was constructed? If so, I can see
that a resistor in parallel with the inductor could tame a peak.

What did the peak look like with the speaker connected and the
resistor disconnected? I suspect that the inductance of the voice coil
may have had something to contribute to the resonant frequency and
even the Q.

But the real question is, why did the tweeter blow in the first place?
With the tweeter load, is the originally designed power rating
adequate for the power rating of the tweeter? If so, then there was
another reason for the initial problem which changing the resistor
wouldn't prevent - most likely a defect with the tweeter or just too
much power sent to the speaker. Replacing the resistor with one of a
higher wattage only protects a replacement tweeter from being damaged
from a resistor failure resulting from a tweeter failure. This is
probably a good thing, but not a complete solution to the customer's
problem.

> Case 2. Emilar horn diaphragm fails after a decade of use - the owner
> fits a new one and it fails on the very next use. Examination shoes no
> visible damage to the diaphragm. The same thing happens with a second
> replacement diaphragm - again not the slightest visible sign of damage but
> the coil is open circuit. Accusations and wild theories abound as to the
> cause. The local dealer, after contacting Emilar, is eventually forced
> admit he has been supplied a batch of faulty diaphragms.

A much more satisfying explanation.

--
I'm really Mike Rivers (mrivers@d-and-d.com)
However, until the spam goes away or Hell freezes over,
lots of IP addresses are blocked from this system. If
you e-mail me and it bounces, use your secret decoder ring
and reach me here: double-m-eleven-double-zero at yahoo
September 19, 2004 11:02:38 PM

Archived from groups: rec.audio.pro,nz.tech (More info?)

Patrick Dunford wrote:

> Of course, now that I think about it, the tweeter would have a lower
> power rating than the woofer. How does the crossover reduce the power
> level in the tweeter?

They dont, its relied on that you have semi-sensable source material with a
normal frequancy/voltage curve that tapers off - a loud 10kHz tone will destroy
almost any speaker.
Anonymous
September 19, 2004 11:02:39 PM

Archived from groups: rec.audio.pro,nz.tech (More info?)

"richard" <rich@ihug.co.nz>
> Patrick Dunford wrote:
>
> > Of course, now that I think about it, the tweeter would have a lower
> > power rating than the woofer. How does the crossover reduce the power
> > level in the tweeter?
>
> They dont, its relied on that you have semi-sensable source material with
a
> normal frequancy/voltage curve that tapers off - a loud 10kHz tone will
destroy
> almost any speaker.


** Ummm - a loud 10 kHz tone may destroy an unprotected tweeter but not a
mid/bass driver.

Tweeters can easily be protected from over current abuse by correctly chosen
fuses, low voltage lamps or a device called a polyswitch. These must be
fitted in *series* with any passive crossover network being used with the
tweeter.




.............. Phil
September 20, 2004 12:31:14 AM

Archived from groups: rec.audio.pro,nz.tech (More info?)

Phil Allison wrote:

> ** Ummm - a loud 10 kHz tone may destroy an unprotected tweeter but not a
> mid/bass driver.

The OP had tweeters blowing

> Tweeters can easily be protected from over current abuse by correctly chosen
> fuses, low voltage lamps or a device called a polyswitch. These must be
> fitted in *series* with any passive crossover network being used with the
> tweeter.

Common in good PA gear where a nasty piece of feedback may go thru, but not so
common in the cheap shitty gear and almost never seen in consumer level gear.

Also, even when protected, a sudden burst will often destroy piezo tweeters if
loud enough. I killed about 4 of them on an old mosfet amp that I had that ended
up to have an issue where it put out about 28kHz at a huge level when the volume
went past about 3/4.
Anonymous
September 20, 2004 12:31:15 AM

Archived from groups: rec.audio.pro,nz.tech (More info?)

"richard" <rich@ihug.co.nz
> Phil Allison wrote:

** Replacing the deliberately snipped context:

" They dont, its relied on that you have semi-sensable source material with
a
normal frequancy/voltage curve that tapers off - a loud 10kHz tone will
destroy
almost any speaker. "

>
> > ** Ummm - a loud 10 kHz tone may destroy an unprotected tweeter but
not a
> > mid/bass driver.
>
> The OP had tweeters blowing


** Shame you said "speaker " isn't it.

No wonder you snipped that bit out.


> > Tweeters can easily be protected from over current abuse by correctly
chosen
> > fuses, low voltage lamps or a device called a polyswitch. These must
be
> > fitted in *series* with any passive crossover network being used with
the
> > tweeter.
>
> Common in good PA gear where a nasty piece of feedback may go thru, but
not so
> common in the cheap shitty gear and almost never seen in consumer level
gear.


** Completely irrelevant to my info - do you see the word "can" ?????

BTW Tweeter protection is common in hi-fi gear, since the advent of
polyswitches.


> Also, even when protected, a sudden burst will often destroy piezo
tweeters if
> loud enough.


** Another red herring from a faker.

Piezos are a special case - and you know that.

They require input voltage limiting rather than protection from "over
current abuse " as I specified.




............ Phil
Anonymous
September 20, 2004 1:55:27 AM

Archived from groups: rec.audio.pro,nz.tech (More info?)

Just to rule out the obvious, I would put a scope across the feed to the
tweeter and make sure that the amp is not feeding out high levels of
parasitic oscillation (i.e.Above the audible range)

A friend lost 4 good tweeters this way, all being fed by home-brew
amplifiers.

Regards

Ron
"Patrick Dunford" <patrickdunford@nomail.invalid> wrote in message
news:MPG.1bb6be468fe5a83298a475@news.paradise.net.nz...
> We have a wedge speaker (stage foldback) which has twice had the tweeter
> fail. It is a two way speaker with passive crossover.
>
> The failure has been detected as distortion from the tweeter.
>
> Since it happened only in that speaker out of four identical ones coupled
> in two parallel pairs to the same amp, is there any possible crossover
> fault that could cause such a failure to occur? It has been connected in
> parallel with another speaker of the same type on the same channel of the
> amp as was the case previously, but there have been no problems with the
> other speaker.
>
September 20, 2004 3:34:40 AM

Archived from groups: rec.audio.pro,nz.tech (More info?)

Phil Allison wrote:

> "richard" <rich@ihug.co.nz
> ** Replacing the deliberately snipped context:
>
> " They dont, its relied on that you have semi-sensable source material with
> a
> normal frequancy/voltage curve that tapers off - a loud 10kHz tone will
> destroy
> almost any speaker. "
>
>
>>>** Ummm - a loud 10 kHz tone may destroy an unprotected tweeter but
>
> not a
>
>>>mid/bass driver.
>>
>>The OP had tweeters blowing
>
>
>
> ** Shame you said "speaker " isn't it.

Speaker is a system involving drivers, which the tweeter is one of. I'm
nitpicking, you started it. Sure, a nasty high frequency whine wont do anything
to a subwoofer, but thats irrelevent here.

> No wonder you snipped that bit out.

Cropping previous conversation is general netiquite

>>>Tweeters can easily be protected from over current abuse by correctly
>
> chosen
>
>>>fuses, low voltage lamps or a device called a polyswitch. These must
>
> be
>
>>>fitted in *series* with any passive crossover network being used with
>
> the
>
>>>tweeter.
>>
>>Common in good PA gear where a nasty piece of feedback may go thru, but
>
> not so
>
>>common in the cheap shitty gear and almost never seen in consumer level
>
> gear.
>
>
> ** Completely irrelevant to my info - do you see the word "can" ?????
>
> BTW Tweeter protection is common in hi-fi gear, since the advent of
> polyswitches.

I havent seen it on the ones I have pulled to pieces recently, but I must admit
I was not looking too hard, and they too were pretty bottom end devices hence
the need to open them up, in that case it was replacing capacitors.

>>Also, even when protected, a sudden burst will often destroy piezo
>
> tweeters if
>
>>loud enough.
>
>
>
> ** Another red herring from a faker.
>
> Piezos are a special case - and you know that.
>
> They require input voltage limiting rather than protection from "over
> current abuse " as I specified.

The OP did not specify what the tweeters were, so this is hardly a red herring.
I am aware that piezos are a special case, and I am also aware that they are
often just connected across the input terminals directly with no protection
whatsoever against abuse in order to save a few dollars on the cheap bins that
are oh-so-popular among the garage band industry.

I would suggest that the OP checks every component in the crossover, including
the other driver.

As I often do not have access to a capacatance meter, what I usually do in those
cases is observe the responce on my analog multimeter and compare it with the
one in the other one that is not suspect. Not ideal, but beats driving across
town to check a capacitor out.

In addition I would check the internal cabling to the input jack that the
connections are to the correct place on the crossover if its one where the
internal and external connections share some terminals. I have seen one
conversion to a speakon jack where the connection was made to the wrong point on
the crossover bypassing part of the crossover. In that case it was low tweeter
but I can see that a misconnection may cause something else to be bypassed.
Anonymous
September 20, 2004 3:34:41 AM

Archived from groups: rec.audio.pro,nz.tech (More info?)

"richard" <rich@ihug.co.nz
> Phil Allison wrote:

> >
> > ** Shame you said "speaker " isn't it.
>
> Speaker is a system involving drivers, which the tweeter is one of. I'm
> nitpicking, you started it.


** I really was not nitpicking - just making sure readers did not get
the false idea that high frequencies damaged woofers.

Besides their generally much higher power capacity, the inductance of a
woofer's voice coil's protects them from high fequencies.



> > No wonder you snipped that bit out.
>
> Cropping previous conversation is general netiquite


** Snipping the actual words in dispute from view is a usenet crime.




............. Phil
Anonymous
September 20, 2004 3:34:42 AM

Archived from groups: rec.audio.pro,nz.tech (More info?)

Phil Allison wrote:


> ** Snipping the actual words in dispute from view is a usenet crime.


A "usenet crime"? Damn, quick, someone call the usenet police.
September 20, 2004 3:42:27 AM

Archived from groups: rec.audio.pro,nz.tech (More info?)

Ron McNulty wrote:

> Just to rule out the obvious, I would put a scope across the feed to the
> tweeter and make sure that the amp is not feeding out high levels of
> parasitic oscillation (i.e.Above the audible range)
>
> A friend lost 4 good tweeters this way, all being fed by home-brew
> amplifiers.

I had that happen on a cheap commercial amp, only happened into 2 ohms, and only
when the front panel gain was above about 3/4s the way around. Seen a tweeter
glow before? I sure think this must have before smoking.
Anonymous
September 20, 2004 4:25:07 PM

Archived from groups: rec.audio.pro (More info?)

"Mike Rivers"
Phil Allison
>
> > [the crossover network] - it has a 10
> > dB peak in the output at 1.4 kHz ?? I notice that the wood surface
next
> > to a 33 ohm, 10 watt resistor is scorched - this resistor is in
parallel
> > with an inductor so has to be disconnected to do a test - it tests
OPEN !!
> So I replace the resistor and the response is now as expected. But how
did
> > it ever burn out like that ??
> >
> > Running a 1.4 kHz sine wave into the x-over with NO load connected
provides
> > the answer - when driven at even moderate power levels at or near 1.5
kHz,
> > the 33 ohm resistor starts to SMOKE !! With no load the x-over
becomes a
> > series resonant circuit that *magnifies* the drive voltage by several
times
> > at 1.5 kHz so that 33 ohm resistor cops about 10 times more power than
it
> > ever would normally.
>
>
> It would help to have a schematic for reference.


** Help who ? Not you that is for sure.


> The first question
> that came to mind was "how can you have a circuit with the load
> disconnected,


** Errr - happens whenever the diaphragm fails open.


> but then I thought about variations in crossovers. I've
> seen crossover designs which have a series L-C circuit


** With no load a basic LC (ie 12dB/oct ) crossover becomes a *series
resonant circuit*.

The input Z drops to near zero ohms at the resonant frequency while the
input voltage is magnified a great many times at the output.



> What did the peak look like with the speaker connected and the
> resistor disconnected?


** Errr - as stated above it was 10 dB.


>
> But the real question is, why did the tweeter blow in the first place?


** Separate issue entirely.

The original diaphragm had lasted for many years of live PA work and so did
the replacement after the resistor change.




............ Phil
!