The gear you get to work with sometime... I have a couple of Ashton PB15
two way PA speakers that have Motorola / CTS KSN1142 piezo tweeters
(probably mounted on separate horns) and 15" woofers.

The high frequency end of the audio spectrum reproduction is very bright
and sharp. Without knowing very much about the speaker design, I know
that piezotweeters are often used without a crossover and it seems likely
to me that not enough attention has been paid to matching the speakers in
the bin so the tweeter probably has higher sensitivity than the woofer
resulting in the sharp HF tones.

I'm just looking at output EQ and whether the highs can be brought down a
little in a 1/3 octave graphic equaliser or whether the effort would be
better expended on new speakers because cheap tweeters will never sound
as good as the more expensive ones.

Mainlander wrote:

...two way PA speakers that have Motorola / CTS KSN1142 piezo tweeters
... piezotweeters are often used without a crossover and it seems likely
to me that not enough attention has been paid to matching the speakers in
the bin so the tweeter probably has higher sensitivity than the woofer
resulting in the sharp HF tones.

The Motorola division that makes the piezo's had an app note
about crossovers to tame the tweet; I'll see if I can dig it up.

Mainlander wrote:

The gear you get to work with sometime... I have a couple of Ashton PB15
two way PA speakers that have Motorola / CTS KSN1142 piezo tweeters
(probably mounted on separate horns) and 15" woofers.

The high frequency end of the audio spectrum reproduction is very bright
and sharp. Without knowing very much about the speaker design, I know
that piezotweeters are often used without a crossover and it seems likely
to me that not enough attention has been paid to matching the speakers in
the bin so the tweeter probably has higher sensitivity than the woofer
resulting in the sharp HF tones.

I'm just looking at output EQ and whether the highs can be brought down a
little in a 1/3 octave graphic equaliser or whether the effort would be
better expended on new speakers because cheap tweeters will never sound
as good as the more expensive ones.

Those Motorola piezo's are horrible, but you have worse problems. If
you're going straight from a 15" woofer to a tweeter, you're missing the
all-important midrange. The highest your woofer should be doing is about
200Hz. From there to about 2.5kHz isn't being handled by anything.

Rob

Mainlander *@*.* wrote:
The gear you get to work with sometime... I have a couple of Ashton PB15
two way PA speakers that have Motorola / CTS KSN1142 piezo tweeters
(probably mounted on separate horns) and 15" woofers.

I am terribly sorry.

The high frequency end of the audio spectrum reproduction is very bright
and sharp. Without knowing very much about the speaker design, I know
that piezotweeters are often used without a crossover and it seems likely
to me that not enough attention has been paid to matching the speakers in
the bin so the tweeter probably has higher sensitivity than the woofer
resulting in the sharp HF tones.

Piezo tweeters are high-Z devices that are very capacitive. So, to make
a high pass filter with one, you can use a series resistor and let the
device capacitance be the rest of the filter. So that series resistor
you see in front of it _is_ a first-order crossover. It's about as
effective as any other first-order crossover (like the capacitor in
series with the horn on an Altec A-7).

I'm just looking at output EQ and whether the highs can be brought down a
little in a 1/3 octave graphic equaliser or whether the effort would be
better expended on new speakers because cheap tweeters will never sound
as good as the more expensive ones.

The real problem with the piezo tweeters is that they have a lot of narrowband
resonances, and you can't easily EQ those out. Get an oscillator and sweep
back and forth in the high end and listen; the response plot looks like a
porcupine, so it'll sound nice and clean until you raise the frequency a
little and it starts to squeal when it hits a resonance. You could try
using a parametric EQ set up as a bunch of notch filters although I don't
know how much good you can really do.

I have never heard any piezo system that I liked, but Marshall Leach has
and he tells me I shouldn't just discard the whole piezo thing out of hand.
I'd suggest replacing your horn assembly with something that will cross over
at about the same place.
--scott


--
"C'est un Nagra. C'est suisse, et tres, tres precis."

Piezo tweeters are capacitive. So, to make a high-pass
fitler, you can use a series resistor and let the tweeter's
capacitance be the rest of the filter.

Scott, you seem to be succumbing to the getting-things-backward syndrome that's
been affecting me and a lot of other people the last week or so.

Sticking a resistor in series with a capacitor produces (with respect to the
voltage across the capacitor) a low-pass filter.

Bill's right. A series resistor with a piezo produces a low-pass filter.

The way to put a crossover on a piezo is to shunt it with an 8-ohm resistor,
then design a crossover like you would for any 8-ohm tweeter. That will
slightly mitigate the awful sound of the piezo (by keeping bass frequencies out
of it), but only slightly. You still have the sharp resonances and the high
distortion to deal with, not to mention the inbuilt problem of a 15" 2-way
system.

So ditch the things and get a decent set of speakers. These are not worth the
effort to (slightly) improve.

Peace,
Paul

In article ,
William Sommerwerck wrote:
Piezo tweeters are capacitive. So, to make a high-pass
fitler, you can use a series resistor and let the tweeter's
capacitance be the rest of the filter.

Scott, you seem to be succumbing to the getting-things-backward syndrome that's
been affecting me and a lot of other people the last week or so.

Sticking a resistor in series with a capacitor produces (with respect to the
voltage across the capacitor) a low-pass filter.

Sorry, too little sleep. I meant shunt and not series. If you check the
back, you will find a shunt resistor. There is a really nifty Motorola
app note about these things which is probably on their web site.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

I'm don't think a shunt resistor would produce a high-pass response. If the amp
has no trouble driving the shunt, then there's no resistance-reactance
interaction to alter the response.

I suspect the shunt is to make the impedance of the tweeter look essentially
resistive -- an ultra-crude Zobel network.

The Motorola tweeter was probably intended to be used without a crossover. To
get a high-pass response, you'd hang the tweeter across the resistor in an RC
high-pass network. If the network's impedance is much lower than the tweeter's
impedance (the tweeter is basically a small capacitor, so that would be easily
done), the tweeter would have little effect on the RC high-pass response. QED.

Jon Dahlquist is a friend. If you guys insist, I'll ask him how he handled the
piezo tweeter in the DQ-10.


Scott Dorsey wrote...

William Sommerwerck wrote:

Piezo tweeters are capacitive. So, to make a high-pass
fitler, you can use a series resistor and let the tweeter's
capacitance be the rest of the filter.

Sticking a resistor in series with a capacitor produces (with
respect to the voltage across the capacitor) a low-pass filter.

Sorry, too little sleep. I meant shunt and not series. If you check the
back, you will find a shunt resistor. There is a really nifty Motorola
app note about these things which is probably on their web site.

We, Gidney and Cloyd, wrote:

The Motorola division that makes the piezo's had an app note
about crossovers to tame the tweet; I'll see if I can dig it up.

The appnote I was thinking of was written by Jon Bost, but
I can't find a copy of it ... maybe on a cold disk somewhere.

The Moto ceramics division was sold to http://www.ctscorp.com/.

Did find bookmark for http://members.misty.com/don/pzfix.html.