Piezo phisics question

[Pbassred]

I read that a piezo tweeter is a capacitive rather than a resistive load, and it has no affect on the overall load. That said, there is no such thing as a free lunch. How many piezos can you drive from (say) a 10W amplifier? What is the formula?

[Keith Taylor]

Being a capacitor, piezo transducers present a similar load to the amplifier to the case of ESL's (electrostatic speakers). There is not much real (resistive) component to its impedance so considerations of power do not have much relevance. The impedance is a pure reactance meaning it varies with frequency and the current will lead the voltage by 90 degrees resulting in a "wattless" load. The crunch for the amplifier comes at high frequencies when the impedance drops to a low value; the more so with parallelled tweeters. Attributes of a power amp would include an ability to source plenty of current and to maintian stability with capacitive loads. An amplifier designed for operation into 8 ohms and above would not be a good choice.
Keith

[Conrad Hoffman]

Keith summed things up very succinctly. I might add that I like to use a resistive pad with piezo tweeters. This isolates the amp and lets any crossover see something more resistive. Crossover design has to take the load into account and piezo tweeters are anything but conventional. IMO, they have an undeserved reputation for poor sound but can be quite good if properly integrated in the overall design.

CH

[gpapag]

You may find these links useful

Pulsar Developments Ltd - Piezo Tweeter Application Note

AudioKarma.org Home Audio Stereo Discussion Forums - View Single Post - Using piezo tweeters wisely: a "how to"

Regards
George

[LineArray]

Originally Posted by Conrad Hoffman

Keith summed things up very succinctly. I might add that I like to use a resistive pad with piezo tweeters. This isolates the amp and lets any crossover see something more resistive. Crossover design has to take the load into account and piezo tweeters are anything but conventional. IMO, they have an undeserved reputation for poor sound but can be quite good if properly integrated in the overall design.

CH

I agree to that, i once had a cheap piezo from a motorola horn tweeter.

I used it for fun without the horn and put a resistor across the terminals,
so i could integrate it in a 2-Way crossover.

The quality improved mainly from the integration to the crossover and
had nothing to do anymore with the "sissle" sound common to this
kind of tweeter, when applied without proper crossover.

[Godzilla]

>>> IMO, they have an undeserved reputation for poor sound but can be quite good if properly integrated in the overall design.

I agree Conrad.

Piezo Super Tweeter

I enjoy their sound and their price!

Godzilla

[norman bates]

ideally, you run a transformer, a zobel for the transformer, maybe a notch on the amp side, and a resistor on the piezo side (matchs the impedence).

Do a search on audioasylum.com for piezo under the high efficiency asylum.

I ran a piezo with an 24db Linkwitz Riley active crossover (6db down at crossover point) crossed at 5khz. The piezo had a peak at 4.5khz that was apparant as I move the crossover point lower. And I ran a resistor in series with the piezo (can't remember what value).

The results were astonishing, unbelievable to people who hate piezos.

Run wide open (like a tweeter with nothing) most have a climbing response and a peak or two, sounding spitty. Some have luck with a resistor.

Oh yea, the cheapy ones are junk, look for the older motorolas such as the 1038a, the 1016a.

CTS / Motorola KSN-1016a Surface Mount Piezo Electric Wide Dispersion Horn Tweeter</font>

they used to be a couple of dollars, now the best ones are rare.

a quote from abuddy
"The piezo requires a very special network to even sound listenable.
Blind tests always leave 'audiophile' listeners stunned when they find out they have been listening to a $5 piezo. They get red in the face, make self-serving remarks like yours, and then try and back-pedal.
It's always a howl."

and

"Use a 25V line transformer (backwards). You now have a choice of 4 or 8 ohm primary and the secondary is 62.5, 125, and 250 ohms on the 10W, 5W, and 2.5W taps. You must use the 62 ohm resistor in series with the 0.13µF KSN1016, it will generally be best on the 5W tap. I use a 100µH choke with a 5~10 ohm resistor as a zobel on the primary side. Adjust the resistor for the flatest output. This now looks like about an 8 ohm load to the amplifier so you cap put an L-pad in at this point."

http://db.audioasylum.com/cgi/m.mpl?...t=piezo+djk&r=



http://db.audioasylum.com/cgi/m.mpl?...t=piezo+djk&r=



Norman

[Saturnus]

If you think of a piezo as a losy capacitor then you're pretty much hit the spot.

Seeing it like this is also the key to calculating and using it correctly, so prepare to unlearn everything you thought you knew.

First you must measure the capacitance of your piezo. Just like it wouldn't make much sense to drop a normal tweeter into a filter without knowing it's impedance so is it for the piezos capacitance.

A single series resistor makes a first order high pass filter when combined with a piezo. You just use the standard formula for 1st order high pass and instead of calculating the capacitor value, you calculate the needed resistor value to fulfill the purpose.

Note that many if not most piezos have a peak a the lowest end of it's usable range and are somewhat linear above that. So a good strategy is to put the highpass x-over around the middle of that peak.

This makes them somewhat linear and cuts away what the piezo can't reproduce anyways..

Now, the series resistor will also have parasitic inductance especially wirewound ones. This will make a 1st order low pass filter. So this is both good and bad. The bad news is that even the parasitic inductance of "non-inductive" wirewounds have high enough inductance to affect the audible range. So if you want your tweeter to work as intended you must use metalfilm or old fashioned solid carbon resistors. I vastly prefer the later for it's sonic qualities with piezos.

If you use these technique most piezos can actually sound pretty well. And if you at the same time dampen the most often crappily made plastic horn, and disassamble it and smooth out the rough edges most often present inside. You can actually get pretty decent performance that rivals even very good regular tweeters.

Just try it. Most piezos cost next to nothing so experimenting with them is almost free. And you will be amazed at the results.

[Saturnus]

I forgot to mention that piezos require burn-in time. No matter what you do they'll not sound good the first 20-100 hours.

[David_Web]

What happens if you DC "bias" a piezo? (SE class A maybe)

Is a resistor needed if you run active filter with EQ?
Or does something happen to the piezo itself when run at high Z.