I need to deliver about 40 watts to a piezo tweeter at 30 khz. The tweeter is an CTS ksn1016A and is rated at 45 watts and is good for 40 khz. The amp is an LM7828 rated at 40 watts into 8 ohms. The problem is the tweeter has an capacitive reactance of 84 ohms at 30khz. The impedance mismatch is so bad I`m only geting a few hundred miliwatts into the tweeter.http://www.national.com/ds/LM/LM2876.pdf.
Any sugestions are welcome.

zkt

 

Piezo tweeters are voltage driven, not power driven - if it's rated at 45W at 8 ohms, that means it's suitable for use in a 45W 8 ohm system, not that it will take 45W.

Stick a scope on it and measure the voltage across it, 40W into 8 ohms is about 51V p-p. If you're getting that, everything is working fine.

 

as I clearly stated Xc= 84 ohm at 30 khz. Checking the datasheet on thelm 2876 output power vs load resistance graph utput power is a few miliwatts which I also determined experimentially.

 

From what I can determine, piezo speakers are rated for power handling per EIA RS-426 standard (See CTS Piezoelectric Tweeters, down near the end). This apparently specifies power as if an 8 ohm load were being driven with the same voltage as is being applied to the piezo speaker. 40 watts into 8 ohms requires about 18v RMS (are speakers rated for average power?). If all this is true, you can apply a maximum of 18v rms. God only knows how much acoustic power this will deliver - that is, unless we have an audio guru in the audience.

I tried to find a copy of RS-426 with Google, but EIA wants you to pay money for it - can't blame them.

 

Output of 45W into an 8 ohm load is about 18V RMS, 18V RMS feeding an 84 ohm load gives 3.8 watts RMS output.

As I mentioned above, have you checked with an oscilloscope (or AC millivoltmeter), what the actual output is.

As your are working in ultrasonic frequencies, is your amplifier designed to do so? - it's possible it's high frequecny cutoff is lower than that. Which is why I keep suggesting you measure it.

 

18v across 84 ohms gives 3.8 volt-amps. This causes the amplifier to dissipate power, but without knowing the phase angle between voltage and current in the speaker, we can't know how much real power the speaker dissipates. Having to move air must add a resistive component to the capacitive nature of the piezoelectric element, but how much?

Perhaps a similar argument could be made for electromagnetic speakers regarding the inductive nature of the voice coil.

 

I don't think it really matters!. Although the tweeter is 'rated' at 45 watts, it isn't really - as I said before, it's really rated at a voltage, the original Motorola piezo horns were actually specified as that.

In general it doesn't make any difference, as long as you remember the 45W is the system power the tweeter is designed for, and not it's own power.

I've always been impressed with piezo horns, nice and cheap, very easy to use, and sound 'reasonably' decent - a lot of people don't like them, but they are very hard to beat for the price.

Not really, they usually specify the power the driver will take, although it is quite common for tweeters to also specify the system power - like this :
'maximum power 10W, 50W system power with 4KHz crossover'.