• TalkBass has been independent since 1998. Add your voice.
    Create a free account to reply to discussions, view embedded media, and browse with fewer display ads.
    Join free Log in
    Want zero display ads or expanded classifieds tools? Compare plans.

DIY tweeter top box 2 - piezo units.

JimChjones

JimChjones

Aug 6, 2017
2,147
3,569
SE England
#1
Tailing on from the other thread I'm also considering a top box... but budget is a concern. Trying to get my head round the right circuit for a piezo tweeter based installation... My understanding is there are three desirable components...
Firstly a capacitor to filter LF, and I think 4.7uF for 4k 'crossover' in 8 ohms is about right.
Secondly a lot of sources talk about the desirability of a series resistor in the 20ohm range in piezo installs, and
Thirdly a volume control of some kind, presumably an l-pad.
What's making my head hurt is that it seems to me the series resistor fights against the others . Is this a pick any two situation?
If I put it the crossover side of the l pad then it seems to me impedance on that side is no longer 8ohm, does this affect filter frequency? Alternatively if put it next to the tweeter then won't the impedance of the system vary with the load, plus an 8ohm lpad will presumably then take the lions share of the power anyway.
The whole business of driver impedance and crossovers confuses me anyway, since at 4K the spec sheet for the main driver shows its impedance to be around 30ohm, not the nominal 8...
So this makes me wonder if maybe piezo and 8 ohms lpad aren't a good match? Would one be better off with a much larger value lpad in front of piezo tweeters, or even not an lpad at all, just an adequately rated vr or pot? Does the nature of a piezo tweeter reduce the advantages/desirability of an lpad delivering constant impedance? But it seems to me if the piezo is a very high impedance device a lpad, designed to maintain a constant low impedance, is not doing anything very useful. Seems to me essential to have some kind of volume control to be able to dial back harshness and match up sensitivity.
 
Last edited:
#6
BFM cabs use piezos, very successfully. But the minimum number is four and they have to be in a vertical array. Also, they're not resistive loads, they're capacitive, they're not 8 ohms, they can't be used with crossovers or L pads made for resistive loads. If you're going to use them you have to be not only a skilled designer but one skilled with piezos.
 
  • Like
Reactions: JimChjones
#7
Crossovers and piezos do not go hand in hand because a piezo driver increases in impedance with frequency at the same 6dB/octave as the series capacitor.

The resistor in series is generally used to insure stability with sensitive amps.
 
  • Like
Reactions: AstroSonic, basscapes and JimChjones
#9
Rick James said:
Piezos work just fine with a purpose designed crossover that includes a parallel resistor to provide a resistive load. That's what BFM uses. AFAIK there are no off the shelf piezo specific crossovers.
Click to expand...
Why can't the L-pad in series after the crossover provide a relatively consistent resistive load? Sorry if it's obvious--I'm completely unschooled in piezos and crossovers!
 
#10
Gents, those were the sorts of conclusions I was reaching as I wrote the post: I couldn't see how an l-pad could work, and I couldn't see how a conventional crossover would behave usefully.

Its interesting that piezo tweeters are very common, but expertise in their optimal use appears to be very rare. Maybe people working on a real budget in general put up with the limitations of a crude installation, and people designing higher value systems stick to the compression driver because it is well understood and is considered a higher quality installation by the customer. I haven't found what I consider a safe source for a volume control for a piezo tweeter on line. I've seen a circuit diagram of a voltage divider in front of one, which seems logical, but no provenance to give me confidence it not just someone drawing what seems logical. It also didn't have anything useful like component values.

It seems reasonable that a circuit for controlling piezo level would start with a series capacitor to reduce the power the circuit needs to dissipate, then a small series resistor for amp stability, and then an adjustable voltage divider. But what the component values of the voltage divider would be, and whether appropriate inductors or capacitors would be needed to control frequency response in the voltage divider is utterly beyond me.

But it does seem to me that without some sort of volume control the chances of matching the sensitivity of the full range driver and the piezo tweeter approximate to zero.
 
#11
bobcruz said:
Why can't the L-pad in series after the crossover provide a relatively consistent resistive load? Sorry if it's obvious--I'm completely unschooled in piezos and crossovers!
Click to expand...
It's too low a value. The BFM piezo high pass filter uses a 30 ohm resistor. Less than that and there's too much insertion loss.
I haven't found what I consider a safe source for a volume control for a piezo tweeter on line.
Click to expand...
You won't find a good piezo high pass filter either. BFM wanted one so he designed one. My Jack 12 with a full array of piezos works great with no separate volume control on the tweeters, but that's because the woofer is horn loaded.
 
  • Like
Reactions: bobcruz
#12
When you understand how piezo tweeters work, you realize that there are 2 ways to implement them. The cheapest way possible, which works "ok" and putting more effort and cost of parts which works much better but by the time you do that the appeal of a compression driver becomes compelling.
 
  • Like
Reactions: BassmanPaul, AstroSonic, Al Kraft and 2 others
#13
JimChjones said:
Gents, those were the sorts of conclusions I was reaching as I wrote the post: I couldn't see how an l-pad could work, and I couldn't see how a conventional crossover would behave usefully.

Its interesting that piezo tweeters are very common, but expertise in their optimal use appears to be very rare. Maybe people working on a real budget in general put up with the limitations of a crude installation, and people designing higher value systems stick to the compression driver because it is well understood and is considered a higher quality installation by the customer. I haven't found what I consider a safe source for a volume control for a piezo tweeter on line. I've seen a circuit diagram of a voltage divider in front of one, which seems logical, but no provenance to give me confidence it not just someone drawing what seems logical. It also didn't have anything useful like component values.

It seems reasonable that a circuit for controlling piezo level would start with a series capacitor to reduce the power the circuit needs to dissipate, then a small series resistor for amp stability, and then an adjustable voltage divider. But what the component values of the voltage divider would be, and whether appropriate inductors or capacitors would be needed to control frequency response in the voltage divider is utterly beyond me.

But it does seem to me that without some sort of volume control the chances of matching the sensitivity of the full range driver and the piezo tweeter approximate to zero.
Click to expand...


There used to be a lot more info in the wild, but piezos are generally considered taboo these days. I remember reading an in-depth paper on piezo crossover design that I found online years ago, but I can't find it any more.
Maybe this page will provide some useful ideas: Invalid Link Removed

Perhaps a better solution would be to save up for an Eminence APT80, so you can use a traditional crossover and L-pad. My GK 700RB combo runs an APT80 biamped at 5K, but I think the APT80 works well down to 3.5K in bass amp applications with a proper crossover.
 
  • Like
Reactions: AstroSonic
#16
Rick James said:
It's too low a value. The BFM piezo high pass filter uses a 30 ohm resistor. Less than that and there's too much insertion loss.
Click to expand...
Thanks, that's helpful. There are 16 ohm L pads; I wonder if the insertion loss would be acceptable in an application where the piezo was likely to be considerably more efficient than the woofer. It would of course somewhat defeat the use of the L pad since the insertion loss would make the piezo start out already "dialed down." Just musing here--I agree that a compression driver and regular crossover are the best solution for the bright box.
 
#17
The thing that makes me wonder about the suitability of the lpad is that it's intended to provide a constant impedance, and I'm not sure that's needed in a piezo installation. It seems to make more sense to me to have a simple voltage divider. I suppose the cheapest way to prototype something would be a rotary switch and a series of resistors with taps. Would one be able to get a ballpark by extracting the spec chart impedance at say 5, 10 and 15k for both devices and calculating net impedance current and voltage through ohms law? That's just about the limit of my capabilities...
 
#18
Last edited:
#20
I have not used a piezo tweeter in a long long time. My P.A. cab company experimented with them back in the early 80's building arrays of them (up to 30) and they all failed miserably. They sounded like scrunching up cellophane.
I don't know if piezo technology has changed (the price hasn't). Good Luck, I hope they work for you.
 
  • Like
Reactions: BassmanPaul

Latest posts

Top Bottom
Back