help with woofer/piezo tweeter design

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hi all, just found this forum, and it looks like just what i was looking for! I want to desing a set (2) of speakers. Nothing fancy, i hope. I'm not too worried about high fidelity. I'm more interested in simplicity of design and compactness.

Here's what i want: a Peerless 10" or 12" XLS subwoofer (see: http://www.d-s-t.com/peerless/index.htm ), and a Piezo horn. No crossover - simple. But is it possible???

If so, what sort of enclosure would be most suited to get the widest range of frequencies out of the subwoofer? As i said, i'm not too worried about fidelity, but i would like the frequency response to be as even as possible over the whole range.

The other specification is i would like to get it all into a small package. Not necessarily ultra-small, but no larger than it needs to be to get good bass frequencies response.

What advice do you have? Should i include a built-in amplifier? I'm not even sure what i will be using them for yet, but i know i want them to be portable to be taken to friend's houses or even maybe outside functions (BBQ's or whatever).
 
Well, i haven't got any replies to my post yet, so i thought i'd post a more precise problem:

I downloaded a sealed speaker box program, and plugged in the parameters of the 10" Peerless. I got some really strange results, and i'm wondering why! The software is AJ Sealed Design.

Here's what i put in: Qtc-0.707 Qes-0.18 Qts-0.17 Vas-89.7 litres Pmax-600W cone diametre-21.2cm Xmax-12.5mm (from PartsExpress). Here's what i got out: Vb: 5.50448 litres Fb: 78.6018Hz F3:78.6106Hz.

The Power Response graph doesn't look too bad, but the Output Response graph looks terrible! Why? Is it because this speaker isn't designed for use in a sealed box?

I was looking for an F3 in more like the 20's - is this reasonable? What should i do differently? Or is my software just crap? And what about the Vb? I don't want it THAT small!!! What would happen if i increased the volume of the Box? Would that force F3 down? My software won't let me change that parameter.
 
diyAudio Moderator Emeritus
Joined 2001
Have some bad news and some good news for you.

The bad news is-forget about using the XLS series from Peerless for a 2 way system. They breakup badly at 500 Hz. The lowest piezos can go is 2,000 Hz.

The good news is that there is a speaker available to you that might fill the bill. It is by Peerless too-the CSX 10". I have not used it-I have used it's older brother, the CCC 10". It is apparently the same woofer with the same low frequency response, only the newer model has a sandwich layer cone which makes it amazingly smooth all the way to 1500 Hz.

The CCC 10" was a pretty fair subwoofer in it's day, and for the price is still a good bargain. But this composite sandwich cone on the CSX version makes it acceptable for the midrange, too.

According to the chart, the CSX is smooth to 1500 Hz, then dips down and is about 7 db down at 2,000 Hz. For a 12 dB crossover, it should be about 6 dB down, so it is right in the range-it seems to have a relatively smooth natural crossover at 2,000 Hz.

I have a collection of piezo response graphs sent to me by Motorola years ago, and I believe that there is a model with relatively smooth response that extends down to 2,000 Hz. I will find it and let you know the model tomorrow.

So yes, I do think it is possible to do what you are trying to do.

I should point out that 10" is a little large to be crossing over at 2,000 Hz. There are plenty of mini speakers that have an 8" that does so, though, so you really are not too far behind there.

I should also say that Peerless, it's sister company Vifa, Audax and others make some pretty fine dome tweeters for not too much money. Yes you have to use a crossover, but crossovers for 2,000 Hz are not so expensive. The crossovers for low frequencies-200 Hz, 100 Hz, etc, are the expensive ones.

Besides, you can just let the CSX 10" just roll off, saving you the cost of half the crossover.

But if you have your heart set on a 2 way crossoverless system with excellent bass response for not too much money-less than $100 a speaker-it can be done.

I will find the Motorola piezo number and give it to you tomorrow.

Here is a chart of the CSX response in free air, (unenclosed), and the bass response in a 2 cu ft box tuned to 27 Hz. The latter was derived using the WinISD free software program.
 
diyAudio Moderator Emeritus
Joined 2001
Here is the chart:
 

Attachments

  • peerless csx 10 incher.png
    peerless csx 10 incher.png
    10 KB · Views: 604
Yes. David Weems in one of his books recommended adding a small value cap, say 0.47uF in series with the peizo to tame the top end. I tried this once on a speaker I repaired for a friend and it did work. Hi Fi World on one of their published DIY Audax kits using the Audax gold dome piezo used a similar solution to smooth it's top end as well...
Hope this helps
tomcat
 
Sounds like i should go with a vented box. Is bandpass or isobaric any good for a wide frequency range, or are they more suited for subwoofers?

I have some very old but well-built speaker cabinets that are 2.18 sq ft. Will this be big enough after i take out the volume of speaker, tweeter, padding, and port?

If not, how would it go as a sealed enclosure? If i went sealed, i would want to fill the box with padding. That reduces the volume by about 10% doesn't it? Depending on the padding used, of course.
 
diyAudio Moderator Emeritus
Joined 2001
Still looking for the chart of that piezo tweeter that extends down to 2,000 Hz. It is around here somewhere-I will find it.

Incidentally, Motorola piezos are now made by a company named CTS. I believe everyone still works at the Motorola factory in Albuquerque, New Mexico though. CTS has a website, but does not have freuency response charts there. They might respond to an Email.

Thge midrange horn Motorola makes is meant for pro sound and has an irregular response.

The rule of thumb is that speaker stuffing is to be counted as the thickness you get when you squeeze it as tight as you can with your fingers. In most cases, an inch or two of stuffing squeezes down to virtually nothing. Do not fill a vented box with stuffing-just line the walls with an inch or so of the stuff.

2.18 cubic feet sounds just right for the woofer. For a 3" inside diameter port, use a port length of 10 3/4" long. Add your tweeter, and you are all set. All for under 100 bucks a speaker.
 
diyAudio Moderator Emeritus
Joined 2001
Found it. It is the KSN 1165A. If you get it mail order, they might have their own model number, but tech help will probably tell you the Motorola/CTS number.

The tweeter will match up, but it has a nasty narrow peak at 2500 Hz and a broad one at 8000 Hz. CTS corp has a downloadable Application Note which tells you about crossovers, etc. Bottom of page, http://www.ctscorp.com/pzt/ffpzt-home.htm

If these were magnetic tweeters, there are inductor/capacitor parallel networks that would tame the peaks, but you cannot use them for piezoelectric speakers, which are capacitors themselves.

From what I can get from the Application Note, with networks for piezoelectric drivers, resistors act like capacitors and capacitors act like resistors. But nobody tells you what inductors act like.

I have several frequency response charts for various piezoelectric drivers but no scanner. Motorola sent them to me, but CTS did not when I requested them years later. Hopefully, they changed their policy there. Maybe they can Email the response charts. I might be able to scan the 1165A for you from another place.
 
crossover???

You've almost talked me into using a crossover for my project. Thanks Dave for your information, it's very helpful.

I had another idea, too. Could i use an equaliser to reduce some output in the overlap area? If i have a frequency chart for both the tweeter and the woofer, can i find out what the 2-way speaker will sound like without a crossover by simply adding the charts together (ie, adding the dB values of the overlapping frequencies)?

If so, i would then have a chart that would show me what frequencies i would need to dampen and what frequencies i would need to boost to even the curve out.

I have a catalogue here with some CTS piezo horns, but not the 1165A. What i have in this one are the KSN1142 and the KSN1188. I wonder how much the differ. the 1142 also looks pretty good down to 2kHz, and the 1188 is a "midrange" piezo that extends down to 800Hz. I am now thinking that if i do end up building that crossover, i may go with the 1188 to take some of the strain off the large 10" woofer.
 
I hate to wake the dead :zombie: :skull: , but I'm kinda interested in the CTS Piezo's.

I have a question though. The CTS drivers themselves are listed as 92dB 1W/m, but is that just the driver itself? Once you couple it to the lense does that increase the sensitivity even more?


Wish I could find some frequency responses on these things....
 
The CTS piezos don't really work that low; I've used them extensively for years and the best you can hope for is going to be more like 2.5kHz. That, combined with the tendancy for even a ten to beam badly at 2kHz, would make me lean towards an MTM crossed at 2.5kHz.

You could use a pair of 6.5s or 8s to get the required bass extension and cone area, and wiring them parallel will get the sensitivity up to where padding the tweeter won't be required. The 1165a will do nicely. You still will have to put a lowpass coil on the woofer section to prevent IM problems in any case, unless you can find a driver that will get -6dB at 30 degrees off axis at 2.5kHz and then acoustically roll off immediately above that.

www.billfitzmaurice.com
 
I'm going to be using 4 of the Vifa Infinity 7" buyouts from PE. They look to be usable out to 3Khz...so thats probably where I'll go. Probably a 12dB/octave on the woofers. So I can use a 20 Ohm resistor is parrellel with the Piezo and then apply a 12dB/octave HP to it (based on a 20 Ohm impedence), right?

I'm leaning towards the KSN1142A. That way I can use a different lense then the smaller 1165 you mentioned. It says it extends down to 1800Hz, so I think a 3Khz XO will be safe.
 
I'm going to be using 4 of the Vifa Infinity 7" buyouts from PE. They look to be usable out to 3Khz...so thats probably where I'll go. Probably a 12dB/octave on the woofers. So I can use a 20 Ohm resistor is parrellel with the Piezo and then apply a 12dB/octave HP to it (based on a 20 Ohm impedence), right?

Using paired Viva's will get you quite close to the 92dB SPL of the piezo, and once you account for room gain in the bass you probably won't have to pad the tweeter at all.

I've used the 1142A, it runs flat to about 2.5kHz. There is no advantage to it over the 1165A, however, as even if you go with a larger lens it still won't do any better within your intended passband, and the MTM configuration will give you all the pattern control that you need.

You may be able to get away with an inductor only on the woofs. I wouldn't do any crossover work until you build the box. Hook it up woofers only and measure the response to see where the SPL begins to rise in response to the baffle-step; that may well be around 1.5kHz. An inductor there to counter the rising response may be adequate combined with the driver acoustic roll off, but if not you can always toss a cap on afterwards for the additional slope you need above the crossover point. The acoustic roll off of the piezos may prove adequate without additonal components.
 
Ive got some of those motorla piezo's in my system -- the KSN1005A.

Crossed at 450hz with a single 22uF cap.
With 2 6" Acoustic research cones for the rest of the bottom end.

They sound great.. and have been running rock steady for 7 years.


So where does the "piezos' only go down to 2k hz" come from?
 
So where does the "piezos' only go down to 2k hz" come from?

Your 1005a tweeters are only working down to about 3kHz at best; below that it's your woofers that are doing the job.

The 22 mfd cap you have isn't doing anything. That is what you'd use with the resistive/inductive load of a dynamic tweeter, and 22mfd with an 8 ohm load would give a corner frequency of about 800 Hz.

Piezos aren't resistive/ inductive, they are capacitive loads, and they have an impedance of at least 500 ohms at 450 Hz, which means that they aren't getting any power into them at all down there.

You can read about the properties of piezos at the CTS site.
 
Any ideas what the difference are between the KSN series and the GT series? Like the KSN-1142 and the GT-400CD

Description of KSN-1142
Bullet type piezo tweeter horn driver. Standard 1-3/8"-18 TPI throat for a variety of lens designs. Piezo elements do not require crossover networks. Excellent power handling capability.

*Power handling: 400 watts max
*Frequency response: 1,800-30,000 Hz
*SPL: 92 dB 1W/1m
*Manufacturer model number: KSN1142A
*Dimensions: diameter: 2-7/16", depth: 1-1/2"

Price = $11.90

Description of GT-400CD
Similar to KSN1142A
* Power handling: 100 watts RMS.
* Frequency response: 1800-30000 Hz
* SPL: 92 dB 1W/1m * Mounts to any 1-3/8", 18TPI horn
* Designed for use with our 15-3/16" x 5-1/8" x 6" horn
* Net weight: 1/4 lb.
* Manufacturer model number: GT-400CD
* Dimensions: diameter: 2-7/16", depth: 1-1/2".

Price = $2.80

The cheaper one specifically mentions using a certain horn....plus its 80% cheaper :)
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.