After reading (and re-reading) Dr. Jim Griffin’s “Design Guidelines for Practical Near Field Line Arrays” it seemed reasonable to narrow the list by considering only tweeters that are capable of crossing over below 2,000Hz. You would think that this would be easy to objectively compare using manufacturer and vendor supplied information. I officially give up my solo attempt after a week of free time spent trying to reconcile conflicting information.

The following ribbon and planar tweeters seem to be easily acquirable. The Frequency Response and Minimum Cross-Over frequencies are the best I could derive from manufacturer and vendor supplied data. The Cost listed is just a general reference based on what I jotted down in my notes. I didn’t note where I found the pricing info, so I’m sure it is inaccurate. It may, however, be beneficial in comparing relative prices (i.e. early elimination of the gorgeous Raven piece).

THE LONG LIST
Aurum Cantus G2Si _ FR: 1.4K-40K _ Min XO: 2.5K _ Cost: $90
Aurum Cantus G2 _ FR: 1.4K-40K _ Min XO: 2.5K _ Cost: $170
Aurum Cantus G3 _ FR: 1.3K-40K _ Min XO: 2.3K _ Cost: $320
Aurum Cantus G3Si _ FR: 1.4K-40K _ Min XO: 2.3K _ Cost: $210
Aurum Cantus G1 _ FR: 900-40K _ Min XO: 2K _ Cost: $400
Bohlender Graebener Neo3(-PDR) _ FR: 2K-28K _ Min XO: 1.2K (line array) _ Cost: $45
Bohlender Graebener Neo8-(PDR) _ FR: 700-20K _ Min XO: 350-700 (line array) _ Cost: $60
Dayton PT2B-8 _ FR: 2K-26K _ Min XO: ? _ Cost: $35
Eton ER-4 _ FR: ? _ Min XO: ? _ Cost: $280
Fostex FT7RP _ FR: 3K-45K _ Min XO: ? _ Cost: $80
Fountek NeoCd1.0 _ FR: ? _ Min XO: ? _ Cost: $60
Fountek NeoCd2.0M _ FR: 1.2K-40K _ Min XO: 2K _ Cost: $120
Fountek NeoCd3.0M _ FR: 1.4K-40K _ Min XO: 2.5K _ Cost: $80
Fountek NeoPro5i _ FR: 850-40K _ Min XO: 1.5K _ Cost: $280
HiVi RT1.3 _ FR: _ Min XO: 2.5K _ Cost: $40
HiVi RT1x-A _ FR: 3K-21K _ Min XO: 3K _ Cost: $40
HiVi RT1II _ FR: 3K-21K _ Min XO: ? _ Cost: $40
HiVi RT1L _ FR: 3.5K-40K _ Min XO: ? _ Cost: $30
HiVi RT-2 Pro _ FR: 2.3K-30K _ Min XO: ? _ Cost: $150
HiVi RT2x-A _ FR: 1.7K-25K _ Min XO: 1.8K _ Cost: $80
HiVi RT2II _ FR: 1.7K -21K _ Min XO: 2K _ Cost: $90
HiVi RT8-II _ FR: 1.5K-20K _ Min XO: 1.7K _ Cost: $80
LCY 106 / LCY 109 _ FR: 2.2K-60K _ Min XO: ? _ Cost: $200
LCY 108 _ FR: 2.2K-60K _ Min XO: ? _ Cost: $190
LCY 110 _ FR: 2K-60K _ Min XO: ? _ Cost: $200
LCY 130 _ FR: 1.8K-60K _ Min XO: ? _ Cost: $240
LCY K100 _ FR: 7K-100K _ Min XO: ? _ Cost: $200
Mundorf AMT2340 _ FR: 2.2K-23K _ Min XO: ? _ Cost: $390
Raven R-1.1 _ FR: 2K-50K _ Min XO: 3K _ Cost: $180
Raven R3.2 MMX _ FR: 650-50K _ Min XO: 800 _ Cost: $1790
Silver Flute _ FR: 3K-20K _ Min XO: ? _ Cost: $40

That should leave the following short list based on the absolute maximum cross-over frequency of 2,000Hz.

THE SHORT LIST:
Aurum Cantus G1 _ FR: 900-40K _ Min XO: 2K _ Cost: $400
Bohlender Graebener Neo3(-PDR) _ FR: 2K-28K _ Min XO: 1.2K (line array) _ Cost: $45
Bohlender Graebener Neo8(-PDR) _ FR: 700-20K _ Min XO: 350-700 (line array) _ Cost: $60
Fountek NeoCd2.0M _ FR: 1.2K-40K _ Min XO: 2K _ Cost: $120
Fountek NeoPro5i _ FR: 850-40K _ Min XO: 1.5K _ Cost: $280
HiVi RT2x-A _ FR: 1.7K-25K _ Min XO: 1.8K _ Cost: $80
HiVi RT8-II _ FR: 1.5K-20K _ Min XO: 1.7K _ Cost: $80
Raven R3.2 MMX _ FR: 650-50K _ Min XO: 800 _ Cost: $1790

Are there any that should be added to or deleted from the above list based on cross-over frequency? Not to skip ahead to criteria 3 (cost), but I know the Raven has already been eliminated…

Bohlender Graebener Neo3(-PDR) _ FR: 2K-28K _ Min XO: 1.2K (line array) _ Cost: $45

Zaph Audio gave them a rave review. At some point I will upgrade my current tweeter section in my Calipso Line Array, and the only one I would use is this tweeter.

Marlboro

• Calipso Full Line array

I'd delete BG Neo8 from your short list, HiVi RT8-II, HiVi RT2 {if PT2B is lower in cost}. You should be able to use G1-G3 ribbons.

I operate my PT2B @ 1700hz, 8th order. Works fine.

>>I need help choosing a tweeter for a line array project.
I would use PT2B in the low priced range, maybe Neo3 in the mid-
higher priced range, and Fountek NeoCD2.0 for a high end design.

Lets assume a 65" tweeter line. [one channel]

1. Eight PT2 = ~ $200
2. Eightteen BG Neo3 = ~ $700
3. Eight Fountek NeoCD2.0 = ~$950

I would probably bypass BG Neo3 and pay extra for the Fountek.
Neo3 is a lower sensitivity and lower impedance driver than
hinder maximizing line array SPL performance.

You might be able to operate the $4 Dayton dome tweeter at 2khz
with a 8th order or greater crossover. Worse case, 2.5khz will work.
You could add that to your list.

If you want an esoteric line array, then use Fountek NeoPro5i,
the higher end AC ribbon, or R3.2.

I read through John Krutke’s “Battle of the Non-Domes” (http://www.zaphaudio.com/nondomes/). John is a huge resource for the DIY speaker community and I am thankful he took the time and effort to share his findings. Unfortunately he only had the resources to compare a narrow set of tweeters.

Is your expressed preference to use the B-G Neo3 in the future is based on entirely on John’s objective test results? Is it also based on some of the subjective web reviews of systems using the Neo3? I assume you haven’t experimented directly with them yet.

Slightly off topic:
I love the Calipso Array! It’s nice to see an atypical application. I know you’ve taken some shots over the driver spacing. The originality of the design and effort required to complete it are very commendable. Meeting all the goals of your design criteria and being able to listen to those every day must be very satisfying.

Thanks for your help,
Dave

Fountek NeoCD2.0
This seems to be a favorite for higher end (more expensive) DIY projects. Looking at John Krutke’s test results from his Zaph article (http://www.zaphaudio.com/nondomes/), the NeoCD2.0 Frequency Response looks like it could be crossed at +/- 1.5KHz. However the linear distortion measurements expressed in the Cumulative Spectrum Decay plots explain the recommended 2,000Hz cross-over. This tweeter seems to have a huge distortion problem under 2KHz.

It would be great to hear from somebody that has used these in an array. If somebody has experience with this driver, did you attempt to cross them lower than 2KHz? If so, what were (are) the results?

BG Neo3
It is interesting that you would spend the extra $500 (2 channels) for Fountek NeoCD2.0 rather than using this unit.
Neo3: 93dB, 4 ohm
NeoCD2.0: 98dB, 7 ohm
The Fountek CD2.0’s certainly look a whole lot cooler than the BG Neo 3’s. I’ll have to read up on the implications of sensitivity and impedance.

Dayton PT2B and its cousins
I had originally ruled out these drivers based entirely on John’s distortion test results. Assuming the tweeters in your array are similar to the ones John tested, how does this distortion (1700-2500Hz) express itself in musical playback? Do you notice it? If so, is it a “liability” or a “feature?” I know this is asking for a purely subjective analysis (oxymoron alert), but I’m interested in your view on the matter.

BG Neo8, and HiVi
Is deleting these based on their distortion measurements or something else you may have experienced or read? The Neo8 looked like a cross-over champ.

All dome tweeters
There is no way I’m going to mess with boxes and boxes of dome tweeters. I’m too old to be cutting down mounting brackets.

Esoteric solutions
The R3.2’s are insane. No way.
The AC G2 or the Fountek NeoPro5i would both be stretching the budget thin. Would they be a real performance upgrade from the NeoCD2.0, or just cooler looking?

Slightly off topic:
I love the control panel and impedance switch on your arrays. I plan to steal your terminal block idea for easy impedance selection, but with somewhat sane impedance choices. You are insane…driving 0.8 Ohm loads. I’ve followed your theories (or mad ramblings) across various forums on the use of bridged pro amps driving multiple kilowatts into micro impedances. I will continue to look for a very bright blue flash followed by lots of white smoke from your direction.

Thanks for your help,
Dave

Thanks for you comments on my array. The spacing is 5 inches, which is clearly within the range of Jim's paper, especially since I'm using active crosses at 2500 and 24db/octave.

People who hit me on it haven't heard it, and haven't built one of these to MY specs.

Listening is fabulous.

Thanks again.

Marlboro

I favor the Fountek CD2.0 or the Aurum Cantus G3 true ribbons for most line array applications. Their higher priced/larger brothers in each line can be used for severe power handling applications. The Fountek tweeters have plastic cases while the A-C tweeters all have cast metal housings. The housing difference accounts for most of the cost difference between these two brands. The ribbon tweeters have higher sensitivity ratings (upper 90's dB SPL) so they are easier to pad down a little to match the output from a line of mid-woofers. The B-G Neo 3 and 8 series have low impedance (4.0 ohms) and are low enough in sensitivity that designing to a passive crossover from a line of midwoofers get difficult as the woofer line is often more sensitive than the tweeters. The B-G Neo 8 and the longer B-G planars roll off over the 15-20 kHz area so they need a super tweeter to hold up the response.

Distortion tests don't reveal the true possibilities for ribbon tweeters used in line arrays. That is because you have power distributed to 8 or 9 devices so power distortion levels are very, very low vs. single driver measurements. For example you can successfully cross the Fountek Neo2.0 and A-C G3 ribbons as low as 1500 Hz in lines of 8 or 9 devices without audible distortion for home listening conditions.

By the way, an OEM version of the A-C G3 called the G3i-130 maybe available from some suppliers at a lower cost than the regular G3 without significant performance degradation.

Jim

Unless you are bi-amping with an active crossover the lower sensitivity planars restrict your choice and/or number of woofers.

People make decisions based on tests, people should buy sample
drivers and do their own listening tests as it offers peace of mind.

Keep in mind that Zaph tested the Silver Flute Yag-20 which is
similar to the PT2B, but as Rick said the PT2B has a better magnet
as is evident by the different in sensitivity ratings. Yag-20 is
rated at 90dB, PT2B is 94dB. It would be premature to rule out PT2B
if you look at Yag-20 tests. Zaph also makes a note about drivers used in line arrays.

For a non-line array design I would choose a dome tweeter instead
of PT2B. But for a big line array with eight - tweleve PT2B per tower, that's hard to beat for the money. You can go to home audio
stores and not find that sound for $20k.

You should be able to get those tweeters for $25 in bulk pricing
and for the money, they do a good job. If you use 48dB/octave active
crossover {DCX2496, etal}, you can stretch the crossover frequency lower. You can also keep more distortion from manifesting when you
push the line array to higher levels. You can also change the crossover frequency for different listening situations. The active
crossover has advantages where passive crossovers are fixed.

re: BG Neo3
There is also BG Neo3PDR, an upgrade to BG Neo3.
Sensitivity is lower; 90dB/2.83v, 4 ohm

re: Esoteric solutions
I didn't mention it before, but I can think of more tweeters to
add in this category, but you'd be spending $700 - $2000 on each
tweeter. lol Is it worth it? Sure it is if you want to push the
SPL level higher and keep distortion lower. I'm going to guess
and say you can probably generate at least 120dB[c] of treble at
12-20 feet away with these expensive tweeters. The highest end
tweeter is rated for 145dB, 3kw peak power, 200mS. lol ...

Based on what I've seen, it seems the logical progression would
be; $1 mylar, $4 dome, $25 PT2B, $120 Fountek {or equivalent that
may cost $200}, then the $300 - $500 ribbons. Just figure out what
your budget is and find the tweeter that fits the wallet.

If you can't choose, buy one of each driver candidate and hook it
up and just listen to the sound.