Tweeter to match JX92S

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Jump to Latest
A

AntM

Member
Joined 2005
I'm looking into building a MLTL design using the Jordan JX92S

http://documents.jordan-usa.com/Enclosures/JX92S-31-MLTL-MD.pdf

The JX92S is an excellent driver when used in a fullrange, but as is typical with many full-range drivers, has a poor off-axis response. I'm planning on adding a tweeter which can be optionally enabled with a simple (ideally first-order ) crossover. I will then be able to listen to them either full-range for superior imaging when listening directly in the sweet-spot, or as a 2-way.

Because of WAF, I don't want to increase the height of the enclosure too much. I'm also keen on mounting the tweeter as close as possible to the fullrange, as I would probably keep the crossover frequency high (maybe 6kHz) and I want to avoid lobing effects as much as possible.

Another factor is that the Jordan has a rising on-axis response, which means that it is better used off-axis. This means that the partnering tweeter should have a very good off axis response.

The question, as always, is which tweeter to use. I've shortlisted several models including
Vifa's XT25SC90-04, NE19-VTS-04, XT19TD00-04 and SB Acoustic's SB29RDNC.

Has anyone had experience of using any of these units?
My requirements are:
- small faceplate to allow close mounting
- good off-axis response
- excellent THD
- Good subjective performance.

I may also consider cutting the tweeter faceplate with a router to allow close mounting.

Any opinions gratefully received.

Ant
 
GDO

GDO

Member
Joined 2009
jerome69

jerome69

Member
Joined 2008
GDO said:
Is this your personal experience, or are you just quoting Zaph ?

Have you built all the other models you mention, or are you still quoting?
Click to expand...
Personal experience with loudspeakers in particular with fullrange. Never built the models given as choice example . I didn't only quote Zaph .

A fullrange covering 30Hz-20kHz : diameter 5"
Designs
These new designs are smaller than the first generation designs yet provide output to, or below, 30Hz thereby restoring the bottom octave.
Click to expand...
Yesssssssssssssssssss !
I don't need to listen this to predict it will never work.
 
A

AntM

Member
Joined 2005
Thanks for your comments. Actually the design is somewhat inspired by the Aurousal VS.

Aurousal VS VirtualSource? Floorstander*

This uses two JX92S drivers operating without a crossover in a MLTL cabinet (except for baffle-step compensation). There is an additional tweeter which just helps with the top-end. The crossover frequency and order is unspecified but I suspect it has a high crossover frequency to just help in the top octave where the Jordan driver's dispersion characteristics are lacking. The design has had some outstanding reviews, so I am convinced there is merit in the idea:
Aurousal VS review from TechRadar UK's expert reviews of Hi-fi and AV speakers
Aurousal VS review from the experts at whathifi.com
Aurousal VS loudspeaker - Review [English]

I have designed a few 2-way speakers with varying success, then recently I built a sealed box design built around the JX92S. I must say I was astonished by the performance that could be achieved by a single drive unit in a simple enclosure. I put his down to several well-known disadvantages in two-way designs:
- loudspeaker impedance varies with temperature, therefore crossover characteristics will also vary, both long-term and instantaneously
- the dispersion characteristics of a 2-way will change suddenly around the crossover frequency from narrow to wide, This will alter the frequency response in a way that is dependent on room characterisitcs. This has inspired designs that attempt to match the dispersion characteristics of the two drivers around the crossover frequency (using waveguide-loaded designs such as The DXT® Tweeter - a waveguide lens for dome tweeters
- The harmonic content of natural instruments is difficult to re-create correctly in a 2-way design due to magnitude and phase variations.
- Vertical lobing (i.e. spatial comb-filtering) can occur due to interference around the crossover frequency.
- Real difficulty in making a good design built around the necessary engineering compromises, particularly for an audio DIYer without having good test equipment and modelling tools.

Of course full-range designs are far from perfect and I'm intrigued by the idea of providing a little extra help at the top-end.
the Aurousal loudspeaker uses a polymer/silk dome Neodymium unit, possibly quite cheap. It also occurs to me that if the crossover is above 10kHz, 2nd-harmonics and above are out of the hearing range (and likewise above 6.6kHz crossover, 3rd harmonics are unimportant), so maybe THD is not that important as long as IMD is OK. ). Also the ear should be pretty insensitive to frequency response variation in the top octave.

I've considered the JX92S/ribbon design but then it starts to looks like a conventional 2-way with the above issues.
Any thoughts/comments?
Ant
 
jerome69

jerome69

Member
Joined 2008
The Aurousal VS VirtualSource™ Floorstander is a 2.5 ways speaker.
The tweeter looks to the Dayton nd20fa
Dayton Audio - ND20FA-6 Specialty-Drivers
Put a tweeter without a serious crossover is for me very laughable : take "advantage" of fullrange but with a tweeter to help :rolleyes:. In fact without doing the job correctly (good phase tracking).
I think it is possible to make a clone of this loudspeaker.

Jim Griffin design seems to me a good design. Look the 15° measurement, no problem of directivity near the crossover frequency.
http://documents.jordan-usa.com/JX92SG2siDesignPak.pdf
The G2Si could be replace by the Fountek JP3.

Hope this helps.
 
A

AntM

Member
Joined 2005
Thanks for your input. I'm concluding that using a super-tweeter with the JX92S will not suffer the usual vertical lobing issue too severely as the off-axis response of the JX92S is so poor. This means that the interference patterns/(spacial aliasing if you relate this to beamforming) will be less of an issue even if the tweeter-woofer distance is more than the half the crossover wavelength. I should model this in Matlab if I get the time. This is all assuming that the speaker is used off-axis. The on-axis response will be somewhat problematic, still. I'm accept some of your points, Jerome, that this is an uncoventional approach, but, hey, that's what DIYing is all about. Otherwise I can just buy the next off-the shelf 2-way that appears in my local hi-fi shop whilst a small part of me (and my wallet) dies.

So, in conclusion, I've ordered the Vifa XT19, which has an excellent off-axis response, a smooth distortion profile, with third and higher-order harmonics below -70dB if I crossover above 3kHz

http://www.zaphaudio.com/temp/Vifa-XT19-HD.gif

My gut feeling is that this tweeter will sound great, and give me the option of using it either as a supertweeter or with a conventional crossover.

I've just had some birch plywood cut, ready to pick up this afternoon. I may post the project if anyone is interested.

Ant
 
A

AntM

Member
Joined 2005
Hey,
I completed the project last year - I'll post some pictures if you are interested.

I played around a bit too with "parallel" tweeter options.

My experience of using an additional tweeter rolled in above 8kHz or so is that the image focus reduces, in line with increased dispersion at HF and increased room interaction; and the listening position sweet spot widens, more or less as expected. However I've noticed that the JX92S has a pretty zingy top-end which can also be seen in the response plots. On some music material it becomes difficult to live with. Using a parallel tweeter does not improve matters and so I'm tempted to think that a full crossover may work better to reduce top end resonances.


By the way, the bass response of the design is really good - deep and dry.

Anthony
 
A

AntM

Member
Joined 2005
Hi Cal,
No unfortunately not - I didn't have the foresight to include different mounting options. I've seen though that top firing tweeters and the like have become pretty popular on high-end designs these days - I think the tweeter is crossed over very high.

I think the design philosophy of a top/back firing tweeter probably fits in better with a full-range design where the dispersion is already quite wide.

Anthony
 
S

smallangryboy

Member
Joined 2004
Ant, post up some pictures!

I'd be interested in an A-B comparison of the Jordan with and without tweeter and perhaps with and without tweeter but with BSC!

I added a BSC and it was an odd sensation, though not bad just different, I think it adds something and takes something away, it's all about balancing I guess.
 
T

Toaster

Member
Joined 2004
I'm not using them presently, but FWIW, Jim's design with the AC ribbon is the best sound I've had from the JX92S. No matter what the toe-in angle or use/ non-use of BSC I never could get on with them fullrange. Crossing higher could work, depending on implementation, but the blend is fine at 3kHz.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Top Bottom