Which high-xo tweeter to use with a Jordan JX92S?

[planet10]

Re: Re: Re: Answers to Your Questions

'm mystified also. When I started building speakers, a sub had an Fs of 20hz or less, hence the term sub, as in subsonic. Not sure when it changed but nowadays it seems to mean woofer. I guess we just have to roll with the punches and accept it.

Quite probably due to the misappropriation of the term for multimedia systems....

dave

 

[EC8010]

Don't get me started!

And computer loudspeakers where the "sub" is a box the size of a loaf of bread and has a >6dB peak at 100Hz before rolling off. Oh, and the system is rated at 200W.

 

[planet10]

Re: Don't get me started!

And computer loudspeakers where the "sub" is a box the size of a loaf of bread and has a >6dB peak at 100Hz before rolling off. Oh, and the system is rated at 200W.

I'm currently salvaging woofers out of a passel of Monsoon woofer boxes. The woofers are actually quite nice (Eastech) -- i guess for MM systems even hi-end --but crammed into too small a box in every case -- even in an optimal box maybe 30 Hz solid. except for the 4" they are really using midbasses. The 4 is really optimized for bass, but getting a 4 to go real low in a small box is a real challenge....

(if interested in seeing more you can follow the links to eBay auctions where pictures & T/S are)

dave

 

[tcpip]

Someone else used a real small tweeter (different thread, you can do a seach with JX92) and seemd to like it. I think it was only about an inch in diameter. It was used in the ceneter channel of an HT system.

I don't know about that one, but rljones used a JX92S in a TL floor-stander, with an ESG2 ribbon in this design and a lot of my inspiration is from there. I don't intend to let the JX92S go all the way down, unlike him, but the upper end will probably be something very similar to what he's done or what Jim Griffin's monitor is doing. They use closely related ribbons (G2 and G2Si). Unlike Jim's xo, rljones' design has a purist 1st order xo. One inductor on the Jordan, one cap on the tweeter. I'm totally inexperienced with xo design, but whatever I've read makes me a little uneasy about first-order xo on tweeters. Even purists like Lynn Olson seem to have had trouble with those, but then I guess one can get away with it if one crosses over high enough.

 

[soongsc]

http://www.diyaudio.com/forums/showthread.php?s=&postid=656980&highlight=#post656980

 

[soongsc]

I would look at some kind of phase compensation for the JX92S before putting a tweeter with it. If you look at the frequency phase diagrams at the Jordan site, I would believe that most preople would thin they need a tweeter because of what they hear caused by phase lag. I have done this with other drivers, but not the JX92S yet.

 

[EC8010]

My reasoning for using a tweeter with the JX92s is that the Jordan is a little rough in the extreme treble compared to its excellent sound elsewhere so a tweeter will enable the Jordan (and its attendant roughness) to be rolled off with a 1st order crossover.

 

[navin]

I don't know about that one, but rljones used a JX92S in a TL floor-stander, with an ESG2 ribbon in this design and a lot of my inspiration is from there. I don't intend to let the JX92S go all the way down, unlike him, but the upper end will probably be something very similar to what he's done or what Jim Griffin's monitor is doing. They use closely related ribbons (G2 and G2Si). Unlike Jim's xo, rljones' design has a purist 1st order xo. One inductor on the Jordan, one cap on the tweeter. I'm totally inexperienced with xo design, but whatever I've read makes me a little uneasy about first-order xo on tweeters. Even purists like Lynn Olson seem to have had trouble with those, but then I guess one can get away with it if one crosses over high enough.

I would not worry about 1st order as long as the tweeter is designed to handle it (common examples are the Dynaudio and Morel tweeters). In most cases what you have to really worry about is excursion. if you cross the tweeter high enough (which can easily be done when the midrange has very wide response like the JX92) the tweete should be ok.

You could even consider a 1st order series network for such an application.

 

[tcpip]

My reasoning for using a tweeter with the JX92s is that the Jordan is a little rough in the extreme treble compared to its excellent sound elsewhere so a tweeter will enable the Jordan (and its attendant roughness) to be rolled off with a 1st order crossover.

I agree. The JX92S can be very fatiguing, specially if you listen on-axis. I have seen others vehemently disagree with this, so I guess the difference may be due to the listening room or electronics I use versus they. I only listen to the JX92S boxes about 15-20-deg off-axis.

 

[soongsc]

I agree. The JX92S can be very fatiguing, specially if you listen on-axis. I have seen others vehemently disagree with this, so I guess the difference may be due to the listening room or electronics I use versus they. I only listen to the JX92S boxes about 15-20-deg off-axis.

If we take into consideration the ear sensitivity curves at different angles from the ear, we may be able to understand the effects of toeing in more.

 

[tcpip]

If we take into consideration the ear sensitivity curves at different angles from the ear, we may be able to understand the effects of toeing in more.

I'd have thought that by toe-in, we'd change the angle of emission of the sound "ray" from the driver, but wouldn't change the angle of incidence of that "ray" on our ears. So how does the polar pattern of our ear's sensitivity come into the picture?

 

[soongsc]

I'd have thought that by toe-in, we'd change the angle of emission of the sound "ray" from the driver, but wouldn't change the angle of incidence of that "ray" on our ears. So how does the polar pattern of our ear's sensitivity come into the picture?

The FR of the ear varies as the sound source moves from 90 degrees to minus 90 degrees where 0 degress is beside your head.

The JX92S has a slight rise in FR, taking into the fact that speakers are placed about at the 70 degrees location, this means the high frequencies are enhanced way beyond normal live performance. Toeing in the speaker around 45 degress causes reduction in the high to a level closer to normal live performance level.

 

[Jim Griffin]

Danger of First Order Crossover for Ribbon Tweeters

Let me tell you that I do not recommend a first order crossover for a ribbon tweeter like the Aurum Cantus G2 or G2si or for that matter any ribbon. My analysis of the Jones first order crossover design with the JX92S and G2 ribbon does not convince me that it would enable the ribbon to long term survive plus he didn't take baffle step compensation into account either.

A tweeter, if properly designed into the design, can make an impressive speaker with the excellent imaging of the Jordan JX92S and the crisp highs of the G2si. Again try out my design and you'll know from where I am coming. See:

http://www.creativesound.ca/pdf/JX92SG2siDesignPak.pdf

 

[navin]

one of the advantages of using fullrnage drivers is that one can make a speaker that in effect has one driver covering almost the entire freq. spectrum. sometimes witha little bit of help at either one or the other or both extremes.

if we are using the JX92 between 100hz and 4kHx with higher order crossovers should we be comparing the JX92 to midrange or midbass drivers that are also capable of this such as the Scan Speak 15M4531 or SEAS W15LY or something as conventional.

how does the JX 92 compare. what advantages does it provide over traditional midrange drivers?

 

[tcpip]

one of the advantages of using fullrnage drivers is that one can make a speaker that in effect has one driver covering almost the entire freq. spectrum. sometimes witha little bit of help at either one or the other or both extremes.

if we are using the JX92 between 100hz and 4kHx with higher order crossovers should we be comparing the JX92 to midrange or midbass drivers that are also capable of this such as the Scan Speak 15M4531 or SEAS W15LY or something as conventional.

In theory, full-range drivers do the magic trick you described... cover the wide, wide range from sub-woofer xo levels (80-100Hz) to super-tweeter levels (greater than 6K). In practice, I'm beginning to get the nasty suspicion that the only people who actually believe this about "full-range" drivers are those who look at just SPL curves. If you go beyond SPL curves, you'll see that full-range drivers have serious changes to polar patterns as you go up the freq scale (laws of physics) and often have serious distortion at both ends of their range (limits of material science and mechanical design of the drivers). Hence, if you want to build speakers which fill the room with clean sound (as against near-field monitors), all full-range drivers need assistance earlier than the Purists believe. At least that's my nasty suspicion. You (and others, specially the Purists, also known as the Illuminati) can call me whatever you like, and I'll happily take it. After all, I'm a newbie, what do I know?

Therefore, to my ignorant brain, the JX92S should not be used above about 5-7KHz, and should not be used below about 100Hz (maybe 150Hz, I don't know). And the steep slope for the tweeter xo is not because the JX92 needs it (though I guess it does). It's primarily because a good ribbon tweeter needs it, as Jim (Griffin) explains so well in an earlier post. Even if the tweeter didn't have any such requirement, it may be a good idea to roll off the JX92 with a 2nd order or sharper xo slope, because its high-end behaviour is a bit funny (rising on-axis and some cone breakup distortion, etc). Therefore, cutting it out of the picture rapidly once you go above its xo frequency and handing the baton over to a good tweeter may be a prudent idea. In fact, given my listening experiences with my little JX92, I feel that even a good $30 dome tweeter will be better than allowing the JX92 to run unaided right to 20K.

However, given all these limitations, the JX92 probably still is a better midrange than most midranges. This is because its less-than-full-range is still wider than most purpose-built midranges.

how does the JX 92 compare. what advantages does it provide over traditional midrange drivers?

A wider frequency range than most midranges. The JX92S can go up to 5K or higher quite comfortably, allowing the use of small super-tweeters which can't come down to the usual 2K that other tweeters can. Similarly, with the JX92 you don't need a woofer xo'd at typical 200Hz or so... you can directly use a sub-woofer in the omnidirectional frequence ranges, eg. at 100Hz or below, specially if you do a 4th order xo there (thus preventing too much of the low freqs from hitting the JX92 cone). The JX92 has a very large Xmax for a four-incher.

Other than the wide(r) frequency range, the other good thing about this driver is a high level of detail reproduction with an extremely natural midrange, without any of the early cone breakup problems that metal-cone midbass units face. I haven't heard any of the good metal-cone midbass drivers... does the JX92 compare with them in terms of detail and accuracy in the midrange?

I'd really like to hear what others feel about the raw midrange sound quality of the JX92 compared to blue-blooded midbass drivers like the Seas Excels. I'd also like to see any distortion measurements anyone has done with the JX92. (In fact, one of the things I've seen, more by omission than commission, is that the full-range advocates rarely talk about the distortion specs of the drivers they use.)

If I'm shooting my keyboards off without any sense in what I've written, please feel free to rip 'em up.

 

[ChuckT]

To me, the jx92s sounds rough, even in the upper mid. I just couldn't get it sound right with a Gainclone and a Sony D amp.
It sounds ok if I didn't compare it with a 2-way Scanspeak Revalator, which is quite a bit better, in all area.
So yes, the Jordan needs a tweeter to smooth out the sound.

 

[soongsc]

To me, the jx92s sounds rough, even in the upper mid. I just couldn't get it sound right with a Gainclone and a Sony D amp.
It sounds ok if I didn't compare it with a 2-way Scanspeak Revalator, which is quite a bit better, in all area.
So yes, the Jordan needs a tweeter to smooth out the sound.

I once recommended a compensation circuit to someone, and the response was "too smooth" and too much bass (rediating into half space), so I would imagine it's a matter of compensation. What enclosure do you have and do you use any compensation circuits or not?

 

[Dumbass]

To me, the jx92s sounds rough, even in the upper mid. I just couldn't get it sound right with a Gainclone and a Sony D amp.
It sounds ok if I didn't compare it with a 2-way Scanspeak Revalator, which is quite a bit better, in all area.
So yes, the Jordan needs a tweeter to smooth out the sound.

Have heard more than once that Jordans need break-in, is this a possible factor?

The time I listened to Jordans, heard no "roughness" whatsoever. Female vox were sublime.

 

[navin]

To me, the jx92s sounds rough, even in the upper mid....
It sounds ok if I didn't compare it with a 2-way Scanspeak Revalator,...

I would be comparing the JX92 to a Scan Speak 18W 8546-9900 system above 100hz. I understand that the ScanSpeak woofer will be capable of higher SPLs but my interest lies in tonal accuracy, coherence and the speakers ability to involve me in the music.

 

[ChuckT]

The Jordan sound ok by itself, I have played over 2 months in a 30cm MLTL, but I must admit I didn't tweak it that much other than using a deq2496 to eq the responds but still couldn't get it sound right.
I think part of the fault is due to my box design but I couldn't what was the problem.
So I went back to my 2-way scanspeak and it was good.
I put in a Morel MDT40 tweeter with the Jordan and I like that better. I think the respond breakup at 8k above can really be heard and therefore the improvement with the tweeter.