JXR6 design analysis for near field setup

[deandob]

I still dont have the JXR6 HD drivers yet, there has been a delay with my local supplier. In the meantime I have been doing more planning to complete the design.

Interesting in WinISD there is only a marginal difference between a 2.6l cabinet (ideal Q of 0.71) and a 4l cabinet, in fact the -3db points dont change at all, its still about 120Hz. I assume that the spiderless design of the JXR6 is not taken into account by WinISD? The real differences between varying enclosure sizes is the slope of the low end response curve (Q). As I'll be filtering below 100-150Hz the extra volume size is not important.

The active equaliser that I use is setup to lower the total system Q to 0.5 using the tangband 871 driver, to match the subwoofer Q of 0.5, and as the tangband enclosure has a Q of 0.75, I dont need the larger enclosure to lower the enclosure Q when using the JXR6. Also the current TangBand enclosure is 6L and does has slightly worse low frequency response compared to the JXR6 at least when modelling in WinISD.

Regarding the cabinet design, I'll be making a mini version of the popular jordan wide baffle design, as attached, 400mm high, 200mm wide, 75mm deep size 2.6l, sitting on a plinth. The drivers will be rebated into the front baffle and I wont be using any felt surrounds, but the edges will have a 45 degree rebate.

I'm a little concerned with such a small enclosure and a wide back panel but not very deep that back panel reflections will be a problem with this design, especially as the reflected wave will pass through the thin Jordan cone, adding an out of phase component and possible cone distortion. As the 75mm distance to the back panel is not that large, any reflection that passes out the front will not be too much out of phase, although as its only the lower frequencies that will pass through through this should not matter much.

Is this really a problem with such a small driver, thin cone and this cabinet design? Is there a cabinet stuffing or lining that would work well with such a small depth?

Regards,
Dean

[Colin]

Hi - I'm also cautious about reflections from the back panel. What works for the JX92 may not be as effective for the JXR6.

I attempted a TL design for my JX53s which had a cross-section of 4" wide by 3" deep. The line was tuned to 120Hz. It always sounded very thin, so I abandoned it. I tried the JXR6 in it and whereas I was now getting output to 120-ish, it still sounded thin. Putting the 6 in my 2.4 litre ceramic cube has confirmed that it was the line which was the problem, not the driver. I can only conclude that the depth of the enclosure wasn't enough. (The width may have something to do with it - 4" vs 6" in the sealed enclosure.)

I'd go for something more balanced in CSA and if a wide baffle is required, integrate some kind of frame around the enclosure. Alternatively, a wide baffle with curved back to give it some depth (it would be possible to do this using the bendy MDF which turns up in DIY stores). A triangular cross-section would also work well.

Hope this helps.

[deandob]

Hi Colin,

That sounds like a good plan. As the driver will be mounted high in the front baffle, it would be easy to make a curve to point the back radiation down into the box, where there is a lot more room and the sound wave will be dissipated by the stuffing. Also as the JXR6 does not radiate that much due to its small size this should be less of a problem. I found a couple of interesting points on the Jordan web site about this.

Q: Will the shallow VTL enclosure cause audible reflections behind the JX92S?

A: On the contrary, it will raise their frequency and make them easier to absorb through use of acoustic damping material placed behind the JX92.

One suggested design for a stand-alone system is shown below, made from 12mm ply or MDF. Two panels are cut for front and back - the sizes shown below are external. Then a matching frame (A) is cut to the same external dimensions. The hole inside the frame should be 15cm (h) x 19cm (w). Five frames, clamped between the front and back panels, will give approximately 2 litres. The number of frames can be altered to match the volume required. The hole inside the frames could be cut in an irregular shape, to help break up internal reflections.

The JXR6 HD should be inset to be flush with the front panel and the outer edges of the cabinet rounded. Place four small pieces of dowel at irregular points between the front and back panels to brace the enclosure and the inside should be lined with felt, BAF wadding or similar absorbant material.

Does anyone recommend any highly absorbent stuffing or lining that has reasonably constant sound absorbtion properties across the frequency range?

I'd like to pass the final design by Ted Jordan himself but his web site mentions that he wont deal with individuals - although its great to have everyone contribute their thoughts to my design, thank you!

Regards,
Dean

[Colin]

I think long fibre wool gets good results. I also use BAF wadding which seems fine at mid and HF.

I doubt Ted will have time to look at individual designs but if you post it here, either he or one of his sales agents may see it. I wouldn't worry too much - all the JXR6 requires is a simple sealed box around 2 to 4 litres and there are plenty of people here who can offer comments.

[deandob]

After travelling for work for a few days, I got home today and the JXR6 HD drivers have arrived. I have a test 4L cabinet from a previous speaker project that I have mounted the JXR6 HDs in and am using a high quality Tripath Class D amp (20wpc) and PC source (lossless + pro audio soundcard).

Impressions compared to TangBand W3-871
- More air & better resolution in the upper registers. A cleaner sound.
- more aggressive, faster dynamics, a touch grating with some recordings.
- mid bass sounds hollow, possibly due to differences in the cross over.
- Instrument placement is more accurate but the soundstage does not seem as wide or enveloping. Possibly due to poor cabinet setup (just a box with not all leaks sealed).

Is it wise to burn in these drivers?

This stuff is a bit like wine, its not better in some ways, in others not so sure - just different. I'll listen to them for a couple of days and swap the TangBand back to determine the difference.

Regards,
Dean

[Colin]

Try sealing the box and making sure it is well-stuffed with absorbant. Felt round the drivers will probably sort the imaging. It may also be worth reducing the internal size of the enclosure to around 2 litres or so...

Yes, it takes a day or so to burn in/get used to the sound.

[soongsc]

The better the drivers are, the more difference the XO makes.

[deandob]

I have been letting these speakers run all day and occassionally listening to them. I'm starting to appreciate them more than the first impressions (which re-reading my post above seems a little negative - but was not meant to be).

The lack of distortion at high frequencies is what I think benefits these drivers; they are like quality tweeters. Cymbals etc sound more "delicate" and the natural echo/decay around an instrument much better defined.

The midbass is still hollow, which I'm sure is more a case of the temporary cabinet I'm using not the drivers (size, not properly sealed, BSC etc). I use a quality active crossover to remove the signal below 100Hz, which possibly needs adjustment, and this driver may need BSC more than the Tangband W3-871 (which I use a notch filter on).

I hear an improvement in soundstage and reduction of a slight agressiveness if I dont toe the speakers in, which is contrary to the feedback that I've had about Jordan drivers. I also tried the damp towels wrapped around the sides of the drivers to simulate using felt and reduce any front baffle effects, and the soundstage reduces, possibly as I have the drivers almost flush with a LCD monitor which may provide additional surface reflections with almost no delay, adding to the soundstage.

I'll concentrate on the cabinet now to try and bring the midbass body back. The current cabinet is about 3L, I think I'll need to make a larger cabinet (4 - 5L) for the finished product to ensure I can get reasonable 100Hz - 200Hz response.

In my collection the two tracks that I notice standout with these drivers are Rhiannon from Fleetwood Mac "The Dance" (an excellent sounding disk) and En AranJuez Con Tu Amor from Il Divo "Ancora" (complex vocals). I have also noticed that the speed of these drivers allows good articulation of higher frequencies even on complex rock tracks which is hard to do with a wide range driver. Removing <100Hz from the driver also helps a lot I think.

Regards,
Dean

[Colin]

Hi Dean

I think you may need to go smaller, rather than larger with the cabinet. This may reduce the hollow sound. I'm using 2.4L cabinets and they sound, with good response to below 120Hz. As an experiment, how easy would it be to put something solid inside your 4L boxes to reduce the internal volume?

Regarding toe-in, this may be a function of using them alongside the monitor panel. They are not as sensitive to toe-in as the JX92. To test the imaging, try them away from the monitor (if you haven't already), with nothing between the speakers. They seem very sensitive to their surroundings (baffle reflections included) and this may operate over quite a wide range. (Normal dome tweeters have a fairly sharp cutoff from wide dispersion to the narrow dispersion of a matching mid-bass cone, whereas the Jordans will be a more gradual effect.)

I've been experimenting with matching the bass, still using a single passive inductor to JX92s. The JXR6 is definitely more sensitive than the JX92 but this could be the difference in impedance (the JXR6 is a 4ohm rather than 8 ohm load). I tried a couple of JX92s in parallel but ended up with slightly too warm a sound. Next step is to pad the JXR6 with a resistor and run it with one 92 again. I think I'm going to have to go active to successfully match these to bass drivers.

Used on their own, the JXR6s made a good pair of movie speakers last night, easy to tuck away out of sight either side of the movie screen. Four of these for surround wouldn't be hard to accommodate.

[deandob]

Hi Colin,

I tried stuffing the boxes with socks to reduce the volume size, and it sounds even more hollow. Some of the mid bass returned when I left one of the port vents open (previously stuffed with a sock - not ideal) so the problem is definitely the box. I can't see how having a smaller box will improve the bass.

I'm pretty happy with the soundstage with the speakers in line with the monitor and with no toe in. I tried putting a damp towel over the monitor to acoustically remove it and the soundstage (sound envelope) reduced slightly but the imaging (instrument location) improved slightly, but it could have been due to the lack of light coming from the monitor (which makes the brain concentrate on the auditory sensations - critical listening is best done with low or indirect light). I'm pretty happy with the effect of having the monitor in the middle, it seems to add to the soundstage, as Ted Jordan notes these speakers are best against a wall or flat surface.

Regarding bass matching, my sealed peerless XLS subwoofer with active filter / linkwitz transform seems to match pretty well for speed, although I still have problems in the 100 - 300Hz range that needs sorting. I would not run these speakers without bass support.

I'm not sure about using a resistor on the driver to reduce its sensitivity, how will this affect its attack when the amp wants to move the cone fast (higher currents will produce a larger voltage drop across the resistor) and affect damping?

I had a friend over tonight and he was amazed at the quality of the sound & the bass from such a small driver, but he didn't spot the subwoofer

Definitely better than the TangBand W3-871 but not by a long shot.

I've setup my saw bench for wood cutting next weekend....

Regards,
Dean