Adapting the TLb for the Jordan JX-92s

[dana]

I recently purchased a pair of Jordan’s JX-92s full range drivers for use an a transmission line project. I have constructed two different boxes to date, that were designed using Augspurger’s alignment tables and stuffing guidelines. I am really pleased with the sound of the Jordans, especially after playing them for a few hundred hours.

I am planning to build a third project, using the TLb as a guide and need a little advice scaling it for the 92s.

I have studied the TLb portion of http://www.t-linespeakers.org/projects/tlB/index.html and think I generally understand the concepts presented there, but I am having a little trouble understanding the author’s derivation of line length and box volume, especially as it relates to using the 92s in a similar design.

It appears that the TLb author’s derivation of volume could have been started as the calculation of a critically damped sealed enclosure for the Vifas. The line length seems to be based upon the requirement for a specific volume and density of fibrous material that could predictably produce a reduction in rear wave air speed in 3 feet, rather using a full quarter wavelength of line to do so. I suppose it is also possible that the author just wanted a loudspeaker that was a meter high and adjusted material, density and volume to do so.

Looking at the specifications for the Jordan JX-92s and the Vifa P13WH-00-08, reveals some striking similarities. Given that the Jordan is built by Vifa on their 5” platform it is not a complete surprise. However, the two specifications that are quite dissimilar are Fs and Vas.

Would it be reasonable to assume that using a pair of 92s drivers in the same box configuration as the TLb, stuffed with a more acoustically resistive material, such as Miraflex would work? The author notes in http://www.t-linespeakers.org/projects/tlB/appendix/density/dense_Fig2L.gif that 54” is the shortest length that can be used for the TLb, aiming for an F2 figure somewhere around Fs (45Hz). Since 54” is a little tall, could the drivers be mounted some distance from the closed end, rather than at the closed end? And if so would it be wise to use the 1/5 line length figure that Augspurger recommends, or could 1/3 of the line length be used effectively?

Dana

 

[pinkmouse]

Hi Dana

Well if you have the boxes already why not just try it!!

Ausberger's work is presented a little differently in the new edition of The Loudspeaker Cookbook, and it might be worth having a look at that to clarify things.

Since 54” is a little tall, could the drivers be mounted some distance from the closed end, rather than at the closed end?

Yes, but this does raise the frequency of the line,as the space above the driver acts as a chamber, rather than as part of the line.

IMHO, folded lines tend to work better than straight lines anyway, folding seems to further attenuate mids, and if you place angled baffles to smooth the airflow around the line, turbulence noise does not seem significant.

So maybe use the long line length, and just put one fold in it to make the boxes stand a reasonable hight off the ground.

 

[planet10]

Don't bother to try to figure a line based on how he did it in the TLb. You already have a line that you know works. Simply double the cross-section everywhere and mount the drivers in a bi-polar fashion. You may need to rejig the folding to make it into an apropriate platform.

Do you have a drawing of your existing line?

dave

 

[planet10]

Since 54” is a little tall, could the drivers be mounted some distance from the closed end, rather than at the closed end? And if so would it be wise to use the 1/5 line length figure that Augspurger recommends, or could 1/3 of the line length be used effectively?

Have you read Martin King's papers at t-linespeakers.org? The 2nd paper goes a long way towards explaining what happens when you place a driver in an offset position and the 3rd paper -- the ML-TQWT -- shows hoe offset & restricted terminus can be used to good effect.

A rough rule-of-thumb is that the driver should be placed at about a 1/3 of the volume from the closed end.

I would like to point out that Augspurger & King developed models simultaneously that give pretty much the same results. Augspurger uses an electrical analog, King a mechanical analog. Unfortuneatly Augspurger's sw is not readily available and we only have crude alignment tables. King's sw on the other hand is downloadable and can be used to create virtual t-lines with pretty good accuracy. This gives one the opportunity to play with combinations of geometry that affect the lines performance -- taper, offset, restricted terminus, stuffing placement & density.

dave

 

[dana]

Al and Dave, thank you both for your replies. I have read most of the documentation available from both Augspurger and King concerning TL design and construction. The t-line.org site invaluable in this regard.

To be honest, I was interested in the TLb because it seemed to be approaching the problem from a different angle. But given your feedback, the lack of coherent process data from the author of the TLb and the wealth of available information from Martin King, it seems more reasonable to explore a bipolar design using King’s tools.

Thanks again for the help.

Dana

 

[planet10]

To be honest, I was interested in the TLb because it seemed to be approaching the problem from a different angle. But given your feedback, the lack of coherent process data from the author of the TLb and the wealth of available information from Martin King, it seems more reasonable to explore a bipolar design using King’s tools.

The TLb is a brilliant design, but the author is, shall we say, somewhat eccentric. He & Martin cooresponded -- i got the feeling if they had been in the same room, blows may have ensued. 3 months of putting that article together were at times quite trying, but it was the 1st BIG article on my site, and if one can get thru the sometimes unwieldy prose (& catch the odd error, which i should really add some editor comment too) it has a lot of useful info in it. One of the biggest advanatges of the design -- no bafflestep -- was totally lost on the author when i pointed it out to him.

After doing some quite un-nice things on a couple of the forums he was soundly flamed and has dissappeared from any visible presence on the internet. I got the feeling he was very much a recluse before that and he may be more so, or have even, sadly, passed away.

One can correlate some of the stuff on the TLb with the 3-part article he did for Speaker Builder, but both King & Augspurger have provided much more understandable & accessible models for development of quarter-wave designs (Martin's sw can be used for all sorts of designs including the ones we pidgeon hole as TLs, Voigt pipes, and horns).

dave

dave

 

[planet10]

it should also be noted that martin now has his own web-site at http://www.quarter-wave.com/ with the most up-to-date stuff and i will be modifying that section of my site with content that is more of a commentary that is complementary to Martin's site.

dave