Anyone know anything about Transmission line(speakers)

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

strangeandbouncy

Well-known member
Joined
Aug 8, 2004
Messages
2,112
Location
West Sussex. UK
Hi,


  I am about to launch into my latest project.
  Some background. I have been using a pair of (loaned!) BBC LS3 5a speakers. I absolutely love them to bits, but am very wary of damaging them. The are as rare as rocking-horse-hens-teeth . . . . I always work at about 83dB(Fletcher-Munschen curves n all), so they are never strained. I desperately need to replace them, just for peace of mind! For those who dont know, they were designed by the BBC for use in OB trucks. They are very inefficient(80dB@1W@1m), are incredible for anything vocal, speech or singing, and possess bass far beyond the normal reach of a 6 litre sealed box, thanks to some clever crossover trickery. Effectively, there is a bass boost, or rather a gentle roll off against the rising response of the driver at the expense of efficiency. There is also an eq to cut the lump at @1kHz.
  When I worked for Roger Quested in the 80s, it was obvious the advantages of a 3 inch softdome handling the all important vocal range - 350Hz to 3kHz5 or so. True, male vocals extend much lower, but it is the upper reaches that are so important. One of the reasons that I think I love the LS3 5a is that one driver runs all the way up to this crucial point - approx 3kHz, and integrates wonderfully with the tweeter. At Quested, we used to make a nearfield with 4 x 5 inch drivers running into a 1 inch softdome at approx 1kHz6. Cool!
  With modern digital crossovers, this kind of design is really childsplay, especially with all the eq built in. I am about to inherit a really good analogue crossover to get the ball rolling. Mnay Many thanks Ruariri! You are a super star.
  I have found a number of potential bass-mids, most of which are Old fashioned, 1 inch voice coil, low power, smooth response devices. None of the 3 inch voice coil units look any good on paper! lumpy response curves, early cone break-up, and high distortion. Not worth the power trade off. Nah. I am very keen to explore the possibility of both closed box and maybe transmissionline. I want to avoid ported cabinets, due to the problems of port noise. I figure that if I am going to eq the bass to get more out of the system, eliminating, or suppressing the extreme impedance rise at the resonant frequency of the driver will be highly beneficial. Transmissionline could be the way to go . . .
  I am very aware that there is no proper equivalent of Thiele-Small for transmission line theory! Amazing that no-one has yet pinned it down! Truly Heath-Robinson I feel . . .

so my question is this. Why quarter wave . . . could eighth wave make a difference . . . It would be quite easy to build such a box, and not much bigger than the 6 litre cabinet of the LS3 5a.

And given that size is not necessarily important(!), why not half, or even full wave, (other than the sheer real-estate required, not to mention complication of design of such a long cabinet!)


  forgive the ramble, any input very gratefully recieved, Gentle People!


    Kindest regards,


      ANdyP
 
> at approx 1kHz6. Cool!

Yuck. Critical vocal area. I'd rather have 600-6KHz all in one.

Arguably you want a Full Range driver, one that sounds great alone, then fill the bottom and peak the top with boomer and zinger.

I like the old-old Fostex FE-103. It's just getting beamy at 6KHz, you might let a dome slump to 4KHz to fill the reverberant field. It's dead-flat {edit} to about 200Hz but at high levels must be crossed higher to avoid strain and IMD. Many of the large-coil 3" mids carry more power but have oddness at 6KHz then drop quite quick.

> I want to avoid ported cabinets, due to the problems of port noise.

Use a BIG port, bigger than cone, no noise. (This also leads to a VERY BIG box.)

> eliminating, or suppressing the extreme impedance rise at the resonant frequency of the driver will be highly beneficial.

How? Your amplifer does not care if the speaker takes current or not.

> Why quarter wave . . . could eighth wave make a difference . . .

Because the back-wave is out-of-phase with the front wave. Either lose it (sealed), or turn it around. For a single frequency at constant amplitude), you can turn-around with a half-wave delay re-entering the room. A closed-end quarter-wave pipe can return the backwave to the cone a half-wave later, which aids the cone (at that freequency). An eighth-wave won't do that.

All "wave length" resonators resonate. You have a comb of big resonances. Awful. You fix this by adding damping. Fortuitously, damping works better at high frequencies. If you get just-right, you hardly-damp the lowest mode and it reinforces that frequency, while the next-up resonance is sorta-damped and sorta-inoffensive.

And if your pipe is as small as a small-box port, it has the same vent-noise; OTOH if as big as the cone and tuned low the total bulk is far bigger than vented or sealed of similar performance.

Transmission line to infinity avoids these problems. It is infinite bulk. It also does not allow you to trade-off response below one frequency to favor response above that frequency, the major design choice of sealed or vented design. (You put a 20Hz Q=0.4 woofer in a box that shifts it to 50Hz Q=1, you get a lot more 45Hz-80Hz and less 20Hz-45Hz.)

You can spend your life computing and stuffing lines. And then you still have to integrate the main part of the spectrum, crossover and mid. If you have EQ and ample cone area, and want uncolored bass, do a large sealed box, EQ it, be done with it. A Fifteen in a 10 cubic foot box can be reasonably efficient and reasonably uncolored 30Hz to 600Hz, and the slump below ~~70Hz can be easily lifted without pipe or port resonances. (See Small QB3; a one opamp 2-pole filter is all you need.)
 
I did quite a bit of reading on transmission line cabinets some years ago.  I seem to remember that the ideal transmission line should completely kill the rear wave from the speaker  But in practice slightly reinforces the front  wave.    I also seem to remember that a speaker has 2 resonant/impedance peaks, & that the transmission line cabinet pushes them lower somehow, so that the lower one is below where it's a problem.

All this makes the bass response fantastically smooth.  It's kind of deceptive at first, because most people are used to hearing the one note bass where cabinet makers boost at the low frequency cut off point to give an impresssion of great bass extension, when in reality it's not that low.  the response went so low that you could tell if you forgot to use the HPF filters due to serious thumping from the bass drivers.

At the studio I was involved with there were a pair of IMF TLS80.  They sound fantastic, & were jaw droppiing to the clients.  The TLS50 is also a good cabinet, but uses a round woofer rather than the flat oval kef driver that the tLS80 uses.

I also used to own a pair of B&W DM2 which are an 1/8 wave transmission line cabinet.  They also sounded good, but not really in the same league as the IMF's.  I appreciate what Paul is saying about the 1/8 wave cabinets, but the DM2 sound very good for the size, I guess it's a case of theory & practise !

If you're not too worried about the cosmetic condition of the cabinets you can get the IMF transmission lines for 100 quid upawards, they might be a great way to get something quite cheap where you could experiment with some more modern tweeters.  The DM2's I've seen go for 25-30 quid.
 
jericho_rendering.thumbnail.JPG


Here's a monitor if you want to listen a little louder, or from a 1/4 mile away  (150 dB spl ).

http://www.danleysoundlabs.com/pdf/JH90%20spec%20sheet.pdf

Actually Danley is doing some innovative stuff with multiple drivers feeding a common horn structure, while this one is clearly designed for large scale sound reinforcement.

JR
 
Hi Guys,


  thank you so much for your replies. It is SO appreciated, and I mean that with all sincerity, Folks!

  Jackies, thanks for the link. I bin Googling for England, believe me! Very little info out there, and I was wondering if any of the Learned Brethren had any further insight. . . .

  John, Id love to see the meterbridge that would accommodate THAT as a nearfield!
 
PRR, Thank you so much for your insight. Most helpful. Incidently, you Yuk at the 4 x 5 Inch turning over at 1k6 . . . just goes to prove that there is no such thing as a perfect loudspeaker! multiple drivers gave a staggering transient response, high power handling, and actually sounded great. - Really! there is always a trade off somewhere. But in this regard I am with you. I want to run up to at least 3kHz with MY speaker.
 
Rob, I remember being floored by a pair of TLS80s myself! I fear that the days of 100 pounds or so are gone . . . there is a pair on Evilbay right now for . . . . 1300quid. I am really interested in a 2 way with a small, lightweight, responsive 5 inch like the KEF B110 in the LS3 5a that can be forced to eek out a slightly extended bottom rather than something quite so fullrange!(and huge)
 
Hey Andy,

I've got the LS3/5's bigger brothers, the LS5/9's. They cost me all of... oh... nothing.  :)

I've used the 3/5's and agree that they are awesome. Very clear image / sense of space. These were Harbeth-made versions at work.

One of my work friends has cloned a few sets of PMC transmission-line jobs. I'm not sure if he'd have kept any measurements or not. They sound pretty good though.
 
strangeandbouncy said:
Rob, I remember being floored by a pair of TLS80s myself! I fear that the days of 100 pounds or so are gone . . . there is a pair on Evilbay right now for . . . . 1300quid. I am really interested in a 2 way with a small, lightweight, responsive 5 inch like the KEF B110 in the LS3 5a that can be forced to eek out a slightly extended bottom rather than something quite so fullrange!(and huge)

Disagree, the ones on ebay at the moment are really expensive, and that's probably the reason they've been on ebay for about 4 months.  They should be maybe 400 quid.  My mate got a pair of TLS50 for about 130 quid just 4 odd months ago, in very good nick with good veneer.  The bass driver on the TLS50 is 5-6" & round as well, which is roughly what you're after.
 
While the example speaker from Danley was too big to be serious, they have an interesting technology where multiple drivers feed a common horn without the on/off axis phase issues associated with typical multi-way boxes. There is also another company using DSP to correct for throat distortion in horns.

I suspect a hifi version of this technology could be pretty interesting.

JR
 
JohnRoberts said:
While the example speaker from Danley was too big to be serious, they have an interesting technology where multiple drivers feed a common horn without the on/off axis phase issues associated with typical multi-way boxes. There is also another company using DSP to correct for throat distortion in horns.

I suspect a hifi version of this technology could be pretty interesting.

JR

Interesting, but surely with the SPL levels achievable in horn speakers multiple drivers are a bit of an overkill for studio monitoring.    I remember being listening to a mono 8' 1930's theatre horn with a single Lowther full range driver & being absolutely staggered by the volume considering the meagre 3-4watt amp being used.  I would also say that they are some of the very best speakers I have heard.  Weird when you consider that they are a square section & flat pack portable, the sides being held together by bolts & wing nuts.  The main problem is the length the horn needs to be for really low bass, I seem to remember that 18' gets you really low.


I have a wireless world magazine somewhere that has some really neat folded horn cabinet designms in it.  Again the trouble being the 4'x3' front fascia.  WHo has room for that !!
 
Hi Rob,



  Wierd . . . Whilst looking at the Fostex full-range that PRR mentioned, Ive just been looking at some Lowther drivers, and some Audio Nirvana that are really similar. I have never heard any. Unbelievably efficient, run very low. I simply cannot believe that they can sound any good at the top end! COne breakup is very visible in the frequency plots. They look amazing up to 2kHz or so. Also, with that degree of efficiency distortion must be running pretty high . . . . Still, I am very curious to try some out.


           http://www.commonsenseaudio.com/


 has the Audio Nirvana. I am only interested in the Alnico ones. That degree of rising response in the other versions is a no-no for me. Cabinet size is not too much of an issue for me, as a transmission line could work perfectly in a studio situation, (assuming they are happy in a transmission line) - no need for stands!


    Damn, another avenue to follow . . . .


  Kindest regards,


      ANdyP
 
The thing with the Lowther drivers is that they have very lightweight cones.   They are almost cartridge paper.  Years ago there used to be a vintage hifi specialist who had a workshop on the platform of Worthing Station of all places.  I remember being there when he was replacing the rubber surround on a pair of  Lowther drivers.  I couldn't believe how floppy they were without the support of the surround.  

Like you said they probably haven't got massive top end, but they do sound great, especially on stuff like Blue Note Jazz.  Couple them with a QUAD II system ...peachy !

In my experiences, Transmission line speakers like a big sized room, so they can stretch their legs a bit.  This may be a function of the smaller rooms I've heard them in not being very "sorted" in  the bass department.  But, in a big room the bass seems to be better coupled to the room.  Not as well as a theatre horn couples it but damn good !
 
Hi Rob,



    Merry Noo Yer . . .


    Trouble is my goalposts keep moving! I still want a speaker with a small, fast, smooth driver. I wanted to play with transmissionline to facilitate a smooth eqing of the bass into a 5 inch driver rather than make fillings fall out and trousers flap . . . PRR is I am sure correct in his assertion about using a sealed box in this case. If I can achieve realistic low bass ( not that I believe too much about manufacturers curves!) with something like a Lowther, then maybe I should give it a go! Can always run it into a tweeter. Expensive driver to buy and discard if I dont like it, compared to the 40 pound Monacor 5 inch I had lined up. Also looking at Beyma and Tang Band, both less than 40 pounds! I will also talk to AudioTechnology, whom it seems, tailor a driver to your needs. I have gone right off all the 3inch voicecoil stuff. Powerhandling at the expense of just about everything else it seems.

  Ho Hum


    ANdyP
 
JohnRoberts said:
While the example speaker from Danley was too big to be serious, they have an interesting technology where multiple drivers feed a common horn without the on/off axis phase issues associated with typical multi-way boxes.
Not really a new concept. In the 80's EV promoted their Manifold trade-mark, used in the MT4 system, where 4 compression drivers were bolted to a single horn via a 4in-1out contraption, in the high-mid and treble sections. Even then, they couldn't patent that because there was ample evidence of prior art (I think Altec had 2-into-1 adapters from immemorial times).
There is also another company using DSP to correct for throat distortion in horns.
The one I'm aware of is from Klippel, a company which produces the most complete loudspeaker design, test and QC suite. AFAIK, the distortion correction software has never been successfully implemented in commercial equipment, due the the enormous DSP power it requires and the subjectively disappointing results.
 
abbey road d enfer said:
JohnRoberts said:
While the example speaker from Danley was too big to be serious, they have an interesting technology where multiple drivers feed a common horn without the on/off axis phase issues associated with typical multi-way boxes.
Not really a new concept. In the 80's EV promoted their Manifold trade-mark, used in the MT4 system, where 4 compression drivers were bolted to a single horn via a 4in-1out contraption, in the high-mid and treble sections. Even then, they couldn't patent that because there was ample evidence of prior art (I think Altec had 2-into-1 adapters from immemorial times).
I need to be careful as I am outside my comfort zone. If this is indeed the same physics as compression driver manifolds, that is indeed old news. It sure looks different to me.  That said, it doesn't mean some wingnut didn't throw different bandpass drivers onto a horn years ago, with less than optimal results.
link to horn white paper http://www.danleysoundlabs.com/pdf/danley_tapped.pdf

Danley has other technology to manage the summing of multiple same bandpass drivers that would be competing with CD manifolds ("Paraline" & "Shaded lens").  I don't know enough details to discuss intelligently. But I do know Danley is all about documented measured results, so not smoke and mirrors.

There is also another company using DSP to correct for throat distortion in horns.
The one I'm aware of is from Klippel, a company which produces the most complete loudspeaker design, test and QC suite. AFAIK, the distortion correction software has never been successfully implemented in commercial equipment, due the the enormous DSP power it requires and the subjectively disappointing results.
IIRC Gunness (David) had to develop his own hardware to perform the math involved, but his "Gunness Focussing" works and has been commercially merchandised by EAW for several years. http://www.eaw.com/technology/proprietary/gunnessfocusing/
gunness.gif

David has since moved on to another company (Fulcrum Acoustic), but i doubt he is finished innovating. 
======
FWIW these guys are working with big SPL speakers. This is the heavy lifting for loudspeaker design and I apologize for the thread veer since this is not a path to studio reference monitors where efficiency is low on the list of concerns.

To make this veer complete, here is a link to a fireworks file (using one of Tom's experimental mics)... Living in the deep south where new years eve fireworks is a tradition practiced by my neighbors this seems appropriate. However caution, this wav file has something like a 40dB crest factor (Very loud peaks), so be careful about setting your playback level too loud and trashing your speakers if they can't keep up.

http://www.danleysoundlabs.com/Finale4.ZIP

Happy new year.

Sorry if it seems like I'm pimping these guys, I just respect their work.

JR
 
JR,
I'm familiar with the Danley Synergy horn; the patent refers to the way diferent types of loudspeakers (LF, MF & HF) are coupled to a single horn. Again there are a number of examples of prior art.
The tapped horn concept is interesting. From the description, I don't see how it is different from the well-documented "dual-interactive chamber" concept that was used commercially in the 1990's. The real-world advantages were largely exceeded by the constraints it put on design briefs. I don't think many manufacturers use that today.
I don't want to diss this company, they have their own set of design choices, as any other manufacturer, and they try to make them attractive to their customers, but I hardly see any real innovation there.

The Gunness focusing system started as an answer to Wolfgang Klippel's white papers in the AES journal. Nexo had bought from Klippel the rights for implementing the results in their first digital loudspeaker processor. It never happened due to the outrageous complexity at the time, and the tepid reaction of sound engineers. Klippel's algorithms are extremely complex in that they map frequency, time and level.
EAW have implemented somewhat simpler algorithms, which require less DSP power. I must say it's a long time now I haven't heard an EAW rig, but I'm convinced that, if the difference was really that significant, every manufacturer would be pushed to incorporate such a technique in their products.
Pre-distortion is a very difficult matter, in vinyl cutting it just didn't catch, neither in tape recording.
 
Multiple identical drivers on one horn is very old, older than Altec because they got it from Western Electric, and thed'd been doing it a long time.

Different drivers on one horn, I want to say is not new. I want to say "J W Davis" made some, but that line is not shown on their current website. There is of course the bad example of piezo horns mounted in the sidewalls of a midbass horn.

Danley's plan does seem novel. It would take a LOT of computer power or elaborate testing to show it is not a nest of phase-snakes; they seem to have done the homework, and it seems "simple" in hindsight.

Danley's tapped-horn does seem novel. Basically they broke the rule that all 50+% efficient speakers at the same F-3dB are the same size, but box speakers can be compromized smaller and horns don't compromise... Danley found an effective cheat. It could be argued that it is a bass-reflex below and horn above, an old-old concept (older than electric speakers)... but a *careful* patent disclosure can focus on the angles which make it work well and are novel.


> looking at the Fostex full-range that PRR mentioned... Unbelievably efficient, run very low.

?? The FE-103 is not efficient and does not run deep. 1.3% at 200Hz-2KHz.

> I simply cannot believe that they can sound any good at the top end! COne breakup is very visible in the frequency plots.

Plots, schmotz. Speaker measurement is full of artifacts and always incomplete. A 3" cone shows peak/dip at 2KHz _ON_ axis and nowhere else in the room (even 2" off axis a foot away). OTOH few plots show off-axis, only at a few angles, and nobody shows the total acoustic output (which has a strong influence on in-room sound).

IMHO, an Eight is OK over the vocal range only if you want Projection. Clarity in a too-reverberant space such as a cement lecture hall. That's too much for almost any domestic-size critical listening. Midget-radio Fours are killer vocal sources; they just don't have ballz. Any $49 woofer can do ballz up to where a Four can manage. The Four narrows past 4KHz-6KHz, have fun selecting a dome.

> with that degree of efficiency distortion must be running pretty high . . . .

Huh?

True, nasty speakers can be loud and distorted.

But if you get efficiency honestly, distortion should be down to air-overload. WEco and Meyer proved that long ago.
 

Latest posts

Back
Top