question about TL speakers for TLS Guy

[Swerd]

First of all I’d like to welcome TLS Guy to Audioholics. I’ve recently noticed a number of your posts around here and have been impressed both by your knowledge of audio and by your obvious DIY skills. Although it isn’t the most active subject here at Audioholics, I have been beating the drum for the benefits of DIY speaker building for some time. I always try to tell prospective builders how important a good crossover is. Your presence here will make a welcome addition.

I have a question for you about transmission line speakers and their response relative to a wall behind them. Conventional wisdom has it that when speakers are close to a wall behind them, their bass response is louder due to reinforcement from bass reflected off the wall. When speakers are moved further away from a rear wall, the bass response becomes less loud, but often (depending on the design of the speaker) more clear sounding.

I have some new MTM speakers that are in quarter wave transmission line cabinets. See my post about them. The opening on the lower rear of the cabinets is about 4" in diameter, not the full area of the pipe, and as a result they are considered mass loaded TL cabinets. They were designed according the method of Martin King.

These speakers were also designed to compensate for the midrange baffle step response that typically comes with an 8" wide cabinet, and as a result, the maker recommended placing these speakers no closer than about 18-24" away from the wall behind them. I tried moving them closer than 8" from the rear wall, and I was surprised to hear the bass sounding more solid and better in general than it was farther from the wall. This defies the conventional wisdom that I had thought applied to all speakers independent of the cabinet bass alignment.

Because you have some experience with TL speakers, have also you noticed this? Is this observation common to other TL speakers, or specific to mine? And, do you have any thoughts or comments about the “different bass sound” of TL speakers?

 

[Seth=L]

I know that the purpose of the transmission line is to act like a bass reflex in terms of how the driver reacts, but the difference is what happens at the port. Bass reflex output audibly at the port where as transmission line enclosures reduce output at the port to the minimum. If done right the port should not output hardly any noise. The speaker itself may be reacting to the placement in the room instead of the port causing its own reaction.

I know that probably doesn't make much sense, but that is the extent of my knowledge (or lack there of) on transmission line enclosures.

 

[TLS Guy]

TLS placement

Martin Kings knows TLS speakers. I suspect you have a good set. First of all Seth=L is not correct. The output from the port should be significant, and generally augment the bass 4db from the F3 point to around 100Hz.

In a good line the enhancement is uniform and more extended than for reflex designs. The TLS enclosures do not advertise their bass. It appears effortless and full, with no boom, when called upon by the program material. People who have good TLS speakers, seldom part with them and if they do they regret it.

As I explained in a previous post, a TLS is a highly specialized organ pipe. Like an organ the port output couples to the room exhibiting the acoustic phenomenon of encircling. That is to say it couples to the room much more uniformly than other designs. Therefore TLS speakers are much less likely to set off room resonances. I have no experience with designs with the reduced port size you describe. It is possible this may alter the room coupling to a degree. As you can see from my pictures the ports in my designs are not restricted.

http://mdcarter.smugmug.com/gallery/2424278#127083295

If in your room the speakers sound better with a little bass reinforcement from the corner, then that is where to place them. This is not uncommon.

 

[Swerd]

Thanks for your comments. These speakers are the first TLs I've spent any time with and I am often surprised at what I hear. They are different enough to make me question every thing I thought I had known about bass sound from standard sealed or reflex designs.

...Like an organ, the port output couples to the room exhibiting the acoustic phenomenon of encircling. That is to say it couples to the room much more uniformly than other designs. Therefore TLS speakers are much less likely to set off room resonances. I have no experience with designs with the reduced port size you describe. It is possible this may alter the room coupling to a degree...

I think that describes what I had originally asked about. It does seem to couple to the room more uniformly. I haven't experimented extensively, but they do seem to be forgiving about room placement.

In a good line, the enhancement is uniform and more extended than for reflex designs. The TLS enclosures do not advertise their bass. It appears effortless and full, with no boom, when called upon by the program material.

That says it very well. TLS don't advertise their bass. That is why, when the material does call for it, the bass seems so surprising.

People who have good TLS speakers, seldom part with them and if they do they regret it.

You got that right!

 

[TLS Guy]

Thank you for your words of welcome

I should have thanked you for your kind words of welcome in my previous post. They were very kind and most appreciated.

From your comments, you are obviously one of the fortunate few with a good set of TL speakers.

I know we have our work cut out to encourage more speaker building. We need to reincarnate Gilbert Briggs. It seemed to me that GAB had everybody building speakers back in the 50s. He got me started age 7! As I have stated, now could not be a better time for people to catch the speaker building bug.
Best Regards, Mark.

 

[WmAx]

A TL speaker couples to the room just as any other monopole sound source. If one wanted to draw a comparison to an organ pipe; this is a Hemhlotz resonator. The very same device called a 'port' on a standard ported bass alignment.

Specific room coupling trends as noted by subjective statements of people in regards to various monopole bas speaker types appears to primarily based on perception of other variables which may typically be a trend for a specific speaker, but are not often what the people assume. For example, a constantly reducing, smooth roll off taper as frequency lowers can help do two things: (1) reduce room mode interactions (2) alter frequency response in such a way as to be perceived as 'tight' - so it seems the common response appears to be based on the response of people on forums.

One can modify the response curve of a ported system with sufficient extension to mimic the 'sound' of any monopole bass alignment with digital signal processing.

-Chris

 

[Seth=L]

First of all Seth=L is not correct. The output from the port should be significant, and generally augment the bass 4db from the F3 point to around 100Hz.

I read it in a snake oil book, go figure.

Oh well.

 

[TLS Guy]

TL speakers do behave differently in practice

A TL speaker couples to the room just as any other monopole sound source. If one wanted to draw a comparison to an organ pipe; this is a Hemhlotz resonator. The very same device called a 'port' on a standard ported bass alignment.

Specific room coupling trends as noted by subjective statements of people in regards to various monopole bas speaker types appears to primarily based on perception of other variables which may typically be a trend for a specific speaker, but are not often what the people assume. For example, a constantly reducing, smooth roll off taper as frequency lowers can help do two things: (1) reduce room mode interactions (2) alter frequency response in such a way as to be perceived as 'tight' - so it seems the common response appears to be based on the response of people on forums.

One can modify the response curve of a ported system with sufficient extension to mimic the 'sound' of any monopole bass alignment with digital signal processing.



-Chris

Dear Chris,
That is all true, however lines do seem to be much less likely to sound boomy in a given room when others do. I do be believe that the tapered lines broader frequency range of bass enhancement, versus the narrow range from reflex ports contributes to this. Also the port area of lines is generally significantly larger than that of reflex ports. This allows for free unimpeded air displacement. However I agree that Auspuger's work has shown much more in common between line and reflex loading than we once thought. However the bass of lines and other designs do sound different, and to me much more natural, and worth the effort.
And yes you definitely can get 3 to 4db of bass lift from line loading all the way down to the 20 Hz range, with only 1 db of ripple. Others have found that also. That is not snake oil like exotic cables.
Best Regards, Mark.

 

[WmAx]

Dear Chris,
That is all true, however lines do seem to be much less likely to sound boomy in a given room when others do.

And it appears to all come down to frequency response.

I can replicate the 'signature' of any bass alignment, given the correct tool(DSP equalizer) and an extended response subwoofer of low non-linear distortion.

-Chris

 

[TLS Guy]

I doubt it

Dear Chris,
I don't think you can. There is just something spacious and effortless about a good TLS. It just can't be replicated that easily.
I have a close friend with the latest B & W 200s. The sound stage is similar, but he also agrees the bass of my rig is more realistic and natural.
Best Regards, Mark.

 

[WmAx]

Dear Chris,
I don't think you can. There is just something spacious and effortless about a good TLS. It just can't be replicated that easily.
I have a close friend with the latest B & W 200s. The sound stage is similar, but he also agrees the bass of my rig is more realistic and natural.
Best Regards, Mark.

This is an easily measurable/quantifiable area. I know of nothing inherently unique that could cause any special coupling or interaction as compared to any other monopole topology. Now, when you move to something such as cardioid or dipole bass systems, then unique room interactions(usually less room interaction due to the perpendicular hemispheres cancelling - thus reducing sources of room excitation) do occur, as compared to monopole systems.

I can (and do) adjust the target curve of the bass response of speakers to replicate the 'signature' sound that other bass alignments are typically known to have. In addition, I have carried out extensive ABX tests on electronically modified signals in order to determine audibility of various factors relevant to this issue.

-Chris

 

[jaxvon]

The most natural bass sound I've ever encountered was reproduced in a small, ported enclosure. There is nothing special about a TLS enclosure that ensures natural bass. The best way to achieve "natural bass" is flat frequency response, low distortion, a wide listening area, and low power compression. All of these things are easily achievable in a ported enclosure, or a sealed enclosure, or a passive radiator. The means is simply proper design.

 

[annunaki]

I think the "natural bass" is, in effect, a subjective term, is it not? Natural bass to one may be completely different to another's perception of natural bass. "Natural bass" cannot be measured. Although a generally accepted or averaged "natural bass" response curve could be measured and therefore implemented regarless of the enclosure type.

I would be willing to bet that a "natural bass" curve would probably start rolling off at around 40hz or so at -6db an octave. This could easily be achieved with an eq or, in the case in point, enclosure design. It could also be simply a lowering of the bass range (20hz-100hz) in overall output level below the reference level.

 

[TLS Guy]

Controversy

I seem to have started quite a controversy. I try to keep things as easy to understand as possible in a forum like this. However we have got to the point where we have to put some very hard facts on the table.

The issue is complicated because the sound of large transmission line speakers with f3 points below 40 Hz have been heard by so few people. The last person to manufacture large lines was John Wright of TDL. He has been dead for about 20 years now.

The other problem has been variability in the performance of lines. In the mid seventies Thiel and Small published a working and correct mathematical model of the relationship of the driver properties to closed and vented boxes. Until the work of G.L. Auspurger and his papers of 2000, we have had no such model for the TL speaker. I know his work is controversial to some, but I believe and have evidence for the fact his model is correct.

I'm going to take the famous KEF B139 bass driver to illustrate the differences in closed box, vented and TL alignment according to the model of Auspurger and his published tables.

I chose the B139 for the following reasons.

It has a good reputation, and in its later versions was very consistent.

I have been familiar with the driver for about 40 years.

Successful commercial designs have appeared over the years using this driver in closed box, vented and Transmission line enclosures.

I have personal experience of the driver in all three alignments. Further I have in regular use in my home speaker systems which use this driver in a vented QB 4 reflex and a transmission line.

The KEF B 139 for the purpose of these comparisons, has a free air resonance of 25 Hz (Fs). It has a VAS (equivalent volume) of 127 lt which 4.48 cu. ft. It has a Qts of 0.37.

For the closed box and vented alignments I have been assisted by Box Pro.

Let us take a look at the closed box alignment. A box volume of 0.923 cu.ft gives us the lowest possible F3 of 79.66 Hz with an acceptable Qtc of 0.75. The roll off below 79.66 will be 12 db per octave (second order). So the 12 db frequency is 35 Hz and the 24db frequency is 17.5 Hz.

Now let us look at the Qb 4 vented alignment.

A box volume Vb of 3.538 cu. ft. with a vent three inches in diameter, 6.307 inches long with two flared ends gives us an F3 of 55.6 Hz. Roll off is 24 db per octave below F3 (fourth order). So the 12 db point is 20 Hz. the 24 db point is 15 Hz.

The lines for closed and vented intersect at the minus 30 db point of around 12 Hz.

Now lets take a look at the transmission line alignment.

A reverse tapered pipe, 6.37 ft long with a volume of 4.48 cu. ft will give us a pipe resonance of 44.4 Hz, and an F3 of 36.1 Hz for the B 139. There is controversy as to whether a line rolls off second or fourth order. My data shows a properly constructed line rolls off second order to the Fs of the driver. So this line will only be 12 db down at near 20 Hz Also this alignment has a maximum 1db of ripple to the F3. So from there the output is boosted by 3db as far up as 110 Hz. At 60 Hz there is a lift of 3db over the vented and 6db over the closed box.
So here already you can see a highly significant and audible improvement in the TL alignment over the other two.

The only downside is that we have an enclosure significantly larger than the others, that is more complex to build. Otherwise the the TL is favored in all other respects.

You could conceivably equalize the closed box to the TL, but it would take huge power, stress the driver and greatly increase distortion. You can not equalize a Qb 4 reflex alignment. It decouples from the box and you get large useless cone excursions.

The other issue is that because of their construction lines tend to be self bracing. Also there are only high pressures at the closed end of the pipe. TL speakers seem to excite cabinet resonances less than the other alignments, which may be a factor in their perceived low coloration.

Now the port area issue. The loudspeaker cone has to provide the entire output in the closed box example. In the reflex, the port area is 4.78 sq inches. The port output augments the driver output at and below F3 and Fb.
My KEF line has a port area of 56 sq. inches!. Pipes have an anti node of air displacement at the open end, and you can feel silent air flow at F3 and below. I believe this does couple to the room like an organ pipe.

Here is my reflex with the KEF B 139.

http://mdcarter.smugmug.com/gallery/2424105#127080849

Here is the TL which uses two B 139s

http://mdcarter.smugmug.com/gallery/2424008#127077020

These big bass lines have an F3 of 27 Hz. Even the little center channel line has an F3 of 47 Hz

http://mdcarter.smugmug.com/gallery/2424008#127077317

So what is the effect of all this.

I went to school at Downside Abbey, Stratton-on-the-Fosse, Somerset UK.

The first time I ever felt my internal organs vibrate was when Father Cyprian really opened up the huge Compton Organ in the Abbey Church.

This organ is faithfully recorded in volume 3 of the complete works of Percy Whitlock. Priory PRCD 942. Percy Whitlock had a particular fondness for this organ. With these speakers the organ does the same thing again in my studio, and it only needs to play at a realistic level. You really can hear the 32 ft pedal stops move and articulate. It sounds so close to how I remember the sound sitting next to Father Cyprian turning pages as makes no difference.

There is a beautifully recorded SACD, Motette MOT 13254. This is about a perfect a recording as you could get of the Klais organs of Cologne Cathedral.
The organ notes hang high in the distance, the huge bass pipes articulate and sound with full natural authority. Now this is something when played to visiting audio buffs makes their jaws drop and their eyes pop open wide.

Now I have built just about every type of enclosure known to man for moving coil drivers, and I can tell you nothing comes close to well set up transmission lines. Only they can come close to letting a moving coil driver reproduce sound with such power realism and authority. Unfortunately very few will ever have the opportunity to confirm that for themselves. Unless just may be I spur a new fervor of speaker building activity.

What about chamber and solo instruments, such as voice and guitar you may ask. Never a fear these are reproduced most naturally without chest emphasis or over rich low end bloom. As I have stated before unless the program has powerful bass, good lines don't advertise the fact they have the goods a plenty.

I admit that to build large lines you need a tolerant spouse. However my wife of 38 years tolerates it on grounds of the superior results.

In summary do you need big speakers to have a good bass? The honest answer no, but it helps. If you ask the question, are large structures essential for the powerful bass authority I have described, than the answer is yes!

 

[annunaki]

Transmission lines more or less seem to be ported enclosures with a large tapered vent with high amounts of surface area. The essentially seem like extended bass shelf designs with a large port surface area. Because of the extra port area, they are able to move more air as it is an extra, large, radiating surface.

 

[TLS Guy]

True pipes

Transmission lines are true stopped pipes. The physics is identical to the stopped pipes of the pipe organ. Unlike a Helmholtz resonator the pressure is not uniform in the structure. The pipe has an anti node of pressure and a node of displacement at the closed end. At the open end there is a node of pressure and and anti node of displacement. And believe me there is air displacement.

Now the organ builder wants harmonic richness. The speaker designer does not. The organ builder wants a narrow fundamental. He wants one note. The speaker designer wants a broad fundamental with minimal ripple, to uniformly augment the driver output over as much of the last two octaves as possible.

The taper widens the fundamental, and makes it broad unlike the narrow one of the organ pipe.

Stopped pipes generate odd order harmonics, the third harmonic being the most prominent. This is controlled two ways. First by placing the driver at the node of the third harmonic to lessen its intensity. The line is uniformly filled with Dacron, so it just fills the space without compressing the fiber.

What is new from recent work is the understanding that there is a relationship between the Vas of the driver and the volume of air in the pipe. This relationship is not nearly as critical as with the reflex enclosure, and within reason liberties can be taken.

The credit for first conceiving the idea of using a pipe to load a speaker has to go to the great audio pioneer Paul Voight. He published the principle of the quarter wave pipe for loading a speaker. He is most famous for his corner horn developed in the thirties. This is generally regarded as the worlds first hi fidelity loudspeaker. Voight"s ancestry continues today in the line of Lowther horn loaded loudspeakers. The late Paul Chave picked up the mantle from Paul Voight.

 

[WmAx]

I seem to have started quite a controversy.

The only issue here it seems, is your subjective statements/interpretations made with no proper controls of perception bias. That is, claims of what is clearly an omnipolar bass source, having some special room loading property as compared to other omnipolar bass sources. The only quantifiable end result difference(s) come(s) down to the frequency response difference(s) and how a different frequency response is audibly different.

You could conceivably equalize the closed box to the TL, but it would take huge power, stress the driver and greatly increase distortion.

Today, we enjoy the wide availability of highly linear woofers that have large thermal dissipation and extremely linear motors/suspension. It is trivial to design a small boxed sealed subwoofer today with low distortion down to 20-25Hz range. Amplifier power is also dirt cheap today. A high quality 2000 WRMS amplifier with low distortion can be purchased for under $400 today. One could argue it is a waste of driver potential -- but if the smallest size enclosure for an application is needed -- this is the optimal configuration.

You can not equalize a Qb 4 reflex alignment. It decouples from the box and you get large useless cone excursions.

You can equalize a ported or passive radiator system. You simply can not push up energy below the port tuning frequency. [But there is no need to do such, based on what I could determine, based on control ABX experiments simulating various cut off slopes after a predetermined frequency, simulating different alignments, in a controlled scenario.] It is again, trivial, to design a woofer in a ported or PR system that has a flat response to 20 or 25 Hz. You can taper down/roll off the existing energy distribution to replicate that of a critically dampened bass alignment.

The other issue is that because of their construction lines tend to be self bracing.

I suspect that the amount of inherent bracing of a line is still far under the required level to prevent audible panel resonances.

Also there are only high pressures at the closed end of the pipe. TL speakers seem to excite cabinet resonances less than the other alignments, which may be a factor in their perceived low coloration.

Stereophile is useful in some cases. In this, it is interesting in that it has in it's archive, panel resonances amplitude measurements from centralized points on the cabinets of two TL speaker systems(PMC IB-1S and MeadowLark Swift). The panel resonance amplitude is no different than would be expected from a standard cabinet system with moderate bracing. I would theorize, that more likely, the reduced perceived coloration comes from the required high level of acoustic dampening materials used in any TL, as compared to the usually insufficient acoustic damping material used in most standard boxed designs. The standard materials as are commonly used do not have a sufficient bandwidth with high co-efficient. The TL requires a high high amount of material, removing this as one problem. The panel resonances are likely worse in a TL, due to the relatively large cabinet radiation surface area. This is likely partially mitigated as the bracing(not intended to be bracing usually - but required as the line channels) is increased over most designs. Yet, I highly doubt it is anywhere near sufficient in creating a true non-audible cabinet surface. I have experimented heavily in this area as well, and gone to the trouble in certain cases, ABX models recorded in a small semi-anechoic chamber under different conditions and to assess the specific types of coloration involved under blinded conditions.

I have come to a point where I found a nearly perfect non-resonant reference device that I compare under sighted conditions for quick/rough assessment. Now that I have trained myself under proper controlled conditions, it makes sighted comparisons more reliable -- though sighted conditions are still not substitute for proper blinded evaluations. The sighted conditions as I said, are only intended to make efficient/quick assessment when more controlled conditions are not practical in a given situation.

-Chris

 

[WmAx]

Transmission lines are true stopped pipes. The physics is identical to the stopped pipes of the pipe organ. Unlike a Helmholtz resonator the pressure is not uniform in the structure. The pipe has an anti node of pressure and a node of displacement at the closed end. At the open end there is a node of pressure and and anti node of displacement. And believe me there is air displacement.

There are both stopped and un-stopped pipe organs. The stop, when used, is primarily intended to reduce the length of pipe required to arrive at a specific frequency. Yes, a stopped pipe does change the behaviour internally somewhat, but it is essentially the same end result.

-Chris

 

[TLS Guy]

Different types of pipe

Organ pipes are generally open pipes, and radiate even harmonics. an organ builder calls a pipe open at both ends, an open pipe. When an organ builder refers to a stopped pipe, he is not talking about the organ stops, but wooden bass pipes that are closed at one end. That is what an organ builder means by a stopped pipe. Only the most expensive large instruments have stops that draw on open 32 ft pipes. The wooden stopped pipes can do it in half the length. The downside is that they are not as harmonically rich.
The TL speaker is a version of the organ builders stopped base pipe, i.e a wooden pipe closed at one end.

 

[Swerd]

This thread seems to have drifted away from my original question. While I can't comment intelligently on Chris's various points, I do not see how they are relevant to my question. Chris, feel free to convince me otherwise. And I also object to the pointless barb about TLS Guys previous statements:

The only issue here it seems, is your subjective statements/interpretations made with no proper controls of perception bias.

Of course they are his subjective observations. At this point, I am not interested in a statistically significant demonstration whether transmission lines and other bass allignments are audibly similar or different. When that is the case, the proper controls of perception bias will be relevant and necessary. Until then, it is an unnecessary diversion.

Transmission lines seem to augment a woofers bass response in a broader range, with a lower Q, than vented woofers do. I think TLS Guy has demonstrated that in his example of the KEF B139 woofer. Is it because of this broad pipe response that there seems to be no narrow area of resonance when a TL speaker is placed near a room boundary?

I understand that few others may have listened to TL speakers. Listening to my new SongTowers has challenged some of the general ideas that I had believed were true about speakers' bass response. It seems that what may be true for sealed or reflex designs is not always the same for TLs. It does get difficult to describe. That's why I'm asking others for their thoughts on this.