Bass Driver for a transmission line speaker

ryanosaur

ryanosaur

Audioholic Spartan
Are you good at designing XOs? I have no experience in that field at all.
If I have it my way, I will be working my first XO design later this year. I am far behind, though, in other work/projects... and the Honey-Do list is as long as my leg! (Unfortunately Audio is not a value the Lady and I share.)
My only experience thus far has been the Mod I mentioned earlier, where I had the schematic and just had to lay out the parts in a logical order. Fortunately, this is not all that difficult and was a great experience building the XO itself from scratch.
Designing one will be a whole other game!
 
P

Paul N

Audioholic Intern
The crossover is the absolute heart of a speaker. You don't find a cheap filter. Off the shelf crossovers do not work ever. You have to design them custom from the ground up. That is probably the most difficult aspect of speaker design. All crossover have to be custom to the T/S parameters, acoustic responses, and power bandwidths of the drivers.

The first issue is driver selection. That is the first step. If you pick drivers out of thin air, you are very likely to find that the design and build of a crossover for that combination is not even possible.
What I am definitely going to is place the XO outside the cabinet, so that if necessary, it can be altered or replaced. With cheap I meant the individual components. Maybe I am wrong, but I am not planning to pay a lot of money for a resistor, a capacitor or a coil. o_O
 
Eppie

Eppie

Audioholic General
Well not first hand, but from what I have read, like https://techtalk.parts-express.com/forum/free-classifieds/54892-jenzen-seas-er-ca-troels-gravesen-tl-design his designs are worth looking at. I might consider building the Jenzen Next cabinet, but no way I am going to spend so much money on the crossovers that he designed. Bi-amping and finding a cheap filter for the mid and high is what I am thinking about.
TLS Guy, what do you think of the cabinet design of the Jenzen Next?
Ah, now I remember where I heard that name before. Those are some of the most expensive XO parts out there. You could probably cut the cost in half or less going to parts with a higher tolerance. The Dayton caps are 5%. I wonder how critical the max voltages are? His four 22uF caps are rated at 800V but a single Dayton 90uF is only rated at 250V on PartsExpress. You should definitely be able to reduce the cost there even if you stick to the same XO design.
 
Swerd

Swerd

Audioholic Warlord
I looked at the Paul Kittinger design you presented a link to and do not understand why he calls it a TL and not a bass reflex.
Paul Kittinger designs transmission line (TL) cabinets using a digital application developed by Martin J. King using MathCad.

According to Martin King, there is a small but important difference between what he calls a mass-loaded transmission line (MLTL) speaker cabinet and a bass reflex cabinet. The MLTL cabinet is tuned the same way any quarter wave length TL cabinet is tuned – by altering the line length. The longer the line, the lower the tuning frequency. When an appropriately sized & located port tube is added to the cabinet, it acts to lower the frequency of the cabinet's tuning. This allows a lower tuning for an otherwise shorter cabinet.

Imagine a flexible metal ruler. One end is clamped to a table, and the other end is allowed to move freely. If you pull & release the free end, it vibrates. The longer the ruler the lower the vibrations. Next, attach a weight to the free end. It lowers the vibration frequency without lengthening the ruler. Martin King says that his port-vented quarter-wave cabinet, acts similar to the mass loaded metal ruler. And he dubbed it a mass-loaded transmission line, a MLTL.

So a MLTL cabinet must have a defined inner volume and line length, whereas, a standard bass reflex cabinet must only have a defined inner volume.

King, an experimental physicist, described how he first arrived at a working math model of his idea:

For most of the 20th century, TL speaker cabinet design remained more of an art than a science. It required extensive trial & error with real wood. In contrast, by 1975, math equations could easily predict the acoustic behavior of sealed or ported cabinets, without any trial or error. As a result, the TL cabinets were rarely, if ever, fully understood or optimized. The high construction and shipping costs for large, heavy cabinets made commercial TL speakers extremely rare.

By the early 21st century, mathematical models that seemed to approximate the behavior of real-world TL speakers and cabinets, began to emerge. According to the website t-linespeakers.org, this led to an understanding that what Dave D'lugos termed the "classical" speakers, designed largely by "trial and error", were a "good job" and the best that was reasonably possible at those time, but that better designs were now achievable based on modeled responses.

However, in the early 21st Century, Martin King and George Augspurger both produced models which show these previous TL designs were generally "less than optimal", but "did a good job of approaching what was possible in their day".

Physicist Martin King based his model starting with mechanical parameters, and audio engineer Augspurger modeled his TLs starting with electrical parameters. Both find their models agree closely with each other.
 
Last edited:
P

Paul N

Audioholic Intern
Paul Kittinger designs transmission line (TL) cabinets using a digital application developed by Martin J. King using MathCad.

According to Martin King, there is a small but important difference between what he calls a mass-loaded transmission line (MLTL) speaker cabinet and a bass reflex cabinet. The MLTL cabinet is tuned the same way any quarter wave length TL cabinet is tuned – by altering the line length. The longer the line, the lower the tuning frequency. When an appropriately sized & located port tube is added to the cabinet, it acts to lower the frequency of the cabinet's tuning. This allows a lower tuning for an otherwise shorter cabinet.

Imagine a flexible metal ruler. One end is clamped to a table, and the other end is allowed to move freely. If you pull & release the free end, it vibrates. The longer the ruler the lower the vibrations. Next, attach a weight to the free end. It lowers the vibration frequency without lengthening the ruler. Martin King says that his port-vented quarter-wave cabinet, acts similar to the mass loaded metal ruler. And he dubbed it a mass-loaded transmission line, a MLTL.

So a MLTL cabinet must have a defined inner volume and line length, whereas, a standard bass reflex cabinet must only have a defined inner volume.

King, an experimental physicist, described how he first arrived at a working math model of his idea:

For most of the 20th century, TL speaker cabinet design remained more of an art than a science. It required extensive trial & error with real wood. In contrast, by 1975, math equations could easily predict the acoustic behavior of sealed or ported cabinets, without any trial or error. As a result, the TL cabinets were rarely, if ever, fully understood or optimized. The high construction and shipping costs for large, heavy cabinets made commercial TL speakers extremely rare.

By the early 21st century, mathematical models that seemed to approximate the behavior of real-world TL speakers and cabinets, began to emerge. According to the website t-linespeakers.org, this led to an understanding that what Dave D'lugos termed the "classical" speakers, designed largely by "trial and error", were a "good job" and the best that was reasonably possible at those time, but that better designs were now achievable based on modeled responses.

However, in the early 21st Century, Martin King and George Augspurger both produced models which show these previous TL designs were generally "less than optimal", but "did a good job of approaching what was possible in their day".

Physicist Martin King based his model starting with mechanical parameters, and audio engineer Augspurger modeled his TLs starting with electrical parameters. Both find their models agree closely with each other.
Thanks for the information on and explanation of the mass loaded TL.
I just found out there is another, more recent TL model by Andrea Rubina. Like George Auspurger’s model, it translates what happens acoustically in the way of an electric circuit. He has a nice website, where you can download it www.transmissionlinespeakers.com/en/home-eng/
 
TLS Guy

TLS Guy

Audioholic Slumlord
Paul Kittinger designs transmission line (TL) cabinets using a digital application developed by Martin J. King using MathCad.

According to Martin King, there is a small but important difference between what he calls a mass-loaded transmission line (MLTL) speaker cabinet and a bass reflex cabinet. The MLTL cabinet is tuned the same way any quarter wave length TL cabinet is tuned – by altering the line length. The longer the line, the lower the tuning frequency. When an appropriately sized & located port tube is added to the cabinet, it acts to lower the frequency of the cabinet's tuning. This allows a lower tuning for an otherwise shorter cabinet.

Imagine a flexible metal ruler. One end is clamped to a table, and the other end is allowed to move freely. If you pull & release the free end, it vibrates. The longer the ruler the lower the vibrations. Next, attach a weight to the free end. It lowers the vibration frequency without lengthening the ruler. Martin King says that his port-vented quarter-wave cabinet, acts similar to the mass loaded metal ruler. And he dubbed it a mass-loaded transmission line, a MLTL.

So a MLTL cabinet must have a defined inner volume and line length, whereas, a standard bass reflex cabinet must only have a defined inner volume.

King, an experimental physicist, described how he first arrived at a working math model of his idea:

For most of the 20th century, TL speaker cabinet design remained more of an art than a science. It required extensive trial & error with real wood. In contrast, by 1975, math equations could easily predict the acoustic behavior of sealed or ported cabinets, without any trial or error. As a result, the TL cabinets were rarely, if ever, fully understood or optimized. The high construction and shipping costs for large, heavy cabinets made commercial TL speakers extremely rare.

By the early 21st century, mathematical models that seemed to approximate the behavior of real-world TL speakers and cabinets, began to emerge. According to the website t-linespeakers.org, this led to an understanding that what Dave D'lugos termed the "classical" speakers, designed largely by "trial and error", were a "good job" and the best that was reasonably possible at those time, but that better designs were now achievable based on modeled responses.

However, in the early 21st Century, Martin King and George Augspurger both produced models which show these previous TL designs were generally "less than optimal", but "did a good job of approaching what was possible in their day".

Physicist Martin King based his model starting with mechanical parameters, and audio engineer Augspurger modeled his TLs starting with electrical parameters. Both find their models agree closely with each other.
There are important differences in the results though. The mass loaded TL does save cabinet volume. The size of the cabinet is the biggest resistance to TL take up by the public. However, it comes at a cost. The frequency range of support for the driver is substantially reduced. This is in no small measure due to the mass loaded TL having fourth order 24 db per octave roll off below F3, against the second order 12 db roll of for the more traditional TL. In addition port output is also more extended on the top end with the traditional design

Also the mass loaded design can not be made truly aperiodic, that means non resonant, as like a ported design, that kills its operation.
The biggest attraction of the traditional TL is that reproduction can be truly aperiodic. That is what really makes those speakers give such non resonant uncoloured bass. It makes them sound in many aspects like an electrostatic loudspeaker, but extend deeper. I know of no other way of loading a loudspeaker that gives that "like live" in the lower octaves.

If I were a speaker manufacturer, I would certainly find the mass loaded approach attractive for the above reasons. I think that is why traditional TLs have very much been the preserve of the home constructor.

The TLs, actually Labyrinths listed above use very small drivers. They are basically souped up Bose Waves. As you lower driver F3, and increase the Vas of the driver, the size of the cabinet really starts to increase.
 
haraldo

haraldo

Audioholic Spartan
The crossover is the absolute heart of a speaker. You don't find a cheap filter. Off the shelf crossovers do not work ever. You have to design them custom from the ground up. That is probably the most difficult aspect of speaker design. All crossover have to be custom to the T/S parameters, acoustic responses, and power bandwidths of the drivers.

The first issue is driver selection. That is the first step. If you pick drivers out of thin air, you are very likely to find that the design and build of a crossover for that combination is not even possible.
@TLS Guy What are the main parameters you look for when you are looking for an optimal driver for a sub-bass TL system?
There seem to be a lot of confusion out there on design of TL!
Would the following be somehow within the correct playing field?
- Low Qts, between 0.20 and 0.30?
- Low mass, what is low mass is relative I guess?
- Fs 18 - 25 Hz
- Vas ????

What I would possibly be looking for is something that better the bass from any commercial system, goes flat in room below 20Hz (using peq is fine if required) and that really has this "kick" you only get from a proper TL. I am afraoid though, that this system may be too large to fit into the living room of a moderate size flat :D

Would these ones fit the bill?
Scanspeak revelator 28W/4878T01

ScanSpeak Revelator 32W/4878T11

I heard first hand from DIY´er who used some of these Revelator sub drivers and who claim they are possibly the best drivers he ever seen, claim they are simply astounding. There are 4 and 8 ohm versions of all these drivers and they have pretty differing T/S parameters.

If building something ourselves, why not go with the absolutely best drivers there is, I am not sure if I see any somewhat reasonably priced commercial products out there that employ this level of driver quality :rolleyes:
 
Last edited:
TLS Guy

TLS Guy

Audioholic Slumlord
@TLS Guy What are the main parameters you look for when you are looking for an optimal driver for a sub-bass TL system?
There seem to be a lot of confusion out there on design of TL!
Would the following be somehow within the correct playing field?
- Low Qts, between 0.20 and 0.30?
- Low mass, what is low mass is relative I guess?
- Fs 18 - 25 Hz
- Vas ????

What I would possibly be looking for is something that better the bass from any commercial system, goes flat in room below 20Hz (using peq is fine if required) and that really has this "kick" you only get from a proper TL. I am afraoid though, that this system may be too large to fit into the living room of a moderate size flat :D

Would these ones fit the bill?
Scanspeak revelator 28W/4878T01

ScanSpeak Revelator 32W/4878T11

I heard first hand from DIY´er who used some of these Revelator sub drivers and who claim they are possibly the best drivers he ever seen, claim they are simply astounding. There are 4 and 8 ohm versions of all these drivers and they have pretty differing T/S parameters.

If building something ourselves, why not go with the absolutely best drivers there is, I am not sure if I see any somewhat reasonably priced commercial products out there that employ this level of driver quality :rolleyes:
The sweet spot for TLs, tends to fairly narrow. About 0.35 to 0.42, may be 0.45. You want Fs as low as possible. As Vas rises so does pipe volume.
You really have to model a TL though as you get surprises. The F3 of a TL design is usually slightly higher than a standard ported design, for the same driver, however roll off below F3 is only 12 db per octave compared to 24 db per octave.

Those drivers you showed, are not optimal for a TL at first glance.

Really good drivers are the KEF B 139, still many to be had. Probably the best of all are the SEAS Excel range with the magnesium alloy cone. The 10" driver of that range really hits the sweet spot. My TL bass lines use two in each and they really shake your chair and liver.
 
haraldo

haraldo

Audioholic Spartan
The sweet spot for TLs, tends to fairly narrow. About 0.35 to 0.42, may be 0.45. You want Fs as low as possible. As Vas rises so does pipe volume.
You really have to model a TL though as you get surprises. The F3 of a TL design is usually slightly higher than a standard ported design, for the same driver, however roll off below F3 is only 12 db per octave compared to 24 db per octave.

Those drivers you showed, are not optimal for a TL at first glance.

Really good drivers are the KEF B 139, still many to be had. Probably the best of all are the SEAS Excel range with the magnesium alloy cone. The 10" driver of that range really hits the sweet spot. My TL bass lines use two in each and they really shake your chair and liver.
Thx A LOT!
 

Latest posts


newsletter
  • RBHsound.com
  • BlueJeansCable.com
  • SVS Sound Subwoofers
  • Experience the Martin Logan Montis
Top