Transmission line damping question.

[Old'n'Cranky]

Hi there....

I'm in the process of building a transmission line (Troel's Jenzen).
I understand that the damping is fairly critical to the performance.
One part of the damping he's specified 30mm 350gsm polyester ( http://www.troelsgravesen.dk/Jenzen-CAB_files/damping-1.gif )
I'm having great trouble locating this material in Australia.
I can get 30mm 200gsm easily.

My question;

Can I double the the 200gsm into sheets of 60mm 400gsm to 'simulate' the 30mm 350gsm ?
The design calls for 3 layers of 350gsm (1050 over 90mm)
I'm thinking about 5 layers of 200gsm (1000 over 140, thats the width of the first turn)
Or will that stuff (haha) it up ?

Alternatively I have found a source for 30mm 300gsm, but I have to buy a 20mtr roll of it......

Unless any readers from Australia care to share supplier info ?

I have asked Troel's, he's of the opinion that i should use 350gsm, but as I said, thats far easier said than done in our little back water..

Any and all help/info much appreciated...

 

[pkitt]

I know I'll get flamed for my comments but I've yet to see a Troels-designed TL that was good performing from a TL standpoint. I've modeled several of them and they are very, very poor. In this Jenzen series of "TLs", the woofer is at the worst possible place in the TL, the very beginning of the line, leading to substandard performance (a low crossover frequency from woofer to midrange will mitigate this to a large extent). As far as I can tell Troels does not use any of the readily available and GOOD TL modeling software that exists. If he did, he would quickly see how poor his "TL" designs are. So, to answer your questions about the "stuffing", IMO it won't really matter what you do, it can't really hurt a poor-performing speaker.
Paul

 

[Old'n'Cranky]

well im NOT going to flame you.. 😀
Ive had close to zero to do with t'lines.
the design I'm looking at is the D mk2 with the audio tech 10c77 driver.
(oh yes, $$$$$$$$$)

I believe he also has a std ported version of the cabinet.
So do I understand that the std ported version could be a preferable option than his T'line ?

Seen near bottom of page - http://www.troelsgravesen.dk/Jenzen-Accu.htm -

(edit - i have the drivers and xover parts, but i havent yet had the cabs built, so all thoughts much valued)

Or what do you believe could be done to improve/rectify the design ?

As I said, ANY and ALL help or info appreciated 😉

PS, IF the design was 'acceptable' would that then mean the 350 vs 200 argument makes a more valuable question ?

 

[pkitt]

I looked over Troels' Jenzen series of TL designs and saw he uses a pretty low crossover frequency of 200 Hz. That will definitely mitigate most of the problems resulting from having the woofer at the very beginning of the line fortunately. Because Troels neither uses stuffing per se nor apparently models the TL's predicted performance with accurate TL software, it's really difficult to model what he creates with any real accuracy. I suspect, but don't know for sure, that he makes the total line length such that its 1/4-wavelength resonant frequency is ~equal to the woofer's Fs, which may or may not be an appropriate choice, depending on the woofer's Qts. I further suspect he tweaks the line's performance entirely by ear, which also may or may not be the best choice. I have more confidence that using the vented box version would probably be a good choice. I wish I could provide more help rather than just being a critic. The best advice I can give is to be very, very sure before spending big bucks!
Paul

Edit: I just took a look at the T/S values for the AT10C77 Troels provided. The average Qts after burn-in was 0.27. From my experience I've seldom found a driver with a Qts below 0.3 to perform at all well in a TL other than an ML-TQWT, which this design is not, lending more confidence to the standard vented box. Other designers/builders may have had better success, however. Please note that I haven't heard any Troels-designed speakers, just going on my personal TL modeling and building experience.
Paul

well im NOT going to flame you.. 😀
Ive had close to zero to do with t'lines.
the design I'm looking at is the D mk2 with the audio tech 10c77 driver.
(oh yes, $$$$$$$$$)

I believe he also has a std ported version of the cabinet.
So do I understand that the std ported version could be a preferable option than his T'line ?
Or what do you believe could be done to improve/rectify the design ?

As I said, ANY and ALL help or info appreciated 😉

PS, IF the design was 'acceptable' would that then mean the 350 vs 200 argument makes a more valuable question ?

 

[planet10]

I'm with Mr Kitt on this one.

I'd immediately question the design based on the damping scheme. I don't think that he is using any of the modern design simulators for TLs and has not taken any advantage of the tricks common to modern TLs, driver offet, line taper, or restricted terminus. I'd never build a TL without verifying it in MJK software before cutting wood.

dave

 

[Old'n'Cranky]

Thank you very much gentlemen.
I do indeed value your knowledge and information.
I do believe I shall have a much closer look at the ported design.
Slightly smaller cabinet, and will make it much easier to brace it to my satisfaction.

To everyone else, please chime in and give me your thoughts..

I do understand I could make them sound like rubbish just with bad positioning and room conditions.
But it will be nice to know the issues are the room, not the speakers. Expensive speakers at that 😀

 

[Keriwena]

Well, for a given woofer, I've found I can get the same response from either a BR or TL, without stuffing. The TL will be larger, and will have "pipe resonances" to deal with. In short, given a choice, the BR will give you a smaller box and clearer mids.

Why use a TL? Sometimes you can't get what you want with a BR design. Then the TL offers options, as since the "port" is a greater percentage of the total volume, you can damp the port with a precision not practical in a BR. That is, a well designed TL is sort of an aperiodic BR.

In my view, TIFWIW, YMMV, etc.

 

[pkitt]

I do disagree with part of what you say. In any ML-TL I've designed, the port volume has always been a smaller percentage of the total volume than for the port in an equivalent BR because the 1/4-wave resonant frequency of the line's length is contributing to part of the overall system tuning, leaving less to be contributed by the port. The longer the line, the more this is true.
Paul

Well, for a given woofer, I've found I can get the same response from either a BR or TL, without stuffing. The TL will be larger, and will have "pipe resonances" to deal with. In short, given a choice, the BR will give you a smaller box and clearer mids.

Why use a TL? Sometimes you can't get what you want with a BR design. Then the TL offers options, as since the "port" is a greater percentage of the total volume, you can damp the port with a precision not practical in a BR. That is, a well designed TL is sort of an aperiodic BR.

In my view, TIFWIW, YMMV, etc.

 

[Keriwena]

To be clear, when I refer to the "port" of a TL, I meaning the opening of the pipe. If you close the port off, and then add a duct, well, then you simply have a poorly-formed BR. 😉

 

[pkitt]

Then "terminus" would be more proper to use than port. If the TL and its terminus are properly designed, one can't simply close off the terminus, substitute a port (duct, as you say) and expect good results. The terminus is a very short port, of course, whose cross-sectional area usually mimics the area at the end of the line (pipe, as you say). But, a properly designed ML-TL, especially one that is reasonably long, will have better performance than a ported box (BR, as you say) of the very same total volume with the port's volume a much smaller percentage of the total volume.
Paul

To be clear, when I refer to the "port" of a TL, I meaning the opening of the pipe. If you close the port off, and then add a duct, well, then you simply have a poorly-formed BR. 😉

 

[rcw666]

I question the whole concept of the transmission line in the first place.

You can show that they go on a continuum from overlarge leaky sealed boxes, to over complex reflex ones, and apart from some appeal to some mysterious esoteric property of some sort, have no redeeming features at all.
.rcw

 

[pkitt]

Without knowing otherwise, I can only imagine you've never heard a well-designed and well-executed TL. Still, they're not the only or most appropriate choice for all drivers in all situations.
Paul

I question the whole concept of the transmission line in the first place.

You can show that they go on a continuum from overlarge leaky sealed boxes, to over complex reflex ones, and apart from some appeal to some mysterious esoteric property of some sort, have no redeeming features at all.
.rcw

 

[mannish boy]

Can't you just build a few of prototype cabs and see (listen / measure) what you like best? If I were spending that type of money on drivers and crossovers wasting a few dollars on some MDF would be the least of my worries...

 

[DrBoar]

TLS can sound very good. Damping overall requirement and placement is highly dependent on taper and folding gemetry

An externally hosted image should be here but it was not working when we last tested it.

.
Note that negative taper lowers the Q and frequency of the fundamental frequency but not of the harmonics so they are no longer odd multiples of the fundamental

Foldning the pipe greatly affect the the exitations of harmonics.

With some clever stuffing you can reduce the harmonics and still maintain the increased radiation resistance of the fundamental

An externally hosted image should be here but it was not working when we last tested it.

Overstuffing the pipe and basically getting a accoustic resistance box is ending up in to complicated woodwork to get that result. Stuffing in the wrong place can wreck havoc as well. I agree with mannish boy, designing in silico is clean but tests in sawdust is were theories meets reality to soar or crash.

 

[pkitt]

All of the TL cabinet diagrams you show have the same inherent design defect; the driver is at the very beginning of the line, a far from optimum location unless it's strictly for subwoofer use where the typically low crossover frequency will minimize most of the response deficiencies.

If you use a good TL modeling program, like one of Martin King's, you can model exactly the configuration of the line (straight, tapered, folded, etc.) along with the exact amount and exact location of stuffing to get optimum performance (within whatever cabinet size and shape constraints you impose). If you then build exactly what you model, you will get the predicted results (minus room effects) and you should not need to tweak anything (at least that's been my experience). No need, therefore, to build various configurations to decide which sounds best (unless you don't correctly build what you modeled).
Paul

TLS can sound very good. Damping overall requirement and placement is highly dependent on taper and folding gemetry

An externally hosted image should be here but it was not working when we last tested it.

.
Note that negative taper lowers the Q and frequency of the fundamental frequency but not of the harmonics so they are no longer odd multiples of the fundamental

Foldning the pipe greatly affect the the exitations of harmonics.

With some clever stuffing you can reduce the harmonics and still maintain the increased radiation resistance of the fundamental

An externally hosted image should be here but it was not working when we last tested it.

Overstuffing the pipe and basically getting a accoustic resistance box is ending up in to complicated woodwork to get that result. Stuffing in the wrong place can wreck havoc as well. I agree with mannish boy, designing in silico is clean but tests in sawdust is were theories meets reality to soar or crash.

 

[rcw666]

You can show that the first illustration, that of an open pipe is in fact a Helmholtz resonator in which the neck length is reduced to zero because its area is the same as that of the enclosure, this is an open ended pipe, and in this context can be looked at as a Helmholtz resonator with its q reduced to zero.

If you make the dimensions of the pipe at its open end less than the cross sectional area of the pipe, then you start to make a Helmholtz resonator, and you introduce a second, second order term, i.e. it starts to resemble a reflex box.

You can look at all transmission line enclosures this way, i.e. they are reflex boxes with zero to increasing q, and those for instance advocated by King can be shown to be reflex boxes, but over elaborate ones.
rcw

 

[planet10]

To be clear, when I refer to the "port" of a TL, I meaning the opening of the pipe. If you close the port off, and then add a duct, well, then you simply have a poorly-formed BR. 😉

Martin King, using ANSYS, showed that this is not the case. A BR assumes a box not too far from a cube, if you start changing the aspect ratio so that one dimension becomes larger then the others the BR transitions to a mass-loaded transmission line with quite different behaviour.

It is very important thet one does not confuse the Classical TLs which were typically quite crude and often far from optimal with the much larger design space of the modern TL opened up by the contemporaneously released work of King and Augspurger (only MJK has continued refining and extending the model).

From your comments it is clear you do nothave a solid understanding of a modern TL.

dave

 

[Jonathan Bright]

pkitt, placing the driver at the beginning of the line/pipe is NOT "an inherent design defect". It is the classic position since Bailey's two papers published in the UK some 4O odd yrs ago (and others see those as revivals of still earlier concepts). There is a mountain of material out there about TLs. I know some drivers are placed in other positions and people build DALINES etc but the diagrams shown here are not "wrong" or merely designed for subs.
If you want to follow it up, the early TLs were called "transmission lines" because of the perceived analogy with electrical TLs. Again, it is true the strict correspondence or one to one relationship has been argued about. In the same way with the Bass Reflex enclosure Thiele's (a great Aussie!) work was generated after he saw the fact that there was a direct parallel between what the BR was doing at resonance and what electrical filter theory predicted.
Cheers, Jonathan

 

[planet10]

pkitt, placing the driver at the beginning of the line/pipe is NOT "an inherent design defect".

It is a case of not taking advantage of known powerful design tool (althou it was not explicitely know at the time of the Bailey) to reduce ripple. It is not a defect, but it doesn't make sense not to take advantage of it unless some other design constraint trumps using a driver offset (ie the my modern tribute to the microTower, or the use of an aperiodic midTL), and in the Bailey (or the Radford, large IMFs and in the Swedish paper above) it is easily accomplishable.

The mention of the Daline is appropo as the Bailey like is, as we define them today, an example of a Daline. A daline is simply a box where the TL volume is divided into a prechamber + a line. This is another geometry toolk that improves the low pass filter characteristic of the line.

dave

 

[planet10]

... There is a mountain of material out there about TLs...

And the amount of quality R&D that has been done after valid models became avaialble in 1999 dwarfs everything that preceeded that date.

dave