Oh, I see...
Isn't this that make a TL has low sensitivity? Its close to BR vs sealed analogy.
The reason I asked is to find out if you can hear/perceive the issue with sensitivity.
We cannot know exactly how a speaker will sound before building it, and there is always compromise, so trying out multiple schemes is necessary sometimes. And I found that there are more knowledge gained from tweaking or experimenting with the same circuit or driver (trying out all possibilities), than building more but only one design for each circuit or driver.
I have never used flat impedance as an objective because I don't use "small" amps. With "TL" I always expect bass performance, which is very good and preferable with certain kind of music. But for most recordings (with less LF contents), I found TLs lose its charms.
I have given up TL, and would prefer horn, but is too difficult for me (the "mistakes" even if small will be amplified).
I like a tapered? TL a-la Thor, or use a back panel very close to the driver to reinforce the mid-range and details, which often "missing" in TL design. Or MLTL. But nowadays I prefer sealed (with open possibility for active).
This has been answered several times above; you may want to go back and read through the replies again. Assuming you are referring to cabinet gain rather than system sensitivity however, then short version: an aperiodic TL won't produce much LF gain, but since it isn't meant to, that is hardly a criticism. Not all the boxes that are generally lumped under the 'transmission line' name are aperiodic however. Most are not (MLTLs being the most obvious example).
Since there is no 'issue with sensitivity', as cabinet output depends (or it should depend
) on design goals, by definition, no I can't 'hear/percieve' any such issue. If you compare the LF gain of, say, a QB3 aligned BR to that of an aperiodic TL using the same drive unit, the former will naturally produce more. Hardly surprising, since self evidently the cabinets do not have the same alignment, nor (one would hope) will they be sharing the same design objectives. As a matter of basic common sense, you would not use an aperiodic transmission line if you require a TL that provides considerable box output. You would use one that does produce the desired gain. I think the issue is that you appear to have a specific idea of what a transmission line is, whereas in fact, as has been pointed out, the term is generally applied to a vast array of boxes with a near-infinite variety of possible alignments. Perhaps you would prefer 'quarter-wave line' since it has fewer connotations, and is technially more accurate in the majority of cases.
Depends what you call 'detail' since it could be almost anything. It could for e.g., be high 3rd order harmonic distortion, or indeed it could be very low distortion across the board.
A panel in close proximity to a drive unit needs to be carefully contrived to prevent reflections back through the cone. Either way though, if something is obviously missing in a QW box that is not supposed to be, then it has been badly designed.
~55 L is tiny for a TL with good bass out of a 6.5". You can always reduce its net volume [Vb] to make it smaller though, but of course the bass will roll off higher up.
GM
That's what I was trying to say. To you it is a common sense, to the OP it might not. I kind of "warn" the OP that the effect (from that point of view) is not nice. To me personally QWTL is history (I'm against it but that's another story).
Actually, that one was what I had in mind, the one I have been most familiar with.
I think it is a common sense (using your words) that something will obviously be missing in a QW box. But we have to experience it to know/feel how severe it is. Analogy is like bookshelf vs floorstander. It is obvious what will be missing in each topology. But again we have to experience it (in each design) to decide whether we need to drop the topology or find a solution.
I use close distance from driver to back panel to reinforce midrange. The distance is not calculated but trialed during prototyping (and so is the angle to avoid backwave against cone). I think this is one important feature of LS3/5A box style.
I just read this thread, and may I ask you a question about aperiodic loading?
I have, so will use, 4 KEF B139s, "racetrack" shape. Isobaric, 'cos I have liked their sound in the past. The rear driver can be closed box or aperiodic; my question is, would the thick polystyrene cone make these drivers a likely candidate for aperiodic, or does the closed box back load the cone just as much, making the comparison moot? I realise the original point was TL versus aperiodic loading, so have I just answered my own question? Lol!
Then you needed to be clearer. To make a sweeping generalisation does the OP no favours at all. I'm intrigued to know why aperiodic alignments are not 'nice' and presumably other enclosure types are 'nice' though. You appear to be assuming once again that all boxes described as TLs are aperiodic, which as pointed out several times is far from being the case.
Really? What will be 'missing?' You've claimed this several times now, without stating what you mean by it. Please be specific.
No, I didn't explicitly say that aperiodic alignments are not "nice". I said that TL (I meant QWTL) will "kill" speaker sensitivity. This is a basic common sense (as you said), but I tried to reinforce it, saying that this specific effect is not nice.
Of course, apart from its LF performance, QWTL is not nice for me, but I didn't make generalization because I know that many will be in love with what I think is not nice. I can call 15 people and ask their opinion about my TL and non-TL versions using the same drivers, and I can be sure that 100% of them will prefer my TL
To a greater extent, it is impossible to explain it by words, because it is about listening experience, and may be taste. May be it is just something microscopic (exposed itself as musicality for example) that for most is unmeasurable or even non-exist but for me is audible and important.
Take one simple example: can you make a TL "disappear"? Or gives sonic transient and mid-range quality of a (sealed) bookshelf? Again, I think it is about something microscopic of which importance is different to each person.
BTW, I believe I'm not the only one who "gave up" on QWTL.
Oh, I don't have experience with OB, but my instinct tells me that there is something wrong there also (for the majority of OB designs). May be box contribution is required, not something that should be missing. You cannot ask me to back up my argument with numbers, because I don't have
nice TL scott, but to us 'lesser mortals' what the heck is an alpha? In my opinion 1.5 lb per cuft is an EXTREME amount of stuffing. Of the 1 or 2 crude flatER Z type TLs ive messed with, there was no advantage going past 1lb...it just becomes SOLID, very little air left. One 400g pack, 2 'layers' along each path in a single fold TL for eg. Putting more in is stuffing in the taxidermicalogical sense, read PACKING.
Jay, you're still absolutely wrong. A TL, of any configuration, will not inherently kill speaker sensitivity but like all other box configurations, some drivers will not work as well in a TL as they may in other boxes. Some drivers work well in a vented/BR box but not a sealed box, and vice versa for instance. And, I find your attempt to make a distinction between "your" TL and QWTL to be exceedingly confusing and nebulous. Further, making a speaker "disappear" has more to do with the quality of implementation than the box configuration; there's absolutely no inherent reason for a TL to not be able to pull off a disappearing act.
Paul
Paul
No, I didn't explicitly say that aperiodic alignments are not "nice". I said that TL (I meant QWTL) will "kill" speaker sensitivity. This is a basic common sense (as you said), but I tried to reinforce it, saying that this specific effect is not nice.
Of course, apart from its LF performance, QWTL is not nice for me, but I didn't make generalization because I know that many will be in love with what I think is not nice. I can call 15 people and ask their opinion about my TL and non-TL versions using the same drivers, and I can be sure that 100% of them will prefer my TL
To a greater extent, it is impossible to explain it by words, because it is about listening experience, and may be taste. May be it is just something microscopic (exposed itself as musicality for example) that for most is unmeasurable or even non-exist but for me is audible and important.
Take one simple example: can you make a TL "disappear"? Or gives sonic transient and mid-range quality of a (sealed) bookshelf? Again, I think it is about something microscopic of which importance is different to each person.
BTW, I believe I'm not the only one who "gave up" on QWTL.
Oh, I don't have experience with OB, but my instinct tells me that there is something wrong there also (for the majority of OB designs). May be box contribution is required, not something that should be missing. You cannot ask me to back up my argument with numbers, because I don't have
I'm sorry, I'm struggling to comprehend how you could claim this, given that is is contradicted by so many different examples that are rather easy to discover.
A quarter-wave cabinet does not reduce overall speaker sensitivity. That is a function of the drive unit. It only operates, like any back-loaded enclosure, over a very narrow region at the bottom of the total BW. See the attached example 1/2 space frequency response plots of a handful of different QW enclosure types with various drive units. Please explain how these are 'lacking sensitivity.' Please also explain the physics from which you seem to be inferring that a quarter-wave enclosure reduces the sensitivity of a drive unit (which is contrary to the laws of physics BTW). You appear to be attempting to compare apples with oranges. Basic scientific methodology, to say nothing of common sense, dictates that the only time comparisons are meaningful is when a single variable is changed.
As regard 'disappearing', as Paul points out, this is a matter of implementation quality. The simple answer however is 'yes.' Incidentally, you may be interested to know that sealed TLs are highly regarded for loading midrange drivers & the suppression of their entire backwave while also flattening the impedance.
This seems self-contradictory to me. You place the rear panel in close proximity to 'reinforce midrange' but 'avoid backwave against the cone.' Hmm.
How is it reinforcing the midrange? The only mechanism I can think of is that you're encouraging the backwave out through the cone again (cracking), but you say you don't want that. Are you trying to say you want high internal pressures & a high box Qtc?Attachments
Last edited:
Sorry, only just seen this.
Sounds like the old Linn Isobarik to me.
Aperiodic is essentially a leaky sealed box, or at least, that's what most people mean when they talk about it. Sticking with this, the usual reason for employing it is that, well implemented, it essentially acts like a well damped closed box of larger internal volume. So essentially it's a means of lowering the effective box Qtc. It also usually lowers the impedance peak at Fb in the process. This page might be of interest: WD25A Design Pg2An aperiodically damped TL is not quite the same thing. An undamped QW tube is highly resonant; an aperiodic TL essentially referrs to a QW line that has been stuffed until you have completely supressed the back-wave, and largely flattened out the impedance. If the latter is the design goal you can optimise Fp, Vb and damping to almost completely flatten the impedance -as close as you're going to get to a 'perfect' load. That's where the Transmission Line moniker came from, although the term is almost never used so narrowly.
Alpha TL referrs to a specific TL type per Rick Schultz's AudioXpress article a few years ago. Short version: it's a very simple untapered TL based on a set of formulas for length, cross section and stuffing density that Rick derived from repeatedly running George Augspurger's TLwrx software. Basically, it's a more detailed version of the alignment tables Augspurger provided in his AES paper & three AudioXpress articles a couple of years before, albeit just for untapered pipes.
In the example I mentioned, I tapped the driver into the pipe at the optimal location based on its length & cross section, and set Fp to achieve the flattest possible impedance. It is a high level of stuffing / damping / packing, because it's supposed to be. The sole object of that particular example was to flatten the impedance, which leads to the almost total suppression of pipe output. In the Alpha TL, as in George Augspurger's alignments, stuffing density is based soley upon Fp (and the type of material employed); both were aiming for 1dB dip at F3 and less than 0.5dB deviation for F5 & above.
Last edited:
That's only in cases where it is soundwise preferable to do so. LS3/5A boxes for example, you can mount the driver on the narrow side if you prefer. You can also use thick material for enclosure if you prefer. What I'm saying is, box coloration is sometimes wanted. An honest effort to avoid the effect is the OB, or TL to some degree.
You're almost repeating the OP question, to find out what driver parameter suitable for TL design. So what is it? The most common (theoretical) answer is Qts, but I tried to put sensitivity, an uncommon answer, based on (listening) experience.
"My" TL. No, I will not so decided with conclusion if it is just my TLs or yours.
Sure. The indirect analogy of your statement is that a bookshelf speaker can have better bass than a floor-stander, which is correct.
I didn't say that. It is speaker "sensitivity", not driver sensitivity. And nothing is "reduced", but we can usually get more with different alignments.
I believe a driver (or speaker) in an anechoic chamber will sound different, and it will NOT suit my listening preference.
Also, human ears are more sensitive to midrange. Remember that I mentioned all this sensitivity issue strictly based on listening experience, not measuring experience (tho I'm not surprised if measurement will reveal the same thing). It means, it is the midrange that would have been first recognized by my ears.
As for flat impedance, we have lived for so long with "standard" impedance curve, that I don't believe it critically affects my solid state amplifiers (and thus the resulting sound).
If I understand you correctly, then yes, that is (loading midrange?) one feature that I make use in TL speakers.
Drivers that have terrible midrange peaks and distortion (BTW, I don't understand how people can live with those Fostex drivers/designs), can be tamed using TL enclosure.
As opposed to bookshelf design, crossover can be made simple (without introducing peaky sound), and relative to bookshelf enclosure, midrange should better be "overexposed" (with the over-scientific way of thinking this will invite more "questions" I know).
You want colouration? To each their own.
You appear to be talking about cabinet gain. I notice you have inadvertently forgotten to answer the question I asked above, regarding the 4 example FR plots attached to my post: namely 'please explain how these are "lacking sensitivity."' Those were flat, or near-flat alignments. Example FRs of QW boxes with a damped LF alignment, or indeed, one that is peaking, could be easily provided.
Not everybody runs your amplifiers. Ergo, in some situations, said flat impedance is useful.
Hmm. You seem now to have reverted back to the notion that a/ all QW boxes are functionally the same, which is not the case, and b/ they somehow attenuate total system output outside their operating BW. Since back-loaded boxes of any kind, be they reflex, quarter-wave or whatever only have a very narrow functional passband, that is obviously ruled out as a cause. Based on your posts above, I suspect you happen to like the colouration caused by internal reflections within a cabinet which pass back out through the cone, and the absence of these bothers you.
They can certainly do that if you design them to.
How does a TL or QW enclosure tame terrible midrange peaks and distortion? Regarding the point on Fostex, presumably because not all Fostex drive units have terrible midrange peaks and distortion.
Not over-scientific, since I'm afraid the above makes no sense to me. Why should the crossover design for something which you have not specified be necessarily simpler than that of a bookshelf enclosure? Crossover design is crossover design. You design the crossover for the driver-cabinet combination in hand and the design goals you wish to attain. I can think of many high quality standmount / bookshelf speakers with very simple crossovers. I can also think of just as many equally high quality speakers that are not bookshelf / standmount types with extremely complex XOs.
That the driver has no box loading or any reflections that can affect its operation. For a woofer this requires it to be suspended dozens of feet off the ground out in a large field, so in reality it usually just means it's not in a box or mounted to a baffle.
Yes, you can do this, though with 'only' a ~0.81 Qts it will take a relatively large baffle to get any decent output down to anywhere near Fs, though rolling off one of the drivers to make it just a ‘helper’ woofer would reduce baffle size; but again, Scott’s TL seems your ‘smallest’ option, at least WRT to baffle size without resorting to mass quantities of boost like SL uses, which would severely restrict dynamic headroom of a small, low power handling driver.
Still, if it’s a small room and/or you will be relatively close to them, then this may indeed be your best option.
GM
How can a TL kill speaker sensitivity?
A TL is pipe that can work as a quarter wave resonantor. If you cram if full with resistive damping material. it will absorb close to 100% of back radiation and be very close to a closed box filled with damping material.
With no damping material it is similar to backloaded horns and tapped horns by giving a substantial increase in radiation resistance (good) and often nasty resonances (bad).
By using smart placement of folds and damping material much of the fundamental resonance can be used and a lot of the unwanted harmonics supressed.
Many TLS like IMF and many DIY used KEF bextrene drivers with low sensitivity that often needed extensive passive eq in the filiters lowering sensitivity even further. Then you could reach 20-25 Hz in "small" boxes. But you could equally well use JBL PA drivers in TL and get very high sensitivity TLS. Then a 20 Hz TL for a 15" drivers would be about 3 times the volume of a 20 Hz TLS for a nominal 10" like the KEF B139.
So I really do not see low sensitivity as intrinsic for TLS per se.
A TL is pipe that can work as a quarter wave resonantor. If you cram if full with resistive damping material. it will absorb close to 100% of back radiation and be very close to a closed box filled with damping material.
With no damping material it is similar to backloaded horns and tapped horns by giving a substantial increase in radiation resistance (good) and often nasty resonances (bad).
By using smart placement of folds and damping material much of the fundamental resonance can be used and a lot of the unwanted harmonics supressed.
Many TLS like IMF and many DIY used KEF bextrene drivers with low sensitivity that often needed extensive passive eq in the filiters lowering sensitivity even further. Then you could reach 20-25 Hz in "small" boxes. But you could equally well use JBL PA drivers in TL and get very high sensitivity TLS. Then a 20 Hz TL for a 15" drivers would be about 3 times the volume of a 20 Hz TLS for a nominal 10" like the KEF B139.
So I really do not see low sensitivity as intrinsic for TLS per se.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.