1 or 2 Titanic MKIII 15 kits or PC Ultra?

[sivadselim]

----------------

On 4/21/2005 11:10:18 AM Frzninvt wrote:

Gotta say something about Mike he sure responds with the facts and data! No BSing. Hell of job Mike!

----------------

yeah, unlike some others, who insist on spouting mealymouthed gibberish in an effort to impress.

thanks for the succinct dose of reality, mike.

[dragonfyr]

Don't worry, I'm sure you don't let facts confuse you!

[sivadselim]

no, but i am baffled by bullsh!t.

[dragonfyr]

Sorry the SynAudCon data is so confusing!

I guess we just have to be careful not to present any ideas that disagree with your emotional preferences...

And Mark Ureda will be quite amused to hear that!

[formica]

----------------

On 4/20/2005 3:56:22 PM dragonfyr wrote:

Just the assorting interesting pieces that don't tell me much about the actual performance. And as I have mentioned!, I would LOVE to know more about it as I have been intrigued by its potential!

----------------

Given the long wave lengths, I feel subwoofers are much less mysterious than mains especially given the available math models (Hoffman, Thiele, Small). If anything, I've been curious about SVS's actual performance given the lack of available data.

Posting some TEF/MLS/THD measurements would be of interest to many members here, especially after the rumours of high THD on some units (actually a couple SVS subs were tested in anechoic chamber by a third party, but results were not posted due to conflict of interest).

Although I don't have a Titanic... I don't mind eventually posting some MLS curves of my DIY subs once I'm setup.

Rob

[DrWho]

----------------

Comb filtering is the result of two or more apparent non-minimum phase sources that are superimposed! The net displacement of the medium at any point in space or time, is simply the sum of the individual wave displacements. In its most extreme form it is effectively active noise cancellation (to misuse the term), whereby the 180degree out of phase signals completely cancel, and in more common complex forms, this includes a mixture of phase and amplitude that sum, resulting in a modification of the individual component waveforms into a resultant waveform ...superposition. And what we commonly refer to as comb filtering are the nulls resulting from signals that are 180 degrees out of phase.

----------------

Einstein once said that if you can't explain it to an 8 year old, then you don't know what you're talking about. Not to say that you don't know what you're talking about, but you seriously could have worded that a lot simpler. I've always been taught that comb-filtering is when two sources are playing the same material, such that the physical location of the two sources causes relative shifts in phase (yay for constructive and destructive interference). Perhaps I misused the term in the sense that I was referring to the difference in phase information between the right and left speaker and not so much the 180 degree destructive interference. Nevertheless, whatever you want to attribute to the concept of ideal stereo reproduction, it still demands stereo subwoofers for the perfect ideal.

That article about polar response was very much geared towards live sound where lobing is an issue because you are trying to obtain a good sound at very large numbers of locations (polar response also has a lot to do with feedback control as well). A home setting typically will have only one ideal/main listening position, so any possible lobing issues can quickly be solved by positioning. Also notice that the polar shifts listed were on the order of half a wavelength and further apart...two stacked subs are inches apart, where the wavelength of 80Hz is like 12feet. Notice how little 1/4 wavelength affects the polar. Even at 3 feet apart, two subs are going to present minor alterations in the polar response (and 80Hz would be where the max change occurs).

Really it's pointless to play with all the numbers because the fact is that two sounds better than one. If we were to measure a large enough test population and could somehow determine whether or not two stacked subs sounded better than a single sub, then the science is going to have to somehow correlate with the observations. It seems to me that the general consensus based on reviews online is that two sound better than one (i'm yet to read a review that shows otherwise). Harmon Kardon has also provided some testing to show that two will provide a better frequency response and with more SPL.

[dragonfyr]

This is crazy!

Again, the Luddite response to objective measurement that DOES happen to correlate to hearing is amusing!

I am glad you can find some amorphous statistical sample to justify your point (really?), just as someone could find more people who like the rap over Beethoven. And the majority of polled high school seniors a few years ago couldn't locate Chicago on a map nor identify to which continent Mexico belonged. So I guess that's it for Chicago and Mexico!

But I had a fantasy that this was not an episode of Street Smarts and that there might be discussions based upon more substantive measures.

Its nice that you have chosen to reframe your statement, as my issue was not about the nature of comb filtering, but rather that comb filtering is NOT a fundamental element of aural localization. Its still not!

And whether 2 drivers reproducing the same spectral space create anomalies. The point is moot! They do! If you don't feel that the effect is significant, that is your choice!

Now if you want to reframe your statements and pretend that that was the point you meant to make but didn't, practice on your proverbial eight year old. As my point was rather fundamental and has been verified by far too many more knowledgeable and experienced that either of us.

What you might considering doing that would (seriously, as in 'sincerely') prove beneficial to you, is to do a bit of research into the basis of the article, and how it came to be published. If only I could claim to be the sole originator of the ideas! All I can claim is that I have been involved in a small portion of the research and in the verification process for a small portion of claims.

I have simply tried to present some ideas, many of which I mention PRECISELY because they do reflect recent modifications of long held beliefs, I will be glad to play the silly name drop game of those who were involved in empirically verifying such ideas. And if you would like to attend the next Loudspeaker Design Seminar, I will be glad to introduce you to many of those individuals who are responsible for the research and findings presented and you can personally tell them that they are idiots too! I am sorry if you feel that I don't explain things to your liking, but it seems that you have some issues with that regard as well, so I would suggest that we attempt to focus on the ideas rather then the means of expressing them!

But as you feel I write too much anyway, there is no way for me to present the entire presentations to which I was treated by the principal parties themselves! And I have literally notebooks of such material! And I hate the fact that I am limited too often to words rather then to diagrams that represent the concepts much more clearly!!

So since you feel you can find a small sweet spot where the effects are minimized (for only a small range of frequencies!;-), may I suggest you have a seat and enjoy, as some of us who can't simultaneously occupy that amorphous spot might want to be aware of, and to try to discover how to effectively mitigate the anomalies that exist within the real world.

I am very sorry if the complexities of the real world don't simply lend themselves to stacking boxes and pretending the phenomena don't exist!

But if you don't feel the trade off is significant, by all means, do whatever you want! Ignore-ance is indeed bliss.

But, quite frankly, I am tired of the few who prefer to focus on 'presonalities' rather then the ideas at hand.

[dragonfyr]

Regarding the published specs of the speakers, I too agree that the published specs are far too incomplete, to the point of not being very useful. They seem to be oriented more for inflated marketing claims then they do actual useful purposes.

My only issue with the Thiele/Small parameters is that while they are useful for designing and optimizing the speaker enclosure from a first principle design perspective, they do not provide much as far as a useful measure of the final unit - neither in an acoustical space nor in combination with other units.

{The only use the Thiele/Small parameters might provide is for someone such as yourself to reverse engineer a product and to possible determine the trade-offs that might have been made in the manufacture (which are quite common for a variety of reasonable cost-benefit purposes such as fragility, weight, material costs, etc. that contribute to service issues, shipping & material costs, etc. where the business aspect determine that the ROI does not warrant the additional labor, material, or shipping costs.) And I think we can reasonably understand why few manufacturers would be desirous of that process.}

I have attached what little information is found on the SVS website and I am in the process of scanning a few additional useful stats from the owners manual. This forthcoming information should be adequate for most applications.

I will do some additional checking as I believe DBKeele also did a review of the SVS unit a few years ago. I have his review of the Velodyne HGS18, but not for the SVS. Assuming he did a review, it should provide useful data...I will look.

As far as actual measurements, I cannot agree more regarding their usefulness!

As I have mentioned to a few in private posts, I am anxious to obtain a series of TDS/MLS measurements for a host of speakers and drivers. In particular the KHorn, La Scalas, a variety of individual drivers, and my REAL focus of attention, the Jubilee! (I am anxious to build a few of these and hence my interest in the de-construction' of the cabinet elsewhere on the site!!!) And barring the seemingly inevitable scheduling hassles, when my TEF is returned from Nashville in about 2 months - (complete with a NEW Earthworks M30!!, as the one sent with the TEF has been abducted by aliens, or some other strange set of circumstances, the exact details of which colleagues have not yet fully revealed! :-0 )

Having this data would greatly enhance acoustic modeling in such programs as EASE and CATT-A. Although, I must admit, the demand for this data outside of the commercial theatrical/industrial market is pretty limited.

Nevertheless, Klipsch (and others!) should be providing the libraries for EASE, CATT, etc., and augmenting their use by acoustical designers.

So, hopefully I will gather this data within the next several months and make it available for all.

[dragonfyr]

The primary difference in large and small acoustical spaces resides in the lack of a reverberant field in the small acoustical space.

Not whether it is live or Memorex.

[dragonfyr]

Here is the SVS chart I think you will find somewhat informative, along with excursion limits...

A few additional thoughts 'out loud'...

If you were considering corner placement and stacking the Titanics, you might consider placing the lower unit upside down and separating the two units with a piece of 3/4" plywood, thus allowing for a tighter packing density of the two. Just a thought as I have primarily played with the arraying of bass units that were horns, manifolds or forward firing units to the exclusion of downfiring units.

Or, if they are corner placed, you might even consider laying them on their sides and thus stacking them as 'forward firing' units without any loss in sensitivity but availing easier 'handling'...just a thought. It might justify some experimentation to determine any real benefits/losses.

[jacksonbart]

Does anybody sleep around here?

[dragonfyr]

(And as I mentioned before, I would Very much like to hear the Titanics, or at least have some meaningful tech data in lieu of the first hand experience...as I have been staring at them in the PartsExpress catalog for over a year wondering! For their price, they have the potential to be a real 'sleeper' of a deal!)

Here are a few links to SVS reviews with measurements:

There are plenty more...but some of these may provide some additional technical insight.

http://www.hometheaterhifi.com/volume_9_2/svs-cs-ultra-subwoofers-5-2002.html

PB2-Ultra (same driver):

http://www.hometheaterforum.com/htforum/showthread.php?s=&threadid=201221

http://www.hometheatersound.com/equipment/svs_pb12_ultra2.htm

Also, the SVS site has links for quite a few reviews. And quite frankly, it's hard to find a negative review anywhere!

http://www.svsubwoofers.com/reviews.htm

[formica]

----------------

On 4/22/2005 4:53:42 AM dragonfyr wrote:

My only issue with the Thiele/Small parameters is that ... they do not provide much as far as a useful measure of the final unit - neither in an acoustical space nor in combination with other units.

----------------

And as I'm sure you will agree, the space (the actual room, placement, and grouping) has a greater effect on the output than the variations of the theoretical versus actual enclosure performance.

----------------

... Thiele/Small parameters might provide is for someone such as yourself to reverse engineer a product and to possible determine the trade-offs that might have been made in the manufacture

----------------

Which can provide an interesting (perhaps not for the general public) window on which compromises were deemed most important to the potential customers. Note that I can understand the designer who is trying to provide a product which will appeal to a wide customer base where size and cost seem to dominate most designs.

SVS on the other hand has opted to cater more of a niche market and trade size for cost/performance... which is a respectable compromise from an audio fanatic viewpoint.

Unfortunately the published SVS FR curve is a typical marketing curve which tells very little about the subs performance... but there are a couple of interesting pieces of data on the tuning chart. We see that both the designs are EBS, which involves placing the driver in an oversized enclosure and lowering the tuning. I'm not sure what the drivers Fs is but I don't like seeing drivers tuned much lower than it... so 16Hz seems a little low. I'm guessing this is the source of the harmonic distortion measured in one of their other subs.

We also see that the driver has 28mm of maximum excursion... which means the Titanic woofer has about 33% more displacement. What you do with this added displacement is up to the enclosure design and in that respect, the Titanic Kit and SVS-Ultra 16Hz are very different subs.

If we were to design our own enclosure around the PartsExpress driver, it should outperform the SVS even without the Ultra's rumoured nice motor design. I'd also like to add that I don't think two subs is inherently worse than one (esp given the 80Hz and less), but a single unit is usually easier to setup for most people.

Velodyne servo subs are quite different from an engineering aspect given their capability of adapting. In theory, it has a lot of potential... but unfortunately with the added complexity, durability seems to an uphill battle.

----------------

As I have mentioned to a few in private posts, I am anxious to obtain a series of TDS/MLS measurements for a host of speakers and drivers. In particular the KHorn, La Scalas, a variety of individual drivers, and my REAL focus of attention, the Jubilee!

----------------

Actually Klipsch used to have some more accurate FR curves (pro and home) and polar responses (pro only) of their products. Unfortunately they stopped publishing them as they never looked as good as their competitors whom allowed their marking departments more liberty on "smoothing" them. Don't get me wrong, these aren't laboratory plots, but they do give a more honest representation than most other competitors. Check out some of Gil's (William F. Gil McDermott) posts of old literature including some sales brochures. No Jubilee... as the publishing change preceded it's design.

Later...

Rob

[DrWho]

I don't think klipsch is off their rocker with this system:

http://www.klipsch.com/product/product.aspx?cid=774

----------------

And whether 2 drivers reproducing the same spectral space create anomalies. The point is moot! They do! If you don't feel that the effect is significant, that is your choice!

----------------

That was my point, two drivers do present anomalies...the thing is, it takes two speakers to make stereo, and it only makes sense to have stereo subwoofers in a straight up two channel system. (to be more precise, the subs should be located just under the mains, or as close as possible to maintain a perfect stereo image...in the ideal world of course). If nothing else, I find low notes are very localizeable and I find it very annoying when I hear bass coming from other parts of the room that don't correlate to the stereo image coming from the mains.

----------------

What you might considering doing that would (seriously, as in 'sincerely') prove beneficial to you, is to do a bit of research into the basis of the article, and how it came to be published. If only I could claim to be the sole originator of the ideas! All I can claim is that I have been involved in a small portion of the research and in the verification process for a small portion of claims.

----------------

Directly from the ariticle:

"This Tech Topic is an attempt to quantify the theory and application of low frequency arrays for sound reinforcement applications, and is essentially a repackaging of an existing knowledge base using some modern tools for presentation and clarification. The Tech Topic does not address the modal response of the room, and makes the assumption that the spaces in which these arrays might be implemented are very large with respect to the array"

We have modal issues inside our home environment and our rooms aren't large with respect to the array, and we're not doing sound reinforcment (aka, we can aim for a smaller target audience). Basically, there are other larger factors to be considered when applying these principles to a room (at least the article tells us this). Also note that the article claims optimum coupling for any array occurs with these parameters:

"1. The number of devices

2. The spacing between the devices

3. The frequency (wavelength of interest)

4. The time offset between the devices (delay)

5. The polarity of the devices

6. The relative levels of the devices"

we've got two devices, very close, playing at low frequencies, no time offset, polarity equal, and the relative levels can be equal. sounds to me like an optimum situation for arraying multiple subs. (in fact, if using multiple is so bad, then why does the article discuss how to maximize their use?)

So even though this article wasn't directed for a small home environment, we still meet it's criteria for arraying a loudspeaker. I have had plenty of experience with LF arrays for live sound and the lobing there is extremely audible in larger venues. As the rooms got smaller, so did the effects of lobing (thanks to the modal response of the room.) Again, I call attention to the Harmon International white papers:

http://www.harman.com/wp/index.jsp?articleId=1003

Full White paper (PDF document)

My point about taking a poll was obviously missed. If you use theory to claim something has a negative result on the sound, then it should correlate that a negative result on the sound is observed. It is observed that adding an additional sub in a room seems to always have a beneficial effect and therefore there must be something wrong with the theory that says otherwise.

"Again, the Luddite response to objective measurement that DOES happen to correlate to hearing is amusing!"

I'm not saying that measurements don't correlate to what we hear. However, that doesn't mean someone isn't trying to apply anomalies that aren't applicable to the situation at hand. You're such a fan of measuring things. Why don't you go get your fancy testing equipment and compare the measurements between one and two subwoofers in a few rooms in your house, instead of presenting information about situations totally unrelated. Have you even read the Harmon article?

[BobbyT]

WOW! What did I get started?

Would it be better to stack the 2 Titanics? Or place 1 by each main speaker? This will be in a multi channel setup and not stereo altough it will be used for stereo.

[dragonfyr]

As is too often the case, we seem to be dealing with a moving target.

First, the issue of stereo versus LFE; aka music vs. theater

Yes, I agree completely that you can localize bass and that in music, including stereo, this becomes a significant factor. Hence my stressing in multiple places that the issue of localization and the non-directivity of low frequencies be better clarified, as TOO many confuse the non-directionality of low frequencies (referring to the more common low Q polar responses) with the inability to localize the origin of the sound. Hence many think that you can simply place the Sub anywhere in the room (in an attempt to deal with the acoustical anomalies) and that this will not have a detrimental effect upon localization and the perceived soundstage. And while this is minimized with a single LFE output as opposed to stereo, it still personally drives me crazy, as I prefer to establish an accurate soundstage and address the acoustical anomalies more directly.

My issue with the stereo topic related to 2 factors.

One, the LFE from a multi-channel receiver designed for home theater use treats bass as being non-directional/non-localizable or at least to the degree that they do not feel it significant enough to provide multiple channels of the lowest frequencies! Thus the use of multiple sources covering the same acoustical spectrum does present problems. And from the initial description of the task at hand, to provide bass for a mixed use multi-channel music and home theater, it seemed to imply the use of the non-stereo LFE bass feed.

The second issue, comb filtering, is an acoustical aberration and not a method of localization, and while present, it is not the basis for the creation of the stereo soundstage image. And especially with a single LFE feed (with the notable exception of one configuration, the Bessel array, which by the way, is not necessarily suitable in many cases to small spaces with limited distance from the source), comb filtering is a moot point.

So it seems when I speak of one, you focus on the other, and when you address an aspect, I have been focused on the other as well, in too many cases.

But with regards to the orientation of the article towards large acoustic spaces, the point was made for the express purpose of eliminating a rooms modal response from the focus of the issue. Instead, the focus of the article was upon the superposition of the signals and the resultant comb filtering and polar anomalies which are present regardless of the room size. These aberrations do become much more greatly apparent in large acoustic spaces where one has the opportunity to move farther away from the sources until their presence (in the form of vacuous nulls!) cannot be ignored (I say this despite the fact that they frequently are! And the amazing thing is that most attendees of an event in a large acoustical space are either completely ignorant of the massive acoustical aberrations or they simply attribute them to the wonderful concert experience! And while we can laugh about them afterwards, I know, I have mixed or teched quite a few very prominent and absolutely amazing technical train wrecks on both ultra-large scale stadium events as well as on television, only to have the whos it record company execs come up afterwards and offer congratulations and even bonuses for the best show they have ever experienced! (When I fully expected heads to roll! But if one learns to simply act as if it was planned, far too many seem all too quick to marvel at the ingenuity! So needless to say we simply looked at each other and rolled our eyes in amazement!)

I do however find it interesting that we often find it acceptable to obsess over, say, nuances of electrical design on the one hand, but are so quick to dismiss nuances in the acoustic realm that have a more substantial impact on the total listening experience.

Many of the aforementioned anomalies are often considered acceptable. But may I suggest that until you have experienced environments where these issues HAVE been well addressed, you may not be aware of the significant audible difference that can be obtained. And my focus is upon this aspect.

And to this end, may I suggest that we all attempt to define our frame of reference more carefully so as to increase the likelihood that we simultaneously address the same issue at the same time!? Its a thought!

[dragonfyr]

For a single channel feed, close coupling the units by stacking them vertically will afford the best horizontal response while also acting to mitigate some of the vertical response as well. And since additional gain is the ultimate goal here, corner placement would also be optimal.

I would also anticipate substantial modal issues, that can be addressed with absorption with Helmholtz resonators.

And to this end, may I suggest a website that you may or may not be aware of (it seems that we are always reinventing the wheel here - I sure wish we had separate areas to post reference material so as to build a searchable reference library!)

http://web.archive.org/web/20010211090156/www.headphone.com/EditorialHeadroom/RoomTubesChapterOne.asp

Not only does this present an excellent visual portrait of room modes, but it presents an excellent step by step process for building Helmholtz resonators (Tube Traps)!

OK, pile on!

[BobbyT]

Thanks great read.

[michael hurd]

Re: Dumax testing of raw woofers

"1) I am not quite sure what this testing would afford one, as it primarily determines the maximum excursion, which, along with the Thiele/Small parameters simply aids in determining which order alignment would be optimal for enclosure design....

As we are dealing with a 'finished' design, I think this information would be moot.

What would be useful is the total system response measured in more complete and useful terms other then the max spl and wattage listed on the parts express site. SVS does provide a reasonable (but still incomplete) spl vs. frequency response curve for 2 tunings on their website."

Well, for starters, how about reduced DISTORTION, due in part to an improved motor design that reduces major variations in motor strength and suspension action, throughout the VC travel?

"2) Additionally, combining subwoofers is not as simple as some here suggest. Simply placing them next to each other does not necessarily result in some magic coupling without aberrations. "

Again, with the long wavelengths we are dealing with, coupling is not really a problem. A 30 hz signal has a wavelength that is 37.56 feet, exactly how far away are you going to have problems in summing? Somewhere around 1/4 wavelength, so we'll say roughly 9 1/2 feet away. So you say, you have a crossover of 80 hz at 24 db/octave. No problem, 3 1/2 feet should be quite fine. Most consumer subwoofers can be stacked, and still have mutual coupling without abberations. The lower the crossover point, the farther away they can be.

As the article deals with LF directivity in the sound reinforcement world, and " subwoofer " means any device that plays lower than 150 hz, then you will have problems summing, without creating " power alleys " and huge nulls.

[michael hurd]

"http://www.hometheaterhifi.com/volume_9_2/svs-cs-ultra-subwoofers-5-2002.html

PB2-Ultra (same driver):

http://www.hometheaterforum.com/htforum/showthread.php?s=&threadid=201221

http://www.hometheatersound.com/equipment/svs_pb12_ultra2.htm

Also, the SVS site has links for quite a few reviews. And quite frankly, it's hard to find a negative review anywhere!

http://www.svsubwoofers.com/reviews.htm"

Hard to find a negative review on the SVS? Hmm. Hard to find a negative "review" on any product, even the Bose 3-2-1 system. I agree that the SVS CS-ultra is a very nice subwoofer, I have heard a pair over a year ago. If you have trouble believing that, then I will dig up the link for you.

I like that " speculation based on cone diameter, wattage and enclosure size does not a valid comparison make "

Lets see: The Ultra has a nice motor, underhung with a very large magnet and thick top plate. Linear excursion has it pegged at +- 12 mm, according to one of the charts.

The Titanic 15" has a larger cone, allowing it to displace the same amount of air, with less excursion. Again nice motor design.

We have a DUMAX sheet on the Titanic, showing linear behavior over at least +-12 mm of excursion. We have no such info on the Ultra. Short of testing the drive unit in question, I would have to say that the Titanic can sweep more volume ( cleanly ).

So you say that it still is not a valid comparison, because we are trying to compare a " complete system ", to a raw driver? Well, build a Dayton, and test it.