 
|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
|
|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
66.81.43.100
it seems many feel that the front baffle/cabinet edges and corners should always be rounded or at least beveled to a great degree to reduce diffraction issues. are such rounded or beveled edges important for optimal and superior sonics by reducing diffraction or is it more a philosophical issue of design or one without siginificant audible credibility since many high end speakers do not incoporate such rounded or beveled edges and corners and do not use other diffraction tweaks such as felt pads around tweeters etc.
Follow Ups:
I first became interested in minimizing diffraction when modifying a pair of NSM 10S Ultra Definition Mini-Monitors for nearfield listening. Rounded cabinet edges are said to help with diffraction, but very few speakers incorporate this into their design, and if your speakers weren’t manufactured this way, it is a difficult modification to make, and beyond what the average tweaker is willing to get into. Therefore, you are left with the option of some sort of diffraction minimizing device around the tweeter.I first tried a pair of the “Felt Defraction Rings for Tweeters” available from Madisound.com. They had a slight effect of making the high end a little smoother, so I knew I was on the right track. But there are two problems with these simple rings. First of all, these rings are very thin and are not thick enough to fully absorb the reflections. Research has shown that that in order to be of maximum benefit, they should be made of real wool felt in the proper density and of substantial thickness. Secondly, in order to help eliminate the problem with diffraction at the cabinet edges, it is recommended that the diffraction material extend all the way out to the edges of the baffle. (This may not be practical, or even necessary with large speakers / baffles, but with small mini-monitors like the NSM’s, it doesn’t make sense to stop with the diffraction material before you get to the edge.) The Madisound Rings help with diffraction immediately surrounding the tweeter, but do nothing for the diffraction at the edges of the cabinet.
After more web searching, I discovered that a diffraction pad in the proper density, thickness and dimensions was available through a company by the name of ‘Diffraction Be Gone’ and are marketed as ‘Touchstone Tweeter Surrounds’. I contacted Jim Goulding at diffractionbegone.com/and gave him the make, model, & dimensions of my speakers and shared with him my plans to optimize them for nearfield listening. He assured me that the elimination of diffraction provided by his tweeter surrounds would be most beneficial in this particular application. Since I was not going to be using the grills on the NSM’s, I was a little concerned about how the diffraction pads would look. (The natural felt color of the pads are a grayish color.) Jim offered to dye the pads a brown color which would blend in better with the walnut wood finish of the speakers. I accepted his gracious offer, and within about a week, I had the pads and was ready to give them an audition.
My first reaction after hearing my speakers with the Touchstone Tweeter Surrounds in place was “Good Gosh… these things are killing the highs!” But how could they? The pads weren’t in front of the tweeter, so there’s no way they could have been blocking the highs. Then it dawned on me - the highs that were now missing were not the primary highs emanating directly from the tweeter dome; they were the highs that had been reflected off the baffle area around the tweeter. These pads were doing exactly what they were intended to do.
With the Touchstone Tweeter Surrounds in place, the highs were much cleaner sounding, a result of subduing the reflections which were arriving at my ear slightly behind the direct sound and smearing the top end. The imaging of the NSM 10S (already acknowledged as being one of the best in this regard) was enhanced with the use of these diffraction pads. With the tweeter reflections removed, the time and phase alignment of the drivers is preserved, leaving the subtle cues that assist with the pinpoint imaging of individual instruments within the soundstage. Needless to say, I am very impressed with this mod.
The stark difference and apparent loss of highs that I noticed when I first installed the pads was because after hours and hours of listening to these speakers without the pads, THAT had become my ‘reference’ sound. But now, after spending a considerable amount of time listening with the pads in place, I prefer the way the speakers sound with the Touchstone Tweeter Surrounds, which is more natural and relaxed. The sound with the pads installed has now become my ‘reference’ sound, and removing them causes just as stark a difference, but not for the better. Without the pads, the speakers are brighter, harsher, fatiguing.
Some have questioned why they should pay what Jim is asking for his Touchstone Tweeter Surrounds when you can buy the felt and cut them out yourself. Go ahead - buy a sample piece of 3/8” felt and try to cut it, but I’m telling you now, even with the sharpest knife or scissors, this stuff is almost impossible to cut and get a nice clean edge; and trying to cut out a clean hole in the center for the tweeter - forget it! (I think Jim uses a modified die cutting press to cut the felt which results in a nice precision cut pad.) Considering the quality look of these pads, along with the improvements they made in the sound of my speakers, I consider the Touchstone Tweeter Surrounds to be a bargain.
Below is a photo of my NSM 10S Ultra Definition Mini-Monitors with the Touchstone Tweeter Surrounds installed. A complete rundown of the modifications I have done to these speakers can be seen in my personal email to Erol Ricketts which has been posted on the homepage of the NSMT website at nsmt-loudspeakers.com/.
Ken Harris
Nashville, TN
Edits: 04/28/12 04/28/12 04/28/12
Well, here's what I know: When I had my Dahlquist DQ-10s, the assortment of little baffles inside the cage was an assortment of little diffraction egdes. I wrapped the array from behind with some absorbent material, having the material curl around the baffles a little to the front. The result could be described as "cleaner" sound -- the music no longer had that low-level random zizzing in it.
Virtually 0 edge diffraction issues. These speakers are amazing, they do soooooo many things right!
-RW-
The enclosures are also very quiet. No dreaded single standing-wave coming out of the port, either.Very few audiophiles have ever heard a quiet-enclosure, clean-diffraction loudspeaker, including those responding below to your post who just don't know what they are dismissing. The effects on timbre, expression and attacks are considerable.
IME the effects are most audible on real stereo recordings of acoustic music. Typical popular genre recordings - multiple close mono miked mixdowns - have a lot of phase and timing errors and cancellations built in already, which could easily reduce the importance of diffraction. Tongue in cheek!
I have not heard a single sharp-edged speaker I can live with, since buying the spheres. Not even mini-monitors plus subs where edge-shape shouldn't matter, but IMO still does.
Even back then the 'spheres were 'bleeding-edge audiophile,' using good sounding Philips yellow-block film capacitors in ladders for each C in the circuit, air-core coils, and low L resistors. No low-pass just R and C to Eq the mid-bass* driver's Z-rise. * It rolls off after 3.3 Khz.
Foster, Japan drive units. Easy 8 ohm load, 91db/w/1m, 35-13k +/- 3db.I bought them instead of a pair of almost new black-mesh QUAD 57s (among some other lesser options). No down sides, almost as good at 3D, open throated on singers - which I've been for over 40 years, deeper bass, and went louder while still clear. They have revealed and assessed every other change I have made to the system since.
? e.g. Polystyrene capacitors versus teflons for passive RIAA Eq? A mixture is best, as for MF R's, but bulk foils are slightly nicer, and less selecting needed.
They have confirmed my preference for real stereo recordings of acoustic music (?simply-miked IE NOT multiple mono.) I listen to acoustic music and sometimes sing in it, at least once a week.
They can be aimed very precisely as their stands are dished, which doesn't hurt at all ;-)!
John Dunlavy 'respected' them, his own words to me when he was based here.
Note that a post in response is preferred.
Warmest
Timothy Bailey
The Skyptical Mensurer and Audio Scrounger
And gladly would he learn and gladly teach - Chaucer. ;-)!
'Still not saluting.'
Edits: 04/18/12 04/18/12
.
I do remember Apogees on Krell MBs on demo in Melbourne and picking that the absolute polarity was out. Much surprise when it was switched and was right.
Ported speakers' Q at fb is more important than that they are ported, imo. A critically damped or over-damped alignment works best IME and the corpus of reviews seems to confirm that.
I have heard quite a few 63s over the years and can usually get them a lot better in a given room. I've done two set-ups with those Scandinavian dipole woofers, and they needed to be pointed almost at right angles to that used for the main's midbass and on up. Which kind of made them useless as stand-subs. Both owners were a bit peeved but bought new stands, ;-)!
I'd only be guessing but in much larger rooms this might all be easier.
Distributed subs near corners and under my mains is where I am going, with the tower of spheres plan.
Note that a post in response is preferred.
Warmest
Timothy Bailey
The Skyptical Mensurer and Audio Scrounger
And gladly would he learn and gladly teach - Chaucer. ;-)!
'Still not saluting.'
This need an explanation? Something that works. Hearing is believing. It'll stand measurements, too. For when your cabinets don't have seriously rounded edges. Waveforms shorter than the width of your cabinets are not diffracted, they are absorbed with the use of what you see on the speaker baffle above. That just leaves the pure signal (i.e. what the microphones captured). Imagine how diffracted waveforms could be mucking about with the delivery of everything upstream of them, what your system is working in earnest to deliver. Well, remedy that and then you won't have to imagine. Your recordings and your components will thank you.
Edits: 04/20/12 04/20/12 04/20/12 04/20/12 04/20/12 04/20/12 04/20/12 04/20/12
Brought to you by.....
(eyeroll)
Actually, I am surprised he left out the full contact information in bold letters, a link to a youtube infomercial, current pricing, and ordering information.
For those who feel the need to put felt on their speaker baffles - there are a number of easy supply choices (in addition to the one continually re inserting himself into this thread) to assist your creative urges to modify your speakers.
Here's a popular supply choice: www.mcmaster.com
Felt on speaker baffles is not rocket science. With a little dye and an F10, F11, or F13 sheet of felt, one can experiment to their heart's content.
Hopefully, the next unsolicited post in this thread about BAFFLE ROUNDOVERS...HINT...HINT...HINT...(JIM)... will be on topic and not focused on someone's semi lucrative side job... (massive eyeroll)
And a massive raspberry to you, sir. Just kidding, I thoroughly enjoyed your post, someone should get your dander up more often, and I don't mind being called an opportunist. Plus, the info you provided for DIY'ers is solid. That's kinda like the way I started with my own speakers.
I'm wondering, are you and Tim taking about the same thing? Frequency response and/or time and phase issues? Diffraction is insulting to both, right? What, then?
Timbo said:
"I have not heard a single sharp-edged speaker I can live with, since buying the spheres. Not even mini-monitors plus subs where edge-shape shouldn't matter, but IMO still does."
Subs suffering from audible effects of sharp edged baffles????
Really????
Looks like audiophool tin foil hat time once again....(eyeroll)
Perhaps you've been hanging out with Dennis Murphy a bit too much....LOL!
Descending to ad hominem is unnecessary.
mind how you go!
Note that a post in response is preferred.
Warmest
Timothy Bailey
The Skyptical Mensurer and Audio Scrounger
And gladly would he learn and gladly teach - Chaucer. ;-)!
'Still not saluting.'
"I was talking about whether diffraction matters with mini-monitors!"
There's no such thing as a "mini-monitor". If there were, all studios would use them (or be required to use them) instead of the giant energy-sucking monitors they use, because they would save thousands of dollars and/or help to save the planet.
If you're talking about small 2-way speakers, such as the plethora of 8" 2-ways that are everywhere in the marketplace, supplemented by a sub (which actually upgrades them from a so-so 2-way to a poorly designed 3-way), I'd still argue that cabinet diffraction is among the LEAST of the concerns.
In any case, cite 3 examples of the "mini-monitors" to which you refer.
I thought that large speakers as a rule were -more- efficient than small ones. Am I still in the Sixties?
And they did have diffraction problems. They were audible as sources when compared to the baby single driver spheres.
Have you ever heard a speaker where diffraction had been designed out of equation, say the Gallo spheres. In particular the coaxial versions?
Note that a post in response is preferred.
Warmest
Timothy Bailey
The Skyptical Mensurer and Audio Scrounger
And gladly would he learn and gladly teach - Chaucer. ;-)!
'Still not saluting.'
Chocolate lover, to try and answer your query, the way that diffraction effects our end game is twofold. When diffracted waveforms sum in, they are late arriving and out of time and phase. They can be additive to our frequency response unless the designer recognizes this and has compensated for it. Secondly, because they are late arriving and out of time and phase, they will slightly confuse proper imaging and depth of field. So, the speaker with diffraction effects will benefit from a smoother, more wholesome sound of instruments and voice with truer imaging and separation in space, AND a more 3D soundfield . . sans diffraction. Whatever the recording tho noticable on some more than others. I listen to a lot of acoustical music recorded on location myself and want ALL the information as truthful as I can get it. Rounding away a speaker's cabinet edges is a step in the right direction regardless if the frequency response is linear or not. Cheers.
Edits: 04/17/12 04/17/12 04/17/12 04/17/12 04/17/12
I go by the results. Which sound the best - then correlate back to what the designers say. If Speaker A sounds better than speaker B by a significant margin then what the designer of speaker A says carries more weight than the designer of speaker B - even if there are 350 manufacturers making similar speakers to speaker B and only 5 who make similar speakers to speaker A.Three of the top 5 speakers I have heard are hard edge speakers and several speakers would get honorable mentions.
Audio Note, Trenner and Freidle's Ra Box, and recently the ATC SCM 150 pro studio monitor are 3 of the 5 in my top 5 thus far. Acapella is not totally applicable here and the other is a Manger driven single driver horn.
Pictured is the ARC SCM 150 -
Edits: 04/16/12
Personally, I wouldn't hold the ATC speaker you pictured as the poster child of a great sounding speaker. To me, they sound like a decent box speaker. To a lesser degree, the AN speakers too.
I guess it all depends on your frame of reference and your level of expectations...
Well it depends on whether it can emotionally involve me or not - part of the AN speakers prowess may very well be owed to their digital - nothing else in the audio industry is the same and nothing else in the audio industry that I have heard is as as good at CD replay. Since this is still the largest current medium of music and since it is usually played through AN speakers it has a tremendous advantage right out of the gate.
I'm not terribly interested in large loudspeakers - the boxed designs don't sound cohesive and panels unfortunately always sound like panels (ie; they sound not quite right) which isn't to say I don't like some of them or even that I would no consider owning a set. But the ones I consider good are either too big or too expensive or both. And if a SET can't drive it - it is disqualified as being considered a high end product on that merit alone since you have to drive it with inferior sounding equipment right off the bat.
The AN E took me a couple of times to get them - kind of like A Clockwork Orange.
The ATC was driven by an all tube set-up which runs counter to what would normally be used - which helped my opinion of them. And then it depended on the tube type. This is always fascinating thing with tubes.
The trick with boxed speakers is when you hear real full bodied bass - realize that that is real full bodied bass and not distortion - the fact that non boxes don't have any real full bodied bass - may make it seem "clearer" but in reality - it's clearer cause it hacked out a massive amount of what was on the source disc and has no semblance of what a live drum kit or piano low notes sound like in real space. Baby out with the bathwater.
While it is true that how a speaker sounds is ultimately all that matters, the OP's original query on the legitimacy of baffle edge round over usage deserves a complete and accurate response. The pictured example you gave is further evidence of what I was talking about. Notice the wide baffle and offset tweeter position. These two factors alone are often enough to eliminate any noticeable diffraction effects on the speaker's response. The staunch proponents of large round overs or heavy felt will always try to ignore how some of the world's "top rated" speakers do not possess such features yet they exhibit the kind of exceptionally flat response that would suggest a lack of diffraction problems. No one is saying that round overs can't help or that proper use of felt is a waste of time. The reality is that like a lot of things in the audio business, the effect of these methods is often heavily overstated and much of the details that actually explain what is measured or observed are left out.
I think what largely happened is that style took over - fat wide baffle speakers like the ATC, Audio Notes, Snells (the early good ones the AN's are based upon, Harbeths, Devore Orangutan (pictured), Trenner and Freidl etc. There are other good ones that I have not heard yet or not enough to comment too much on like the Shindo Field Coil, and Classic Audio speakers.The appearance is needed in order to sell sexy sleek looks to mainly male buyers - marketing guys make their living on knowing which music to play at a shopping mall and at what volume to get people spending and they know that "thin is in." To make such inferior designs sound decent they do all sorts of work to get the square peg to fit into the round hole and spend lots of cash on marketing science spin.
It never ceases to amaze me that some boring old rectangle box with a paper woofer and silk dome tweeter can so utterly embarrass speakers that spend massive amounts of advertising on state of the art (cough) bullet proof drivers and diamond tweeters with special separate top loading tapered technologies and others with their dual drivers, specially curved boxes or snail shaped enclosures and or pistonic motions this or that blather on top of blather. Sure there is probably some real truth in there - the compromised skinny ass box with puny plasticy sounding drivers probably really does sound significantly better if you curve the cabinet - but so what? You started with 3 day old crappy meatloaf and then you slathered a nice tasting gravy on top and you made the truly disgusting into something somewhat edible - thanks but I'd rather the thing that doesn't need to be "fixed" to be passable - even if the thing looks uglier.
Edits: 04/17/12
Isn't it true thst diffraction effects contribute to time and phase error as well as frequency response anomaly? Correcting the frequency response is one thing, but time and phase error is another. Please tell how this aspect can be dealt with without correction. Offsetting tweeters will make it less of an offense, but what about speakers whose drivers are stacked in a straight line with narrow cabinets as are fashionable these days?
Edits: 04/17/12
I'd add that the ATC grills fit around the baffle flush with it and are angled on the sides which also reduces diffraction like a rounded corner making the ATC line a poor example of a hard edged baffle.
And they make damn good speakers which was why ATC were the last speakers purchased by Gordon Holt.
The pro version of the ATC 150s differ somewhat besides being active: No grille but rounded edges.
Btw people who compared passive to active larger ATCs to the real instrument very much preferred the active ones by a fair margin. In their case the amps are rack-mounted and do not share the cab with the drivers.
Almost agree - except they sound better without the grills.
I am also glad to see that ATC puts a premium importance of driver matching and pair matching of their speakers - few speaker makers bother or advertise the fact or for that matter even seem to understand it. AN also gives a hoot about it as did Snell.
This topic has been kicked around numerous times over the years on the Madisound and PE DIY forums. The consensus of the "gurus" seems to be that at LEAST a 2" roundover is required on the baffle edges to measurably and audibly improve diffraction effects. This of course complicates cabinet construction, and adds cost.
The "ideal" baffle shape, from a diffraction POV, is a perfect sphere. I believe this knowledge goes back to F. Alton Everest, or earlier, but only a few manufacturers (like Gallo) have every really implemented it in a commercial design.
Felt is also good to absorb early HF reflections from the baffle. A nice thick blanket of natural wool felt covering the entire baffle surface around the drivers (as used by Spica, Vandersteen, and others) will of course be of more benefit than just a donut around the tweeter, but would probably work best if "designed in," along with the crossover, as opposed to being applied as an after-market "fix." Which is NOT to say that after-market products such as Jim's will be of no benefit. Even the tweeter donuts sold by Madisound can result in some measurable and audible smoothing of the response at the highest frequencies.
Caspian said:"This topic has been kicked around numerous times over the years on the Madisound and PE DIY forums. The consensus of the "gurus" seems to be that at LEAST a 2" roundover is required on the baffle edges to measurably and audibly improve diffraction effects. This of course complicates cabinet construction, and adds cost."
Firstly, there aren't any "gurus" lurking about on any of the forums you listed - guru wannabees maybe but not real "gurus". Most of them are novices who have day jobs that are not related to the audio industry. And much of the advice given is either wrong, useless, or misleading. Consistent with that observation is the quote shown above. You will not find any professional in the audio business who'll support the assertion that "at LEAST a 2" roundover is required to measurably and audibly improve diffraction effects". As I noted earlier, there are a number of factors involved that determine whether or not a given speaker's response will be significantly affected by roundover at the baffle edge or lack thereof. Among these are baffle width and tweeter placement. In 30+ years of examining my own data and that of others, I've not encountered a single professional who has either offered or endorsed such a claim about 2 inch round overs. This is just one more example of bad advice from non professional wannabees.
Edit:
Btw, like most of the "advice" found on internet forums that are trying to sell speaker products, this statement by Caspian:
"The "ideal" baffle shape, from a diffraction POV, is a perfect sphere. I believe this knowledge goes back to F. Alton Everest, or earlier, but only a few manufacturers (like Gallo) have every really implemented it in a commercial design."
IS TOTALLY WRONG. The spherical speaker shape has the highest response drop off associated with diffraction precisely because of the equal distance from the "baffle edge" to the speaker driver over the entire speaker's circumference. This accentuates the affect of pressure drop at the precise frequency tied to the quarter wavelength that terminates at the circumference edge. For every accurate post on such internet forums, I've seen several stuffed shirt posts by people who are trying to impress others with all that they don't know - a lot of hot air from very insecure individuals. Unfortunately, they have good reason to be insecure because their "knowledge" comes more from reading a few old text books or trade articles rather than hard core engineering experience and testing that comes from working in the industry day to day for many years.
Edits: 04/16/12
Dude, I'm not looking for a fight. And guys like John Kreskovsky, John Krutke, Jeff Bagby, and Dennis Murphy -- all of whom have done design consultation for high-end manufacturers, and have decided to share their knowledge with the DIY community out of love of the hobby -- are hardly "novices."
Please see the linked article, published by Harry F. Olson in the JAES way back in 1950. Graphs 6-17 show the measured effects of diffraction on frequency response, caused by different baffle configurations. The sphere and the rectangle with deeply chamfered edges (he didn't test any roundovers) clearly show the best responses. This is not new information, and I doubt if the laws of physics have changed in the interim.
You can easily model the effects for yourself, with cheap/free software such as The Edge, or Baffle Diffraction Simulator from the FRD consortium, or more advanced professional software like Sound Easy.
If you look carefully at the linked paper, you'll find that the responses in figure 15,16, and 17 (the rectangular box) have the best response. The sphere with offset tweeter (figure 6) is not a practical design and you didn't specify an offset driver which makes a significant difference as the centered driver on a more practical circular baffle (figure 8) is the worst case (which is what I was referring to).Additionally, I'm familiar with most of the people you cited above. Most of them are novices who routinely make ridiculous and inaccurate statements online with the exception of Mr. Krutke who is the closest person to being a professional in the industry and has actually designed and had manufactured an excellent mid bass driver. Anyone can work up a passive crossover and give the plans to a cabinet maker as some of the aforementioned "gurus" have done. I've heard their products and they all pale in comparison to those designed by true industry professionals at PSB, Revel, Canton, KEF, NHT, Vienna... etc - in some cases at a fraction of the cost. The more experienced companies routinely produce products with a much better balance of sensitivity, dynamics, spectral accuracy, and imaging coherence. Unfortunately, with modeling programs that generate transfer functions and response estimates based on rough dimensional parameters and raw driver responses - virtually anybody can build a speaker who has access to basic carpentry tools. Buying expensive Accuton or RAAL drivers doesn't make up for the lack of knowledge in acoustics or electrical engineering which is why some of the aforementioned manufacturers do a heck of a lot more for a whole lot less.
Edits: 04/17/12
You are correct that offset, baffle width and driver's bandwidth/coverage play a role in diffraction - for flat baffles on rectangular boxes. And that it can be managed to some extent using those factors. I disagree that it can be made to disappear."you'll find that the responses in figure 15,16, and 17 (the rectangular box) have the best response. The sphere with offset tweeter (figure 6) is not a practical design and you didn't specify an offset driver which makes a significant difference as the centered driver on a more practical circular baffle (figure 8) is the worst case (which is what I was referring to). "
You have made a significant error about spheres and the drivers in the Olson paper. I suggest that you read the referenced 'OLD' paper by Olson, all of it, a couple of times. If you had read it you would not have made this rather revealing error.
The reported tests used the same very small high quality driver, to ensure that it was a pistonic source. The pictures of the sphere and the hemisphere throughout the paper are offset so that you can see the single driver clearly. So Fig 6 is not a special case with an offset tweeter to give that very smooth FR.
Put a single driver on a sphere - anywhere - and it will be centred on the sphere by definition (as would any infinitesimal dot be,) no!? And, you would measure it on axis, as with all the other shapes covered.
Fig. 6 - the sphere with a single - centred - driver, has the smoothest FR / best diffraction behaviour. Despite the driver being - inevitably - centred. So that rule about centred drivers falls away - in the case of a spherical enclosure only.
Because a sphere is not a circular baffle as it does not have any edge. It will have a very smooth BDS dip, as predicted by the spherical diameter formula, so that compensation for it is easy to build for, or you can forget about it, as my 2-way spheres do. Room-gain helps.
Even a truncated sphere with a small sliced (circular) baffle is measurably superior to a 90-degree edge box and audibly so.
In the late 1970s. Enclosures of the same Vb, same tuning, stiff laminated walls for the box (3mm hardboard pressure glued to the inside walls, which took days) same amount of BAF. Gave the box every chance by offsetting the driver (not done on the sphere) and making the baffle area narrower than that of the sphere, and maximally dissimilar dimensions, sharp edges but. The mass of the enclosures were matched using a concrete block glued underneath the box encl, so motor reaction effects were at least similar. Same screws, same plastic rawl plugs. The same WR 5inch Coral Flat 5 was used.
Measured FR? no contest. Listening? not in the race. Measurements done at ANU. More than 30 years ago.
Offsetting the driver on even the small circular baffle on the sphere did help a bit more, so the production model 2's had that as well.
Now, the audible superiority back then may have had an additional cause - that curved enclosures are quieter and are self damping. The test did try to minimise the differences there, and the enclosure talk of boxes just is a given.Practical / real spheres then were truncated and thus did not have the predicted single/audible internal standing wave either. Just in case you want to bring that up. Neither did Gallo's spheres. Very good tight bass though.
It is plain to me that wide radius curved baffle edges are best. Probably not practical. Curved edges are better than sharp, and felt helps, a lot.
LBNL John Dunlavy 'respected' both the Model 2 Audiosphere with the single Coral Fl5, and the 2-way Model 3. Told me so himself.
Mind how you go.
Note that a post in response is preferred.
Warmest
Timothy Bailey
The Skyptical Mensurer and Audio Scrounger
And gladly would he learn and gladly teach - Chaucer. ;-)!
'Still not saluting.'
Edits: 04/18/12 04/18/12 04/18/12 04/18/12 04/18/12 04/18/12 04/18/12 04/18/12 04/18/12 04/18/12
For the record - the paper by Olson was designed to prove his assumption that the spherical enclosure would produce the "smoothest" or most "ripple free" response. In attempting to suggest I have been in error about spherical enclosures, Timbo conveniently ignored this aspect of Olson's paper and study:
"The axial response-frequency
characteristic thus obtained was corrected
so that the volume current produced
by the mechanism was inversely
proportional to the frequency, as previously
described."
The energy lost to diffraction (bending around the spherical enclosure into free space) is maximal with the sphere enclosure - it has the highest pressure drop off. The curves presented in the linked paper are deceiving in that this heavy drop off at the lower frequencies is "corrected". In the real world, the one in which we actually build speakers, we often don't have that luxury of either "correcting" the drop off in response via a paper mathematical computation, via electronic equalization, or via the sacrifice in sensitivity that happens when high frequency rise is compensated for passively. In point of fact as should be obvious to anyone with the slightest aptitude for physics, the response of spherical shaped loudspeakers is the most dominated by the phenomenon known as diffraction. Without heavy equalization, that fault or problem would likely be the most evident to the average listener of all the enclosure types listed in the linked paper.
There is no magic bullet or "perfect" speaker design. Some designs incorporate more rigorous research and development or engineering finesse than others and that's about it.
MYTH ABOUT SPHERICAL SPEAKERS BEING THE BEST POSSIBLE DESIGN - BUSTED!
:)
that a sphere has the smoothest diffraction behaviour and the smoothest frequency response.
'Heavy Eq' needed to compensate? By the time the curve has flattened again about a 3db shelf is enough IME and that of most others who have built BDS shelf Eq.
Where did I deny that BDS losses occur with spheres or that it is theoretically more lossy? It does but slowly and smoothly. In real rooms by about 3db by where the FR response flattens again. Heavy Eq?
I am not so much concerned with FR, but with the effects on timing, interplay, expression and nuance which are what acoustic music is about. Villa knows most of the market doesn't much care for acoustic music.
Villa is shrieking to the rooftops that BDS loss is a FLAW of spheres, yet it is by definition present in almost all speakers, but far less smoothly than with a sphere. Right in there between 200 and 400 hz where much that is fundamental in acoustic music happens.
Eq'ing for BDS is built in to almost all 'practical' dynamic speakers even the big ones, 2-ways, even 3-ways, no matter what the shape of the enclosure.
Viz. A '2 and a half way' column with two or more woofers, one covering bass and mids, and one doing bass below the BDS only. OR The tweeter and mids get padding resistors up to -6db worth.
The 2.5 way might seem to be a more transparent way, than using padding R's but doesn't always get good reviews.
The easy way to compensate for BDS - from a sphere - is not heavy Eq or a parametric, but a shelving response circuit in place of or in addition to a bass tone control. Sitting in the near-field and playing a bit louder works for me. So, I'd rather have less circuitry and not bother.
With boxes the FR is no longer smooth, and the compensation ought to be more complex. But most mfrs don't bother to fully compensate the peaks and dips, if at all.
I'd rather have good diffraction behaviour with a smooth BDS curve below about 200hz, than use a box with poor diffraction behaviour.
;-)!
That I happen to have chosen to use spherical enclosures is an educated choice, and I did it because I found that they got out of the way of the music. As predicted by the science.
Note that a post in response is preferred.
Warmest
Timothy Bailey
The Skyptical Mensurer and Audio Scrounger
And gladly would he learn and gladly teach - Chaucer. ;-)!
'Still not saluting.'
Well, you can always treat the box. That's what I do.
Using a variety of techniques - some of which were outlined earlier in this thread, diffraction based response anomalies associated with the use of "boxes" can be essentially eliminated. The measurements don't lie. And one of the smoothest measuring speakers ever made for home audio is NOT shaped like.....(gasp)... a sphere.
http://www.stereophile.com/floorloudspeakers/jbl_synthesis_1400_array_bg_loudspeaker/index.html
It's not often you hear John Atkinson say things like "Good grief!" about a speaker's ruler flat response.
" Note also that the JBLs are generating full output down to 25Hz. But, good grief! Look at the in-room response above 200Hz: It is extraordinarily flat and even, falling within superb ±1dB limits other than a very slight boost at 2kHz. "
This is clear, unambiguous, and convincing evidence that the issue of diffraction related anomalies can be eliminated to the point of inaudibility in NON SPHERICALLY shaped loudspeakers. The same cannot be said of true spherical loudspeakers which in the current highest expression of the art form, still exhibit uneven response despite the assistance of full digital equalization. Theory and practice are not always on the same page.
http://www.stereophile.com/content/cabasse-la-sph232re-powered-loudspeaker-measurements
Building a limited bandwidth and limited dynamic range loudspeaker and equalizing its response to create a smooth graph and prove a point is one thing. Building a real high fidelity loudspeaker that exhibits truly exceptional performance in every respect is quite another.
The description of the first case in the linked paper is not very explicit in terms of the driver's position. It could very well be offset with the center axis of the driver not being co-linear with the sphere's radius. If one relies on both the pictures and the description - it certainly is a valid perspective and not clearly WRONG as you assert."ig. 2. The sheet
metal sphere shown a t ( A ) is 2 ft. in
diameter. The loudspeaker mechanism
Is mounted with the cone approximately
fludh with the surface. The sheet metal
hemisphere shown a t (B) is 2 ft. in
diameter with the back closed by a flat
board of hard wood. The loudspeaker
mechanism is mounted upon the zenith
of the hemisphere with the cone of the
loudspeaker mechanism approximately
flush with the surface."
Offsetting the driver could actually improve response by reducing radiation into segments of 4pi space and similarly focusing that radiation into other segments.I read this paper many years ago. And I stand by what I said. Sphere's are not practical designs. The severe drop off caused by centered placement of drivers on a sphere requires heavy equalization to produce an acceptably uniform response. More importantly, it complicates the process of integrating multiple drivers successfully. Cabasse's Sphere is a very obvious case in point. Heavily equalized coaxial speakers are not for everyone. Like anything else, there are strengths and weaknesses. And there are technical and practical challenges to getting it right. If it were as you suggest - that spherical enclosures are head and shoulders above all other types of enclosures, then the high end market would have been flooded with spherical shaped speakers not long after Olson published his paper 60 years ago. The traditional box with offset drivers and rounded edges can readily match the sphere's level of performance from a diffraction perspective (as evidenced by the last three graphed responses) WHILE incorporating a higher performance, multi driver design at reasonable cost. I've heard Cabasse's Sphere and they are certainly impressive but for what they cost - they ought to be. Given the limitations imposed upon them via the spherical shape - there are other speakers that outperform them in several respects that don't cost anywhere near as much - flatter response, deeper bass, greater dynamic range...
Edits: 04/18/12
I gotta think that the resonance -inside- a sphere would be insurmountable.
Consider a situation where the sphere is sized such that the motor of the driver occupies the central space of the sphere- then you'll have a complex set of modes caused by the varying dimensions to the motor from the inside of the sphere.
Bass is supposed to sound big. 6.5" is not a woofer size.
Thank you 3rdRock for your input. As I have been tenaciously reminded, this thread is about something other than an aftermarket treatment to remove diffraction effects so I will bugger off. The flash made them look quite pink. They are not pink. Again, thank you. Listen well.
To state as you did in earlier posts that it is 'WRONG' that a sphere gives the best FR is plainly and utterly incorrect.
You were wrong and I am asking, given that you were also quite rude to another poster in that part of this thread, that you acknowledge that at the top of this thread. I also think you should do so to the relevant poster.
Next? With a single driver on a sphere 'offseting' the driver is clearly unnecessary because the FR is as smooth as it gets on axis.
You still have not got that as one of the central facts the paper established, for all time.
I can not believe that you ever read the paper or absorbed it, your speed and narrowness of response here and now simply serves to confirm that view.
Gallo showed that spheres are practical enclosures and that they do sound very quiet. My spheres are heavy, but far lighter than many current 'high-end' wide range arrays, and single piece enclosures. But, they are easier to aim / adjust to very fine tolerances in azimuth, elevation and rotation, by eye. That's a pretty practical idea I think. And, I started treating our difficult room DIY and learning about speaker positioning in the 1980s. Also very practical, and cheap to boot.
That both sides of the high end market haven't taken them on, simply confirms for me that one of the central propositions of market theory - that consumers and suppliers are rational often fails to be true in reality. Mostly untrue is more accurate IME. Further I am no longer persuaded that the short term interests of consumers and investors should be the ONLY compass and rudder for humanity.
I pay attention to the hobby to learn, and find ways of DIYing those ideas that stand up to examination. I didn't get in to the hobby to spend lots of money but to do the very best possible job of reproducing real stereo recordings of acoustic music, within a budget. Especially of choral and orchestral music in large spaces. I hang out here to convey what I've learnt and am learning.
So I'm not very interested in the $$$$ and conspicuous consumption aspects of home audio. The high end and HT are to me good examples of a hobby that has lost its point. The high end is focussed on selling expensive low ROI upgrade components, and even lower ROI cables, as often as possible, to a market who want to reproduce recordings that I couldn't / wouldn't use to make sensible judgements, about where I'm going in the hobby. Don't wave the 'high-end' at me, it isn't convincing, put it away.
As your typical response to being corrected is to duck it and then use ad hominem both here and elsewhere within the same thread, you may not last around here. I have, you may not.
I can see why you 'left' DIYAudio. Perhaps you can't help it.
Or maybe it's just a matter of believing that you know more about wave launch than THE wave guide man, John Dunlavy. Wave-launch is what home audio boils down to. And that's a systems judgement.
Mind how you go.
Note that a post in response is preferred.
Warmest
Timothy Bailey
The Skyptical Mensurer and Audio Scrounger
And gladly would he learn and gladly teach - Chaucer. ;-)!
'Still not saluting.'
Tim,
I respect the work of John Dunlavy. But I also respect the work of people like Greg Timbers, Laurie Fincham, Peter Walker, and Lawrence Dickie - to name a few. All are or were great designers in the loudspeaker business and none of them have endorsed the use of spherically shaped loudspeakers. In fact, Mr. Walker proved with his novel approach that you could emulate much that is good about a point source without actually trying to build one. If that isn't a clear statement on spherically shaped loudspeakers, I don't know what else could be. Like so many in audiophilia, you pretty much disqualified yourself with an extreme, ridiculous suggestion that subwoofers suffer from diffraction effects when sharp edged baffles are used. And with that, I see no point in further discussion - much like at DIY Audio, where I felt no need to go any further when people like Dennis Murphy and John Kreskovsky tried to suggest that sound was being reflected back into a tube cavity loaded with a driver on one end. I've been at this too long to take instruction or lessons from total novices in acoustics or electrical engineering.
mini monitors don't, or that it doesn't matter. When I am pretty sure that they do.I perhaps did not punctuate correctly. You could have asked what I meant, but because I had dared to correct you elsewhere in the thread, you decided to be insulting.
In this thread have been repeatedly rude, insulting, smug, and not just to me, and just plain wrong on the subject of the thread. And still refuse to accept that you were wrong, and that your behaviour is unacceptable.
Spheres may be impractical to your mind, my experience is different, that they do work, and for well established reasons.
You know better, eh? The problem is that you don't know any better.
Note that a post in response is preferred.
Warmest
Timothy Bailey
The Skyptical Mensurer and Audio Scrounger
And gladly would he learn and gladly teach - Chaucer. ;-)!
'Still not saluting.'
Edits: 04/18/12
He may not have alliances with the bigger companies such as "PSB, Revel, Canton, KEF, NHT, Vienna... etc " but his work with smaller scale speaker manufacturers such as Dave Ellis and Jim Salk is surely impressive. He may not be a Richard Modaferri (hardly a novice) but he understands crossover networks way beyond what one would consider a "novice" who makes "ridiculous and inaccurate statements online" IMHO. But maybe you weren't including him as well...
You mean the viola player turned speaker designer? LOL!!
You need to get out more and listen to speakers that have a full bottom octave and more than 85 db of sensitivity. I have had discussions with Dennis (some of the most laughable and ridiculous stuff I've ever heard) and I stand by my previous statement. As I said, you don't need to have a solid background in acoustics or engineering to "design" a crossover and hire a cabinet maker to build speakers. To build a loudspeaker with exceptional performance and value as most of the experienced manufacturers I listed have done for years - you need solid experience in the field of acoustics and engineering. Dennis has been putting together crossovers for quite a while and I would hope by now he's at least got that part somewhat nailed down. But that doesn't mean he's capable of building or designing products as good as people like John Krutke, Kevin Voeck, or Paul Barton. I've heard the Salk speakers you've mentioned. I've also heard just about every product put out by Revel, PSB, Vienna, KEF, Canton, and NHT. Without getting into too much of a political debate, I will say that there's a reason a company like PSB sells hundreds of high end speakers to every one Salk sells. You're free to think that all the PSB or NHT owners who have spent hundreds or thousands on their equipment are crazy and I can think what I want to think. In the end, they all build nice looking cabinets with transducers of similar quality. Some have a world wide dealer network and millions in sales annually. The other - well, you know... But it's "high end" audio. Reason or logic need not always apply.
: )
Thanks for the info - musician turned speaker designer...guess that rules him out LOL! Not to beat a dead horse, I was referring to the "novice" status and after speaking with Dave Ellis and Jim Salk as well as Dennis, I never got the impression that any of them were novices, especially Dennis but you obviously know more about him and his work than I certainly do. But I guess that folks like Bobby Palcovic or Jeff Joseph who sell a fraction as many speakers as Revel, PSB, Vienna, KEF, Canton, and NHT are probably in the same novice camp. Oh gee..Palcovic was a musician turned speaker designer so we can rule him out too. LOL! But of course, all things in audio are highly subjective but I appreciate you being the self-appointed expert asylum inmate. We need the data and we often need the comic relief. Thanks also for letting me know that I need to get out more and listen to speakers that have a full bottom octave and more than 85 db of sensitivity since I've probably only heard a few hundred in the 50 years I've been involved and lord knows I need to get out more. By the way, I don't think that all the PSB or NHT owners who have spent hundreds or thousands on their equipment are crazy (some NHT's even sound halfway decent) and yes, you can think what you want to think so please go ahead and continue to feel free to post what you think regardless of what anyone else thinks...:) Your reply to my original post was very informative and I appreciated reading it! As in life as in audio, pay attention to everything, assume nothing and DON'T TAKE ANYTHING PERSONALLY!
Not every high performance speaker uses heavily rounded baffle edges simply because not every design needs them. Understanding the basics as to why response anomalies occur in the first place goes a long way to resolving the question for a particular design. Baffle width around a particular driver, the wavelengths the driver is producing, and whether or not the driver is centered on the baffle also affect the speaker's "diffraction signature" - not just the amount of radius at the baffle edges. One needs to keep in mind that a portion of the driver's wavefront travels along the baffle face and in doing so is reflected outward off the baffle. When the edge is encountered, instead of being largely reflected, it wraps around towards the back of the cabinet - producing a response drop off. The larger radii edges soften this transition and that is all they do. It is equally important to understand that with shorter wavelengths, the wavefront will have undergone several complete pressure/rarefaction cycles by the time it reaches the baffle edge. The phenomenon known in acoustics as "relaxation" causes a more noticeable drop off in amplitude with these short wavelengths so that by the time it reaches the baffle edge - there is very little "reflection" taking place regardless - so the edge diffraction effect is less severe. In addition, offsetting a tweeter from the baffle's vertical center line reduces the pressure drop occurring at a particular frequency associated with a fixed baffle width because the quarter wavelength distances on either side of the driver to the baffle edge are staggered. It's not a simple matter of felt, round over size, or baffle width - but everything taken into account. And as others have suggested, the affect of edge diffraction can be quite small - almost inaudible. If all of the above parameters are screwed up in the design, it can amount to a ripple in response of a noticeable amount ( several db) - particularly off axis. So one shouldn't get overly worked up about it but it should never be completely ignored either. Extreme approaches with extraordinary claims are a sign that someone either doesn't know what they're talking about or they are trying to create something that isn't really there.
I have used thick felt rings on tweeters two occasions with two different speakers and there was audible difference. However, I think the amout it helps depends on the speaker design. The speakers I used them on were box speakers with sguare edges.....
70s vintage Avid loudspeakers had a 3/4" round-over in the edge of the grill. The grill fitted flush to the baffle board.
When you removed the grill it sounded out-of-focus, quite poor actually.
Snap the grill back on, it sounded fine.
Nice offer, effective, too, if I do say so. Customer and reviewer comments can be seen at the site.
I'm sorry, you sound too much like a shill.
Please attach your post to one asking a question about add-on diffraction rings.
My post was about Avid loudspeakers, no mention about foam or felt rings.
I think you are violating the terms of use for this site.
I wasn't responding to your post, ole bean, tho you seem to be always around to say something negative, I was responding to the topic. Nor do I make generic rings. What I DO make has been very well received. TAS' Robt E. Green gave what I make an '08 Golden Ear Award. He ain't no dummy. He's a math professor at UCLA. To quote, "These elegant and inexpensive felt devices will make an improvement all out of proportion to their low cost." He had played around with them for over two months. Thanks.
Miraculously, I actually completely agree with something DJK has posted. Those of us in the trade have to be careful not to cross the line of commercialism when attempting to "help out" others on this site. The rules for manufacturers are pretty explicit and I think in this case may have been broken:"A member of the trade may not volunteer any information about a specific product that he is selling or making, in response to a general request for information about a type of product or in any other discussion.
A trade member may not post announcements, advertisements, sales information or the like about a product he makes or sells."
If the topic introduced was specifically about Jim's stuff, he might have been justified but that doesn't seem to be the case.
Btw Jim, the animated diagram you have demonstrating the effect of diffraction with the use of red colored concentric rings is incorrect. That is not the shape of the resulting waveform when an original wavefront or series of wavefronts encounters a sharp corner of a baffle edge. Some of the images found at the link below are far more accurate. : )http://www.pa.op.dlr.de/acoustics/essay1/beugung_en.html
Edits: 04/16/12
I posted a link to an ad running on AA. Of course, it was meant to be read and generate some interest. I respect your comment and avoidance of inflammatory words. Thanks for the link. I do wish it were active, however.
Edits: 04/16/12 04/16/12 04/16/12 04/16/12
and while bigger is MUCH better, even smaller ones do matter. remember that you don't always encounter the edge treatment head-on, so it can act like a much bigger roundover than it is.
Bass is supposed to sound big. 6.5" is not a woofer size.
Bass is supposed to sound big 6.5" is not a Woofer size ?
I get excellent adequate bass from a transmission line speaker that features TWO 5.5" full range drivers no X crossover of course.GOOGLE ' Acoustic Insight '
Edits: 04/17/12 04/17/12
I've used and built a lot of small drivers, both in TLs and in vented and sealed cabs. Bigger woofers just sound more natural in the bass.
Now, if bass isn't a priority for you, then certainly there are things that fullrange-based systems do very well. But they don't include realistic bass or realistic SPL before significant distortion. I don't always listen loud, but sometimes, I do, and I feel that a loudspeaker should be able to speak- loud :)
It should be noted that most of the instruments I've played have significant lower-register content, including Piano and Tuba, so there's an affinity for the deep stuff in me.
Bass is supposed to sound big. 6.5" is not a woofer size.
...as illustrated in the enclosed link.
When you realize that applying felt rings won't work...
http://www.tolvan.com/edge/help.htm
http://www.pvconsultants.com/audio/diffraction/downloadbds.htm
Properly used felt and baffle treatment makes a huge difference, just like roundovers can.
Bass is supposed to sound big. 6.5" is not a woofer size.
If you've ever measured the effects of felt rings you might feel the same way. OTOH, thick felt applied across the entire baffle (Dunleavy, Spica, etc.) will have a distinct measurable effect. The operative word is "measurable".
What makes you assume that I haven't worked extensively with the use of absorbtive materials in baffle terminations? Did you even bother to read the text of my post? I wrote:
"Properly used felt and baffle treatment makes a huge difference, just like roundovers can."
Rings can make a meaningful difference, though I prefer them with a randomized interior edge and more thick than with a typical "ring" but even a ring makes a difference off-axis. Just not as much as a thicker more thorough treatment.
But your tone could be improved fella.
Bass is supposed to sound big. 6.5" is not a woofer size.
Reread the subject of own post. You said "What's wrong with felt rings?" .
I stand by my response given the fact you seem to imply a felt ring will provide "similar benefits" to either significant cabinet edge roundovers or significant use of thick felt baffle coverage.
That said, I do agree with some of what you said in the body of your reply (the part that pertains to the subject material).
And within your post took a potshot at my experience level.And felt rings do provide measurable benefit, just not a whole lot of it, if it's the right felt and used properly.
If you're not spoiling for a fight, try polite.
Bass is supposed to sound big. 6.5" is not a woofer size.
Edits: 04/17/12
nt
Regards,
Geoff
Post a Followup:
| FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
||||||||||||||
This post is made possible by the generous support of people like you and our sponsors:
Top of Page ]
[ Contact Us ]
[ Support/Wish List ]
[ Copyright © 2024, Audio Asylum, all rights reserved ]
