Enclosure resonances, not a big deal?

[gedlee]

Probably Gedlee considers a 10db difference 'a very small effect', since it would only result in a 1db amplitude disturbance or less.

The amount a 10 dB change in extraneous sound radiation makes to the amplitude response depends on how far down this radiation is in the first place. If its down far enough then a 10 dB change makes no difference at all. I was not able to detect any amplitude response changes that could be linked to structural changes. That IS a data point.

 

[planet10]

A crude but useful way to get an idea of what any given panel is contributing to the overall sound is to, while music is playing, place one end of a pen, pencil or similar lightweight small diameter rod against the the portion of the panel in question and press the other end near the ear, closing off the air channel to the eardrum. This will use bone conduction to selectively filter for the panel contribution. MDF sounds awful using this technique in my experience.

A mechanics stethoscope works REALLY well.

dave

 

[planet10]

The amount a 10 dB change in extraneous sound radiation makes to the amplitude response
depends on how far down this radiation is in the first place. If its down far enough then a 10 dB change makes no difference at all.

I'm not exactly sure what you are saying here.

But, i'm not overerlly concerned with anything that would affect the typical measured FR, of concern to me is any extraneous noise that will affect the signal 30-40-50 dB down. It is this delicate part of the signal that is important in giving the music greater reality and subtle imaging clues.

thoriated mentions "MDF sounds awful using this technique in my experience... tend to 'drone'"

This is exactly the continuous ouzing of delayed energy that obscures low-level detail. The prime (sonic) reason i stay away from MDF.

dave

 

[thoriated]

How's this? For audiophiles, MDF is the woodworking equivalent of the electrolytic capacitor.

 

[gedlee]

But, i'm not overerlly concerned with anything that would affect the typical measured FR, of concern to me is any extraneous noise that will affect the signal 30-40-50 dB down. It is this delicate part of the signal that is important in giving the music greater reality and subtle imaging clues.

dave

Hi Dave

This point is a guess on your part that is unsubstantiated and generally not in line with data regarding masking etc. If this IS your belief then I understand your position, however, you must understand that little to nothing in psychoacoustics would support this kind of conclusion.

 

[gedlee]

How's this? For audiophiles, MDF is the woodworking equivalent of the electrolytic capacitor.

No, better would be "MDF is the woodworking equivalent of Oxygen Free Copper." Electrolytic problems are a real effect and I don't use them for that reason.

 

[planet10]

How's this? For audiophiles, MDF is the woodworking equivalent of the electrolytic capacitor.

Good one. To be avoided if at all possible.

dave

 

[badman]

Hi Dave

This point is a guess on your part that is unsubstantiated and generally not in line with data regarding masking etc. If this IS your belief then I understand your position, however, you must understand that little to nothing in psychoacoustics would support this kind of conclusion.

Have you done masking tests with music/dynamic content rather than overlaying frequencies upon a steady state wave?

Also, when the input ceases, if there's stored energy resonating in the enclosure but not the driver output to mask it, couldn't that be audible where it might be masked during the conditions that start it resonating?

"It's too low in level to be audible" is an excuse to be lazy in addressing known issues. That might have been said of HOMs, eh? But you found the ear-brain mechanism have some variability in its sensitivity, did you not? Much more sensitive at higher amplitudes, etc? Perhaps box resonance is something that we're sensitive too in a variable fashion and the tests performed did not emulate our sensitivities effectively. You of all people should be attuned to the fact that microphones and ears act differently.

 

[gedlee]

You of all people should be attuned to the fact that microphones and ears act differently.

I am quite attuned to this, but I'm also not prone to jump at obscure issues that are obviously of minimal impact just because some audiophool claims that he "hears them".

IF, and that is a big "if", cabinet resonances were to be audible then it would be the tail of the resonance that would be the most audible. This is identical to the HOM situation (which, by the way IS completely consistant with psychoacoustic studies NOT contrary to them) since it is a group delayed effect which is not masked effectively. But, UNLIKE HOM, the enclosure resonances are likely to be much lower in level that the primary sound. HOMs can be nearly equal in level to the main signal in a worst case, and would always be greater in level than enclosures resonances. Hence, I cannot see the case for enclosure resonances being anything at all on par with HOM, especialy since, as I said, I tried to measure them and could not. I HAVE measured HOMs!

The entire side of this discussion which believes in "box effects" is without anything of support except for "beliefs". The real data is all on the other side.

 

[Pano]

What you hear is what YOU hear, that is not data.

Indeed it IS what I hear. It could be "data" if compared to what other people hear.

When I hear differences that are super subtle and ephemeral I don't get too interested. But differences as big as box resonances, or the lack of them? Those are too obvious to ignore. The difference between 3/4 MDF and 3/4" Baltic birch? I don't know. But the difference between cheap 5/8" ply and 1" super dense marine ply? That I DO know. Ditto sand filled walls.

The question should be of "how much stuffing, bracing, damping is needed to get the coloration down to a level where they don't matter?" That level is going to be different for different folks. And the techniques used to get there will be different. But it can be done.

<rant>
Trouble is (for me, anyway) that this forum has become almost as useless as A-A. Ask a practical question about building, then get an endless argument telling you what you hear or how you approach the problem is useless, misguided, unscientific, not measured right, whatever. How the hell is anyone supposed to build anything? Real, practical experience seems to count for nothing.

Reading many threads on this forum one might conclude that any changes or improvements you make are useless, bogus, the result of an over active imagination, stupidity or voodoo. Apparently everything sounds the same and it's easy to build a perfect a perfect speaker, amp, DAC, whatever. “Just do it my way, cause everyone else is wrong.”

Talk about useless..... </rant>

 

[planet10]

This point is a guess on your part that is unsubstantiated and generally not in line with data regarding masking etc. If this IS your belief then I understand your position, however, you must understand that little to nothing in psychoacoustics would support this kind of conclusion.

This is the part where i say that it does make audible differences that have a beneficial effect/improvement to the sonic illusion.

And you say, have you got measurements to back that up.

And i say, no, i don't really care, i trust my ears (and those on my associates)

And you say, you must be imagining it.

At that point we just have to agree to dissagree...

... or send me 2 sets of your kits, We'll build one to spec, and one properly, and then set them up for a panel to evaluate (VI diyFEST is coming up end Aug)

(i've already done that enuff with my own speakers)

dave

 

[planet10]

... And you say, have you got measurements to back that up.

And i say, no, i don't really care, i trust my ears (and those on my associates)

And you say, you must be imagining it...

PS: If the measure has to proceed the hearing for it to be true then there would be no progress...

dave

 

[xpert]

But the difference between cheap 5/8" ply and 1" super dense marine ply? That I DO know. Ditto sand filled walls.
...
Reading many threads on this forum one might conclude that any changes or improvements you make are useless, bogus, ...

Hi,
I'm new and I feel addressed. In this thread the original question was whether or not resonances could be equalized. The answer was given early.

In reply to that someone argued that equalization would take away subtle micro/macro information. That - of course - would never be measurable but a true audiophile will ever hear that as a tremendous degradation of soul of music. It would once again besides so many other causes wreck music to chaotic nonsense along the objective attempt, just some tones, rationalized and therefore nothing. That was the "AA"-part of the story. If audio discussion is about loading the topic with tons of irrational fears - well done!

The rational part is much easier. There is no monster under the bed. You don't have to speculate how it could look like. No need to imagine what it could do with You. There is none. That simple. All participants were invited to convince themselfs. Some didn't, because maybe the fear has grown to strong already.

The bigger part in evaluating a loudspeaker is knowing what to measure. If one has to conclude that cabinet issues affect a design, then accleration measurements of walls may be of use. But if a speaker is consistent in its output, no bigger wiggles in amplitude, no excessive group delay, not even off axis (backside?) then why bother with woodwork?

The last argument against that ever lasting fears is again simple if not catchy. In spite of overdoing the particle board / plywood thing use lead, marble the like, or some more recent material as specialized plastics. Do it and be finished. To iterate over growing totalisations of compromized extremes looks a bit silly to me.

cheers!

 

[thoriated]

Gedlee probably has the resources to substantiate or disprove his assertions. He could take a speaker system (UUT) minimally built to his standards, place accelerometers on all the relevant facets of the enclosure, using a mix of musical material emphasizing transient and spatial information, mixdown and amplify the resulting aggregate signal and play them back on a more or less co-located second speaker (along with the UUT) versus the 'control' of playing the UUT by itself. Then it should be easy enough for him to determine to his own satisfaction just how much margin below measured audibility his concepts give. I would suggest a minimum of 30db for a high quality enclosure construction.

It occurred to me that the reproduction of the accelerometer responses could be allocated to an otherwise unused widerange driver in a subenclosure of the UUT to minimize variables or, alternatively, summed into the input signal to the actual UUT itself, if the proper control measures are taken.

 

[gedlee]

Gedlee probably has the resources to substantiate or disprove his assertions.

And he has to his own satisfaction. That is exactly what I am trying to report on here, but people just don't want to hear it. They want to believe what they want to believe and no "data" is going to change there minds. They "know what they hear".

Thats fine, nothing new.

And to Panomaniac - there is a wealth of really good information that comes out in these forums. You just need to listen and not simply expel it all because it does not agree with your "perceptions".

I guess that I've said my peace, take it or leave it. I did not come to my position lightly and nothing that I've heard here changes it at all. So I'll just bow out at this point.

 

[Pano]

I did not come to my position lightly and nothing that I've heard here changes it at all.

Nor did I come to mine lightly. That's the point.

 

[Geoff H]

Dave, thanks for the reminder about the stethoscope / screwdriver test. Just deep bass off the steel. None of that pipe sounding echo chamber off the front or sides to muddy the mids I had with the same drivers on MDF.

Equipment used: 1 Stanley model 65-522, my left ear and Tony Joe White CD.

Time taken: ~ 40 seconds.

I could set a moving coil phono cartridge on a pendulum and a balanced instrument amp to confirm my test. I'd rather listen to music and contemplate WAF on these ugly things.

 

[xpert]

Dave, thanks for the reminder about the stethoscope / screwdriver test. Just deep bass off the steel. None of that pipe sounding echo chamber off the front or sides to muddy the mids I had with the same drivers on MDF.

Equipment used: 1 Stanley model 65-522, my left ear and Tony Joe White CD.

Time taken: ~ 40 seconds.

I could set a moving coil phono cartridge on a pendulum and a balanced instrument amp to confirm my test. I'd rather listen to music and contemplate WAF on these ugly things.

Hai,
I was in the mood to measure what is speculated on so eagerly.

(1) box made of cheepest particle board, 16mm resp. 5/8" thick, front/backpanel 420 x 300 mm^2 resp 16" x 11". Absolutely no bracing, damping whatsoever, very lousy glueing too.

(2) The box was placed with about 3cm resp 1" spacing to the wall. The circimference of the box was loosely tightend with a blankett. So in between box and wall a damped cavity was build. The mic was placed in that cavity, nearly centered.

a) measured amplitude vs. frequency with back against the wall
b) same with front against the wall

My assumption is that the measured levels represent what the speakerchasssis (b) and the cabinets back wall (a) contribute to the total soundfield. The pic below shows what I achieved. Cabinet part is well down 20dB+. One may argue that a stronger seperation of the cavity in which the mic hang would have given even higher dampening figures. No resonances! No hassle. And that was a very flimsy construction!

So that is the monster You have worried for so long. Ain't it funny in the end? Keep heads up,

by


The black curve shows the background (noise etc).

 

[Rudolf]

xpert,

the experiment looks valid for a first guess. 20 dB+ is very modest - you would never accept that from any CDP or amplifier, so there is no need to be ebullient. Granted, results would become better with an improved test setup.

Resonances? I see the major deviations from a smooth curve at just above 100 Hz, above 200 Hz, 400 Hz and 800 Hz. Makes me thinking ...

 

[dlr]

I've been curious about how much factual basis exists in the box debate. My favorite speakers prior to building my own were Apogee Centaurus Major hybrids. The woofer/hybrid part was the weak link, but not due to box resonances. Matching closed box to a ribbon is just not easy, if truly doable. It was a tradeoff. Before that system was my first and long-time Dahlquist DQ-10, so don't think that I have some resistance to open baffles or dipoles. I wished that I'd have the money for the big full-range ribbon Apogees back then.

I have done some of my own experimentation with accelerometers on boxes. It's not as simple as some may think. You can't just stick one on a box at various places and measure the impact acoustically in absolute terms, especially regarding perception. My accelerometer is not calibrated in any way and I doubt that most folks (DIYers) who buy them consider that aspect. In addition, the act of attaching it to a box adds uncertainty. Unless it's glued tightly, the attaching material will itself be inherently a filter of some sort, the transfer function being unknown, especially as frequency rises.

My ULD15-II subwoofer has an accelerometer on the driver that provides <1% distortion at maximum output. However, every system has to be calibrated with the woofer/accelerometer to attain that, there are variances in both. Accelerometers are like microphones, they must be calibrated to be of value for absolute measurements.

I'm still doing some work on a simple 2-way system and have done a few accel measurements. These will be before/after the shelf brace. Walls are 3/4" mdf and front baffle is 1". Curiously, I found a similar response on the side unbraced as I found on the 1", more narrow front baffle with two drivers mounted.

When I am through, I'll have some data for relative comparison, but that's all that anyone can provide with an accelerometer. Box panels are in a bending mode for most energy that they may emit and not just transfer compressionally to the back panel. There's probably no way to measure the SPL response from that, they show only relative changes to bracing and due to the uncertainties of attachment not necessarily reliably for that, either. After my experiences with them, I see a system such as that used by Klippel or laser interferometry to examine baffle vibration in the only meaningful way for examination other than finding resonance frequencies and relative amplitudes.

The other issue that seems to be glossed over in the thread by some is that for most of the box resonances, in fact for all but the front baffle and maybe a side in partial view, box panel emissions can be nothing more than indirect room response. If you have a panel that has a total panel sectional output that is 40db down from the first arrival, what is the magnitude at the listener after it's reflected off of walls and anything else that disperses and/or dampens? It's in the room response region and seems mighty unlikely to me that any of that is provably audible, it if is at all.

Then of course, the two lateral side panels are vibrating roughly equivalently and in-phase, but are facing the opposite direction. How does that affect the response at the listener's spot? And another thought, we're talking a flat baffle. What do you suppose the directionality is for such a large area? The listener is pretty close to 90 degrees in the off-axis of side and top baffles. Another confounding factor having an impact on the direct response.

Front baffle, that has merit, but is not a given. Internal air resonances not sufficiently damped, that has merit, we've all heard that. Damp it well and it can be made to be insignificant. But side walls and back panel? How does one differentiate side and rear panel output from any other system linear and non-linear distortion especially given it's indirect pathway? It's beyond the Haas effect range in time for most of it's output for a speaker out in the room.

I don't buy most of what I read about boxes.

Dave