Open Baffle speakers

[beagleman]

I have heard a few, and while they do "Seem" to offer a few key advantages or seem at least interesting sounding in some ways.....

I have found those that tend to champion them, are very defensive about them in general, and tend to skim over the disadvantages and often are not even totally aware of some of the downsides. Similar to how some Vinyl guys will skirt around built in issues with vinyl records.

When I mentioned "Baffle Peaking" on another forum, I got mostly crickets and told I was making up stuff or just bashing their baby speakers.
The guys that said they were experts on OB speakers seemed blissfully unaware this was even a well known issue.
(Baffle peaking is when the sound meets from front and back and is in phase and creates a 6db boost in the sound, usually between say 200-400 hz or thereabouts)

There is also a cancellation frequency where sound from front to back cancel each other out and result in a reduction of most all bass, as frequency goes lower and lower.
Most make up for this with Drivers with a VERY high Q factor, that partially boost bass at least at the mid and upper bass frequencies.

 

[ctrl]

Once again I will doubt that Klippel is a proper way to measure OB at low frequencies because some space needed to sum front and rear sound, thus nearfield measurements can be confusing.

The Klippel NFS measures the frequency responses in the near field, but by means of a "mathematical algorithm" the frequency responses can be determined for any point in space.
The principle briefly explained:

Therefore, if the nearfield measurements are performed correctly, the design concept of the loudspeaker does not matter and the measured data can be used to create the free field measurements of the DUT in, for example, 2m distance which is the default measuring distance recommended by the CTA-2034 Standard for a typical loudspeaker.

So Erin's measurement of the (part) OB speaker LxMini with the NFS worked without any problems (AFAIK)

Linkwitz LXmini Speaker Review

Linkwitz LXmini Review

www.erinsaudiocorner.com

 

[olegtern]

Therefore, if the nearfield measurements are performed correctly, the design concept of the loudspeaker does not matter

I knew the principles, but doubt how it will work with big dipole or passive cardioid. Hope someone we'll see such measurements

So Erin's measurement of the (part) OB speaker LxMini with the NFS worked without any problems (AFAIK)

Linkwitz LXmini Speaker Review

Linkwitz LXmini Review

www.erinsaudiocorner.com

Thnx, interesting. LX Mini doesn't have dipole bass, only mids. But maybe you're right and it can measure it also.

 

[ctrl]

I knew the principles, but doubt how it will work with big dipole or passive cardioid.
LX Mini doesn't have dipole bass, only mids. But maybe you're right and it can measure it also

Yes, the Lxmini is only part OB, but mathematically there is no difference between the calculation for 200Hz or 50Hz (AFAIK).

For another thread I scanned and prepared the gated measurements of the Lxmini by S. Linkwitz some time ago.
Here is the comparison of the data from S. Linkwitz (gated outdoor) and Erin (NFS) as a horizontally normalized sonogram:

Sources: Lxmini, Spinorama DB

The results are almost identical, considering that the measurements of S. Linkwitz show severe smoothing, the measuring distance was 1m SL and 2m NFS and the crossover tuning is slightly different for the measured Lxmini versions.

 

[NTK]

I knew the principles, but doubt how it will work with big dipole or passive cardioid. Hope someone we'll see such measurements


Thnx, interesting. LX Mini doesn't have dipole bass, only mids. But maybe you're right and it can measure it also.

Animated polar response plot using Amir's data.

 

[-Matt-]

Animated polar response plot using Amir's data.

View attachment 244165

Wow! At low frequencies that looks really close to the theoretical dipole.

Edit: Thanks for pointing me to the Magnepan LRS review, here is the measured klippel, horizontal directivity data that I was looking for earlier:

 

[sarumbear]

There aren't many of them, but the fact that they are seems interesting. For example:

ANTEROS-9 OPEN BAFFLE Loudspeaker system - Thivanlabs

thivanlabs.com

Interesting. They are a Vietnamese company. (emphasis mine)

The 21-inch woofers [in a separate module] are controlled by a separated power amplifier and valve pre- amplifier substaintially contributing to the full energy and natural bass.

 

[atmasphere]

Hmm, interesting, however, the next part seems to be a non-sequitur...


Why would a speaker with a figure of 8 radiation pattern excite fewer room modes than one where the sound is predominantly radiated forward (possibly with narrow dispersion)?

'Room nodes' are the result of standing waves. This happens because bass notes are pretty long. At 80Hz, 14 feet. So the bass wave can be reflected by the wall behind the listening position, and may be in phase ('super position'; enhancement of the bass note) or more commonly cancellation (for which no amount of DSP room correction or bass traps can fix, since for the former the more amplifier power used still gets cancelled and for the latter, would have to move about the room dynamically with the bass notes to be effective).

Where there is rear-firing bass, the bass wave can bounce off of the wall behind the speaker, this being at a different phase at the listening chair as opposed to the front output of the speaker. This means less cancellation and super position; the speaker is more able to prevent standing waves.

It takes the full waveform passing the ear for it to acknowledge its existence and a few more to know the frequency. By this time, at 80Hz and below unless the room is very large, the bass is entirely reverberant. So subs can be easily integrated if kept below this frequency (above that they may attract attention to themselves). A nice practice to eliminate standing waves is to simply employ 4 subs asymmetrically placed so as to break up all standing waves. Because the bass is reverberant, a mono bass signal can be used. This type of subwoofer system is known as a Distributed Bass Array (see Dr. Floyd Toole and Duke LeJeurne of Audiokinesis).

The distance from the wall behind the speaker is important for any speaker with rear firing information. The ear/brain system makes a copy of any sound it hears and looks for examples in near time (a few milliseconds). If it finds a near match, the information is used for echo location. But if the secondary sound arrives too soon the ear will interpret this as harshness. This is why side walls can be problematic and also if the OB speaker is too near the wall behind it. The 'magic' number is about 5 feet from the wall behind the speaker; at this distance the reflected information is delayed about 10 milliseconds, allowing the ear to use the echo location information. This has the effect of making the sound stage more palpable.

Note that this works for any speaker with rear firing information, not just OB speakers. So an array identical to the front array but otherwise in a box can have the same effect and no worries about 'time alignment' or any such nonsense because the 10mS issue is what is important.

 

[kongwee]

Can you name such a 100k speaker? All I listen to don't have such problems. And yes, boxy sound = wrong sound, can't understand what you arguing.

Kharma speaker as an example, of course other brands too. Super thick MDF, diamond tweeter, ceramic cones. Still sound boxy in the same category price taking Avalon Acoustic for example. A lot of people love sonus faber over genelec, the amount of boxiness is just right for many people. You can sell $1 million dollar speaker that has all the latest exotic technical part or manufacturing. If boxiness is still your selling point, you would not get rip of it. Boxiness have more diverse preference. Go to shows, you will listen a lot more brand in that cat.

 

[Adi777]

(emphasis mine)

Could this negatively affect the sound?

 

[Rednaxela]

Depends on the kind of emphasis.

 

[olegtern]

Kharma speaker as an example, of course other brands too. Super thick MDF, diamond tweeter, ceramic cones. Still sound boxy in the same category price taking Avalon Acoustic for example. A lot of people love sonus faber over genelec, the amount of boxiness is just right for many people. You can sell $1 million dollar speaker that has all the latest exotic technical part or manufacturing. If boxiness is still your selling point, you would not get rip of it. Boxiness have more diverse preference. Go to shows, you will listen a lot more brand in that cat.

I listened Kharma in Munich, and don't hear any boxy sound (on demo music, of course I can't rule out something on other tracks). It's hard to remember if the top Sonus Faber have this problem, I don't think so, but I don't like them very much and didn't listen too carefully.

And I visit exhibitions quite regularly (Munich, Warsaw, Prague). What exhibitions do you visit?

I have a feeling that you call boxy sound something of your own understanding, so it's hard for me to understand. Harbert, Audio Note has boxу sound, for example, but they don't cost 100k. Maybe it's about the sound of the bass reflex? What I usually hear. But Genelecs always have this flaw. I have long owned Genelecs and heard new 3-coax line.

And I don't like boxiness of course

 

[kongwee]

I listened Kharma in Munich, and don't hear any boxy sound (on demo music, of course I can't rule out something on other tracks). It's hard to remember if the top Sonus Faber have this problem, I don't think so, but I don't like them very much and didn't listen too carefully.

And I visit exhibitions quite regularly (Munich, Warsaw, Prague). What exhibitions do you visit?

I have a feeling that you call boxy sound something of your own understanding, so it's hard for me to understand. Harbert, Audio Note has boxу sound, for example, but they don't cost 100k. Maybe it's about the sound of the bass reflex? What I usually hear. But Genelecs always have this flaw. I have long owned Genelecs and heard new 3-coax line.

And I don't like boxiness of course

Listen more to ESL or Maggie or just own them. I once own ML for five years. I am pretty sure lots of $100k, like Wilson Audio, Focal, Dynaudio, ATC......etc In my own country at retail or show in my tiny country, Singapore. These are the boxiness level I heard below Avalon Acoustic. I have not heard Magico aluminium box which suppose to be much less boxy than Avalon. I can only remember these brands. There are brands that come and goes at $100k tower speaker, I don't remember. Some from home visit.

Avalon Acoustic is my benchmark as pinpoint detail, you can't achieve in panel speakers. Brutally revealing in audiophile term. Their boxiness low enough that never bother me. Of course with ML, it is so easy to achieve "seamless soundstaging that cover the whole system" because these is no box in their marketing term.

 

[olegtern]

Listen more to ESL or Maggie or just own them. I once own ML for five years. I am pretty sure, lots more $100k, like Wilson Audio, Focal, Dynaudio, ATC......etc In my own country at retail or show in my tiny country, Singapore. These are the boxiness level I heard below Avalon Acoustic. I have not heard Magico aluminium box which suppose to be much less boxy than Avalon. I can only remember these brands. There are brands that come and goes at $100k tower speaker, I don't remember. Some from home visit.

Why should I listen to what I don't like? Too many tips...

WA (Wamm, Alexia, Sasha), Focal Utopia — no boxy sound. Magico too, agrees. But not a Maggie/ESL sound, that's true.

 

[kongwee]

Why should I listen to what I don't like? Too many tips...

I do listen what I don't like. And love to understand why people like them. In fact I enjoy what people can craft. In the end. It is a hobby, not a work.

 

[sarumbear]

Could this negatively affect the sound?

As a valve amplifier has noticeable distortion, it’s normally expected to affect the sound. However, I fear the only reason a valve is in the signal path of a sub woofer is to tick an audiphool box.

 

[atmasphere]

However, I fear the only reason a valve is in the signal path of a sub woofer is to tick an audiphool box.

There's more to it than that. While triodes are quite linear, tube circuits in general don't tend to run as much feedback as solid state amps do (mostly due to frequency poles that result in limited phase margins). The result is a significant 2nd harmonic (in the case of amps with a quadratic non-linearity) and 3rd harmonic, which are prodigious enough to mask higher ordered harmonics from the ear. The ear uses the higher orders to sense sound pressure and so is keenly sensitive to their presence (this is easily demo'd with simple test equipment). When the higher orders are masked as in tube equipment, the mids and highs tend to sound very smooth. OTOH lots of solid state amps have unmasked higher ordered harmonics and so tend to sound bright and harsh. A lot of 'audiophools' don't like that- this is what has kept the vacuum tube industry alive 60 years after being declared obsolete. You really don't need to know anything technical to understand this if you understand economics.

The ear assigns a tonality to all forms of distortion (higher orders are assigned 'bright and harsh'; the 2nd and 3rd are assigned 'warmth' and are considered 'euphonic'). What many people don't seem to get is that harmonics lend musical instruments their tone color; if additional harmonics are present, obviously it changes that. This is both measurable (in the form of harmonic spectra) and audible and should not come as any surprise; we've known a lot of this since the 1930s at least (see Radiotron Designer's Handbook, 3rd edition).

Thus lower orders cause coloration as do higher orders if not masked. This is why its so important to get distortion down to a really low level if neutrality is your goal (so the music itself is able to mask it), as we are seeing in such amps as the Benchmark, Soulution and Purifi. Those designers aren't doing that as a race- they realize the impact distortion has on their 'sound' and are trying to avoid any 'sound' at all.

 

[sarumbear]

There's more to it than that. While triodes are quite linear, tube circuits in general don't tend to run as much feedback as solid state amps do (mostly due to frequency poles that result in limited phase margins). The result is a significant 2nd harmonic (in the case of amps with a quadratic non-linearity) and 3rd harmonic, which are prodigious enough to mask higher ordered harmonics from the ear.

I understand why audiophools like distortion but do they have to be real fools not to realise this: For a subwoofer amplifier's sound to be 'euphonic' the fundamentals of the 3rd harmonic that will be affected is lower than 30Hz (at standard 80Hz LP filter)?

 

[atmasphere]

I understand why audiophools like distortion but do they have to be real fools not to realise this: For a subwoofer amplifier's sound to be 'euphonic' the fundamentals of the 3rd harmonic that will be affected is lower than 30Hz (at standard 80Hz LP filter)?

I'm not talking about frequency response error. A sub definately helps reduce harshness in the high end due to how the ear 'tilts' its perception when something is missing, in this example the bass. Similarly too much bass can cause the highs to seem muffled.

The 'euphonic' sound I'm talking about is caused by the 2nd and 3rd harmonic, masking succeeding orders. To do that the harmonics must be on an exponential curve as the order is increased. IME, a cubic non-linearity is preferred because the higher orders drop off at a higher rate.

Just a FWIW dept.: They are not being fools if they are using tubes to avoid harshness and brightness (endemic with solid state amps). Its an unpleasant coloration, especially if the amp was really expensive! Usually its caused by insufficient feedback and the feedback being applied to a non-linear node, such as the base of a differential pair of transistors, the other of which is the input transistor. The transistor isn't linear, so the feedback signal is distorted before it can do its job. This causes higher ordered harmonics to be generated. But in a solid state amp you can typically run enough feedback so that the lower orders are suppressed, leaving the higher orders unmasked. In a tube amp, the feedback node can be a tube that's fairly linear, like a 12AX7 so the feedback isn't distorted quite so much. If the feedback were not distorted before it does its job, it can work much better. This is why opamps can work so well (as long as not too much gain is asked of them), since the feedback isn't applied to them internally. The resistors doing the feedback tend to be far more linear than semiconductors or even triodes!

There is a different problem that is also common - that of rising distortion with frequency. This too causes brightness due to the ear's mechanism. It can happen if the amp has insufficient Gain Bandwidth Product. When the amp is at its limit, the feedback falls off with frequency, thus the distortion rise.

I just now noticed that your comment was in respect to subwoofers. I agree there is no point using a tube amp to drive a sub if its cutoff is only 80Hz!

 

[sarumbear]

I'm not talking about frequency response error. A sub definately helps reduce harshness in the high end due to how the ear 'tilts' its perception when something is missing, in this example the bass. Similarly too much bass can cause the highs to seem muffled.

I have not mentioned a FR error.

The 'euphonic' sound I'm talking about is caused by the 2nd and 3rd harmonic, masking succeeding orders. To do that the harmonics must be on an exponential curve as the order is increased. IME, a cubic non-linearity is preferred because the higher orders drop off at a higher rate.

I think I will bow out here as I have no idea what “cubic non-linearity” even means.