ARCHIVED FORUM -- March 2012 to February 2022READ ONLY FORUM
This is the second Archived Forum which was active between 1st March 2012 and 23rd February 2022
I know you asked for "matching", but the "tights from Tesco" hack for 8000's was so good, I must bring it to your attention: https://archivedforum2.beoworld.org/forums/p/10043/272364.aspx
In addition to the usual speaker DIY firms, consider the incredible variety of acoustic cloth from Guilford of Maine: https://www.guilfordofmaine.com. Their well-known "FR701" in black is available from many places. (Probably not the B&O OEM since it's US-based, yet it was used "everywhere" in decades past.) But just imagine the custom coolness (and WAF) of "Coastline" in Vapor Blue...
trackbeo: In addition to the usual speaker DIY firms, consider the incredible variety of acoustic cloth from Guilford of Maine: https://www.guilfordofmaine.com. Their well-known "FR701" in black is available from many places. (Probably not the B&O OEM since it's US-based, yet it was used "everywhere" in decades past.) But just imagine the custom coolness (and WAF) of "Coastline" in Vapor Blue...
Hi,
It looks to me like their "Acoustic Cloth" options are all for putting in front of absorbers (and have significant absorption coefficients according to the graphs that I've looked at on their site...)
What you need in front of a loudspeaker is transparency - so I'd be looking for an absorption coefficient of 0 at all frequencies in a best case.
Or am I missing something?
-g
Geoff Martin:It looks to me like their "Acoustic Cloth" options are all for putting in front of absorbers (and have significant absorption coefficients according to the graphs that I've looked at on their site...) What you need in front of a loudspeaker is transparency - so I'd be looking for an absorption coefficient of 0 at all frequencies in a best case.
Yes, the Guilford fabrics are intended for office sound-absorbing panels. But panels have their own particular engineered characteristics, so the requirement for their fabrics is that they *pass thru* as much of the sound as possible to the panel behind! The two bugs are: (1) they feel that "a little" absorption across all frequencies is not a bad thing (as you pointed out), and (2) they only really care to measure in the 300-3000 Hz range because that's what office-furniture manufacturers want to catch. That said, the "FR701" looked closer to zero than the rest of their line, and looked (comparatively) flat. Question: Let's imagine that we have a fabric which is miraculously flat across the audio frequency spectrum, but has significant non-zero absorption. Yet if I turn up the volume, am I not simply overcoming that absorption (and generating a bit of heat)? Nothin' wrong with that... Or is there? (No fabric will be flat, but I would guess a sort of band-pass with a slope at one end, which can be EQ'ed away?)
Geoff, are there graphs for the Kvadrat fabrics used by B&O? They look rather "wooly", which I take as frizzy at the surface, which I then take as a proxy for sound-absorbing at mid and high frequencies -- but I'd love to be proven wrong. Maybe the original poster should be buying those!
Yet surely those fabrics are better than metal grilles with little holes punched in them, which must be 50% reflective (for short wavelengths) right back at the source! For that matter, I'm suspicious of lamellas, which, placed at varying angles as they are, intuitively seem to be creating a huge reflective comb interference filter! (Or does 1" by 1/4" put them outside the frequency range of the drivers they're hiding?) Of course intuition is no substitute for science, which itself is no substitute for measurement of the results! Sorry to veer off topic, but I'm curious to know how or why lamellas aren't a really bad idea... [[EDIT: I was thinking about the Beovisions where the metal grilles and lamellas cover full-range speakers. In the BeoLab 18s and 50s, the crossover frequency to the acoustic lens could put them out of the range of any interference. So ignore the "placed at varying angles" bit.]]
Geoff Martin: trackbeo: In addition to the usual speaker DIY firms, consider the incredible variety of acoustic cloth from Guilford of Maine: https://www.guilfordofmaine.com. Their well-known "FR701" in black is available from many places. (Probably not the B&O OEM since it's US-based, yet it was used "everywhere" in decades past.) But just imagine the custom coolness (and WAF) of "Coastline" in Vapor Blue... Hi, It looks to me like their "Acoustic Cloth" options are all for putting in front of absorbers (and have significant absorption coefficients according to the graphs that I've looked at on their site...) What you need in front of a loudspeaker is transparency - so I'd be looking for an absorption coefficient of 0 at all frequencies in a best case. Or am I missing something? -g
I have actually wondered about to what extend you account for the absorption coefficients of the fabrics (especially with the Beolab 50, which have cloth and wood/metal options). I guess with the Beolab 50 you always need to have a cloth front? Could you actually publish the absorption sweep for the cloth used by B&O? I have used cloth from Akustikstoff.com, who actually publish the sweep signal with different kinds of cloth (sadly only in German: https://www.akustikstoff.com/media/products/0900591001442653810.pdf). Would you say this fabric is "close enough=good enough" to use with B&O speakers (not kvadtat)?
P.S. I thoroughly enjoy reading your blog!
I recently rebuilt a pair of Tannoy PBM-8 speakers and used the Jet Black cloth from Simply Speakers. They are US based, but perhaps there is a more local supplier in your area. This fabric is pretty much a "dead ringer" for the original cloth. I can't say how it matches in your instance, but I admit that in my case I like it.
Details at: https://www.simplyspeakers.com/speaker-grill-cloth-fabric-black-a-569.html
Regards,
Roger
B & O dreaming since 1985 - Living the dream …. finally.
Beomaster 3000-2Beogram 3404Beocenter 2500Beovox S2200BeoCom 4Bottle Opener
Welcome Roger - now we have (at least) 2 Rogers on the Forum....one in Norway and you.
MM
There is a tv - and there is a BV
Beosince98:I have used cloth from Akustikstoff.com, who actually publish the sweep signal with different kinds of cloth (sadly only in German: https://www.akustikstoff.com/media/products/0900591001442653810.pdf).
Hmm... Just looked at a Guilford sweep and their explanation of what they are measuring again; I misunderstood their measuring setup. Take their "Felt" #9900 (more transparent than "FR701") with a NRC of .99 -- they say "99% of sound passes thru!" That should be great, right? But they place their mic to measure the *reflected* sound, telling only that 1% of the sound *doesn't* pass. They are not measuring the signal amplitude on the other side of the fabric! It might be absorbing, or the panel behind it might. Also their graph only shows from 200-2500 Hz, over which range the Akustikstoff graph shows us all fabrics they tested are pretty similar. Sigh, can't use that cool blue wavy pattern after all... Oops.
[[EDIT: Just read the (many!) threads on this topic in DIYaudio.com (easiest reading: https://www.diyaudio.com/forums/multi-way/96820-speaker-grill-clothes-affect-sound-quality-debate.html), and Audioholics.com (specifically https://www.audioholics.com/loudspeaker-design/grilles). As far as fabrics go, this quote stood out: "The fabrics might attenuate the sound slightly in very high frequencies, but the effects would be minor compared to the diffraction caused by the grille frame." The Audioholics article goes on to show frequency sweep graphs -- Bless them, data!!! -- for a few commercially available speakers, with and without grilles -- including one metal one.]]
trackbeo: Beosince98:I have used cloth from Akustikstoff.com, who actually publish the sweep signal with different kinds of cloth (sadly only in German: https://www.akustikstoff.com/media/products/0900591001442653810.pdf).Their multiple traces seem to mistakenly have a small positive origin offset after having the no-fabric signal subtracted. Surely their Akustikstoff fabric cannot *amplify* the sound around 6KHz, nor can polyester amplify sound from 500-1800 Hz? (I.e. No-fabric signal is flat-zero because you subtract the Fostex 6301B speaker's actual signal sweep from itself. All others should be completely below that line, shouldn't they?) Hmm... Just looked at a Guilford sweep and their explanation of what they are measuring again; I misunderstood their measuring setup. Take their "Felt" #9900 (more transparent than "FR701") with a NRC of .99 -- they say "99% of sound passes thru!" That should be great, right? But they place their mic to measure the *reflected* sound, telling only that 1% of the sound *doesn't* pass. They are not measuring the signal amplitude on the other side of the fabric! It might be absorbing, or the panel behind it might. Also their graph only shows from 200-2500 Hz, over which range the Akustikstoff graph shows us all fabrics they tested are pretty similar. Sigh, can't use that cool blue wavy pattern after all... Oops.
I wondered exactly the same thing. I wrote them an e-mail. I will let you know if this is a mistake or not when they reply.
Saaay, maybe if you spray pink noise at a stretched fabric membrane, it resonates, frequency depending on the frame size and stretch tightness -- like a drum head! Uhhh... No. Especially not if it's "acoustically transparent". I bet on simple measurement error. Thanks for following up; curious to hear their reply.
trackbeo: Saaay, maybe if you spray pink noise at a stretched fabric membrane, it resonates, frequency depending on the frame size and stretch tightness -- like a drum head! Uhhh... No. Especially not if it's "acoustically transparent". I bet on simple measurement error. Thanks for following up; curious to hear their reply.
My guess is that the fabric always absorbs X amount of sound, so the first reading at 0,5kHz was put as their new 0 (i guess the fabric also absorbs some sound at 5 kHz). Now if they do a sweep and the fabric absorbs less sound at 6kHz the measurement is positive. In any way, this graph seems to have an error.
I received an answer: These anomalies are due to the fabric only hanging and therefore being able to flutter. I don't really understand this explanation, but that is all I got. Quite disappointing that the published measurements don't seem to be that accurate!
Kvadrat are coy about which fabrics were used for the Vifa products ("a special version of 'Clara'" and "a new fabric design with a special weave") and the B&O A6 ("specially engineered" wool-blend): https://kvadrat.dk/in-use/bo-play-beoplay-a6
For two sound *absorptive* fabrics used for acoustic curtains, Kvadrat published detailed test results. Looking for speaker grille fabrics, perhaps one doesn't care, but the layout of the test is very well-described, including a photo of the test room -- which shows the "hanging". Also the designer (of one of the two fabrics) wrote her thesis, which mentions "flutter" of the fabric in the tests. I found it interesting, but there are a lot of details to wade through just to get a vague understanding:
https://static.kvadrat.dk/assets/pdf/collection/sound-absorption/a4/sa-7914-ssound-absorption-airflow.pdf
https://static.kvadrat.dk/assets/pdf/collection/sound-absorption/a4/sa-7915-ssound-absorption-airflow.pdf
https://pdfs.semanticscholar.org/269d/1551ce209d127116b840d9e448c7ba967ef7.pdf
Part of the problem with this discussion is that the topic itself is rather complicated...
Usually, the absorption coefficient of a material, like a cloth is a measurement of how much acoustical energy is reflected from the material. There are different ways to measure this, one way is to send sound (say, an impulse or "click" - or a signal like a swept sine tone or an MLS signal that can be used to calculate an impulse) at the material, and then measure how much sound is reflected. After compensating for attenuation due to propagation distance, the amount of energy that was not reflected is assumed to be absorbed, one way or another.
(I just said "one way or another" because absorption can be the result of different physical effects. One possibly is that the acoustical energy is converted to heat one way or another... Another option is that the acoustical energy is cancelled because the surface it hits can move - and that movement winds up creating a reflection that is out of phase with the incoming signal.)
Another way to do this is to put a loudspeaker and a microphone in a room that is designed to create a diffuse field by having a lot of reflections. You measure the energy level (by frequency band) at the microphone. Then you add a known amount of the material to be measured into the room, and do the measurement again. Any loss in the second measurement must be caused by the material absorbing energy.
Since this discussion is about loudspeaker grilles, then what we should be talking about is the transmission index instead of the absorption coefficient. It's not a question of how much sound is not reflected - it's a question of how much sound gets through - which is a related, but different question. So, you cannot look at a measurement of a fabric's absorption coefficient and assume that this tells you what you need to know about its transmission index.
(To answer one of the questions above: typically, the transmission index of the speaker cloth (say, on a BeoLab 90) has a loss of about 1 dB at 20 kHz. This is very small... The lower the frequency, the smaller the losses.)
Cheers- geoff
Just going back through some of the comments and questions above...
Regarding the effects of grilles and lamellae: Anything put in from of a loudspeaker (or any sound source, for that matter) that has (as a rule of thumb) a size of more than 1/4 of a wavelength of the frequency content will have an effect on the directivity of the source.
If you have multiple things that are causing this effect, then you may/will have a subsequent effect on the magnitude response (the "magnitude response" is what people mean when they say "frequency response") at any one location in a free field (a "free field" is a space that is free of reflections - an infinitely big room, for example...).
If you're building a loudspeaker that has such things (say, an aluminium grille or the lamellae that has an acoustical impact) then these effects are measurable as a function of level vs. frequency vs. angle. (a fancy way of saying "the frequency responses in different directions").
If the multiple things in front of the source are regularly spaced, then the angle-dependent magnitude responses are more likely to exhibit a regular pattern vs. angle when measured in a free field.
So, we then have to consider what the measured effect is:- In the very simplest case, you have no effect. Or, the measured effect is so small that it's less than the difference caused by a cat walking into the room.- In a simplest case, you have an impact on the magnitude response that is the same in all directions. This is unlikely, but possible. (For example, in a case where the grill is so restrictive that it creates a resonant cavity behind it (and in front of the driver) and that resonance is measurable everywhere.)- In the more likely (and most complicated) case, you have an impact that is different at different angles. Then the question is "how big is the impact, and how much does it change as a function of angle?"
After those evaluations are done, then the first question to ask is "although this is a measurable effect, in what cases could it be audible?" If the effect can only be audible in an anechoic chamber while listening to anechoic xylophone music (yes... I have a recording of anechoic xylophone music - on one of the few CD's that B&O has produced...) then there's not much to worry about.
Another question is: "is this something that can be taken care of "upstream" in the signal processing?" As a simple example: if the effect of a grille is to create a single minimum-phase resonance at one frequency with a fairly low Q, then this is something that can be easily "undone" in the signal processing. You just put a filter in the system that has the opposite phase characteristic, and it will cancel the effect of the resonance in the time domain (and also, therefore, in the frequency domain).
Another question to ask is "does the effect matter, given the expectations and use cases of this loudspeaker?" For example: the grille on a portable loudspeaker may cause diffraction that results in a wider directivity (or "diffusion" as some people call it) - but this is exactly what you might want for such a product. (The acoustic lens generally falls under this heading - it's a thing in front of a loudspeaker driver that has an effect on its directivity - but it's on purpose...) This could also be considered to be an advantage for a centre loudspeaker, since the purpose of a centre loudspeaker is to make the centre image more easily locatable for a large group of people. If you had one chair and no friends, you would not need a centre loudspeaker - assuming your chair was in the right place... So, a highly directional centre loudspeaker makes no sense.
So, to take this into the real world: When we make a loudspeaker with a grille, we do the initial measurements without the grille, and then repeat the measurements with the grille - and assess whether we need to react. Maybe the effect is negligible. Maybe the effect is measurable, but not audible in most cases. Maybe the effect is audible, but we can undo it using signal processing. Maybe the effect is audible - and desirable. Or maybe it's really bad and there's no way to fix it. In that last case, we'll have a meeting to start thinking about what we can do about the grille. At least one of those things will happen (probably more than one - since we're talking about different effects at different frequencies...)
Jumping to the end of the development: the final assessment and tuning of the loudspeaker (here "tuning" is done both with measurements and by listening to the loudspeaker in different rooms and situations) is done on the complete final product. So, if you listen to the loudspeaker with the grille/lamellae/whatever OFF, then you are not hearing what we heard when we decided that this is the way the loudspeaker should sound.
However, it could be that, due to your preferences, or your room, or your cat, a loudspeaker without its grille on sounds better to you with the song that you're currently listening to, an the volume you're listening to it at, in the place where you're sitting, today. (As a simple, but highly unlikely example due to the extreme numbers: If a loudspeaker fabric drops the level by 1 dB at 20 kHz, and therefore we have put in a boost of 1 dB at 20 kHz to compensate for this, but you have a slight hearing loss of 1 dB at 20 kHz, then by taking the fabric off the loudspeaker, you'll get a 1 dB boost that will sound better to you...)
A long answer for a simple question.... sorry.-g
daptay1527:All that being said, what would you get and from where lol? I’m not that sophisticated but want to put the right stuff on.
That's a question that someone else has to answer. Unfortunately, I can't. (although I wonder, for example, if you've looked at an online shop like this: https://www.madisoundspeakerstore.com/grill-cloth/)
I don't know anything about how our procurement department buys components or materials from our sub-suppliers, and I don't know anything about after-market products, which is in many cases, of course, a completely different thing.
All I can do is to offer some insights into technical considerations to help you make your decisions. I can't tell you which car to buy or even how to fix your car - all I can do is to explain the difference between horsepower and torque.
Cheers-geoff