You can use this data in Akabak, but be aware that the values will change with physical material size and mounting. There is also no simulation of transference through the material as far as I am aware of.
To make a curve file, the data in the tables on that site is already in ISO octave bands, so just make a comma-separated text file with the band frequency and absorption coefficient value. Ignore the NRC value, as this is the single-value noise rating criteria for the entire passband.
Import that as a Curve file in Akabak, and then assign it to a surface using the Damping type in a Wall Impedance. It's probably worth importing a few varieties as curve files at once, so you can easily swap materials out. Also note the thickness values; you might want to offset a plane for the absorptive material from the cabinet wall to account for the 'block' of stuffing.
Sure, french word, there is also a verb which is tamponer...but it doesn't mean what you meant there, lol.An abrasive Tampon would have quite a different translation in English
It's strange that I have quite some interesting results using some different types of fibrous materials, including mineral wool and fiberglass and natural sheep wool, but no luck at all with "eco felt" which is some kind of felt made of recycled cloth scraps which consists of majority polyester and cotton. I searched quite seriously, hoping to find the answer why this felt doesn't work here? Does anyone know anything about this? Some kind of adhesive is used in production of this felt, wouldn't this contribute? This is it:
I used 30 and 50kg/cubic meter fiberglass with comparable results but with using more or less thickness. What density of fiberglass did you use?
Sheep wool also works but it's hard to keep it in place and lacks the uniformity.
This kind of felt is a very effective absorber, it is most likely too effective to work in this application.
Actually, I just found this image that seems to be the same felt material.
http://www.groundsound.com/GalleryThomasM.php
Attachments
I am using this type. Sorry, I did not find anything in english, but you get the numbers ;-).
The weight of 20 sheets is 11,56kg. The volume according to datasheet is 0,27cubic meter. That gives about 43kg/cubic meter. On the other hand, the volume of non compressed sheets should be 13,44sq. meters x 0,05.= 0,672cubic meter. That gives about 17kg/cubic meter.
I used a turntable to measure the speaker in different angles. I have written a dozen of notes about designing a cardioid. I may collect them all in a PDF maybe someone someday can use them to design something similar.
We know fibrous materials like fiberglass and mineral wool and polyester wool reduce the speed of sound because the sound wave has to propagate in an oblique manner instead of going direct in a straight line, so the traveling distance will increase and the C=D/t ratio will change. This means, if we can route the sound of the back side of the diaphragm to a channel with longer length the wave will travel more distance, so the back wave will be somehow delayed in comparison to the front wave of the diaphragm. What I'm trying to explain is that we might be able to concoct some kind of maze at the back side of the speaker and the other end could end up to speaker sides for cancellation to shape cardioid. Something like back-loaded horns. Folding the maze will eliminate higher frequencies, thus acts as a low pass filter. Any idea about this? Am I missin' something?
Did some more investigation. The volume of 0,27 is a transport volume. The sheets are compressed about 2,5 times for transport, so the correct density is the last uncompressed one (17kg/cubic metre)
^^ dave123, Sounds feasible, since all that is needed from the backwave is delay. But, I suspect there is going to be some resonances which would benefit damping and you'd now be back to damping material land and the maze would add too much delay, depending what kind of a pattern you are after. Just my intuition, which can be wrong and you should try and figure out how to test it. It would be great if it worked though!
Is there any real world products doing it with a maze?
Is there any real world products doing it with a maze?
I see. The lower density means less reduction in speed (more/less delay), hence the material should be thicker. Disparity in thickness will change the STC and absorption coefficient of the resistive material that would result in the low pass filter alteration. Such a multi-unknown equation passive cardioid speaker is.
That's right. It should be a small working area to tune the resistive port to work properly. There are countless combinations of failure and very few ones that would work.
No, there is no speaker on the planet that uses a maze to shape a cardioid polar pattern.
Good stuff, will look into it. That said, I guess a single plane facing inwards towards the driver will suffice since the values are for a given thickness? I am thinking of placing the plane where the stuffing would "start"?
I was wondering if it would be feasible to use a passive radiator at the back side of a speaker and use it to shape a cardioid response. I think the whole internal air inside the cabinet acts as a fluid and the passive radiator works like a resonator, so I don't know exactly if it would be possible to apply some delay using resistive material and make the passive radiator vibrate accordingly. Just crossed my mind, wanted to think loudly.
I was referring to the remark about eco-felt. My observations with blankets of fibrous material is that the absorption is better when sound is exposed to the sides instead of the blanket surface. That is why i asked.
Dear all,
I only recently came across the 8c's. Even though I've never heard them, I am very impressed by the design and the laudatory reviews.
I've built myself a couple of speakers and experimented with DSP. My latest speakers are active 4 way open baffles, DSP is done in Roon outputting to Exasound 8 channel DAC. The result if really OK despite the reverberant room (RT 0,7...) but the speakers suffer from bumpy directivity index over frequency....
I'd really like to prototype something along the lines of the 8C, a 3 way driver :
The challenge is the mids : design the speaker enclosure so that the ports and the insulation actually provide the delay and low pass filtering required for the "cardioid" profile generation.
I have started to play with Excel on a very empirical approach. I have usef a MH Audio calculator for speed of sound, but I am quite perplex, I had bad experiences in the past with their calculators.
Anyway this is really step 0, I guess a lot of experimentation and measurements will be needed ! It's really a challenge to find the material and thickness that provides the right amount of delay AND low pass transparency.
I'd really like to see the 8c opened and check what they have for insulation to filter/delay the back wave of the mid-speaker...
I only recently came across the 8c's. Even though I've never heard them, I am very impressed by the design and the laudatory reviews.
I've built myself a couple of speakers and experimented with DSP. My latest speakers are active 4 way open baffles, DSP is done in Roon outputting to Exasound 8 channel DAC. The result if really OK despite the reverberant room (RT 0,7...) but the speakers suffer from bumpy directivity index over frequency....
I'd really like to prototype something along the lines of the 8C, a 3 way driver :
- low : 20-100 Hz : two 8" woofers in closed vent, this is the easy part
- mid : 100-1300 Hz : one 8" speaker
- high : above 1300 Hz : maybe a horn? to be seen later
The challenge is the mids : design the speaker enclosure so that the ports and the insulation actually provide the delay and low pass filtering required for the "cardioid" profile generation.
I have started to play with Excel on a very empirical approach. I have usef a MH Audio calculator for speed of sound, but I am quite perplex, I had bad experiences in the past with their calculators.
Anyway this is really step 0, I guess a lot of experimentation and measurements will be needed ! It's really a challenge to find the material and thickness that provides the right amount of delay AND low pass transparency.
I'd really like to see the 8c opened and check what they have for insulation to filter/delay the back wave of the mid-speaker...
