your own results show it moving from 70Hz to 67Hz.
And the wider hump shows that Q is lower as well.
Holy 🙂 !
Imagine two drivers with a 70Hz Fs matching this well...
Between units:
Typical Fs tolerance is 15% (combination of Mass and Cms below)
Typical Re tolerance is 10%
Typical BL tolerance is 5-10%
Typical mass tolerance depends on mass, gets tighter with more massive drivers as the glue mass becomes negligible. could vary by a gram or so.
Typical Cms tolerance 20%+
Tolerances are (or can be) tighter within a batch, but when your stuffing effect is less than manufacturing tolerances, it is hardly necessary to make a big deal about it.
Much ado about nothing, I say.
Last edited:
Ron, the setup used the same woofer in the same enclosure on the same position with same air humidity and temperature. The only change was taking out the damping material.
Both are at 67Hz Andrew, the hump you see on the impedance plot which I believe makes you think the actual peak to be at 70Hz was actually caused by the spike pads. Removing the pads removed the hump, never expected this, if I have to be honest especially considering their price ( very good quality spikes ).
I do not have the measurement with non-lined box and no spikes but it was identical - the only difference being the Q being lower asindicated by the wider impedance bump but my point was damping won`t increase volume by the numbers mentioned previously, won`t come even close to them. If you look at the two plots, you`d note that the magnitude is 5 ohms lower. While the Fs would move lower in frequency with increased enclosure volume, it would also increase in magnitude, none are present.
Both are at 67Hz Andrew, the hump you see on the impedance plot which I believe makes you think the actual peak to be at 70Hz was actually caused by the spike pads. Removing the pads removed the hump, never expected this, if I have to be honest especially considering their price ( very good quality spikes ).
I do not have the measurement with non-lined box and no spikes but it was identical - the only difference being the Q being lower asindicated by the wider impedance bump but my point was damping won`t increase volume by the numbers mentioned previously, won`t come even close to them. If you look at the two plots, you`d note that the magnitude is 5 ohms lower. While the Fs would move lower in frequency with increased enclosure volume, it would also increase in magnitude, none are present.
Last edited:
Here is a copy of an old article that used to be fairly widely referred to on the measured effects of stuffing a sealed box. Not large effects despite the enthusiasm of the author but not negligible either.
Mario, your box volume appears to be too great for the intended
woofer. Have you got any more details on the drivers parameters,
box volume and box measures? The standing wave at 150 Hz still
lives after "heavily stuffing" it. Ever measured without the XO filter?
I read that 20 years ago, note, no graphs of amplitude. No pictures...
The question is not whether you can measure anything, it is what you are measuring, and whether it makes a difference at all in comparison to room placement and listener positioning.
If you read some scholarly treatment on the subject (Putland) there is even a question of whether "adiabatic to isothermal" is the mechanism....
Could you even hear it with a simulated headphone test? Doubtful. Damping is worth doing, but not worth stressing over.
Lojzek, its quite the opposite - the Qtc is around 0.81 🙂 The bump at 150Hz ( 156Hz if we have to be precise ) was caused from the speaker being placed directly on the floor. Under normal circumstances, it would not be and will be placed on a 4mm thick steel stand where the spikes mount. Below is a measurement on the stand under the same conditions. There is still some resonance as evidenced by the phase blip but not even near to the floor mounted one. Not that much of a problem as the driver is used to cover only up to 90Hz @ 12db/oct RL.
By the way, I forgot to mention all tests were done with polyester filling only, I haven`t tried other more exotic materials as they push the project above budgetary constraints. I would, however, be interested to test other materials as well ( especially wool ) and using a drivers with known parameters, the closed box method can estimate if there is a change in observed virtual volume by the driver, I think, and how does the filling and material affect it. Maybe a study is worth it 🙂
By the way, I forgot to mention all tests were done with polyester filling only, I haven`t tried other more exotic materials as they push the project above budgetary constraints. I would, however, be interested to test other materials as well ( especially wool ) and using a drivers with known parameters, the closed box method can estimate if there is a change in observed virtual volume by the driver, I think, and how does the filling and material affect it. Maybe a study is worth it 🙂
Attachments
Last edited:
I still have no clue what the box width, height and depth is.
Is that a vented cabinet? You've measured impedance with ARTA.
Then you must have basic parameters like Fs,Qms and Qes of the
driver in the air or you didn't bother to measure. Thanks.
Is that a vented cabinet? You've measured impedance with ARTA.
Then you must have basic parameters like Fs,Qms and Qes of the
driver in the air or you didn't bother to measure. Thanks.
Many studies have been done on stuffing of enclosed woofer boxes (e.g. acoustic suspension). The best improvement in performance is obtained with fine fiberglass fibers like those use to line your home wall for insulation. Box resonance is reduced and the impedance peak at resonance also is reduced. For your boxes I suggest about 250g FG. I have digital charts and tables of those studies. PM me if interested as they are copywrited material.
I thought it is evident from the single impedance peak this isn`t a vented cabinet. 🙂
The driver was custom built, Fs 32Hz, Qts 0.44, Vas 79l, SPL 90db/w, Re 5.2ohm. It also has somewhat acceptable mechanical factor of 2.3 The enclosure dimension and etc. I cannot share (yet), sorry.
The driver was custom built, Fs 32Hz, Qts 0.44, Vas 79l, SPL 90db/w, Re 5.2ohm. It also has somewhat acceptable mechanical factor of 2.3 The enclosure dimension and etc. I cannot share (yet), sorry.
Hi,
Just tables that are easy to read : A better link :
http://techtalk.parts-express.com/attachment.php?attachmentid=63055&d=1452624890
Any argument with the methods ?
rgds, sreten.
It is the same old BS. Measure impedance and play with software. He doesn't give away enough of his methodology that anyone sane should have confidence in his results. There are quite a few degrees of freedom here. The fact that his vented boxes had a dropped (predominantly lower) peak means he interfered with the port action.
The proof is in the SPL response, not the impedance. This is a less than 2 page article, written for a Car Stereo mag, the message is, basically ...use stuffing...stuffing good...
What do you think this article proves, anyway?
Hi,
Oh it is BS is it ? The proof is most certainly the impedance peaks
frequencies not some nebulous assumptions based on SPL responses.
Yes the vented arrangement doesn't really add a lot of clarity, but
it is pointless to say it is BS, it has nothing to do with software.
What it proves depends on how you understand the methods.
rgds, sreten.
Oh it is BS is it ? The proof is most certainly the impedance peaks
frequencies not some nebulous assumptions based on SPL responses.
Yes the vented arrangement doesn't really add a lot of clarity, but
it is pointless to say it is BS, it has nothing to do with software.
What it proves depends on how you understand the methods.
rgds, sreten.
Sorry to pull you up on this one, Andrew, but I too used to get mixed up with those thermodynamics terms! Constant temperature is described as "isothermal", which is fairly close to what you get by stuffing the cabinet (the fibres have a much higher thermal mass than the air, and are in good contact, so stabilise the temperature).
The air in the unstuffed cabinet will of course heat up and cool down during compression and rarefaction (with a little heat transfer to/from the cabinet walls), as you say, and that's an example of an "adiabatic" process.
For the OP:
A reasonable degree of stuffing is said to increase the effective volume of a sealed cabinet by about 20%, and this should be taken into account when designing for a given resonant frequency and system Q.
(As has been said, a ported cabinet shouldn't normally be stuffed, but rather lined with something a lot denser than the fill used for a sealed box.)
If any serious speaker builders who visit here regularly don't own or have access to Vance Dickason's Loudspeaker Design Cookbook (no affiliation), you NEED TO GET hooked up with one. Included in it is a very well done controlled, closed box stuffing experiment using different materials to stuff and combinations of some of them. Additionally, there's a great summary table (already shared with Pankov) and pages of FR and impedance curves for each different stuffing iteration. The best I've ever seen.
Here's just one place you can get a copy....
http://www.amazon.com/Loudspeaker-D...=loudspeaker+design+cookbook+7th+edition+book
Here's just one place you can get a copy....
http://www.amazon.com/Loudspeaker-D...=loudspeaker+design+cookbook+7th+edition+book
Thanks for coming back with your correction.
Are you saying that the normal unstuffed is equivalent to adiabatic and that the stuffed version is equivalent to isothermal?
Yes indeed. It's nice that, for once, Physics has provided a win-win situation: damped resonances AND reduced cabinet size!
The only negative is that you seem to get the best damping with nasty fibreglass.
It is useful if you need to dampen resonances due to standing waves developing in the box. I attached a few measurements from a recent speaker project, where I needed to get rid of such a resonance in the woofer department at about 290 Hz. The effect of the damping material (lambs wool) is quite substantial.
If your woofer is filtered to frequencies below the internal standing waves of the box, damping may not be necessary. Otherwise it's certainly a good thing to do.
Attachments
Bingo, that's what I was suggesting too, look at the impedance plot of your driver (bare) and in the enclosure. Any new peaks in the impedance compared to the bare driver plot are added by the box.
Damping can get you closer to your desired FR plot. I personally use fiberglass insulation and real wool felt (for lining the walls) but the impedance plot was telling me what to use. Twaron's angel hair is something special in that regard. It behaves a bit different from fiberglass and is excellent for midrange(*). Not cheap though.
As I was trying to find a recipe for the entire bandwidth fiberglass was hard to beat. It was the only one that worked very good in the entire spectrum. Even better with the wool felt, which I also used as a guard behind the driver to keep the fiberglass away from the driver's vent.
Here's an impedance test from my test enclosure with a couple of different combinations:
And another one showing driver alone on baffle, driver in empty box and driver in box damped with fiberglass and wool felt:
(*) You can use less of it for the same amount of damping fiberglass brings...
Last edited:
One way to remember which is which: ISOthermal is like ISObar - the "iso" in both cases is from the ancient Greek "equal", so equal temperature and equal pressure.
- Status
- Not open for further replies.