From their spec sheet:
Cabinet recommendations: Sealed 2.0 cubic ft. (net internal) with 1 lbs. of Acousta-Stuf polyfill, f3 of 36 Hz with a 0.707 Qtc alignment
However, when modelling their driver, a Qtc of 0.707 gives a Vb of 107.4 L (3.79 cu ft.). Typo on their site or am I doing something wrong?
Cabinet recommendations: Sealed 2.0 cubic ft. (net internal) with 1 lbs. of Acousta-Stuf polyfill, f3 of 36 Hz with a 0.707 Qtc alignment
However, when modelling their driver, a Qtc of 0.707 gives a Vb of 107.4 L (3.79 cu ft.). Typo on their site or am I doing something wrong?
I analyzed PE's recommended volumes when they seemed smallish. I found that in general, PE's recommended sealed volumes result in a Qtc of ~0.8 for the undamped case when Qts<0.7 or so. Whoever wrote the copy may have assumed it was 0.707.
Simplified approximate damped box calcs here which give roughly the values I quoted:
Sealed Boxes - DIY Loudspeaker Design
You can also use a full model like Unibox with Ql=Qb=10 (or 5 for the combo) for the same stuffed result or Ql=Qb=20 (or 10 for the combo) for the lined result - and Ql=Qb=∞ for the undamped case.
When above I say "for the combo" I mean simply that Ql*Qb/(Ql+Qb)=5 or 10. For example, Ql can be as low as 5 (where Qb would be infinity) and vice versa to get a damping Q of 5.
These resistive damping models essentially act like a notch filter at Fc, they don't predict that the box appears larger, so you will see your F3 go up.
Stuffing models are a tricky business.
After learning a ton about box modeling (that is useful in and of itself for other purposes) I have decided that mostly Qtc doesn't matter much once you factor in room acoustics and the main difference between Qtc's (within reason) is box size and not so much sound. People who try to tell you Qtc=0.5 (or 0.707106 or whatever) is the holy grail or some other such nonsense are feeding you a line of carefully considered scheiße.
Last edited:
^Scheiße.
Yes, sometimes higher qtc sounded better to me, well above 0.707.
You just need to experiment.
On closed enclosures I will design a test box with a "variable back wall" and adjust box volume to taste.
Sometimes you will find a "sweet spot", within a small range it will just sound best.
From there I will go for the final enclosure.
Ron, thanks for the info. Using the calculations on the page you linked I get insanely small enclosure volumes with a reduction of around 50-60% depending on what driver and Qts I am aiming for. For the Dayton UM12-22 I get a stuffed enclosure Vb of only 38% vs. empty, which doesn't seem right.
I guess modelling is difficult as you say.
I do not follow your two sentences referring to the unibox model quoted above. Do you have a link?
I guess we conclude that as soon as you start stuffing sealed enclosures the simulations are no longer accurate, probably due to the complexity of the filling material (density/fiber thickness/thermal properties/etc). We either go with a volume adjustable test cab (as suggested) which we can measure and tweak, or build something tried and tested. I guess in the case of the Dayton, their 2 cu ft stuffed enclosure (with their recommended 1 lbs. of Acousta-Stuf polyfill) is the way to go. I assume this is a design they have tested and it is not wildly off the stated 0.707.
From the calculations, a Qtc of 0.707 on this driver gives:
Empty: 107.35L / 3.8 cu ft
Lined: 70.62L / 2.5 cu ft
Stuffed: 41.15L / 1.5 cu ft
So their 2 cu ft volume falls right in between the lined and stuffed estimated volumes.
I mean any combination of Ql and Qb that give Ql*Qb/(Ql+Qb)=5 or 10 will give roughly the same response in unibox as is predicted by the formulas I linked to..
I highly doubt they have tested the box size, it is just what software spits out and it is roughly the same volume as for a Qtc=0.8 undamped box.
I highly doubt they have tested the box size, it is just what software spits out and it is roughly the same volume as for a Qtc=0.8 undamped box.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.