Made some old fashioned calculations for your LaVoce driver in that box
Box, internal dim. 32"x24"x24" = 306L
Chassis deduction 6L
Brazing deduction 10L
Vb (netto volume) 290L
Volume increase do to stuffing 1,,,,, 1.2,,,,, 1.4
Vab 290L,,,,, 348L,,,,, 406L
Vas/Vab 0.8446,,,,, 0.7038,,,,, 0.6033
Airload factor 0.9,,,,, 0.9,,,,, 0.9
Fsb 27,,,,, 27,,,,, 27Hz
Fc (system resonanc) 36.67,,,,, 35.24,,,,, 34.19Hz
Qesbg (Rg=0, Vcoil t=25°C) 0.378,,,,, 0.378,,,,, 0.378
Qesbg (Rg=0, Voicecoil t=200°C) 0.642,,,,, 0.642,,,,, 0.642
Qec (Rg=0, Voicecoil t =25°C) 0.513,,,,, 0.493,,,,, 0.478
Qec (Rg=0, Voicecoil t=200°C) 0.872,,,,, 0.838,,,,, 0.813
Qmsb 7.444,,,,, 7.444,,,,, 7.444
Qmc 10.11,,,,, 4.858,,,,, 4.713
Qtc (Rg=0, Vc temp=25°C) 0.488,,,,, 0.448,,,,, 0.434
Qtc (Rg=0, Vc temp=200°C) 0.803,,,,, 0.715,,,,, 0.693
With the LaVoce I would not worry about to low Qes. When taking into account amp+cable resistance and the Qes increase do to temperature you will hardly drop below Qtc 0.5. Neglecting amp and cablesistance at Vc temp of 200°C with proper 1.2-1.4 stuffing you have butterworth response, f3 = 35Hz
Sorry, there where 3 rows side by side, they got scrambled. I inserted ,,,, to make reading a bit easier.
First row (1) for the box without stuffing, next row (1.2) the with stuffing practically possible box volume increase. Last row (1.4) the with stuffing highest theoretical possible box volume increase
Box, internal dim. 32"x24"x24" = 306L
Chassis deduction 6L
Brazing deduction 10L
Vb (netto volume) 290L
Volume increase do to stuffing 1,,,,, 1.2,,,,, 1.4
Vab 290L,,,,, 348L,,,,, 406L
Vas/Vab 0.8446,,,,, 0.7038,,,,, 0.6033
Airload factor 0.9,,,,, 0.9,,,,, 0.9
Fsb 27,,,,, 27,,,,, 27Hz
Fc (system resonanc) 36.67,,,,, 35.24,,,,, 34.19Hz
Qesbg (Rg=0, Vcoil t=25°C) 0.378,,,,, 0.378,,,,, 0.378
Qesbg (Rg=0, Voicecoil t=200°C) 0.642,,,,, 0.642,,,,, 0.642
Qec (Rg=0, Voicecoil t =25°C) 0.513,,,,, 0.493,,,,, 0.478
Qec (Rg=0, Voicecoil t=200°C) 0.872,,,,, 0.838,,,,, 0.813
Qmsb 7.444,,,,, 7.444,,,,, 7.444
Qmc 10.11,,,,, 4.858,,,,, 4.713
Qtc (Rg=0, Vc temp=25°C) 0.488,,,,, 0.448,,,,, 0.434
Qtc (Rg=0, Vc temp=200°C) 0.803,,,,, 0.715,,,,, 0.693
With the LaVoce I would not worry about to low Qes. When taking into account amp+cable resistance and the Qes increase do to temperature you will hardly drop below Qtc 0.5. Neglecting amp and cablesistance at Vc temp of 200°C with proper 1.2-1.4 stuffing you have butterworth response, f3 = 35Hz
Sorry, there where 3 rows side by side, they got scrambled. I inserted ,,,, to make reading a bit easier.
First row (1) for the box without stuffing, next row (1.2) the with stuffing practically possible box volume increase. Last row (1.4) the with stuffing highest theoretical possible box volume increase
Last edited:
My 2ct.:
Driver Qts and resulting sealed box Qtc is almost irrelevant in an active system as long as
Series resistor or cap is a cheap trick for EQ (by rising apparent Qts) but it's a compromise. Resistor will increase losses and in both variants damping is reduced, aka less local feedback in the driver to linearize/stabilize itself (unless severely overdrive).
---------:---------
Of importance for "fast bass" with any subwoofer, besides the required good room acoustics, is any group delay of sub<-->mains crossover. Often this is a 4th order XO at 50...100Hz, right where most of the bass power will be. With a linear phase crossover, or preferably an amplitude- and phase-corrected total system via digital room correction that group delay can be avoided.
Driver Qts and resulting sealed box Qtc is almost irrelevant in an active system as long as
- you have enough headroom on the amp for EQ, and the amp preferably being Class-D (no dissipation penalty)
- Qtc is lower than 1 or so but the lower the better
- Fs is as low as possible, which means Vb >> Vas
Series resistor or cap is a cheap trick for EQ (by rising apparent Qts) but it's a compromise. Resistor will increase losses and in both variants damping is reduced, aka less local feedback in the driver to linearize/stabilize itself (unless severely overdrive).
---------:---------
Of importance for "fast bass" with any subwoofer, besides the required good room acoustics, is any group delay of sub<-->mains crossover. Often this is a 4th order XO at 50...100Hz, right where most of the bass power will be. With a linear phase crossover, or preferably an amplitude- and phase-corrected total system via digital room correction that group delay can be avoided.
Lots of Linkwitz Transform info here: https://www.diyaudio.com/community/threads/dsp-assisted-reflex-system.404349/page-5#post-7552138
My guess is that assuming you have a good quality EQ you will likely get the most efficient result with EQ as well as more likely to preserve the character of sound. The one wild card is that the sense of "speed" my be highly correlated to the response curve (i.e. fast = high frequencies).
Indeed that is true. The sim is Hornresp using the manufacturers TS specs. In real life I expect experimentation would be necessary. The nice thing is that for a sub as opposed to a passive woofer EQ is probably the easiest and most effective solution. Of course one has to keep in mind the Xlim of the driver when "forcing" more low end output.
Not at all - after fifty years of building speakers, if I had come across any loading which sounded better to my ears I can guarantee you that I would be building it - I'm not simply being obtuse or head-in-the-sand in my choice! Further, if I thought that cuboid enclosures sounded better than the twelve-sided design I have settled upon, I would be more than happy to say goodbye to pesky fifteen degree mitres...
I am convinced that low LF energy storage contributes to this oft-used but somewhat vague description of bass quality. Fourier, of course, tells us that good HF response is a prerequisite for a fast rise time in transient signals, but does this mean that 'fast' bass becomes subjectively less so with age as our hearing fails?
Well, I certainly wouldn't let Dirac and hope steer the ship.
A woofer in a sealed box is as simple as it gets regarding testing and equalization.
Put your microphone right in front of the woofer dustcap. That will yield a nice measurement and you can EQ it from there.
Dave.
If you look closely at the entire sequence starting here, you'll see that the rise time is pretty much the same in all cases. The difference that contributes to the impression of "fast bass" seems to be the amount of ringing that occurs after the input signal stops.
Can you explain to me what I can do manually that Dirac can't do automatically? My minidsp is my pre-amp. I can take measurements after the Dirac EQ with my laptop and see if it's accurate or not and tweak from there, no? I'll be honest 80% of this thread has flown over my head.
Apologies for the ambiguity of my statement. I was referring to the rise time of the system, not the sub. Obviously I agree to the inference that energy storage is problematic in subjective bass quality; indeed I have used the double-pedal kick drum of Metallica et al argument many times, and find that its reproduction on large resonant enclosures (tapped horn/bass reflex/horn-loaded etc.) to be far from accurate with significant blurring of one beat to the next.
So if we accept 32x24x24 is the max size I can build (it is), then by adding more stuffing, I have more buffer for when the driver heats up and Qec goes up, however earlier in the thread you recommended I keep the driver "cool", wouldn't adding a ton of stuffing make it even warmer? So it's a bit of a catch 22 here? Less stuffing = higher initial Qec, but less heat, vs a fully stuffed enclosure would have lower initial Qec but heat up way more, which would produce a higher Qec?
My only quibble would be that in close mic'ing you have to separately account for room gain so you probably want to attempt to get the rolloff to be the inverse of the room gain.
You could save $200. 🙂
Dirac is for big-picture "room correction" type things. For a simple measurement/equalization task like this, it's completely unnecessary. And you will learn much more by working through the process manually.
Dirac is like the self-driving car of the audio world. Everything is hunky-dory until you take your hands off the wheel and it drives you through a stop-sign. 🙂
Dave.
I gave up on filter wizardry after realizing that Qes of the speaker is highly temperatur dependend.
Yes, in theory you can get lower f3 by adding a filter. At first glance it may appear that the only prize to pay is steeper than 12db/octave fall off.
Maybe people blinded by todays stupiditly high given power ratings easely forget that operatin in the compliance limited region must be increasingly forced with power.
Efficiance drop - increased power demand - voicecoil temp rise - decreased damping - inceased settling time - possible waviness....
However, should driver TS param, box size, SPL requirement or whatever dictate to exchange a given closed box f3 slope into lower f3 with much higher slope but even larger program dependend damping variations, ok, go for it if is a given situations best choice, but please be aware of the temp dependence of driver Qes and its implications on overall performance.
Comparing Qtc at 25° voicecoil temp with the Qtc at 200°C (post #22) is hopefully an eyeopener to at least some of you.
Yes, in theory you can get lower f3 by adding a filter. At first glance it may appear that the only prize to pay is steeper than 12db/octave fall off.
Maybe people blinded by todays stupiditly high given power ratings easely forget that operatin in the compliance limited region must be increasingly forced with power.
Efficiance drop - increased power demand - voicecoil temp rise - decreased damping - inceased settling time - possible waviness....
However, should driver TS param, box size, SPL requirement or whatever dictate to exchange a given closed box f3 slope into lower f3 with much higher slope but even larger program dependend damping variations, ok, go for it if is a given situations best choice, but please be aware of the temp dependence of driver Qes and its implications on overall performance.
Comparing Qtc at 25° voicecoil temp with the Qtc at 200°C (post #22) is hopefully an eyeopener to at least some of you.
This begs the questions:
How much power does one need to feed to a subwoofer, and for how long, in order to reach 200°C?
How likely is it that those conditions will be reached in my living room while listening to In-A-Gadda-Da-Vida?
@gorgon53 My goodness, let's try not to make things way more complicated than they really are.
Driver X in sealed-box Y is a defined system regards SPL capability and power requirements.
Modifying the response with equalization doesn't change either of those factors.
If you plan on heating your voice coil to 200" degrees C, (I would suggest you don't!!) you can do that just as easily with, or without, equalization.
Dave.
Driver X in sealed-box Y is a defined system regards SPL capability and power requirements.
Modifying the response with equalization doesn't change either of those factors.
If you plan on heating your voice coil to 200" degrees C, (I would suggest you don't!!) you can do that just as easily with, or without, equalization.
Dave.