Here's the situation. I have lots of power amps with lots of watts of power each. I have 20 12" drivers. i have 4 cabinets. I want to use them in my livingroom. I can get more cabinets, these are cheap. I have lots of EQ available. Efficiency doesn't matter much.
Active three-band crossover to a dozen 6" scan-speak mid-bass drivers in sealed boxes, and stacked Newform Ribbon (tall skinny line-source planar dynamics).
So how many 12" drivers to load into each of these cabinets? Looking at sealed boxes with the intent of adding EQ and driving them WELL below resonance:
1) I can put 1 driver in each for a Qtc somewhere around .7 but can I live with 20 cabinets in the house, 10 on each side for stereo? Probably not.
2) I can put 2 pairs of clamshelled isobaric pairs of drivers (4 drivers) in each, for the same Qtc around .7, twice the power out of each cabinet(but not twice the volume, and fewer cabinets) and 1/4 the efficiency (which doesn't matter in this particular circumstance when 4 times the power is easy to come by, assuming I put in more 20amp power lines and outlets) and same impedance (2 parallel, those pairs in series). This could make 5 cabinets...more likely 4 cabinets and 4 spare drivers; 2 cabinets on each side.
3) I can put 1 pair of clamshelled isobaric drivers in each, for about a Qtc below .5 which would make 10 cabinets, 5 on each side.
I'm interested in fidelity, low disortion, mostly for stereo listening.
What I'm actually considering is option 3 with lots of cabinets, and using EQ to try to get some usable bass when the speakers are at the -9db point or -24 db point or well into unusual territory for me.
I've got the T-S parameters, but I've just been using the most basic box calculator applet so I haven't really figured out how low they'll go below resonance. But this is interresting territory for me, as for once I don't have to worry about efficiency or output or even flat response; in fact I don't care about any of those. I'll EQ like mad, plenty of power to do that, plenty of cone area with 20 drivers, LOTS of power-handling with 20 voice coils.
So I guess this is nearly infinite-baffle. Territory my dad explored, but I certainly never have. But I'm doing it isobaric with lots of power, so it's not too huge. And 10 smaller boxes are easier to move, stack, etc.
Since I have removable baffel boards, I might build one #3 and one #2 and listen.
I do like really low-frequency output, and it doesn't have to be impressivley loud (though even moderate levels often take a lot of power). Somewhere I have a disc with a huge orchestral bass drum tuned even lower, but I can't find the track because I haven't had speakers that give any hint it's there. You don't hear it, you feel it. Or you don't. entire instruments missing with above-average hi-fi speakers.
Now I need some better free software that will draw me response curves that continue way down the left bottom corner of the graph. And perhaps tell me cone excursion...
As I get lower below resonance, I can't use a reflex or it unloads, I can't use a high Qtc as even with EQ throwing 1000 times the power at it, 1000 times nearly nothing is still nearly nothing.
Or I could plan to add weight to these poly cones, re-measure the T-S parameters, and see what the graphs would look like for a #2. Qtc would still be reasonably low, but would lowering fs get any really low bass. It's an option, as I don't have to care much about efficiency. Or cost, as this is all stuff I have (and obtained thru some generous folks and good deals).
These drivers are from the NHT AR limited:
Vas 269 (I assume that's liters)
fs 19.8 hz
Qts .205
Active three-band crossover to a dozen 6" scan-speak mid-bass drivers in sealed boxes, and stacked Newform Ribbon (tall skinny line-source planar dynamics).
So how many 12" drivers to load into each of these cabinets? Looking at sealed boxes with the intent of adding EQ and driving them WELL below resonance:
1) I can put 1 driver in each for a Qtc somewhere around .7 but can I live with 20 cabinets in the house, 10 on each side for stereo? Probably not.
2) I can put 2 pairs of clamshelled isobaric pairs of drivers (4 drivers) in each, for the same Qtc around .7, twice the power out of each cabinet(but not twice the volume, and fewer cabinets) and 1/4 the efficiency (which doesn't matter in this particular circumstance when 4 times the power is easy to come by, assuming I put in more 20amp power lines and outlets) and same impedance (2 parallel, those pairs in series). This could make 5 cabinets...more likely 4 cabinets and 4 spare drivers; 2 cabinets on each side.
3) I can put 1 pair of clamshelled isobaric drivers in each, for about a Qtc below .5 which would make 10 cabinets, 5 on each side.
I'm interested in fidelity, low disortion, mostly for stereo listening.
What I'm actually considering is option 3 with lots of cabinets, and using EQ to try to get some usable bass when the speakers are at the -9db point or -24 db point or well into unusual territory for me.
I've got the T-S parameters, but I've just been using the most basic box calculator applet so I haven't really figured out how low they'll go below resonance. But this is interresting territory for me, as for once I don't have to worry about efficiency or output or even flat response; in fact I don't care about any of those. I'll EQ like mad, plenty of power to do that, plenty of cone area with 20 drivers, LOTS of power-handling with 20 voice coils.
So I guess this is nearly infinite-baffle. Territory my dad explored, but I certainly never have. But I'm doing it isobaric with lots of power, so it's not too huge. And 10 smaller boxes are easier to move, stack, etc.
Since I have removable baffel boards, I might build one #3 and one #2 and listen.
I do like really low-frequency output, and it doesn't have to be impressivley loud (though even moderate levels often take a lot of power). Somewhere I have a disc with a huge orchestral bass drum tuned even lower, but I can't find the track because I haven't had speakers that give any hint it's there. You don't hear it, you feel it. Or you don't. entire instruments missing with above-average hi-fi speakers.
Now I need some better free software that will draw me response curves that continue way down the left bottom corner of the graph. And perhaps tell me cone excursion...
As I get lower below resonance, I can't use a reflex or it unloads, I can't use a high Qtc as even with EQ throwing 1000 times the power at it, 1000 times nearly nothing is still nearly nothing.
Or I could plan to add weight to these poly cones, re-measure the T-S parameters, and see what the graphs would look like for a #2. Qtc would still be reasonably low, but would lowering fs get any really low bass. It's an option, as I don't have to care much about efficiency. Or cost, as this is all stuff I have (and obtained thru some generous folks and good deals).
These drivers are from the NHT AR limited:
Vas 269 (I assume that's liters)
fs 19.8 hz
Qts .205
Last edited:
...and have 4 drivers left over as spares. Still isobaric each could have a very small footprint. I would make each 'tower' in two stackable pieces to aid mobility and flexibility of arrangement, which would make it the same as option #3. That would allow me to experiment with very-low Qtc. But...I already have 4 enclosures, that would require buying 4 more. And they would not really reach the ceiling.
I'll be stacking them, whatever I do.
I also need to decide whether to center the drivers in the baffle board. Usually I don't, preferring any edge effects be more randomly assymetric, but I'm undecided if I intend to stack them.
Any advice about aiming for very low Qtc?
I'll be stacking them, whatever I do.
I also need to decide whether to center the drivers in the baffle board. Usually I don't, preferring any edge effects be more randomly assymetric, but I'm undecided if I intend to stack them.
Any advice about aiming for very low Qtc?
..You really need to measure your drivers as documented T/S for these is not consistent;Foster-AR12_T-TQWT: b
T/S are consistent
iscard the picture in the post here-above,By a mistake wrong Sd was entered:b
Attachments
Bjorno,
Took me a while to figure out how to view that .jpg bigger, but I saved it to disk and even Windows picture and fax viewer let me enlarge it and scroll around in it.
Now I've got to digest it. I do have the rest of the T/S parameters, these were just htose most immeidately relevant to a sealed box.
Took me a while to figure out how to view that .jpg bigger, but I saved it to disk and even Windows picture and fax viewer let me enlarge it and scroll around in it.
Now I've got to digest it. I do have the rest of the T/S parameters, these were just htose most immeidately relevant to a sealed box.
Hi,
1) versus 2)
A) Same impedance. Same Qbox. If your LTing box Q does not matter much.
B) Max SPL per box will be quadruple, max power handing will quadruple.
C)Efficiency will be the same.
For 3) max SPL is the same, (per box), efficiency halves, power handling
doubles and impedance halves. In this case you may get a response in
room that requires no further EQ to sound really good.
2) and 3) will have the same maximum SPL capability with all cabinets,
1) with all cabinets would be +6dB more, double volume displacement.
Theoretically with EQ 1) will dig the deepest due to the highest SPL,
but requires 4 times the box volumes of 2) and twice that of 3).
(20 boxes in a room go +26db louder than one.)
FWIW 10 drivers (pairs) in a room are +20dB SPL and efficiency
compared to one driver (pair). 2) has the smallest total box
volumes but requires more power and LTing to equal 3).
So IMO your choice is 2) or 3), to LT with more power or not to LT.
Guess it boils down to how many same sized boxes are acceptable.
rgds, sreten.
Boxes are larger than I thought.
So I'm guessing if I mount (adjusting the box volume for the space occupied by the speaker(s):
1 driver in the cabinet it's about 2.2 cubic feet, Q just below .48
2 pairs of 2 clamshelled drivers in two holes in the cabinet it's a similar 2 cubic feet, Q around .49
2 clamshelled drivers in one hole in the cabinet, it's equivalent to 4.4 cubic feet, ridiculaous Q around .38
So I'm guessing if I mount (adjusting the box volume for the space occupied by the speaker(s):
1 driver in the cabinet it's about 2.2 cubic feet, Q just below .48
2 pairs of 2 clamshelled drivers in two holes in the cabinet it's a similar 2 cubic feet, Q around .49
2 clamshelled drivers in one hole in the cabinet, it's equivalent to 4.4 cubic feet, ridiculaous Q around .38
So, I don't care about the -3db point but
Qtc .38 down 14.61db at 20 hz, down 10.26 db at 30 hz
Qtc .49 down 16.49db at 20 hz, down 11 db at 30 hz
Doubling the box size it's a little teensie bit better output below 40 hz, but any noticeable improvement is below 25 hz, if it can make the excursion.
Qtc .38 down 14.61db at 20 hz, down 10.26 db at 30 hz
Qtc .49 down 16.49db at 20 hz, down 11 db at 30 hz
Doubling the box size it's a little teensie bit better output below 40 hz, but any noticeable improvement is below 25 hz, if it can make the excursion.
Hi,
2 pairs per cabinet clamshelled will down low with room gain.
LinearTeam
You can add further EQ, a lot easier than cone mass adjustment,
but TBH critically damped boxes (Q= <0.5) should sound very good.
Having researched this, commercial subwoofers go very low at modest
levels, with bass boost, but back off the boost as you approach the
excursion limits of the drivers, to prevent overexcursion.
These adaptive circuits are common commercially, but not in DIY.
Line level passive EQ bass boost is very simple for Q=0.5 boxes.
see : Baffle Step Compensation
but think 30Hz rather than 300Hz as the centre frequency. Lowering
the lower resistor will increase maximum relative bass boost available.
Though a better solution (in max SPL respects and perceived extension at the limits)
might be a peaking high pass filter set to Q = 2 to 3 set to as low as you want to go.
rgds, sreten.
Might as well point out the drivers with such a low Q are not really suited
to sealed or isobaric ideally, and individually vented very low in boxes you
would end up with a monstrous low end capability, lots of big boxes though.
2 pairs per cabinet clamshelled will down low with room gain.
LinearTeam
You can add further EQ, a lot easier than cone mass adjustment,
but TBH critically damped boxes (Q= <0.5) should sound very good.
Having researched this, commercial subwoofers go very low at modest
levels, with bass boost, but back off the boost as you approach the
excursion limits of the drivers, to prevent overexcursion.
These adaptive circuits are common commercially, but not in DIY.
Line level passive EQ bass boost is very simple for Q=0.5 boxes.
see : Baffle Step Compensation
but think 30Hz rather than 300Hz as the centre frequency. Lowering
the lower resistor will increase maximum relative bass boost available.
Though a better solution (in max SPL respects and perceived extension at the limits)
might be a peaking high pass filter set to Q = 2 to 3 set to as low as you want to go.
rgds, sreten.
Might as well point out the drivers with such a low Q are not really suited
to sealed or isobaric ideally, and individually vented very low in boxes you
would end up with a monstrous low end capability, lots of big boxes though.
Last edited:
Well, I didn't make the best choices but practical ones. I'm stuffing 4 (2 clamshelled pairs) into each box for Q .5 with a brace installed. It's not going to go as low as a tapped horn could, but with 4 boxes that's still 8 12" drivers exposed to room air and they have acceptable xmax, so it's going to make plenty of bass in the more-perceived range. If I want more lower, I might get some more of the same cabinet kits and load them up with 2 15's clam-shelled isobaric in each. Yes, big ported boxes would have a lower -3db point. But I'd hoping these put out something detectable even lower. I will see soon enough. Few more weekends of construction. I sure wish I hadn't sold my table saw.
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.