sbhattac@rnd.gba.nyu.edu (Shankar Bhattacharyya) (09/21/90)
Quite possibly you are asking yourself, "Why does this bozo think a list of
undesirable drivers is of any interest to anyone?" Well, I think I have a
valid reason for putting this on the net. These are drivers which looked
promising at first glance, from the Madisound and Audio Concepts
catalogues, but turned out not to look so good. The manufacturers are all
reputable, so it is not that the drivers are bad. Many of these drivers
come easily to mind when one thinks about choices in drivers, so it
struck me that seeing these numbers might indeed be useful to some people.
They are just not that good for sealed boxes. Many will work very well in
vented designs, and produce excellent results.
Perhaps I should mention that I perceive a certain bias in the Madisound and
Audio Concepts catalogues. Audio Concepts seems to be quite heavily biased
towards sealed boxes, i.e. most of their woofers seem to be designed for
sealed boxes. Madisound seems to be heavily biased towards vented boxes, in
the same sense. I don't mean this in any philosophical sense, just as a
descriptive statement about their lines of woofers. Of course, both sell
the Dynaudio drivers, some of which, eg. the 30W54, work very well in
either. The 30W54, incidentally, seems like a most versatile driver.
(Sanders mentioned a transmission line woofer he built with the 30W54, in an
aside in a recent series on electrostatics in Speaker Builder, and he
sounded as if he thought it was phenomenal.)
As before, the numbers were generated by a trivial little C program, which
I do not guarantee to be correct. I'd be crazy to issue guarantees. So I
suggest that you check numbers out before you believe them completely.
I have included some comments on why I think the driver is not acceptable.
If my reasons make sense, perhaps those who have not designed closed boxes
will find something useful in the comments. If I am wrong, I daresay
someone will correct me.
Actually, I'm surprised that there have been no corrections yet. Hell, I
must have made some mistakes.
If you find errors, please let me know.
For convenience:
1 liter = 61 cu in
1 cu ft = 28.3 liters
All volumes below are in liters, frequencies in hz.
Dynaudio 24W75 $ 50
Qts = 0.83, Vas = 88.00, Fs = 33.0, Qes = 1.11, EBP = 29.73
The high Qts makes lower Q alignments impossible, and, for a little driver,
the box size is huge for Qtc < 1.2. In fact, all the alignments here would
usually get regarded as infinite baffles rather than acoustic suspensions.
Also, the F3 numbers look wrong to me, but it may just be that I am not
accustomed to drivers with Qts so high. I included this driver because
it is a Dynaudio, and is pretty cheap for a Dynaudio. Note that the EBP
does not look good either for vented boxes or for closed ones.
(My program produces odd looking numbers for Qts close to Qtc, and I don't
know why this is so. I don't know that the numbers are wrong, they just
look odd to me. Since realistic designs should not aim at that sort of
alignment, this may be no problem in either case.)
qtc alpha Vb Fc F3
(hz) (hz)
0.50 requires non-positive box volume -- not feasible
0.60 requires non-positive box volume -- not feasible
0.70 requires non-positive box volume -- not feasible
0.80 requires non-positive box volume -- not feasible
0.90 0.16 544.05 35.57 29.50
1.00 0.43 202.64 39.52 31.07
1.10 0.74 119.65 43.47 32.90
1.20 1.07 82.60 47.43 34.90
1.30 1.42 61.80 51.38 37.01
JC12 (from Audio Concepts) paper cone, dual voice coil $ 60
Qts = 0.31, Vas = 339.00, Fs = 18.0, Qes = 0.32, EBP = 56.25
This is actually not bad. Of course, I don't know what to do with the dual
coil. If someone out there does, please post some information on this. What
happens if you hook the coils up in parallel? Does that change alignment?
My major reservation is that a 12" woofer with a Fs of 18 hz should provide
a better F3. The very low Qts accounts for this, although the EBP looks
good. I think this would do better in a vented box, although the Qts is
a bit low even for that. I may well be entirely in left field on this one,
as you say in the American. Caldwell has designed good stuff over time, so
I assume he knows stuff that I do not.
qtc alpha Vb Fc F3
(hz) (hz)
0.50 1.60 211.68 29.03 45.11 (my numbers are different
0.60 2.75 123.45 34.84 42.12 from Audio Concepts's.
0.70 4.10 82.71 40.65 41.06 They get F3 of 36 hz,
0.80 5.66 59.90 46.45 41.67 but I have no idea how
0.90 7.43 45.63 52.26 43.35 dual vc woofers behave,
1.00 9.41 36.04 58.06 45.65 so my numbers may be
1.10 11.59 29.25 63.87 48.33 useless)
1.20 13.98 24.24 69.68 51.27
1.30 16.59 20.44 75.48 54.37
Audio Concepts AC-8 $ 40
Qts = 0.41, Vas = 58.00, Fs = 30.0, Qes = 0.50, EBP = 60.00
Actually, this looks quite good in a closed box. The issue is that since we
are talking about woofers to extend bass in to serious territiory, 50+ hz
just does not do it. But this would be a serious candidate for a two way
system.
qtc alpha Vb Fc F3
(hz) (hz)
0.50 0.49 119.05 36.59 56.85
0.60 1.14 50.81 43.90 53.08
0.70 1.91 30.29 51.22 51.74
0.80 2.81 20.66 58.54 52.52
0.90 3.82 15.19 65.85 54.62
1.00 4.95 11.72 73.17 57.52
1.10 6.20 9.36 80.49 60.91
1.20 7.57 7.67 87.80 64.61
1.30 9.05 6.41 95.12 68.52
Swan 305 (substitute for Precision/Peerless TA305) $ 50
Qts = 0.31, Vas = 167.00, Fs = 24.5, Qes = 0.34, EBP = 71.22
This looks terrible in a sealed box. In a vented box, however, it should
perform very well, with F3 in the 20's. I included this basically because
people will think of the TA305. Note that the EBP is high for a closed box.
Seth Bradley's design uses the original TA305 in a vented box, very
successfully.
qtc alpha Vb Fc F3
(hz) (hz)
0.50 1.54 108.75 39.01 60.62
0.60 2.65 62.99 46.82 56.60
0.70 3.97 42.07 54.62 55.18
0.80 5.49 30.41 62.42 56.00
0.90 7.22 23.15 70.22 58.25
1.00 9.14 18.27 78.03 61.34
1.10 11.27 14.82 85.83 64.95
1.20 13.61 12.27 93.63 68.89
1.30 16.14 10.35 101.43 73.07
KEF B139 $ 108
Qts = 0.37, Vas = 164.00, Fs = 25.0, Qes = 0.40, EBP = 62.50
This driver is something of a legend, particularly in Britophile circles.
I recall seeing it in an early WAMM, and it has been used in many
loudspeakers,, including, obviously, many from KEF. However, if I were
spending $100 on a driver, I would be annoyed if I could get no better than
50 hz bass without equalization. I think this would work much better in a
vented box, yielding F3 in the mid 20's perhaps. It is also a popular
driver in the builder community for transmission lines. Until the Dynaudios
became well-known in the US, this used to be the fancy driver of choice.
Many people still like use it.
qtc alpha Vb Fc F3
(hz) (hz)
0.50 0.83 198.51 33.78 52.49
0.60 1.63 100.63 40.54 49.02
0.70 2.58 63.58 47.30 47.78
0.80 3.67 44.63 54.05 48.50
0.90 4.92 33.36 60.81 50.44
1.00 6.30 26.01 67.57 53.12
1.10 7.84 20.92 74.32 56.24
1.20 9.52 17.23 81.08 59.66
1.30 11.34 14.46 87.84 63.27
Eton 11-580 $ 210
Qts = 0.30, Vas = 180.00, Fs = 23.0, Qes = 0.31, EBP = 74.19
I include this because Etons are well regarded, and people will consider
them. This is an 11" driver. For $210 you get F3 > 50 hz. This is obviously
silly. Note the low Q, and consequent high EBP. This was designed for a
vented box, and should work very well in one.
qtc alpha Vb Fc F3
(hz) (hz)
0.50 1.78 101.25 38.33 59.56
0.60 3.00 60.00 46.00 55.62
0.70 4.44 40.50 53.67 54.22
0.80 6.11 29.45 61.33 55.03
0.90 8.00 22.50 69.00 57.23
1.00 10.11 17.80 76.67 60.27
1.10 12.44 14.46 84.33 63.82
1.20 15.00 12.00 92.00 67.70
1.30 17.78 10.12 99.67 71.79
I will try and put something together about what one needs apart from
drivers, and perhaps on the mechanics, but I have now run out of the stuff
I put together slowly over the last few weeks, and it is time to get on
with life. The rest will take time.
----------------------------------------------------------------------
Shankar Bhattacharyya, Information Systems, New York University
sbhattac@rnd.gba.nyu.edu
----------------------------------------------------------------------