|
THE "SCANSPEAKER"
This loudspeaker design is based on
a speaker by Pirmin Schwitter. I used the same drivers but I built
a differnt case and crossovers. I'd like to thank him for his ideas
and advice!
Drivers
As
the name of the "ScanSpeaker" suggests, all the drivers are made by
ScanSpeak. I used the 8545 woofer
(17cm diameter) and the 9900 tweeter
(aka "The Revelator", 28mm dome). These are are expensive, but very
good drivers. But: I have to admit I have used drivers which are easier
to work with...
|
|
The 8545 woofer's
membrane of woofer is made of a mix of carbon-fibres and paper
and it has an inverse dust-cap. It has ScanSpeak's SD-1 drive
to achieve a very linear "motor". The woofer's Thielle-Small
parameters are:
Qts: 0.27
Qes: 0.29
Qms: 3.69
fs: 28 Hz
VAS: 49 l
Mmd: 21 g
B*L: 8.3 Tm
Rms: 1.0 Ns/m
SPL: 88 dB/2.83V/1m
Le: 0.4 mH
Rdc: 5.5 Ohm
The 8545 has the reputation
of very smooth reprouction of voices. This wasn't true in my
speakers at all until I added impedance-compensation for the
impedance-rise caused by the voice-coil. After doing so the
8545 easily outperfoms the Seas Excel magnesium drivers I used
in my "Tube-Speakers" (ok, I didn't try the impedance-compensation
with the "Tube-Speaker" yet...).
The 8545 has a strange "bump" at about 600-700 Hz. It doesen't
look like a membrane-resonance at the waterfall-diagrams so
it can be corrected with an LCR-network.
|
|
|
TheRevelator
is one of the best tweeters I have ever listened to. It has
a special frontplate which is supposed to produce good lobing-behaviour.
I've been told that this frontplate causes a slight resonance
somewhere between 10 kHz and 20 kHz, but I couldn't see it on
my measurements. The dome diameter is 28mm and the resonance-frequency
is at 530 Hz, so the Revelator goes lower than most other tweeters.
The Revelator has NO ferrofluid in the gap to muddle the sound
or to damp the resonance. Therefore it's a good idea to care
about the resonance-peak in the cross-over network.
The Revelator acts a bit like a "diva" when designing the crossover.
I think this is mainly because of the very high resonance-peak,
but it's also because the Revelator's resolution is very good,
so it's really easy to hear differences of differnt capacitor-types
(by the way: good caps don't have to be expensive!!!!!).
|
|
|
The
Case
The case is as important as
the drivers to make a good speaker! An ideal case does not vibrate
at all - the driver's membranes are the only thing that should
move. To get a "dead" case I built the case from 18 mm particle-board
(inside) and 10 mm MDF (outside). I put ceramic tiles on the
inside (like tiles in your bathroom), then a layer of "Hawaphon"
(lead particles). On top of all this I put 10 mm felt and some
wool to damp resonances/reflections inside the case. All in
all the thickness of the walls is about 4 cm (without the felt).
The I used a bass-reflex construction to get enough bass-SPL:
I put two channels in the bottom of the case, "looking" to the
floor (the speaker stands on spikes, about 2 cm above the floor).
The idea of this arrangement is that the midrange "garbage"
coming out of the bass-reflex channels is damped by the carpet.
The internal volume of the case is about 18 Liters. The front
is 18.8 cm wide which is about the diameter of the woofer. It
is important to make the front as narrow as possible to move
the frequencies where reflections from the corners of the case
appear as high as possible (away from the frequency-range whrere
the ear is most sensitive).
Maybe I'll build a new case some day. I might change the following:
-
Use a "sandwich"-case with
two thin MDF-layers (about 5 - 10 mm) and a thicker layer
of sand in between (about 20 mm). Such a case will be as
"dead" as dead can be.
-
Another good case could
be a ceramic pipe like the ones used in the sewer-channels.
These tubes are very cheap or you can even have them for
free from the next constuction-place. Look at Esben Beck's
"Speakers from the Sewer" at the Speaker-Building-Page.
The back of my case is removable
to allow changes at the inside of the case. I experimented with
different stuffings and carton-plates as deflectors to avoid
standing waves. The following is not bad: I put a piece of carton
at the top of the case with some foam glued to it. The angle
between the carton and the front-plate is about 30 degrees so
standing waves along the height are damped. This is very important
because the bass-reflex-channels are just at the other end of
the standing-wave (at the bottom of the case).
|
The
Crossover
The Crossover changes from time to
time... Sometimes I have ideas how to improve the crossover and then
I start soldering, listening, measuring, soldering, listening, soldering,
measuring, listening, soldering - until I get a completely new crossover-network
or else I return to the old schematic.
There are two points which are important with this speaker:
-
The tweeter's impedance-peak is
very high, so it's a good idea to correct for the impedance-peak.
In my crossover this is done by the L-pad before the tweeter.
The main purpose of the L-pad is to match the tweeter's sound-pressure-level
to the woofer, but it flattens the impedance peak very well, too.
-
The woofer's "hump" at about 600-700
Hz must be treated with a LCR-network. The value of the inductor
must be very high (I used 15 mH) because lower values will cause
a dip which is wider than the woofer's peak which we want to cancel.
a) The woofer's cross-over network
|
|
L1 = 1.8 mH
L2 = 15 mH
C1 = 10 µF
C2 = 15 µF
C3 = 3.9 µF
R1 = 5.6 Ohm
R2 = 4.4 Ohm (tot. resistance of LCR including L2)
R3 = 1 Ohm
|
b) The tweeter's cross-over network
|
|
C4 = 3.9 µF
L3 = 0.82 mH
R5 = 10 Ohm
R6 = 8.9 Ohm
|
|