Wire in
Europe has been sold in Metric form for the last 20+ years.
Wire Gauge
AWG (American
Wire Gauge) or B & S
SWG (Standard
Wire Gauge - UK)
Inches
mm
Inches
mm
Nearest
stock Falcon metric sizes
10
0.1019
2.59
0.1280
3.25
12
0.0808
2.05
0.1040
2.64
14
0.0641
1.63
0.0800
2.03
16
0.0508
1.29
0.0640
1.63
18
0.0403
1.02
0.0480
1.22
1.25mm
19
0.359
0.912
0.040
1.016
1.00mm
20
0.0320
0.813
0.0360
0.914
0.9mm
21
0.0285
0.724
0.0320
0.813
0.8mm
22
0.0253
0.643
0.0280
0.711
0.71mm
24
0.0201
0.511
0.0220
0.559
0.56mm
26
0.0159
0.404
0.0180
0.457
0.45mm
28
0.0126
0.320
0.0148
0.376
0.355mm
30
0.0100
0.254
0.0124
0.315
32
0.0080
0.203
0.0108
0.274
34
0.0063
0.160
0.0092
0.234
36
0.0051
0.127
0.0076
0.193
38
0.0040
0.102
0.0060
0.152
Note
for metric wire the 'usual' sizes in the range we wind are:- 0.56mm, 0.71mm,
0.8mm, 0.9mm, 1.0mm & 1.25mm
*
* * * * * * * * * * * * * * * * * * *
Special
Offers
INDUCTORS
Value
Type
Wire
Approx Quantity
Cost 1-9
Cost
10-50
2.5mH
SSP
1mm
50
£4.94
-10%
3.5mH
SSP
1mm
50
£5.14
-10%
7.5mH
SSP
1mm
50
£6.37
-10%
4.7mH
SP
1mm
100+
£5.05
-10%
0.6mH
SP
1mm
70
£2.66
-10%
9.0mH
SP
0.8mm
600
£4.28
-10%
0.93mH
HP
1.25mm
150
£2.87
-10%
8.0mH
HP
0.56mm
350
£2.00
-10%
4.5mH
HP
0.8mm
200
£2.88
-10%
All
prices include VAT, carriage as normal (See Page1). Quote 'Special Offer' on your
order.
* * * * * * * * * * * * * * *
* *
Tweeter
Sensitivities
Tweeter
sensitivities/resistance/impedance seem to cause much confusion. It is actually
quite simple; it really does not matter….. 'As long as the resistance is
sensible'. It is better to work (with tweeters) by using the Rdc. Nominal
8 Ohm
impedance tweeters usually have a 6 Ohm
Rdc. and 15 Ohm,
a 12
Ohm
Rdc. Some manufacturers with 6 Ohm
Rdc units call them 6
Ohm!
Hence some of the confusion. So for a nominal 8 Ohm
system, a tweeter with an Rdc of 4 to 12 Ohm
would be suitable as long as its sensitivity (output), with 2.83 volts applied,
is equal to or greater than that of the LF unit with the same applied voltage.
Nominal
sensitivities
Audax
TWO25AO 90 or 91dB depending upon source Vifa D27 90 @ 1w/1m, & 91 @
2.83v/1m - greater because it has a Rdc of 6 Ohm
Sensitivities
vary due to: 1. The response not being flat over its usable range. The range
over which it is measured will determine the value. If white noise is used, a
peak in the response, quite common on many dome tweeters, could give an inflated
reading. A sensible sensitivity reading should be from 4-10kHz. 2. There are
two normal methods of determining sensitivity: a)
1w/1metre. This is one watt into the actual unit, i.e. voltage squared / impedance.
b) 2.83v/metre. This is measured using 2.83v (equiv. to
1 watt into 8 Ohm)
and if the impedance is lower than 8 Ohm
it will give a higher reading as more current will be drawn. Conversely lower
if a higher impedance. 3. The sensitivities will also vary when a network
is used as this also affects the impedance etc. What actually matters then is
the output of the tweeter over the range 3 - 12kHz through the network to be used
with 2.83v applied. This is because the same voltage is applied to each unit in
the system. 4. Although the impedance can be usually be 'ignored' if the
sensitivity is quoted with 2.83 volts applied, it must be taken into account in
the design of such things as crossovers and Zobel networks.
We
have a sheet showing the various fitting dimensions for a range of tweeters,
both current and discontinued (for fixing size reference). TWEETER
fitting sizes.pdf (9kBytes)
*
* * * * * * * * * * * * * * * * *
UNIT‘SWAP’
Section
This will be converted to a separate
page if it takes off
Please let us know of any problems
etc.Falcon takes no responsibility
about these items; this is only a contact/swap board.
DO
LET ME KNOW WHEN ANY ARE SOLD SO I CAN TAKE THEM OFF.
We will however withdraw any mention
if problems are reported. Items are unknown to FALCON and items
are at Buyers/Sellers risk Send details by e-mail and we will put them on
the page MUSTS:- E-mail address, Postal address (Falcon use), Unit
description and SP number. e.g. B200SP
1014 Age - usually a code on the unit lege.g. G1378, Usage and Price if selling or buying. Other if swap. Contact
is direct to ‘advertiser’.
* * * * * * * * * * * * * * * * * *
* * * * * *
NEW
No.
Mfs.
Unit
Use
Condition
2
Kelly
Ribbon Horn
?
Good
4
GEC
Metal
Cone Unit
?
Good
Contact weegenaar@clear.net.nz
New
Zealand Notified 15/3/2008
* * * * * * * * * * * * * * * * * *
* * * * * *
WANTED
No.
Mfs.
Unit
SP Number
Use
Condition
1 or 2
KEF
B200
SP 1039 (104 type)
Good
Offer price £10 each
Contact W Hateley wilfros@wanadoo.fr
- Notified 1/2/04 Beds.
DO
LET ME KNOW WHEN ANY ARE SOLD SO I CAN TAKE THEM OFF.
IT WILL ALSO
SAVE YOU THE HASSLE OF UNWANTED E-MAILS !
2
way crossover approx 200-250Hz. 7 components. 2 x 4.7mH SP/LL 1mm wire & HP/LL
inductor. On PCB 10. Works with bookshelf
speakers coupled with 2 LF unit systems
£14.00 + p&p £7.00 + p&p
LINN
Networks
(Linn) Special PCB (Isobarik's
B139B/B110A/ScanSpeak D2008 or Hiquphon) Size 126 mm x 165mm x 40mm These
are suitable up to S/N 6699 / 6700 in
1988.
(Note
that Linn used an earlier numbering system 10,000 to 11,000+ from 1976 to 1980)
After this LINN changed to a high slope 4 Ohm network, which was fitted into the
stand.
These
Crossovers are now sold out.
Each £25.00 + £6.00 p&p
per pair. 8 ohm hence 2 networks
needed per system, (One per set of units). Hence 4 per stereo system pair.
* * * * * *
* * * * * * * * * * * *
Below is the Tips page from the Falcon
booklet, ‘"A Practical Guide to Crossover&
System Design"’
Available from Falcon, 28+ pages,
A5 format @ £3.00 +p&p
Use this in conjunction with our other
booklet "Active Loudspeaker Crossover Filters and Sub-Bass"
This also includes details on types of loading and cabinet volume calculations.
Also available from Falcon, 32 pages,
A5 format @ £3.00 +p&p
* * * * * * * * * * * * * * *
* *
Tips
·Make
a predesigned kit.Even if you want
to play, use this as a starting point.The designer has probably got lots of test equipment and
many years of experience.
·Do
not expect a theoretically designed network to be anything like optimum.It will work - probably, but the performance could
be the difference between a CITROEN 2CV and a Formula 1.Both work,but. . . . . . .
·Make a quick and unfinished
‘pre-production’ test box.
·When initially designing the
system, use the Small-Thiele parameters of the chosen bass unit to determine the
nett volume of the enclosure required.Additional volume should be added
for cabinet wall damping, bracing shelves and battening etc.Then add about 10% (minimum) to the
calculated volume.The volume can
be reduced by adding wooden blocks or bricks, but you cannot easily increase the
box size.
·When working out the box size,
do take into account the Rdc of the inductor(s) in series with the bass unit.They can significantly alter the unit ‘Qtc’.Start with 0.3 Ohm
- 0.5 Ohm or 0.6 Ohm
to 1.0 Ohm if using Air Cored
See Fig.6 p.11
·Determine the frontal width
required - this is the main diffraction-causing dimension and should be kept constant
during the design process. [1]
Is it wide enough for the bass unit to fit in?The height is next determined. This is normally a function
of aesthetics (golden ratios) and whether the cabinet is to be stand or floor
mounted.Test that the units plus
spacing will fit in the height chosen.The depth is then easily calculated.This dimension is the one to change on your final cabinets as it has minimal
effect on the response curve.
·If making a column speaker
it is a good idea to partition a small ‘box’ on the bottom and fill
this with washed, dry sharp sand. Use pointed cones for good LF system
anchorage.
·If modifying an existing system
do not put the MF &/or H.F. units into a narrower box or baffle as this will
have major consequences on the diffraction effects.The crossover will have to be redesigned.
Probably an increase in the value of the LF section input inductor.
·Radius the frontal edges of
the cabinet (quadrant or bevelled edges), and any reflex ports.
·Always flush mount the tweeter
and preferably the midrange/bass-mid.
·Use banked capacitors which
allows variations to be made easily.For a theoretical 5mF use a 3.3mF and some smaller values, such as 1mF
and 2 x 0.47mF and/or 3 x 0.33mF.A few
0.1mF
capacitors would also be useful. These can be connected in parallel to make up
the required value in small steps.This
need not be too expensive as initial tests can be made using Polyester.Do however use 5% components.
·As a general range of capacitors
the following give 1 - 10 mF in 1 mF steps:-
5mF,2mF,2mF,1mF.
For 0.1mF
- 1mF
use:-
0.47mF,0.22mF,0.22mF,0.1mF.This does
not give the exact 0.1mF steps, but near
enough.You can of course use 10
x 0.1mFand 10 x 1mF
·Do not connect inductors in
series to make larger values as this doubles the Rdc. An inductor wound to double the value
will, in theory, have an Rdc 1.41 times the original value.Due to the extra turns being of larger
diameter than the original the factor tends to be 1.5x.Still better than 2x!Tapped inductors are available, and
although a bit expensive they are cheaper and more convenient than having multiple
inductors.
·Some books suggest deducting
the value of the speaker inductance from the series inductor.DO NOT DO THIS.It may make the impedance flatter,
but will mess up the response shape.If
anything the value needs to be added to counter the unit impedance rise!
·Resistive attenuation can
be used on both midrange and tweeter units, but not on bass units (see Unit ‘Q’ in formulae).It is however sometimes possible to
shunt the bass unit with a high-power resistor.33 Ohm to 15 Ohm are suggested trial values.As most midrange / tweeter units are more sensitive than
bass units, this is not normally necessary.
·Computer programs are useful
in determining box sizes and loading.The formulae are quite accurate, or at least as accurate
as the Small-Thiele parameters used, although they tend to be a little optimistic
by a dB. or so.Very useful in trying
out differing types of loading and checking power handling and excursion limitations.
·If you like playing and you
are not just making a ‘one off’ it could be worth buying a small FFT
test set-up.AIRR with a SoundBlaster 16 bit duplex card, a
microphone capsule and a preamp will give frequency response, phase and waterfall
curves.Total price excluding the
sound card is less than £120.Some saving could be made by building your own preamp.
·Whilst ‘playing’
do use PolySwitches.These devices
have a very low series resistance in their cold condition, but become almost open
circuit when the series current heats them above a certain temperature.Available in a range of currents.Suggest:
0.5A for tweeters;
0.9A for bass/mid & midrange;
1.6A for bass.
Very useful for normal use, in particular for tweeters. Much cheaper than
units.
·Zobel networks.
Starting
points for L.F. units
10 Ohm
and 10 – 16mFd.
Starting
points for L.F./M.F. units
10 Ohm
and 10mfd.
Starting
points for H.F. units
8.2 Ohm
or 10 Ohm and 2.2mFd.
·Connecting units ‘in
phase’
The
phase difference between the 18dB. slopes is 90deg. and therefore there is no
absolutely correct polarity connection; the acoustic phase shift in the loudspeaker
unit determining the actual polarity connection.
An easy and quite
accurate method of checking for the 'best-fit' phase is to connect only the two
units you wish to test, the other(s) should be disconnected and the network loaded
with resistors.#These need not be exactly the correct resistance, for an
8 Ohm speaker the resistor should be between
6.8 Ohm and 9.1 Ohm with a rating of at least 5 watts.
A
white noise source-pickup input with the gain near maximum(don't touch the pickup or arm!) - de-tuned FM tuner etc.Move your head between the axes of
the two speaker units at a distance of 18" (½ metre).For correct phase the sound should
gradually increase from the axis of one to a maximum midway between the two and
decrease down to the axis of the other unit.Change the polarity of the 'uncertain' unit and try again.It is usually very obvious which polarity connection gives the correct
phase at the crossover point.Start
with the bass unit as the reference and determine the polarity connection of the
midrange or tweeter, disconnecting the tweeter on a 3 way.Once this has been done, disconnect the bass unit for a
3 way and connect the tweeter.Alter
this to give the correct phase match with the midrange.
Whilst
this method is not infallible, I have found it 90+% accurate and far more accurate
than theory
#This does not apply to series networks as the unit impedance
interacts with all the sections.In this case the unwanted unit should be covered over to
attenuate its output as much as possible.
[1] If
building a kit into an existing box the width is important. The crossover was
designed for the width of the designed cabinet. The Small-Thiele parameters
should then be used to check the suitability of the cabinet volume.
These
Crossovers are now sold out.