This is only a preview of the December 1992 issue of Silicon Chip. You can view 54 of the 104 pages in the full issue, including the advertisments. For full access, purchase the issue for $10.00 or subscribe for access to the latest issues. Articles in this series:
Articles in this series:
Articles in this series:
Items relevant to "A 2kW 24VDC To 240VAC Sinewave Inverter; Pt.3":
|
Diesel sound simulator
for model railroads
For added realism, why not fit a diesel sound
simulator to each of your model locomotives?
It's great to see and hear them rolling around
the tracks with the sound of the motors rising
and falling as they speed up or slow down.
By DARREN YATES
Well, it was inevitable! After we
featured the SteamSound Simulator
in the April and September 1991 issues of SILICON CHIP, it didn't take ·
long for the diesel fraternity to ask
"why don't you do the same for diesel
locos?"
To those of us who prefer steam
locomotives, it's almost a case of"why
would you bother?" The only problem was that it was the boss who
made the final request and after sidestepping the issue for the last 12
months or more, there are only so
many medical excuses you can come
up with for not doing a project. So I
12
SILICON CHIP
came right out and said it: "What a
silly project idea"!
When I regained consciousness, he
reminded me that there are readers
who believe that diesels are actually
far more interesting than steam locos.
So in the interests of remaining fair,
impartial and unbiased in any shape,
manner or form, I proceeded to do the
project.
We published a diesel sound generator for model locos some time ago,
in the December 1988 issue of SILICON CHIP, but this was intended to be
mounted under the layout and could
be used with any locomotive scale.
This new design can be mounted right
inside the loco, provided the scale is
HO (1:87) or larger.
Mounting the unit inside HO locomotives is made possible by using the
27mm mylar speakers that are now
readily available. In addition, we have
designed the PC board so that it can
be cut in half and folded, if necessary.
The design is primarily intended to
be used with the pulse width modulated speed controllers described in
the April 1988 and April to June 1992
issues of SILICON CHIP, although it can
be adapted to other controllers, as
described later.
Circuit diagram
As you can see from Fig.1, the Diesel Sound Simulator contains only
two !Cs and a small number of other
components. Power is derived from
the rails through diode bridge BR1.
This is a 1B04 4-pin DIP device capable of providing up to 1 amp. The
bridge rectifier feeds diode_Dl and a
100µF capacitor to produce smoothed
4.711
TO
TRACK.,.
660k
16
.022
IC1
4046
ZD1
9.1V
400mW
9
vco
IN FREQ
MAX
MIN
11
12
R1
5.6k
.,.
.,.
R2
470k
.,.
SIG
IN
COMP
IN
VCO
OUT
16
11
14
- - - - - - - - --1 038
3
12
- - - - - --1028
9
4
------CKA
04A
IC2
4015
DB
_____0.1._-+---'l--'-IDA
7
COMP 2
OUT
10
15
15k
5
.,. .,.
8
.,.
.,.
B
100 EQc
VIEWED FROM BELOW
.,.
DIESEL SOUND SIMULA TOR
Fig.1: the circuit of the Diesel Sound Simulator uses a phase lock loop IC
connected to an 8~bit shift register to produce the "rumbling" sound. Zener
diode ZDl, transistor Qt and diode D2 monitor the back-EMF from the loco's
motor to make the pitch of the sound rise and fall with the lo,co's speed.
DC which is then regulated to 12V by
a 78112 3-terminal regulator.
The regulated+ 12V rail powers IC1,
IC2 and a power amplifier consisting
of four transistors (Q1 Q2, Q3 & Q4).
IC1 is a CMOS 4046 phase-locked
loop (PLL) which, together with IC2,
a CMOS 4015 dual 4-bit shift register,
provides most of the circuitry to pro-
811
27mm
+
4.7
16VW+
duce the diesel sound. The frequency
range of IC1's voltage controlled oscillator (VCO) is set by resistor R1 at
pin 11 and resistor R2 at pin 12.
By connecting the two clock inputs
(pins 1 & 9) of IC2 together and feeding the fourth output of register A
into the data in of register B, we've
created an 8-bit shift register. The 6th
and 7th bits are fed into the SIG IN
(pin 14) and COMP IN (pin 3) inputs
ofIC1. As a result, the VCO frequency
is constantly forced to change back
and forth to produce a "rumbling"
sound which is taken from pin 2
(COMP OUT) of IC1 via a 15kQ resistor.
To provide the effect of turbocharger
whine, the VCO output at pin 4 is
mixed with the low frequency output
via a 27kQ resistor. The mixed high
and low frequency signals are then
filtered with a 0.22µF capacitor before being fed to the 4-transistor audio amplifier.
Speed monitoring
The PC board can be separated into two halves which can then be folded
together to fit inside the body of the loco. Alternatively, you can mount the
board inside a following carriage.
As noted above, the frequency of
the Diesel Sound Simulator rises and
falls in proportion to the motor speed
and this gives a very realistic effect.
To achieve this effect, we need to
monitor the speed of the model locomotive's motor.
In practice, we can't monitor the
speed directly but we can monitor the
motor's back-EMF. The is the voltage
produced by the motor which opposes
the current flow through it and this
voltage is directly proportional to the
speed.
But how do we separate the motor
back-EMF from the voltage applied to
it to make it go? You might remember
that we said that this sound simulator
we designed especially to go with the
DECEMBER
1992
13
the amplifier is around 350Hz, although the bass really sounds quite
impressive when fed to a large speaker.
Construction
Take care with the orientation of the two ICs when installing them on the PC
board & don't get them mixed up. You can cut the board in half, if necessary,
after all the parts have been mounted in position.
PWM controllers designed by SILICON
CHIP. These controllers produce a
pulse output at about 200Hz and have
a DC amplitude of about 20V peak.
The power to the locomotive is varied
over · the full range by varying the
width of the pulses.
Now since the pulse output from
the controller is usually off for 50% or
more of the time, that gives us a chance
to monitor the back-EMF of the motor
because the motor naturally keeps
spinning between the pulses and
therefore its back-EMF can be measured at these times. The waveform of
Fig.2, taken from a motor driven by a
PWM controller, demonstrates this
effect. You can see the large pulse
output of the controller and in between each pulse is a spiky waveform
which represents the motor's backEMF. The spikes come from the motor's commutator and brushes by the
way.
We can monitor the back-EMF by
using a circuit which ignores the full
pulse output of the speed controller
but measures the voltage in between
~~
~
pulses. This is done using zener diode ZD1, transistor Ql and diode DZ.
ZD1 conducts whenever the voltage across the rails is more than about
11 V; ie, whenever high voltage pulses
are present. When ZD1 conducts, so
does Ql and therefore the monitoring
circuit ignores these pulses. At other
times, Ql is off and so diode DZ conducts, via the 15kQ resistor, to charge
a 4. 7µF capacitor. This capacitor is
proportional to the motor speed of the
loco and is fed into the VCO IN (pin 9)
ofIC1 to control its frequency. Hence,
increasing the motor speed increases
the VCO frequency and thus the diesel sound rises in pitch.
Audio amplifier
The audio amplifier is based around
transistors Q2-Q5. This is a simple
complementary output stage which
has a gain of about 5. 7, as set by the
470Q and 2.2kQ resistors connected
to the emitter of Q2. The low frequency rolloff is set by the 4.7µF capacitor in series with the 470Q resistor. The actual cutoff frequency for
-FEB-1 QQ ? 1? '.'l5
CH1= 500m\/
DC P*1
· win dow
w 2m s /d
rr
"
~
f'l
"'
14
L
r,
.L
&
fl'' ,, f~
SILICON C H I P
lot,
.
(
~T
fl
T., ,~ l
-
~~~
~
Fig.2: this
oscilloscope
waveform shows
the voltage from a
PWMspeed
controller at a low
speed setting. The
pulses have an
amplitude of
about 17V. The
spiky waveform
between the pulses
is the motor backEMF. The hash on
the waveform is
caused by the
brushes & the
commutator in the
motor.
All of the components, except the
speaker, are installed on a small PC
board measuring 102 x 25mm and
coded 09111921. Regardless of how
you are going to finally install the
unit, we suggest that you build the
board ih its supplied form and get it
going. Then, if it is necessary, you can
cut the board in half to fit it into the
loco.
Before beginning any constructional
work, check the PC board carefully
for any shorts or breaks in the tracks.
If you find any, use a small artwork
knife or a dash of solder to fix the
problem as appropriate.
When inserting components, start
by fitting the two wire links which go
under IC2. Next, solder in the resistors. You can use either the standard
0.25W 1 % metal film resistors or
Philips MRS16T 0.4W types which
have a body that's only 3mm long. If
you are using the standard 7mm body
(0.25W) resistors, some of them will
need to be installed "end on" while
the 3mm body types can be installed
flush with the board. Our photo shows
a mixture of both types ofresistors on
the board.
If you can't make out the colour
bands on the resistors, use your multimeter to make sure of the value.
Next, solder in the diodes and the
bridge rectifier. The "+" pin on the
bridge rectifier package should be.adjacent to the end of the PC board, as
shown on the wiring diagram ofFig.3.
Make sure that the diodes and electrolytic capacitors are installed with
correct polarity.
Take care when soldering in the
transistors to make sure that they are
correctly oriented. Q3 and Q4 have
their faces butted together, while. Q2
and Q5 face away from each other.
The two CMOS ICs go in last - make
sure they are correctly oriented.
Testing
To test the Diesel Sound Simulator,
you'll need to hook it up to your layout with a loco ready to run on the
track. This is because the Diesel Sound
Simulator relies on the back-EMF produced by the motor to get the correct
speed information. We also suggest
that you don't use the specified 27mm
SIAS
(Voc)
·i.
."'
I
!
> :
JU? • •I r"!
! ''
JI
i
l
!·!J'"i!_'
GEN ■ ii
~ M E N U S :__ = J
Power analyser
Fully autoranging
for voltage, current
and frequency.
700VRMS
20ARMS
DC-20KHz
0 Accurate measurements to 0.1 %
O Accurate with distorted waveforms
O Accurate with all power factors
O Measures any short term events
from inrush current to burst power
O PC Software and interfaces
provided with most instruments.
I
•
: ,
ii
•
he "smart" digital technology
in these Voltech analysers and
testers, results in measurement
made easy. A touch of a button
gives direct readout of true power,
harmonics and a host of other
useful measurements for:
o Switch mode power supplies
o Uninterruptable Power Supplies
o Lighting Ballasts □ AC Motors
o Electronic Variable Speed
Motor Drive (PWM Drive)
o Generators o Transformers
o Aircraft Electrical Systems
T
Harmonics (V+A}
to99th.
1400VRMS
30A RMS (200Apk)
DC-200KHz
D
Power analyser
Automatic
Transformer
Tester
700VRMS
20ARMS
DC-50KHz
I
Universal Three
Phase Analyser
o PM1200
Fully autoranging
for voltage, current
and frequency,
with interface.
j
A13500
Multi function, built
in printer, Hi-pot test
up to SkVac.
1MHz testing
Safety interlocked
Up to 800% faster.
PICTURED ABOVE
D TF2(XX)
Frequency Response Analyser
Measures Gain; Phase & Harmonics
(to 39th harmonic) 3 microV-300VRMS
0.0lHz-l00KHz with built-in generator.
WESTINGHOUSE INDUSTRIAL
PRODUCTS: Contact John
Thompson, Phone (03) 391 1300
Fax (03) 391 6607. 59 Stephenson
St., Spotswood, Victoria 3015
<at>.
'C!:~~![~9h2~se
INDUSTRIAL PRODUCTS
ro
miniature speaker at tnis stage - it is
rather fragile and should be handled
as little as possible. Use a conventional larger speaker during testing.
When you first apply power, and
with the speed controller set for minimum output, the Diesel Sound Simulator should be in "idle" mode; ie,
simulating the effect of the motor at
idle. As the speed setting is increased
to make the loco move, the pitch of
the diesel sound should begin to rise.
Wind up the speed controller to
maximum and observe that the diesel
sound is still realistic. It should sound
like a diesel running at a good pace,
not like a motorbike doing 8000RPM.
If the maximum diesel pitch is too
high, you can reduce it by increasing
the value ofRl. Similarly, if the maximum pitch is too low (unlikely), you
can increase it by reducing the value
ofRl.
Similarly, if the idle speed is too
CAPACITOR CODES
0
0
0
0
Value
IEC Code
EIA Code
224
104
223
SPEAKER
Fig.3: here's
how to install
the parts on the
PC board. Make
sure that you
install the wire
links before
fitting IC2.
high, you can reduce it by increasing
the value of R2.
If the PC board does not work, the
most likely reason will be shorts between tracks caused by tiny solder
splashes. Careful inspection of the
board under a good light with a magnifying glass will locate these problems.
Mounting it in a loco
As noted above, the Diesel Sound
Simulator is intended for locomotives
in HO, 00 or larger scales. However,
some HO diesel locos are just too small
or are so crammed inside that it will
not be possible to install the sound
components in them. In these cases,
the only practical approach will be to
install the sound components inside
a dummy loco (ie, without a motor) or
inside a carriage or wagon.
Either way, you must arrange to
pick up power from the rails. This
can be done from the loco's wheels
and then via a miniature 2-way cable
and connector. Alternatively, you will
need to pick up power from the rails
via collectors on the wheels of the
dummy loco or wagon.
On the hand, if you can possibly
manage to fit the sound components
inside the loco itself, the effect will be
even more realistic. You can then have
PARTS LIST
1 PC board, code 09111921, 102
x25mm
1 27mm 8Q Mylar loudspeaker
Semiconductors
1 4046 phase-locked loop (IC1)
1 4015 dual 4-bit shift register
(IC2)
1 78L 12 + 12V regulator
2 BC548 NPN transistors
(01 ,02)
1 BC558 PNP transistor (03)
1 BC337 NPN transistor (04)
1 BC327 PNP transistor (05)
1 9.1V 400mW zener diode (201)
1 1804 bridge rectifier (BR1)
1 1N4004 rectifier diode (01 )
1 1N914 signal diode (02)
Capacitors
1 100µF 35VW electrolytic
1 100µF 16VW electrolytic
1 10µF 16VW electrolytic
2 4. 7µF 25VW electrolytic
1 0.22µF 63VW MKT polyester
1 0.1 µF 63VW MKT polyester
1 .022µF 63VW MKT polyester
Resistors (0.25W, 1%)
1 680kQ
1 10kQ
1 470kQ
1 5.6kQ
2 100kQ
1 2.2kQ
2 47kQ
1 1kQ
1 27kQ
1 470Q
2 15kQ
1 4.7Q
0.22µF
0.1µF
.022µF
220n
100n
22n
0
0
0
0
0
0
0
0
0
0
0
0
0
No.
Value
4-Band Code (1%)
5-Band Code (1%)
1
680kQ
470kQ
100kQ
47kQ
27kQ
15kQ
10kQ
5.6kQ
2.2kQ
1kQ
470Q
4.7Q
blue grey yellow brown
yellow violet yellow brown
brown black yellow brown
yellow violet orange brown
red violet orange brown
brown green orange brown
brown black orange brown
green blue red brown
red red red brown
brown black red brown
yellow violet brown brown
yellow violet gold brown
blue grey black orange brown
yellow violet black orange brown
brown black black orange brown
yellow violet black red brown
red violet black red brown
brown green black red brown
brown black black red brown
green blue black brown brown
red red black brown brown
brown black black brown brown
yellow violet black black brown
yellow violet black silver brown
diesel sound all the time, even when
the loco is running "light engine"
(that's railway speak for when it's running by itself).
RESISTOR COLOUR CODES
rn
2
2
1
2
SILICON CHIP
The prototype fitted neatly inside a HO scale Victorian Railways V/Line G Class
Mk2 loco but only after we had relocated the internal wiring for the lighting. To
obtain adequate sound output, the loudspeaker should be fitted inside a 30mm
length of tubing (not shown here).
As a practical exercise, we mounted
the Diesel Sound Simulator inside a
HO scale Victorian Railways V/Line
G Class Mk2 loco. This model is quite
roomy inside although it still requires
some internal surgery to make everything fit.
First, it was necessary to replace
the steel weights with lead. Health
regulations prevent toy and model
suppliers from selling products containing lead but there is nothing to
stop model enthusiasts from putting
in lead after purchase.
It was also necessary to relocate the
diodes and the wiring associated with
the directional lighting. These diodes
were originally terminated to solder
lugs under screws which retained the
steel weights. We relocated the diodes to a small piece of Veroboard
which was then mounted inside the
roof above one of the cabins.
As noted above, the PC board has
been designed to be cut in half if
necessary. This is best done with a
miniature high speed grinding wheel
fitted to a Dremel Moto-tool or an
Arlec Mini-tool. This enables you to
fold the board in half, with both halves
connected together with very short
lengths of miniature hook-up wire.
The bisected board is then squeezed
into the cleared space above the lead
weights. We put insulating tape over
the lead weight to prevent shorts to
the board.
Finally, we fixed the speaker inside
the roof of the loco adjacent to louvres
which conveniently let the sound
through. All that you can see in the
relevant photos.
How did it sound? Not too brilliant
- at least not initially. While it was
satisfactory at idle, once the loco got
going the sound of the motor and following rolling stock tended to drown
the sound from the speaker. It really
needed a baffle. Hence, after the photos were taken, we mounted the
speaker inside a short length of tubing (anything over 30mm long does
the job) which is mounted above the
unpowered bogie. This can only be
done if the top half of the bogie housing (essentially it is a dummy motor)
is removed to make space for the tube.
That done , the sound output is
much louder and is quite satisfactory.
Under the layout mounting
While the circuit was specifically
designed to run with our PWM controller designs, it is possible to use it
with other controllers which simply
vary smoothed cir unfiltered DC. An
example is the simple controller featured in the November 1990 issue of
SILICON CHIP.
In these cases though, it is not possible to install the system inside the
locomotive. Instead it will have to be
installed underneath the layout. You
can then use a much larger loudspeaker and obtain even more sound
output.
The circuit will require a few modifications, however. These are as follows:
(1). Since the DC from the rails can
vary from Oto 12V or thereabouts, the
circuit will have to be run from an
external power supply which can be a
12V DC plugpack.
(2). The anode of diode Dl will need
to be disconnected from the bridge
rectifier BR1 and connected instead
to the positive lead from the external
DC power supply.
(3). You will probably also have to
change the value ofRl to obtain satisfactory operation. We found that 4 7kQ
was suitable.
Apart from these modifications, the
Diesel Sound Simulator is connected
as in the loco except that its inputs
are wired directly across the track (ie,
across the output of your speed controller).
Acknowledgement
Fig.4: this is the full-size etching pattern for the PC board.
Our thanks to Powerline Models
Pty Ltd who kindly supplied the V/
Line G Class Mk2 locomotive used in
this article. They can be contacted at
PO Box 264, Penrith NSW 2750. Phone
(047) 396 204.
SC
D ECE M BE R
1992
17
|