12-to-24V inverter wanted
I am inquiring about a 12V to 24V inverter. I have several
Hella 24V revolving orange lights that I would like to operate on 12V. (M. L.,
via email).
We have not produced a suitable circuit for your
application. It might be easier to change the lamps to run on 12V – the 24V
motors would probably still run at 12V.
Amplifier for sound card
I am looking to build an amplifier to drive a pair of
4W
loudspeakers to reasonable levels from my sound card. I want more level than the
existing multimedia speakers but I don’t need hifi levels. And I would also like
to build the amplifier into the PC itself and power it from the PC as well.
I was thinking of using your "Mini Amplifier For Personal
Stereos" which was pub-lished in the October 1992 issue. Would this be a good
approach? (G. C., via email).
The Mini Stereo Amplifier featured in October 1992
used the TDA2822M dual 1W 8-pin IC amplifier. While it is still a current
design, its output is quite low and probably not enough to suit your needs.
Have a look at the MultiMedia Sound System amplifier featured
in the October 1996 issue. This used three TDA1519A bridge amplifier ICs to
deliver around 9W into two woofers and 1.5W into two tweeters and also included
an electronic crossover. It was designed onto a PC card and was internally
powered by the PC itself. Sounds ideal, eh?
Jumbo clock
is very slow
I’ve built the Jumbo Clock described in the March 1997 issue of
SILICON CHIP. The clock is running very
slow or it doesn’t seems to be running at all. All the LED segments are working,
as well as the hour and minute switches.
Do you have any suggestions? I think it might be related to the
crystal or 4060 chip. (A. W., via email)
The first thing to check is that the colon flashes
at a one-second rate. If so, then the problem will be in IC2 or IC3. Check for
dry joints, solder between IC pins or hairline cracks in the PC tracks. If the
colon does not flash or is very slow, check that the pin 14 output of IC1 does
go high and low at a 2Hz rate. You can use a LED connected in series with a
2.2kW
resistor to check this. Alternatively, use an oscilloscope, logic probe or even
an analog multimeter.
If IC1 is not producing the correct rate, then the problem
could be with the crystal or components connecting to it. Note that it is
unlikely that any of the ICs are faulty. Simply check for other problems either
with the PC board or soldering. Alter-natively, the resistors for the oscillator
could be incorrect or the capacitors the wrong value.
RC speed controller is temperature sensitive
I built the RC Speed Controller described in the May 2000 issue
of SILICON CHIP. I had a problem in that the
ZN409’s 1.5ms reference oscillator varied significantly from day to day and a
hair dryer showed it to be a temperature thing.
The nominal 1.5ms reference, adjusted at "room temperature"
(about 18°C in my garage), varied between 0.5ms and almost 2ms when the PC board
was heated or cooled (between maybe 40°C and 15°C). Even simply placing a finger
on the capacitors caused the rise time to begin to drift.
There’s not much in the way of external parts for the
oscillator, so I replaced the two 0.1mF capacitors with tantalum
types – now it’s rock steady, even at elevated temperatures. Measuring the two
old capacitors out of circuit showed they had a room temperature capacitance of
0.1mF,
which dropped down to under .05mF once warmed.
All components have some kind of temperature coefficient,
though I’ve never seen such a sensitivity before! (The supplied units were those
small blue, non-polarised types, polyester I think?) Perhaps I received part of
a bad batch?
Also, I have a suggestion for people who want to use the
controller with relatively light loads – rather than cut a hole in the case to
fit the loaded PC board, cut off the MOSFET tags. This way, you don’t
need to make any holes in the little case. The tags don’t add much in the way of
heatsinking and with 20A loads or less, the MOSFETs don’t even get warm.
Finally, this mod makes it possible to make the case
water-resistant via a silicone seal around the case lid. (B. L., via
email).
The blue capacitors are monolithic (stacked ceramic)
types intended for supply bypassing on computer and logic boards. They are OK
for that task but as with any high-K ceramic capacitor, they do have quite a
large temperature coefficient and should not be used in any circuit involving
critical time constants.
Improving the Class A amplifier
I have read with interest your recent project on the Ultra-LD
Amplifier (March, May & August 2000) and as I already have a very good class
AB amplifier I am interested in your July 1998 15W class-A design.
Would it be possible to amend this to use the later generation
of transistors used in the output stages of the 100W class AB amplifier? Is it
possible to raise the voltage with these new transistors to increase the output
to about 25W? What changes would be required to increase the output of the 15W
project by using paralleled output transistors? (C. M., via
email).
Since the distortion is already extremely low, there
is little point in going to the more expensive transistors. It would be possible
to increase the supply rails to get more power but this would greatly increase
the overall power dissipation.
We have not done any work along these lines and therefore we
are reluctant to recommend that it be done.
Electric fence
needs extra zap
Some years ago I purchased an Electric Fence kit from Jaycar
that I believe came from your magazine. It used a car ignition coil to produce
the charge. I am having some difficulty gaining any real noticeable "boot" at
the fence and need some pointers regarding what to look for to fix the
problem.
It may be as simple as replacing the coil (which was old) but
that is a $30.00 fix that may be unnecessary. Can you help? (K. M., via
email).
This project was published in the July 1995 issue of
SILICON CHIP. Two errata on this project
have been published. First, we recommended changing the 6.8W resistor to
1.2W to
increase the output to 10kV. If you short out the resistor, the coil will
deliver full output.
Second, use a 500mA fuse as 250mA fuses can have high
resistance.
Shifting a sawtooth oscillator
I’ve made up the Waveform Generator from Jaycar’s Short
Circuits Volume 3 handbook and it works fine. But can I have it go down to say
20Hz (rather than the present 100Hz) so I could use it to test subwoofers? (G.
K., via email).
You can change the minimum frequency to 20Hz by
using a capacitor at pin 2 which is at least five times bigger; try
.056mF
or larger. However that will mean that the maximum frequency will also be five
times lower, at around 4kHz.
The sawtooth waveform is really not suitable for testing any
loudspeaker since it has quite a high harmonic content. You need a low
distortion sinewave oscillator for testing speakers.
VU meter needs auto level control
I have constructed a LED VU level meter for my car stereo,
which is purely for aesthetic purposes; ie, regardless of volume, the display
should work over most of its displayable range.
The only problem is that the stereo does not have a constant
volume output and therefore the input sensitivity of the VU meter must be varied
each time the volume of the stereo is varied. Is there a way to obtain a
constant volume level from the stereo, so that I won’t have to turn two dials
each time I want to change the volume? (J. P., via email).
Short of building our CD Compressor
described in the July 2000 issue, the only way to avoid the need to change the
LED VU setting is to take the signal from across the volume control; ie, you
have to access the signal inside the car stereo.
Rev limiter kills the tacho
I have recently purchased the Rev Limiter kit described in the
April 1999 issue of SILICON CHIP. I installed it as a simple
shift-up light into my Proton M21 coupe. After tapping the signal from the
triggering module, the shift light works but the car’s tachometer went dead.
It seems that no two things can work at the same time. Is there
any way I can get around this? I checked out the impedance from the tacho-meter
to be 150kW and the voltage from the triggering module to be at 5V. Could this be
the problem? (J. M., via email).
The input impedance for the low voltage input for
the Rev Limiter is around 10kW. However, this could be much lower if the
1mF and
.056mF
capacitors plus the 10kW resistor are still on the PC board for the ignition
coil input.
So first remove the above components from the PC board. If you
still have problems running the
tachometer, it is possible to increase the
input impedance by changing the 10kW resistor across ZD2 to 100kW and the 1kW low voltage input
resistor to 10kW.
The .056mF capacitor should be changed to a value of .0056mF (5n6 or 562) or it could be
removed altogether.
