This is only a preview of the June 1990 issue of Silicon Chip. You can view 49 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. Items relevant to "Universal Stereo Preamplifier":
Articles in this series:
Articles in this series:
Articles in this series:
|
You'll get a buzz out of this one
Many years ago, as a factory hand I learned the
difference between field faults and production line
faults. Faults in the field tended to he simpler; at
least one knew that the set had once worked
correctly. But production line faults - well, it was
anybody's guess as to what could have happened.
Although this column will reach
my readers a good deal later, I am
writing it in the month of March.
More precisely, it is March 17 and
that is very appropriate for, as all
good Wearers of the Green will
know, March 17 is St Patrick's Day.
And why is that so appropriate?
Because this story deals with one of
St Patrick's countrymen; one we all
know and love to hate. Yes, that's
right - Murphy!
This is not my first story about
Murphy's antics of course; there
have been plenty of those. But this
was a particularly vicious attack by
that worthy which created a very
weird and potentially frustrating
situation.
It all started when a dealer colleague for whom I do warranty
repairs asked me to stop by his
showroom and check a new set. In
greater detail, it was a National
Panasonic model TC 2969 which
seemed to have excessive hum in
the sound channel. And, as an
afterthought, he added, "It'll probably only take you a few minutes
to fix". He was probably being a
mite sarcastic, but the remark had
more sting in it than either of us
realised at the time.
The TC 2969 is a large screen
model (68cm) and features a stereo
sound system with external speaker
outlets. It has been on the market
for about two years now; time
enough for most of the inevitable
bugs to have surfaced and been duly swatted.
In fact, it has been remarkably
free of such beasties, so it came as
50
SILICON CHIP
something of a surprise when a new
set, just out of the carton, exhibited
such an obvious fault. Still, these
things do happen.
So it was that I duly faced up to
the offending set in the showroom.
My colleague switched it on and I
flicked it through several channels,
expecting an obvious fault. Truth to
tell, I was not immediately aware of
any problem and it was only when I
paid particular attention to the
sound that I detected the background hum. But while it was more
than there should have been in a set
of this quality, it was not at
.
\
~
7)1~
.,111,,,
.
r:
o
d
o
I
--
W~'( ~ M~-1'1
STUFF UP
A, VJ01<.\(~
anything like the level my colleague
had implied when he first called
me.
When I raised this point, my colleague looked puzzled but was adamant that the hum was "much
worse than that" when he first
switched the set on. So that was
puzzle number one; the fact that the
fault seemed to be variable. Or was
it? In a showroom situation, where
ambient sound can vary considerably, subjective assessment
might be suspect. But for now, I had
to assume the worst.
Puzzle number two was something I couldn't quite put my finger
on but the nature of the offending
sound was worrying me.
Before I could think much more
about that, my colleague introduced
a third factor - a second set with
the same fault. It had been sold to a
customer, delivered in something of
a hurry late one afternoon and on
the basis of a quick check, had
seemed OK. However, the customer
soon complained of excessive hum
level and this had subsequently
been confirmed by my colleague.
So not only did we have a strange
(variable?) fault but we had it in
two sets. And this changed the
scene completely. Instead of a oneoff component fault, I was now
thinking in terms of a design or production line fault. But why should
such a fault rear its ugly head at
this time, after nearly two years?
Of more immediate concern was
what to do about the customer's set.
My colleague had left the set with
him on the basis that I could fix it,
as he put it, "in a couple of
minutes". But there was no way I
was going to tackle a fault like that
out of the workshop, with the real
possibility that I would need help
from the National service department.
So my colleague decided to simply replace the customer's set. And
this was another surprise; he had a
third set on hand which was totally
free of the problem. Anyway, the
sets were duly swapped and the
customer's set was delivered to my
workshop.
Frame buzz
In the meantime
in fact, as
soon as I left the shop - I realised
what was nagging me about the
hum. Although my colleague had
said "hum" - and I had accepted
the term - it wasn't hum in the
strict audio sense. Hum in an audio
amplifier can be caused by induction into improperly shielded leads
or by earth loops, etc. This hum will
be at 50Hz and sounds low pitched
and reasonably pure. Alternatively,
it can be due to faulty power supply
filtering in which case it will be at
100Hz and sounds higher pitched
and rather rough.
But it came to me in a flash that
this was neither. It was low pitched
- 50Hz - but harsh. That meant
only one thing; it was frame buzz. I
mentally kicked myself for being
puzzled but frame buzz was the last
thing I had expected to encounter in
this set - even as a fault.
A quick check on the customer's
set confirmed both the nature of the
buzz and that it was quite bad. I
fished out the manual as a first step
and in the process, came across a
number of modification sheets
which the firm had issued. And I
must give National full marks on
this score; they do make every effort to keep their servicemen up to
date on any problems or modifications as they occur (unlike some
other companies, which practise
the mushroom treatment).
Anyway, one of these mod. sheets
caught my eye. It didn't nominate
this particular set but I recognised
the two boards involved; the " E"
board and the "B" board. Both are
used in this set and both needed
modification, although no reasons
were given. The "E" board mod. involved the audio output IC and on
the "B" board, the IC for the earlier
stages. And part of the mod. involv-
ed fitting extra "earth" leads between sections.
In the light of my problem, all this
seemed to make sense. There
wasn't a lot involved and I had the
whole job done in about 20 minutes.
But it didn't work; the buzz level
was exactly the same as before. I
was back to square one - and the
realisation that I needed help.
I rang one of my contacts in the
National service department,
outlined the nature of the problem,
and described the changes I had
made. His response to this latter
point was not what I had expected.
In mock sarcasm he told me to go
back and take another look at the
mod. sheet, because the mods. were
not intended for that model set. In
greater detail, he explained that
they were intended for an M14
chassis whereas mine was an M15
chassis.
I was well aware of the number
of my chassis and said so, going on
to point out that there was no
chassis identification on the mod.
sheet. Initially, he wouldn't buy
this, insisting that there was a
message at the top of the page
clearly stating that these mods.
were intended for an M14 chassis
only.
I eventually convinced him that
there was no such message on my
sheet. How it had been omitted, and
from how many sheets, remains a
mystery. But it was strike one to
Murphy.
The upshot of all this was (1) that
I should remove the mods. as they
had nothing to do with the problem;
and (2) that he was well aware of
the frame buzz problem and knew
how to correct it. The cure was simple but the story behind it was quite
complex. If I cared to listen he
would fill me in. Always anxious to
know the reason why, I said, "Go
ahead" .
Supply circuitry
Fig.1 is the relevant portion of
the circuit showing, in particular,
the network providing the voltage
regulating function for the switchmode power supply. This starts
at IC801 at the bottom centre of the
diagram. As shown, this IC consists
of a transistor, a zener diode, and
three resistors. It is connected to
the 113V rail (pin 1) and a proporJUNE 1990
51
IL.!!£.J
T801
~
p
For 21rnch
L807-"- ET549K512A
TSC937 ~ For 26/29rnch
EXCELDR35C ET849K511A
R813
330K
Jp
-
&
C849
0.01
500V
D806
0-01
TVBESIC
SOOY
®
D827
IIA649RL
Jl18
C818
680P
2KV
I 86
8V~
C814
LBIO ,.&.
TSC925-4
EXCELSA35B
5V
'I'
TBC937~
EXCELDR35C
IW
[Hy]
,
r
0
1.0
1
8
::==================-,---:----,
[..806
& R837
I
113V
B5
r 62
ft,
I 64
C815
330P2KV
L803
,"TSC925-4~
EXCELSA35B
ER 20V
\
+113V
E
C842
E
22P
0850
aRZC10DK621
25V
-~-
[!ill
R35C
825 I
7~P
I
ONI.Y FOR
R8I9
S.8K
RB40
180K
Jt V
0815
El/2
r-
R823
}';II
2SD965
0
0825
IIA4240
-------
R835
0-33
0808
1-5
2W
:Z.'f-V
0806
2S0965R
i
I
=
--------------
--------l
&
CA-2969
- ___ ,
I
R82Q
270
IC801 ~-
s1854LBM-2:
- - - __ .J
0802
25C1685/0R5
2SC331\A/OR5
Fig.1: the power supply circuit for the National TC 2969 TV receiver. IC801 is at bottom centre, with the optocoupler, Q803 and Q806 to its left. The additional diode, drawn in, is at the top right corner, near D827.
tion of this voltage is fed to the transistor's base.
The zener diode in the emitter
circuit provides a voltage reference
against which the base voltage is
compared. The collector of this
transistor (pin 2) is connected to the
18V rail (right of the diagram, half
way up) via D823, R828 and the
LED in opto-coupler D826 (ON3105R).
The transistor in the opto-coupler
is coupled to Q803, which is coupled to Q806, which is coupled to
Q807. And Q807 is the " regulator"
or "switching" transistor in the
switchmode power supply. (Its collector connects to the OUT pin of the
52
SILICON CHIP
switchmode transformer, T801).
In normal operation, the IC801
LED will assume a brightness level
determined by the actual voltage on
the 113V rail. If this rises above
113V, the LED brightness will rise
and its associated transistor will
draw more current. This will increase the current through Q803
and in turn, the current through
Q806. And this will pull down the
base of Q807 and thus bring the
113V rail (and others) back to the
correct value.
(This network is also controlled
by Q802 which, in turn, is controlled by the remote control receiver to
turn the set on or off. But this a
secondary function and is dormant
at all other times).
Well, if you've digested all that,
let's go back to that 18V rail. As
well as supplying IC801, this rail
goes off up the diagram to perform
another function. That other function is to power the audio output
chip. And, as already mentioned,
this set features outlets for external
speakers.
So what happens if John Customer fouls up the connections and puts
a short across the speaker terminals? Well, for starters it takes
out the audio chip, which is bad
enough, but there's worse to come.
When the audio chip fails, it shorts
and takes out safety resistor R827
which shuts down the 18V rail.
And that's when the fun starts.
Since the 18V rail is involved in
controlling Q807, loss of this
voltage causes Q807 to turn hard
on, with drastic consequences. The
resultant over-voltage can take out
transistors and ICs left, right, and
centre. In a couple of cases, sets
had been returned to the National
service department which required
extensive work to salvage them. Not
all sets so abused were as seriously
affected as this but this is what can
happen.
(Yes, the set has protective
devices and circuits designed to
prevent such mayhem but it appears that they do not always work,
or work fast enough}.
Back to the buzz
And what has all this to do with
the frame buzz problem? In case
readers imagine that I've wandered
off along some irrelevant sidetrack,
let me assure you that such is not
the case. Be patient and all will be
made clear.
Because of the problem just
of the diagram, above Q802} and
the idea was to connect this to the
junction of D823 and R828 via a
diode. Normally, this would have no
effect because the diode would be
reverse biased. But if the 18V rail
failed, the diode would then be forward biased and so the 14V rail
would now feed the LED in the optocoupler.
It was not intended to keep the
set working but it would maintain
the opto-coupler circuit sufficiently
to prevent Q807 from running
amok. The scheme worked very
well and had the advantage of being simple to implement; just one inexpensive diode tacked into the
circuit.
TETIA TV TIP
Pye T29 chassis
Symptom : Almost no vertical scan.
Picture' reduced to a 2. 5cm band
across the centre of the screen .
Circuit diagram gives no useful
voltages but those measured appear to be reasonable .
Cure: 0504 (BC638) faulty. The
transistor checks OK both in and
out of circuit but will not amplify
signals applied to its base.
TETIA TV Tip is supplied by the
Tasmanian branch of The Electronic Technicians' Institute of
Australia. Contact Jim Lawler, 16
Adina St, Geilston Bay, Tasmania
7015.
Strike two
outlined, someone hit on a clever
circuit modification which would
protect the rest of the circuit should
the 18V rail be knocked out. I'm not
sure whether this was instigated by
the parent company in Japan or
whether it was a local development
but it was very simple.
There is a 14V rail in the set (top
Enter Murphy for strike two, in
the form of a wiring hand who
couldn't tell one end of a diode from
the other and wired a number of
sets with the diode reverse
connected.
And that was what created the
frame buzz. The fault was not
discovered until some sets had gone
into the field and there was no easy
TV TEST EQUIPMENT
PCB POWER TRANSFO RMERS
(AUSTRALIAN MADE)
SHORTED TURNS TESTER
Built in meter to check EHT transformers including
split diode type , yokes and drive transformers.
$78.00 + $3.00 p&p
HI-VOLTAGE PROBE
Built-in meter reads positive or negative 0-50kV. Can
also be switched to check negative voltages. For
checking EHT and Focus voltages as well as TVs and
Microwaves. Hi-tension voltages.
$89.00 + $5.00 p&p
LOW-VOLT AGE PROBE
Ideal tor checking microwave ovens and TVs. The ~
ranges are from Oto 5kV and from Oto 1 OkV. Double
insulated for safety.
$79.00 + $5.00 p&p
D
Proudly manufactured in Australia by
HARBUCH ELECTRONIC S P J
90 George St. , HORNSBY, N.S.W. , 2077
Phone (02)476-5 854, Fax (02) 476-3231
td
DEGAUSSING WAND
Strong magnetic field , larger than usual coil with
multicore centre. Double insulated for safety with
momentary switch operation. 240VA/C 2.2
amps. As important as having a soldering iron!
$75.00 + $10.00 p&p
TUNER REPAIRS
We repair most tuners on an exchange basis.
From only $1 7 .00 + p&p
Cheque, Money Order, Bankcard or MasterCard
Y.V.TuNERs)
216 Canterbury Road,
Revesby, NSW 2212
Phone (02) 77 4 1154
JUNE 1990
53
\\1~,l/
,~
~ ~\Ii~\
~
;?
,
A
-=--•-
[l,
.
,:;?
. iii)>
~
~
-, '
way to identify them. Fortunately,
there weren't many but of course, a
couple had to land on yours truly's
doorstep.
Don't ask me to explain why the
reversed diode did what it did. I
was happy enough to confirm that
this was the fault in these sets and
to put things right.
All of which carries a clear
message; such modifications,
without the benefit of printed board
markings, are risky. Introduce
something new to a production line
hand and the original Murphy's
Law prevails; if something can go
wrong, it will go wrong.
A nudder story
And now, in lighter vein, here's a
short story from my colleague J.L. in
Tasmania. Remember the story in
the March issue a bout a colleague
who had to re-orient his thinking in
order to service the local sausage
machine? Well, this is along similar
lines, only this time it is a transition
from TV sets and video recorders to
54
SILICON CHIP
the much more mundane subject of
dairies, cows and the associated
milking procedures. Here's how J.L.
tells it.
Most electronic servicemen are
"townies" and spend their time
repairing domestic or commercial
equipment brought to them by other
townies. It's not often that we need
to put on gumboots and solve farmyard problems. This story is about
one of those situations where you
leave dropped tools just where they
fall. I mean, ankle deep in ordure is
no place to go fishing for a $2
screwdriver, is it? (Speak for
yourself, J.L., I come from a long
line of Scottish ancestry).
Imagine, if you can, a modern
dairy farm where all cows carry
electronic identification tags and
their milk is instantly analysed so
that appropriate feed supplements
can be delivered to the computer
controlled manger.
Imagine further an automated
milking line where the farmer has
only to attach the cups to begin the
extraction process. After the udder
is empty, the lack of milk triggers a
removal process which releases the
cups and pulls the assembly away
so that the cow can back out of the
stall, all without further attention
by the farmer.
And one last point before we get
to the story proper. Cows don't let
down their milk immediately. They
need half a minute or so to settle
down before the milk flows, so the
cup removal system needs to be inhibited during this period. This is
accomplished with a simple
automatic timer. So now to our
story.
Strawberry's problem
It's early one wet and windy morning. Strawberry enters her stall
and contentedly munches hay while
farmer Giles washes the udder and
attaches the cups. He moves the
lever to start the suction and the
cup assembly falls off the udder
and is pulled upwards by the
retracting mechanism. Farmer
Farmer Giles rang Arthur to ask
him if he would have a look at the
you-know-what milking machine. It
would be next week before the
manufacturer's serviceman could
reach him and poor old Strawberry
could explode before then!
Arthur knew a little about milking machines but none of the details
of this one. But he was prepared to
try, so it was on with the gumboots
and a screwdriver on a string
around his neck.
Float switch
Giles doesn't get out of the way in
time and is clobbered by all four
cups in turn.
He lets fly with some unparliamentary language (and we all
know what that means, don't we),
then tries again, with the same
result. The cups are retracted with
a fair amount of force, so farmer
Giles is beginning to feel rather battered and in no mood to argue with
the "blankety-blank" machinery.
So poor old Strawberry doesn't
get milked that morning. Cows are
only contented in their own stall
and won't co-operate if put in any
other. Nor will they co-operate if
milked out of turn, so Strawberry
can't be brought back later and
milked by hand.
Enter Arthur, a qualified electronics technician. Arthur is also
one of those people who are happiest with their hands in the soil. He
has a small hobby farm within commuting distance of town and grows
fruit and vegetables to help keep
the family fed.
The cup release mechanism is activated by a float switch in the collector bowl above each stall. When
the milk runs out, the float drops
and closes a switch which operates
a valve to release the vacuum that
holds the cups in place. At the same
time, a catch is released so that a
spring loaded cord can pull the
cups off the udder and up to the
parked position at about shoulder
height.
The inhibit function, at the beginning of the milking cycle, is initiated
by a switch attached to the vacuum
lever. After the lever is operated, a
555 timer begins a 60-second countdown, during which the release
mechanism cannot function.
Our gumbooted serviceman
learned all this by studying one of
the other properly functioning
stalls. So it took him very little time
to realise that the faulty stall was
not producing the inhibit signal due
to a l000µF capacitor on the circuit
board with a broken pigtail. As simple as that!
Arthur's reward
So endeth the technical part of
the story but the best is yet to come.
Farmer Giles was delighted with
the prompt and efficient service
and insisted that Arthur submit his
bill for the time he had spent on the
job. In the meantime, would Arthur
let him show his appreciation with
a small gift?
Which is how Arthur came to
drive home with a bucket of fresh
milk, a side of lamb neatly butchered and packed in a cardboard
carton, and a promise of all the
fresh cream his family could consume. And a week later there was a
cheque for the cash part of the deal
as well.
Why don't I ever get customers
like that?
Dunno J.L., but everyone up here
had a good laugh over that story.
And it brought back memories. As a
callow country youth, when times
were tough, I once earned my bread
and dripping milking cows - and
not with the aid of any new £angled
milking machines either; it was all
very rough and ready.
And I learned about cows; just
how stubborn, contrary, and cantankerous they can be. Nothing in
your story surprised me one little
bit. Truly is it said, they are "cows"
of animals.
~
Fitting a Fax Card to Your PC: ctd from page 21
least cost) to you. Letterheads,
signatures, logos, etc can all be
transmitted provided you have access to a scanner (although you
could fax your signature, etc back
to your PC) and Fax89 supports
most popular graphics file formats.
The fax card, however, cannot be
a total replacement for the fax
machine. If you don't j:i.ave access to
a scanner, there's no way you can
send brochures, promotional material or pages from a manual.
If the bulk of your work is non PCbased, then the fax card is really
not for you. Rather, the fax card
should be seen as an adjunct to the
fax machine. Perhaps the scenario
will be for a cheap lower-featured
fax machine to transmit faxes while
the PC fax card receives all incoming faxes.
Also, the very nature of a fax
suggests round-the-clock access.
Are you prepared to leave your PC
running all day and all night?
However, for the PC user who
works from home or a small office
and mostly communicates via PCbased documents, a fax card probably meets all the specifications
while presenting a major cost saving over a normal fax machine.
The price of our sample fax card,
the Fax89, was $599 from Eskay
Printernet, Suite 3, 124 Marsden
Street, Parramatta, NSW 2124.
Phone (02) 891 1282.
~
JUNE 1990
55
|