This is only a preview of the January 1990 issue of Silicon Chip. You can view 45 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 "Active Antenna For Shortwave Listening":
Articles in this series:
Articles in this series:
Articles in this series:
Articles in this series:
Articles in this series:
Articles in this series:
Items relevant to "Modifying The Studio 200 Amplifier":
|
Sing it again Sam(sung)
My main story this month concerns a brand
of set which is relatively new to the
Australian market. So, as well as discussing
specific technical problems, we will take the
opportunity to present some general
background to it.
The brand involved is Samsung, a
Korean based company represented in Australia by Samsung
Australia Pty Ltd (10/16 South St,
Rydalmere, NSW 2116).
As I said, this is a relatively new
brand on the Australian market,
having come to local prominence
only in the last 12 months or so. And
quite naturally, customers are
tempted to ask just what kind of an
organisation is behind this brand
name.
Fair enough. Well, for starters,
while the brand name is new, the
products have been around for a
good deal longer. Most servicemen
will recognise them as having been
previously marketed under the
58
SILICON CHIP
"Princess" and "Palsonic" labels.
They are a large organisation. As
well as TV sets, they also manufacture video recorders, audio equipment (including CD players), microwave ovens and computers. Their
products are well established in the
United States and they have
manufacturing interests in other
countries as well.
Indeed, the brand has been well
known in New Zealand for several
years, where Samsung has a small
assembly plant. They also have a
very large manufacturing plant in
Wales producing TV sets, video
recorders and microwave ovens for
the British market.
So Samsung is no pip-squeak, fly-
by-night organisation; it is just that
we have only recently become
aware of it in Australia.
Samsung service
My association with the brand
began when one of the local
dealers, for whom I do warranty
and other service work, decided to
market these sets. As part of the
deal, he was required to organise
warranty service so he suggested
that I might be interested.
Well, I am always interested in
more work. At a technical level, my
experience with the Princess and
Palsonic models over the years had
indicated that they were fairly conventional sets. In fact, they were no
more difficult to service than most
other brands.
More to the point was whether
Samsung and I could reach a
satisfactory agreement. As many of
my colleagues will testify, warranty
service has not always been the
happiest arrangement in the past,
at least with some companies.
What has seemed like a reasonable
proposition in theory has not
necessarily proved so in practice.
Anyway, the dealer passed my
name on to Samsung and they in
turn approached me. Alerted by
some of my previous experience
with warranty work, I laid it on the
line that I would need certain firm
undertakings before committing
myself.
First, I needed to be assured of
the ready availabilty of properly
prepared manuals. A lack of
manuals, or manuals that are not
well written, can be more than
simply frustrating; it can be very
expensive, particularly with new or
unfamiliar models. As a result,
what should have been a modest
profit can become a substantial
loss.
The next question is technical
backup. While it is unreasonable to
expect a company technician to
have an immediate answer for
every "funny" that turns up in the
field, there is nothing more
frustrating than talking to someone
who obviously knows less about a
piece of equipment than you do. So
could I bank on adequate help when
I needed it?
Finally, there is the question of
spare parts. Of all the frustrations
we face, lack of spare parts is one
of the most serious. It is often a big
enough problem to diagnose a
tricky fault without then facing the
situation that the vital part involved
is "on back order".
Such situations can mean a delay
of weeks or even months, while the
customer fumes and the servicem~n becomes the meat in the
sandwich. And it doesn't help that
some customers refuse to accept
that this is the true situation,
preferring to believe that it is the
serviceman's own laziness or incompetence (which, I regret to say,
is true in some cases) that is the
cause of the problem.
Anyway, those were the major
points I raised. And the company
didn't quibble; they were eager to
assure me that all those conditions
could be met and so far they have
been as good as their word. I only
hope that they can keep it that way.
But it wasn't a one-way deal.
They wanted some assurances and
background from me. In particular,
they wanted to know how long I had
been in the industry and asked
questions about my workshop
facilities and test equipment.
Since I've been in the game for
more years than I care to think
about, and since my shop is as well
equipped as anyone's, I didn't have
any problem meeting their requirements.
And so an agreement was reached which means that I am now an
authorised Samsung service centre.
Basic models
All this happened several months
ago and since then I have had the
opportunity to come to grips with
these sets at a practical servicing
level. The range includes 34, 40, 45
and 48cm models and all appear to
be designed around a fairly basic
chassis, with only minor variations.
There is also a portable set,
designed for both 240V AC and 1 ZV
DC operation. It uses similar basic
circuitry. The most recent addition
is a 63cm model but I have no data
on it yet.
The basic chassis is the P/50F
and the sets based on it, which I
have encountered so far, are the
CB515F and CB515Z, both 45cm
models. The "F" version uses an
8-channel pushbutton tuner while
the "Z" version is fitted with
remote control.
The "F" version is a very popular
model and my dealer friend has
sold a couple of hundred so far,
many to local motels. Apparently,
motel owners regard the "F" version as ideal - it's a simple set, the
size is right, and the price is
attractive.
One motel alone recently purchased 29 of these sets. In fact, I've
opened a special file on these. Since
they all came on line at the same
time and are operating in the same
environment, they should provide a
lot of valuable service data.
The nitty gritty
Y~'P, NOION~
COMPLAIN-,- ABO
OUR -rv S
Now let's get to the technical nitty gritty. What I am about to relate
covers virtually the same fault in all
cases, producing the same major
symptom but with varying secondary symptoms if neglected.
The major symptom is a bright
vertical line in the centre of the
screen. This is an unusual symptom
in that it obviously indicates a
deflection failure but no accompanying EHT failure.
An extreme side effect of this
symptom, if it is allowed to continue, is a burnt board. This happened to a motel set on one occasion and left a hole big enough to
pass a 5c piece.
Digressing for a moment, this
raises another matter which is
peculiar to the motel scene. Faults
in motel TV sets can often go unJANUARY 1990
59
SERVICEMAN'S LOG -CTD
noticed for days, or even weeks.
Most guests stay only one night and,
unless they are TV addicts or wish
to catch a particular program, a
faulty set tends to be shrugged off
as bad luck. By the morning it is
forgotten and no-one is any the
wiser.
A further complication arises
when the fault is intermittent, as it
often is in this case. It may escape
routine inspection or, worse still,
may be deliberately left in service
on the basis that it works for most
of the time anyway, so why worry.
I am now trying to instill some
sense of responsibility into those
concerned by emphasising the need
to take a set out of service immediately this fault - or any fault
- is observed, to avoid the risk of
further damage.
As already stated, the problem
involves the horizontal deflection
circuit, as distinct from the horizontal output stage. More precisely, it
involves the area around inductor
1401 which is in series with 1408/
C414, side pin-cushion coil T402
and the horizontal deflection coils
(see Fig.1). The problem is basically
one of quality control, because it
boils down to a series of faulty
soldered joints.
And note that I said "faulty"
rather than "dry", because my im. pression is one of fractured joints,
such as occurs when the two components are allowed to move before
the solder has properly set. It is not
an unusual fault and some early
local production runs produced
boards which were literally riddled
with such faults. They were so well
known that we treated them as
routine.
To be fair to Samsung, they are
aware of the problem and have
taken steps to correct it. In the
meantime, those sets already in the
field have to be dealt with. And
again, to their credit, they didn't
hesitate over the burnt board. I
returned the set to them and they
simply replaced it with a new set.
The main offender is the joint at
the" hot" end of 1401. If allowed to
persist, this will destroy R413
because all the horizontal current
60
SILICON CHIP
HORIZONTAL
DEFLECTION
_
___._TO COLLECTOR
OF 0404
R430
150{)
C454
1
R~i3
L401
.,.
TO HORIZONTAL
OUTPUT TRANSFORMER
Fig.1: this simplified diagram shows
where the bad solder joints can occur
in the Samsung CB515F. The
immediate symptom is intermittent
loss of horizontal deflection but
damage to the board and to
components can also occur if the sets
are left in service.
involved must now flow through it.
Another version of the fault involves C414, which may be faulty at
either end. In either case, prolonged use of the set in the fault condition can result in a burnt board in
this area.
Other secondary effects encountered so far include destruction of the horizontal output transistor (Q404), R423 or R420 (both rn
½ W safety resistors), and diode
DHl or its associated resistor, RHl.
The important point to note here
is that the set may be presented to
the serviceman as completely dead,
due to one of the abovementioned
secondary failures, and may appear to be fixed when the faulty
component is replaced. In fact, the
real cause - the faulty joints may simply be lying dormant, ready
to strike again. So be warned.
From southern shores
Talking about vertical lines on
the screen, here 's something from
our southern sub-continent, as told
by J.L.
I thought I had seen just about
every kind of TV fault there is but
one showed up recently that really
had me guessing.
The set was a Sharp CX4846 and
the "funny fault" was not immediately apparent. It came in with
a report that the screen was very
bright, with retrace lines, and that
it sometimes switched itself off. No
mention was made of the fact that
there was no picture but that was
the case.
The excessive brightness was not
all that difficult to solve. Somebody
had been inside the set and had
turned the screen (G2) control to
full on. The excessive beam current
under these conditions was close to
the shut-down level for the power
supply, which explained the occasional switching off.
The final evidence of unqualified
intrusion was that six of the 10
screws securing the cabinet back
were missing and the remaining
four were all loose by several turns.
Mr Nobody
It's funny, though. Quizzing the
owner produced the answer "Oh,
no! Nobody's touched the set". That
Mr Nobody has a lot to answer for.
Turning the screen control down
to about half way produced a
typical snowy screen and the
retrace lines disappeared. It was at
this point that the "funny" appeared. Right down the full height
of the screen, about one third of the
way from the left-hand side, there
was a dense black band a bout
35mm wide.
There was no sign of a picture,
with each of the 10 channel buttons
producing identical snowy screens.
It transpired that Mr Nobody had
not only fiddled the G2 control but
also the channel selectors. My next
job was to retune the various
channels.
The self-search tuning in this set
was very slow and it seemed to take
forever to get the first channel on
screen. When I did, I could see the
fault in all its final confusing glory.
The black band had grown into a
75mm wide strip that had every
sign of being the horizontal blanking bar. I had seen this in other sets
when I had floated the horizontal
scan but I had never seen it rock
steady in an otherwise perfect
picture.
("Floating" involves disconnecting the horizontal sync pulse and
then adjusting the hold control until
the picture "floats"; ie, almost stationary or drifting slowly from side
to side. The trick is a useful one to
This photograph shows the displaced image on the faulty Sharp CX4846. The
black vertical band is the horizontal blanking period and the image at the
extreme left belongs at the extreme right.
determine if horizontal instability is
caused by a bad horizontal oscillator or by faulty sync pulses).
A Philips PM5544 test pattern
from SBS showed just the extreme
right hand side of the test pattern,
on the left hand edge of the screen.
After the right hand castellations,
there was a narrow band of grey
down the picture, then the solid
black bar began.
Right down the middle of the bar
was a narrow green line, then more
black bar before another grey strip,
then the left hand castellations at
about a third of the way across the
screen. The rest of the test pattern
was free of horizontal or vertical
non-linearity although about a third
of the circle was lost.
As a matter of interest, the green
line was the chroma burst. The
sync pulses were buried in the first
half of the black band and, because
they are "blacker than black",
could not be seen on this picture.
Fixing it
So that was the nature of the
fault and it was now up to me to
find out what it was and fix it.
I mentioned earlier that someone
had turned up the G2 control. My
first thought was that they had also
meddled with one of the horizontal
presets and so messed up the
horizontal timing. But there were
only two horizontal presets, a
horizontal frequency and a horizontal shift, and neither of these had
been touched. Nor would any adjustment correct the fault.
This seemed to be a case of
"false line lock", a condition that I
have heard of but never seen. It occurs when something either delays
or advances the triggering of the
horizontal oscillator. If the trigger
was 32 microseconds late (or early),
the blanking bar would be near the
centre of the screen. In this case
the triggering was probably about
42µs late (or about 22µs early).
Horizontal oscillators are invariably of the "flywheel" kind,
designed to prevent noise accompanying individual sync pulses from
causing random triggering and
"shimmering" in the horizontal
plane. These oscillators usually
have an LC circuit that can be adjusted to run at approximately the
horizontal line frequency. This is
then phase locked by a discriminator circuit that compares the
timing of the sync pulse with a
pulse from the flyback transformer.
The whole system is given a
relatively long time constant to provide the flywheel effect.
Until recently most of the components needed to run the horizontal department were discrete and
the horizontal oscillator coil was
easy to find and adjust. But not any
more!
In the Sharp CX4846, the entire
horizontal stage is contained in a
single integrated circuit, an IX
0602CE, and there is nothing that
even looks like a conventional coil
type oscillator. Even the hold and
horizontal shift presets are subminiature pots about 6mm in
diameter!
Even so, unless the designers
have come up with some new type
of sync pulse horizontal drive, the
oscillator still needs a sync pulse
and a horizontal output reference
pulse. These pulses have to be supplied from other parts of the set and
fed into the oscillator chip. So I
tried to find them with the scope.
This is where I encountered the
first major hurdle. This set seems to
be an offshoot from VCR technology. The entire circuit board is
only about 25cm square and is
tightly packed with subminiature
and ultra-miniature components.
In addition, some of the circuit
board tracks are hair-thin and are
often completely covered by the
printed overlay. This made it
almost impossible to trace the
tracks and I found that I simply
couldn't find the sections I needed
without the full service manual. So I
advised the customer that his
repair was going to take some time
and sent away for the needed
publication.
When the manual arrived, I lost
no time in looking up the horizontal
oscillator section of the circuit
diagram. What I saw made me
shudder. There is hardly any part
of the oscillator that is outside the
big integrated circuit mentioned
earlier.
Not only that, but many other
functions are also enclosed within
the chip. There's a complete audio
stage, including IF and output; a
video IF amplifier, synchronous
detector and video amplifier; a sync
separator, vertical oscillator, V and
H blanking amplifiers, burst gate
amplifier and detector; AFC and
AGC detectors; and, of course, the
horizontal oscillator I was trying to
analyse.
With so much going on inside the
chip, I would not be surprised to
find that it was the cause of my
trouble. But by the same reckoning,
I would expect to see some other
JANUARY
1990
61
t
l
0
~
l
now had to consider the possibility
that it was faulty.
It was time to get out the freezer
spray.
With the set warmed up and the
fault apparent, I went over the IC
and surrounding components with
the spray. Nothing happened. I
widened the area of interest, spraying the other chips , various
resistors, capacitors and transistors right up to the horizontal
output transistor. Still nothing happened. The fault remained and
stayed rock steady.
Then, after a few days of this
delayed onset, the fault decided to
revert to its earlier "instant on"
pattern. By this time I'd had enough
and elected to change the chip.
The problem licked
I. 1-\i:\"D 1<.a:::.i;.NT"L..'-< INVE-S'1e.D IN A
VACUUM DE:SOL-DE:.'RI NG s·n::~,TIONoo•
symptoms if it was indeed the chip
that was faulty.
I had no wish to replace an expensive 28-pin IC until I was absolutely certain that it was at fault.
I decided that, before I went that
far, I would check surrounding
components that might have some
connection with the trouble.
These components included two
electrolytic capacitors attached to
two timing circuits inside the chip
the phase detector on the
horizontal oscillator and the coincidence detector on the horizontal
oscillator timing switch.
Both of these are 0.47µF 16V
units and these small value electros
are notoriously unreliable. They
were removed and replaced with
new capacitors but to no avail. The
picture continued to show the
horizontal blanking bar - until
next morning.
At the next switch-on the picture
was perfect, without a sign of the
earlier trouble. I tried various
thump and bump routines but
nothing happened for half an hour
or more. Then, the picture started
to shiver, as if it was about to lose
horizontal hold. Finally, the whole
picture shifted sideways and locked
into the fault condition.
62
SILICON CHIP
Two aspects of this reversion to
the fault condition led me to think
that it might be the chip that was
faulty after all. One was the warming up time taken before the fault
showed. Admittedly, until now, the
fault had been present from cold
but I had changed a few components and I had been poking
around in the circuitry. This could
have led to a slowing down of the
onset of the fault.
The second indication was that
the fault did not snap on but wriggled and shivered for a minute or so
before settling down. I still wasn't
keen to change the big chip but I
TETIA TV TIP
GEC 2213A
Symptom: No sound or picture. All
fuses are OK and there is 320V on
the collector of the chopper transistor, TR502.
Cure: R503 (22k0) open circuit.
This resistor supplies 1 2V to run
the chopper control IC and
disipates 4.9 watts . The original
5W resistor is rather underrated
and a 1 OW replacement will ensure more reliable operation.
My usual suppliers had no IX
0602CE in stock and had to place it
on back order. I had a week off
from worrying about the Sharp
before the chip arrived.
I had recently invested in a
vacuum desoldering station and its
value was really shown in this job.
In fact, I had the old chip out in two
minutes, without the usual struggle
and risk of damage to the printed
tracks.
The new chip went in just as easily and at first switch-on there was
no sign of the fault. Repeated trials
over the next few days seemed to
indicate that the fault was gone. A
week later and I was sufficiently
convinced to return the set to its
owner. There have been no further
complaints.
When you say it quickly, this job
has not been all that unusual. Right
from the start the integrated circuit
was suspect. But its replacement
cost of some $30 plus the difficulty
of de-soldering the 28 pins was
enough to encourage me to try
everything else first.
This has been one of those jobs
where one wastes considerable
time first up but recovers the
wastage next time a similar fault
appears. And if telling the story
here helps you to save some time
when you see this fault, then the
time has not been wasted after all.
Thanks J.L. I think that's enough
to keep our readers on their toes for
this month. Cheers for now.
~
|