This is only a preview of the March 1993 issue of Silicon Chip. You can view 44 of the 96 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 "Low-Cost Audio Mixer For Camcorders":
Items relevant to "Test Yourself On The Reaction Trainer":
Articles in this series:
|
SERVICEMAN'S LOG
The search for the missing link
I hardly need emphasise that the service game
is mostly about diagnosis. Once a fault has been
diagnosed, the repair is usually routine. But
finding the fault is the real job & this one took
months to find & seconds to fix.
The set concerned was a Samsung
34cm colour TV set, model CB 349Z.
The "Z" suffix indicates a remote control version, while an "F" suffix indicates a standard version. Both use the
P50F chassis.
The owner uses this set in two locations: at his Sydney home and at a
holiday cottage down the coast. As a
result, the set has been programmed
for city VHF channels 2, 7, 9 & 10,
SBS UHF channel 28, and south-coast
UHF channels 30, 33, 42, 45 & 48. But
that is more or less by the way, except
that the large number of channels
emphasised the fault's nuisance value.
The fault itself was a tendency to
spontaneously change channels. In
greater detail, it didn't simply jump
from one channel to another; it would
go into a search mode which took it
Our loss is
your gain
Fracarro's
ESVU2. The
really low
loss VHF/
UHF mixer
with areasonable
price tag.
Peter C. Lacey Services Pty. Ltd.
P.O. Box 678 (74Fulton Rd.) Mount Eliza 3930
Tel:03 787 2077 Fax:(03) 787 3460 ACN006893438
42
SILICON CHIP
right through the 10 channels, then·
repeat this procedure, before finally
going into standby mode.
Turning the set off and on again
would usually cure the fault temporarily and it might then work normally for a few hours, several days, or
even weeks. And, as I established later,
it didn't seem to matter whether the
set was hot or cold. Nor was there any
other pattern to the fault. In short, it
was completely unpredictable.
Search function
The search function just described
is a quite legitimate feature in this set.
Basically, it is a self-seeking, automatic tuning function, its primary
purpose being to facilitate the initial
setting up of the channels.
There are two sets of UP/DOWN
buttons in the system: the search buttons and the channel selection buttons. It is important to clearly differentiate between the two.
The setting up procedure is initiated by activating one of the two search
buttons (UP or DOWN). This causes
the set to search until it receives a
signal of adequate strength to lock
onto, while very weak signals are ignored. In the unlikely event that no
signals are detected, the system will
search right through both the VHF
and UHF bands, then go back to the
beginning, repeat the search, and go
into standby mode.
In practice, of course, it would normally lock onto a signal somewhere
in either the VHF or UHF band. When
it does, it stores this channel in
memory, waits for the button to be
pressed again, then continues the
search, stores the next channel it locks
onto, and so on.
When all the available channels
have been stored, they may be selected in two ways: (1) sequentially
by using the UP or DOWN channel
select buttons on the front of the set;
or (2) directly by pressing the appropriate channel number on the remote
control unit.
And how doEls the system know
when it has found a channel? The
relevant part of the circuit is shown
here, including the PWB-CONTROL
board and a section of the PWB REMOCON (remote control) board.
Most of the work is performed by
RIC01 (M50431), the main IC in this
section. Pin 12 of this IC is fed with
sync pulses and these provide the
locking command.
So inuch for the general background. When the customer described
the fault, I immediately took the precaution of warning him that, since it
was capable ofrurining OK for weeks
at a time, it could well take me many
weeks to find the cause. Fortunately,
he had anticipated this and added
that he had another set which would
serve in the meantime. And just as
well as it turned out.
RR65
• RC11
',
UND
:
1
SY
''
-l
RR36
1W,24»<
RC06
1
t60V 1,
''
'
''
RR40 I., 11< RR4 I 4. ?K
sov,1,..
RR4
RR37
10K
~ K5C815-Y
ROOS
l-: 25C1815-\
RC09
2K
RQ04
K~C81S·Y
2SC1815-Y
RC01
Tip, 334 I0334
f tOOV
I
•
-: IOOV
RC08
22
130CH I
R014
ROO?
UPC574J
JN4148
IBfil2l
I START l
1:, _ _ _
"
ROIS
€e ~;;~,22
RC12
30
RR63 10K
C
!1i
0
'
a,
'
,...
82
I
Reos•
103.SOV
~
RR67 30K
RR33 30K
RR32
I0K
RR34 10K
RR35 30K
Fig.1: the relevant circuit section of the Samsung CB 349Z TV receiver. The
microprocessor controller (RIC01) is on the PWB-REMOCON board & locks onto
sync signals on its pin 12 input. The PWB-CONTROL board is to the left of this
board and carries the display unit and most, but not all, of the control buttons.
The fault was hidden under other components.
Knowing how the system sensed
the presence of a signal, my first step
was to check pin 12 ofRlCOl for sync
pulses. Because of the intermittent
nature of the fault, all I could do was
switch the set on and wait for the
fault to appear. It took its time but,
after a couple of days, it eventually
showed up.
I checked pin 12 with the CRO but
there were the sync pulses, exactly as
before. So that ruled out that theory.
I wasn't quite sure where to look
next , so I decided to see whether
Samsung's service department had
any ideas. After all, there is not much
point in spending many hours trying
to track down a fault when someone
else has already been there and the
· details are on record for the asking.
So I rang Samsung outlined the
symptoms, and asked whether anyone else had encountered the problem. The technician had not heard of
any such fault but immediately made
his own diagnosis. "Oh yes, the problem is the microp'rocessor chip, RICOl.
Replace that and it should solve your
problem".
Frankly, I was highly sceptical of
this suggestion. It had all the earmarks
of a snap diagnosis, based on the obvious, but without too much thought
about the fine details of the symptoms. On the other hand, the set was
still under warranty, and if they were
prepared to cover the cost of supplying and fitting a new IC -with 42 pins
- who was I to quibble?
So a new IC was ordered and duly
fitted. And that solved the problem,
or so it appeared. The set ran for over
three weeks and never missed a beat
so, when the customer called one day
to check my progress, I suggested that
he take it away. After all, it might just
as well be tested in his lounge room
as on my workbench. But I warned
him that, in spite of Samsung's suggestion, and my bench testing, I was
still not convinced. If it failed, he was
to bring it back immediately.
Several more weeks went by with
no word from the owner and I was
beginning to think that Samsung had
been right after all. The next thing I
knew, the set was back on the shop
co·unter. "It went real beaut until yesterday. Then it was up to it's old tricks;
really bad".
Control board
So we were back to square one.
Well, almost; at least I could now rule
out the IC. In the meantime, I had
noted one more aspect of the fault; it
always searched up, never down.
On this basis there seemed to be
only one line of investigation left: the
control board (PWB-CONTROL) and,
in particular, the channel UP and
search UP pushbuttons. If one of these
somehow jammed on, either mechanically or electrically, it could create
this effect.
The' PWB-CONTROL board measures about 200 x 600mm and sits at
the front of the set beneath the picture
tube. It is offset to the right, with the
speaker in the remaining lefthand
space. As well as the pushbutton controls just mentioned, it also carries
the volume, contrast, colour, fine tuning and on/off controls, plus the remote control receiver and the channel number display.
MARCH
1993
43
Fortunately, this board is fairly easy
to get at by slipping the main chassis
out and undoing several screws. The
main chassis can then be replaced,
giving a working set with the control
board accessible behind the cabinet.
So, with everything set up like this,
I waited for the fault to re-appear. The
set then ran for weeks without any
sign of trouble but eventually it did
misbehave. When it did, I tried twisting and bending the control board
and this provided the first confirmation that the fault really was on this
board. I could more or less create or
cure the fault in this way.
But I say "more or less" because I
could not establish any exact pattern
that would affect the set's behaviour.
On some occasions, the fault seemed
to be at one end of the board, while at
other times it seemed to be at the
other end. And sometimes twisting
was the most effective, while at other
times bending seemed to work. Whatever it was, I just couldn't pin it down.
But with the board clearly established as the culprit, a replacement
board seemed the logical answer. The
only problem with this was that, when
I called in on Samsung, there were no
boards in stock. Instead, they were
"on back order, delivery date indefinite". How often have I heard that
phrase!
At this point, I had two choices:
either wait, for who knows how long,
for a new board or fix the old board
myself. If it was to be the latter, then I
needed help more than ever. I tackled
44
SILICON CHIP
the service department
again, determined to check
out everyone if necessary
in order to get a clue.
Fortunately, the first
bloke I buttonholed was
quite helpful. He listened
carefully and didn't make
any snap diagnosis. On the
contrary, I could tell by the
expression on his face that
the wheels were turning.
Yes, he had heard of
such a fault. Well, that was
the good news. The bad
news was that it was so
long ago and had reached
him by such a devious
path, that he couldn't recall much in the way of detail. In fact,
he couldn't recall anything initially
but, after much brain racking, he came
up with one thought. It concerned the
wire links on the PWB-CONTROL
board and, in particular, the possibility of two of them touching. And that
was all the help he could offer.
I thanked him and went on my way.
At least it was something but, on the
other hand, it didn't seem to make
TETIA TV TIP
Hitachi VT-640E VCR
Symptom: Machine does not
want to accept cassette. If the
cassette is pressed into the cassette carrier and held there for
several seconds, the machine will
accept it briefly then immediately
eject it. A repeat performance may
finally persuade the machine to
accept the tape.
Cure: Replace the tape end sensors in the cassette housing .
Faulty sensors give the control
microprocessor the wrong information and makes it think a cassette is already installed. A clue is
that the machine can be put into
play mode without a cassette in
place.
TETIA TV Tip is supplied by the
Tasmanian branch of the Electronic Technician's Institute ofAustralia. Contact Jim Lawler, 16
Adina St, Geilston Bay 7015.
much sense. There are several links
on this board but I couldn't imagine
how any of them could touch.
This was confirmed when I checked
the actual board - there was no way
any of the links could touch each
other, or anything else for that matter.
Yet, for all his vagueness, the technician had been adamant that the fault
involved the links.
Display unit
More or less in desperation, I went
back to my twisting and bending routine. The set's behaviour was much as
before except that I gained the im,
pression that the area around the channel number display unit (top left of
Fig.1) was marginally more sensitive.
This didn't make sense. The display should have no electrical connection whatever with the control
functions. Nevertheless, I had to be
sure. The display unit is a dual-inline, 18-pin package, fed independently from the PWB-REMOCON board
via plug/socket CNR03.
This all seemed perfectly straightforward until I took a closer look at
the copper side of the board around
the display unit. It was easy enough
to account for the copper tracks running to the display unit but there were
other tracks as well, with several connections coming through from the
component side that obviously did
not belong to the display.
Well, there was clearly no alternative; the ·display unit had to come out.
It wasn't all that difficult and when I
finally pulled it away, the cause of all
the bother was plain to see.
Hidden under the display unit were
four extra w,ire links. And two of them
had been bent so that, at best, they
were only a gnat's eyebrow apart and,
at worst, were actually touching. In
this situation, almost any variable
(temperature, vibration or attitude) can
change the status of these links, which
is a flash way of describing an intermittent.
The irony was that having found
the fault, it took less than 10 seconds
with a screwdriver blade to cure it.
But that's the way it goes; finding the
fault is the real job.
What caused it?
But how did the faulty condition
come about in the first place? It seems
unlikely that the links were bent in
that formation when they were fitted,
so what happened?
From what I could work out, it
would appear that the links were sitting too high on the board when they
were initially fitted, either because
they had not been pushed down far
enough or because they were bowed
slightly upwards. Either way, they
would be pushed sideways when the
display module was fitted.
It wouldn't have mattered if they
had both moved in the same direction, but Murphy saw to it that they
moved in opposite directions towards
each other. And he also took care to
see that they did not touch each other
while the set was tested; that had to
wait until it was in the field.
Well, that's enough from my bench
for the month. Let's now take a hop,
step and a jump across Bass Straight,
for a story from our colleague J. L. in
Tasmania.
The fishnet VCR
This story comes from a colleague
who lives and works in the country,
100km from the city. It concerns a
Teac video cassette recorder, model
MV307, with a very obscure problem.
The customer complained that
watching a video recording or the ABC
on VHF channel 2 was like looking at
the program through a nylon fish net.
The commercial VHF station and SBS
on UHF showed no signs of the trouble.
In fact, the "fish net" turned out to
be shimmering white lines that flickered across the screen in a diagonal
pattern. They were never still and
moved at a constant speed. In all other
respects, the picture was normal and
was quite watchable if one could only
ignore the white lines.
The customer put up with the lines
while watching video recordings but
took to feeding the TV receiver directly from the antenna, bypassing
the video recorder, to watch channel
2. But this introduced another problem. They were in a remote area and
he had come to rely on the extra gain
provided by the video recorder's frontend amplifier. Without that gain, the
picture from channel 2 was not so
good and so he eventually decided
that the recorder just had to be fixed.
My colleague soon found that the
fault was a fundamental one. It not
only appeared on the off-air TV program and pre-recorded video tape signal, but could also be recorded onto
-WHA1'"'S MORE: 11"" WAS STlt-1IHe: SAME. ~'HEAVY 'RA\N .. .,.
tape and replayed on any other machine!
This particular fault was not an easy
one to diagnose. At first glance, it
looked as though it could be a power
supply problem since similar problems in TV sets had been traced to
this source. However, the nature of
this problem - though not the extentcould be changed by altering the tuning of channel 2 on the recorder. This
suggested a possible fault in tuner.
Since my friend did not have a service manual for this model, he contacted a colleague who does have a
manual and asked if he would care
to take over the repair. This serviceman works in the city but, after the
most exhaustive testing, he could find
absolutely nothing wrong with the
machine. It tuned and recorded channel 2 perfectly. He did reset the tuner
AGC in case of problems there but
otherwise he could do nothing to help.
And so my friend took the machine
home and tried it again. The fault was
still present, which suggested that it
was related in some way to the weaker
signal at his location. For a time, he
thought that the trouble might be due
to interference from noisy power supply lines. However, the trouble was
identical at both the customer's home
and at my colleague's workshop.
What's more, it was still the same
after heavy rain.
MARCH
1993
45
'1k«,, 1:J~ ,l!:euu;e ,li:eteaae
li : i:11::~:1,t:m:1:mi::i i :i l :il i:l :¼JE;l:1: 1;1:1 :1 :11il il i ffl.il !l i! : 1:1: ~11:1;:: : :i :il i ~ ll::~:m:1:1:1:!l il il i: : 1: 1 11
1:1:l:imil il l!;.1:~::i ~
ZI -
I ~
OFFERING
i-"'""'"!'_
11
_ _ _i
EXTRA SIZES
a MORE ACCESSORIES
LOWER PRICES
11 NATURAL or BLACK
ANODISED FRAMES
II FOREST GREY or
BLACK TEXTURE
COVER FINISHES
READER INFO NO. 6
THE UNIQUE ZIP-RACK DESIGN
SIMPLE,FAST & EFFECTIVE THATS ZIP-RACK
Twelve lengths of ZIP-RACK hollow section aluminium extrusion are interconnected using black ABS plastic corner
connectors. This requires only a rubber mallet or similar tool. The assembly is self aligning and can be further strengthened
by 'pop' riviting the overlapping flanges at each corner.
TYPE 46 RACK DATA:
RACK WIDTH 540mm. RACK DEPTH 460mm. 1U (RACK UNIT) =44.45mm OR 1 3/4".
FOR EQUIPMENT/PANEL CAPACITY DEDUCT 90mm FROM TOTAL RACK HEIGHT
ZIP-RACK FRAME
sro'.DEN[IliAPK
lll
1)
:; ~si~j
268
357
490
F04
F06
F09
II
C12
C15
C18
624
757
890
C24
C30
C40
1157
1424
1868
6/01BK
6/02
6/03
6/03BK
6/04
6/04BK
CT2
CT3
CT4
ZIP-RACK is made by "The ZIP-RACK Company" Factory 2, 7 Michellan Court, Bayswater Victoria 3153. Australia.
Under licence to AUTOTRON AUSTRALIA P/L. The ZIP-RACK Designers
Beware of illegitimate copies without the ZIP-RACK name. For product sales, wholesale pricing & product literature
Phone:
{03) 729 7255
or Fax:
{03) 729 6017
SERVICEMAN'S LOG - CTD
So the idea of mains-borne interference had to be discarded. In any case,
the interference pattern was nothing
like that normally caused by faulty
mains insulators and hardware.
Next up, he phoned Teac in Sydney
and asked their technical department
if they had any answers to the problem. He wasn't particularly pleased
when their only suggestion was that
he should check the antenna. I suppose that might be a typical first response to a question posed by an uninformed owner but it's hardly likely
to impress an experienced serviceman.
Anyway, the conversation continued for a time and eventually all concerned came to the conclusion that
the trouble must be in the tuner. So a
new tuner was duly delivered and
fitted but to no avail. The problem
continued.
Another call to Teac produced the
suggestion that perhaps it was a power
supply problem. Nobody could offer
a convincing explanation as to what
was happening but an offer by Teac to
investigate the supply was too good
to miss.
My colleague removed the supply
from the chassis and packed it off to
Sydney. But again it was all to no
avail because Teac could find no fault
of any kind. After this, the machine
sat on the bench for several weeks. In
the meantime, the owners had borrowed another machine. This machine worked perfectly
at their location and finally scotched any suggestion of local interference being the cause of
the trouble.
After that, each time
he came to town, my
colleague would ask
some of our mutual
friends if they had ever
seen the trouble. He had
a recording of the effect
which he would play to
anyone who showed the
slightest interest in the
problem. It almost
reached the point where
he took on the nickname
of "Fishnet". But nobody had ever seen or even heard of
the problem.
·
And so time passed and the machine sat gathering dust, until one
day my colleague was called to another job which really did involve
mains-borne interference.
In this case, the owner had obtained
a copy of the Department of Communications booklet on TV reception
problems. He was into self-help and
only called on my friend when he
reached the stage of needing professional help.
It was while browsing through the
DOC booklet that my friend came
across an idea that he felt might be
worth trying on the Teac recorder.
This suggestion involved winding a
TV receiver power cord around a ferrite rod to minimise mains-borne
interference into and out of the set.
Back at his workshop, he lost no
time in salvaging an antenna rod from
an old portable radio. It was about
13mm in diameter and 100mm long.
He wound the power lead onto the
rod, fitting as many turns as possible.
At switch on, he was amazed·to see
that all traces of the fishnet interference had disappeared. The picture on
all channels was as clean as one could
ever wish.
Just to make sure, he began to remove turns from the rod. The first
three of four turns made no difference
but then a faint trace of the fishnet
pattern became visible. And, as he
removed further turns, the pattern
became stronger and stronger, until it
had returned in all its annoying glory.
He repeated the experiment at the
owner's home and found that it
worked just as well there. So the machine was eventually set up with a
dozen or so turns of its power lead
wound around a ferrite rod, tucked
into the back of the video cabinet.
We have had long and detailed discussions on the problem and the most
likely answer is that it is a power
supply fault. Radiation from the recorder's switchmode supply is one
possibility, perhaps caused by the accidental omission of a suppression
component during assembly. Ferrite
beads are often fitted to the emitter
leads of chopper transistors for just
such suppression purposes , so this
suggestion is not entirely fanciful.
The problem is that the fault doesn't
exist in strong signal areas. This means
that it would never have been detected during factory tests, nor would
it have ever been detected if the machine had gone to a city-dwelling
owner.
At this point my colleague doesn't
feel like recalling the machine and
working over the power supply. If it
ever comes in again, he will look into
that possibility but for the time being
the makeshift mains filter solves the
problem.
Footnote
And now for a footnote. At a recent
meeting of the Tasmanian Division of
TETIA, the fishnet problem was raised
yet again. Discussion circulated ad
nauseum but one member joined in
with more than a little interest. It
turned out that he had an identical
machine with the same problem in
another low signal area.
Our colleague could not wait to get
home to try the ferrite rod exercise.
We've not yet heard whether it was
succei,sful.
So far, on the basis of just one unit,
the idea of a factory error appeals
strongly. However, it could be a design problem - the Teac MV 307 is a
fairly new model, so there may not be
too many of them in country areas
where weak signals . allow the problem to reveal itself. It will be interesting to hear if the fishnet turns up in
other places.
SC
MARCH
1993
47
|