This is only a preview of the December 1991 issue of Silicon Chip. You can view 49 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. Articles in this series:
Articles in this series:
Articles in this series:
Articles in this series:
|
. 'SLOG
SERVIC
When the ghost comes early
In the spiritual sense, ghosts are not usually
expected until after the main event; ie, after
someone has departed this life. It's usually the
same with TV ghosts. Normally displaced to the
right, they are caused by signals which arrive
after the main signal.
I'm not all that well versed in spiritual ghosts but I have heard stories,
from those who delve into such phenomena, about ghosts that appeared
earlier than normal (ie, while the person was still alive). Admittedly, even
the believers concede that this is very
rare - compared with normal ghostly
appearances, that is.
OK, so my cynicism is showing.
And I must admit that I tend to be
trifle cynical also about early TV
ghosts - leading ghosts, that is. Oh
yes , I know they do happen; I've seen
them but, like the spiritual early ghost,
they are quite rare.
So what causes a leading ghost?
Until now, I have seen only one
mechanism ; one where the TV signal
finds its way directly into the tuner.
Since the signal from the antenna takes
a finite time to travel through the
cable, it will arrive later than the signal picked u p directly by the tuner.
And , the longer the antenna cable, the
greater the ghost image displacement.
The TV set will normally lock onto
the stronger sync pulses so, if the
signal from the antenna is the stronger
(as it would be n9rmally), this signal
will produce the "main" visual image
and will be centred on the screen.
The weaker image then becomes the
ghost, whether it is leading or lagging.
I once saw a classic example of a
leading ghost. In TV's early days, a
colleague had been involved in installing a large distribution system in
a motel on Sydney's north shore, almost literally in the shadow of the
transmission towers.
The system had been run with coax
40
SILICON CHIP
cable throughout, so it was a shock
when all the sets produced leading
ghosts. The cause was eventually
traced to the sets' physical layout.
The tuners were mounted at the front
of the chassis, with about 30cm of
soon ribbon running to the antenna
terminals at the rear. And the ribbon
was the culprit; in the shock area, it
was picking up a substantial signal.
Shielded cable cured the problem.
Even rarer than the leading ghost is
a ghost generated by the TV set itself.
I am less familiar with these but a
colleague has encountered a few. The
exact mechanism is unclear but most
TETIA TV TIP
Philips GR100 chassis
Symptom: no sound. Picture is
perfect c:1.nd all other functions work
correctly but there is no response
from the volume control.
Cure: this chassis features a "hotel" mode which allows the hotelier to silence noisy sets in rooms
occupied by inconsiderate guests.
The function can sometimes be
triggered unintentionally by spikes
on the mains. The cure is easy.
Simply select Channel 38 with the
remote control, then press "Store"
and "Control +" together.
TETIA TV Tip is supplied by the
Tasmanian branch of the Electronic Technician's Institute of Australia. Contact Jim Lawler, 16
Adina St, Geilston Bay, 7015.
are caused by a fault in the IF strip.
The suggestion is that it is due to
some kind of frequency dependent
phase shift, whereby the higher frequency video signals (ie, the fine details of the image) reach the video
detector before the lower frequency
video signals and sync pulses.
Another cause experienced by this
colleague was a faulty luminance delay line, in an early colour set. The
delay line was an inductor about 12cm
long and 9mm diameter, over which
was a metal strip, forming one plate of
a distributed capacitor. The fault was
due to a faulty chassis connection to
this metal strip.
And what about an internally generated leading ghost? Well, nobody
seems to have heard of that one. Until
now.
The ghostly Mitsubishi ·
The set involved was a an AWA
Mitsubishi model SC6601, sold by a
dealer colleague about nine months
previously and thus still under warranty. The owner rang me first, complaining that the set had developed
bad ghosting. And it was clear that he
firmly believed that the set was at
fault.
Naturally, I was sceptical, not only
in the general sense but because of
the customer's location. There was a
multi-level department store being
erected at a nearby shopping centre,
with the usual assortment of tall cranes
reaching skywards. This had prompted a swag of complaints, both to me
and as letters to the local newspaper,
about interference to TV reception in
the area.
The construction company, while
sympathetic, could do little to help.
But they did advise that the cranes
would probably be gone in a few weeks
and that, hopefully, this would ease
the problem. So, on this basis, I advised the customer to wait and see
what happened. If the problem persisted, I would get one of my antenna
and I are pretty adept at such jobs, so
together we manhandled it into his
van and delivered it to the workshop.
Initial checks
lI
Q]
\
t
1
! ONC~ SAW A.. C.l.A4S"SI C.
eJ<AM'Pl-E OF ~ C-£ADIN6 GHOST":.,
installation colleagues to have a look
at the setup.
The next thing I heard, about a fortnight later, was from the dealer. Apparently the owner was becoming
impatient and had contacted him
about the problem. Fortunately, the
dealer was about to make a delivery
in that area, so he took the opportunity to call. And he was smart enough
to take another small set with him.
This check was most revealing, because the dealer's set produced a virtually ghost-free picture. And that
meant that it was an internally generated ghost, which put an entirely different complexion on things.
My dealer colleague suggested I visit
the customer's home, check his findings and perhaps fix it on the spot.
"No way", I replied. "No way am I
going to tackle an internal ghost problem in the customer's home. It's going
to be hard enough in the workshop.
We'll have to bring it in".
I used the word "we" advisedly
because I knew why my colleague
had suggested I go to the customer's
home. This is a 65cm set although, in
most cases, I can usually handle sets
of this size alone, albeit with a bit of
puff and grunt. But not this one. It
was housed in an enormous cabinet;
the kind that normally needs a couple
of furniture removalists to shift it.
Well, almost. In fact, my colleague
Two things had emerged from the
visit to the customer's home. One was
his discovery that the problem was
intermittent - which didn't cheer me
up at all. The other was that the two
images were displaced by about 1520mm and that they were of almost
equal intensity. I didn't even try to
rationalise this last effect at this stage;
time enough when I had it on the
bench.
So I set it up and fed in a signal
from the pattern generator, choosing a
simple one consisting of a circle, a
cross and a border. It has no cross
hatch and is ideal for observing ghosts
and also for centring the picture. Now,
with a steady pattern, I realised that
one signal was slightly more intense
than the other and had assumed the
main signal position at centre screen.
But the weaker image was to the left
of the main image; in other words, a
leading ghost.
This was a bit of a shock initially
but, on reflection, I decided that it
was probably of more academic interest than anything else. If the previously expounded theory was to be
believed, it simply meant that the high
video frequencies were arriving ahead
of the sync pulses. This might mean
something to the set's designer but it
meant little to me in practical terms.
In fact, I had very little to go on,
apart from the above secondhand experiences. I didn't even have a manual
for the set and so I phoned AWA and
was assured that one would be despatched immediately.
And they were as good as their word
- except that, somehow, the circuit
diagram had been omitted. It took another phone call to get that. Unfortunately, I found that it left a lot to the
imagination but more of that later.
Three suspects
By now I had mentally designated
three s-qspect areas: the tuner, the IF
strip and the luminance delay line. I
put the delay line at the bottom of the
list. According to my colleague, the
displacement he encountered due to
this cause was minimal and was more
akin to a bad case of ringing than the
marked displacement that I was observing.
DECEMBER1991
41
SERVICEMAN'S LOG - CTD
After that, I tended to favour the IF
strip but decided to eliminate the tuner
first, since it was easy to do. I simply
reset the generator to the video IF
(36.875MHz) and injected this signal
into the IF pack. The result was totally conclusive; the ghost was still
there and the tuner was cleared.
Unfortunately, there were a couple
of problems with the IF strip. The
circuit showed this simply as a box,
with 11 terminals, marked VIF-Pack.
And the chassis itself was just as coy;
the pack was a metal box, about 73 x
50 x 25mm, with 11 terminals. And it
was tucked in•between two vertical
boards on the lefthand side of the
main board.
Getting it out was more tedious than
difficult but it was eventually done.
When I opened the box I found it to be
a typical IF system; a transistor, a
SAW filter, an IC and a few discrete
components, all on a small board.
The intermittent nature of the fault
(which I had confirmed) suggested a
possible dry joint, and this was my
line of attack. I went over the board
with a glass and found not one but
three obvious dry joints, which I repaired. I then put everything back
together, refitted the box in the set,
rubbed my hands with satisfaction
(/)
.
R
42
The right track
I did this and found myself talking
to a technician who was obviously
familiar with the problem. As far as
he was concerned, I was right; the
problem was in the IF strip. Well, that
was some encouragement. Having
checked what I had already done, he
went on to suggest the joints of several other components.
One of these was the SAW filter
and another a small trim pot (the AGC
preset). In addition, there were three
other small components which, without a circuit, I couldn't identify. These
were mounted on the copper side of
the board and were probably small
capacitors. I would have liked to have
discussed the theory of the problem
in detail but there was little more that
he could tell me.
So I pulled the pack out again and
checked these components. The trimpot connections were above suspicion but, in any case, I doubted
whether it could be responsible for
this fault. The joints of the other three
small components were equally above
suspicion. That left the SAW filter.
The SAW filter has
five connections, two
being to the chassis.
And, having been
alerted, a more de tailed check did suggest that some of them
were rather "iffy". I
resoldered them all,
checked the rest of the
board again, and refitted the pack.
I put the set through
another series of exhaustive bench tests
and it performed faultlessly. After about a
week, I concluded that
the ghost had been
finally laid and so the
set was returned to
the
customer. A recent
•
check, two month's
later, revealed that all
oooAN'D INOUl-'D '<OIJ 'B~Uf:\JE. \ , ; 11.\G:'
was well.
However, the phrase
ONLY 'PAN\At:>~ WP.S ~ T\N'< C.\·\\? OF\= mE: CA'e>\Nf;"f"ooo
~!!{p
(fi;
and switched on. You beaut! - no
ghost.
I let it run for several days, switching it off and on occasionally, and it
didn't miss a beat. But then I left it off
for a couple days and, when I switched
it on again, there was the ghost - but
for about 30 seconds only.
I ran it for several more days, switching it on and off, thumping and otherwise abusing it. No ghost. I left it off
for a couple of days and tried again.
And again, the ghost appeared for 30
seconds at switch on but nothing
more.
At this point, I decided that, rather
than waste more time, it would be
better to fit a new IF pack, seeing that
the set was under warranty. So I rang
the company's spare parts department
to order a replacement pack. They
were out of stock but expected new
stocks in about two weeks.
Well, that was OK, the dealer had
loaned the customer another set and a
couple more weeks shouldn't matter.
But the two weeks dragged on into
four and when I rang I was told that
there had been a delay in supplies
from Japan and that it could be several more weeks - or even longer before stocks arrived.
The company was both apologetic
and helpful, suggesting that, if the
worst came to the worst, they would
"borrow" a pack from another set. But
they suggested that I first talk to the
service department, in case there was
still a chance of repairing the existing
unit.
. .,,
SILICO N CHIP
"returned to the customer" in the previous paragraph has a deeper meaning than its simplicity implies. It was
a near disaster.
I've already mentioned the bulk of
this set and the need to have two
people to move it. Well, when it came
time to return it, the dealer who had
helped me bring it in could not immediately spare the time to help again. I
explained the situation to the owner
who was quite happy to help at his
end, so I shanghaied a colleague into
helping at my end.
The snag was, the owner had his
own ideas on carrying the set. He
insisted on going first - which was
OK by me - but also on walking forwards, supporting the set behind his
back.
Years of experience have taught me
that this is a risky practice; it is simply not possible to get as secure a grip
this way as it is when facing the load.
Granted, one needs to make sure that
there is a clear path and to check out
steps, doorways and other obstructions but, on balance, facing the load
is the safest way. I suggested I go first
but this was ruled out.
So we started up the concrete path
to the one step leading to the house.
The path was smooth and level but
the edges had been finished with a
raised roll, about 75mm high. All went
well until he reached the step at which
point, for some unexplained reason,
he stumbled.
What happened next was all over
in milliseconds. He lost his grip on
the left side of the set - the front - so
the face of the picture tube was heading for the raised edge of the path. I
suppose it was pure instinct which
made me do what I did, because there
was no time to think.
I lifted the right hand side of the set
with all the force my arm could muster. And it worked; I rolled the set
through 180° and also pushed it beyond the path. It landed on the top of
the cabinet, legs ·in the air, on the
adjacent grass.
And would you believe it; the only
damage was a tiny chip off the cabinet veneer at one corner. There was
no internal damage. But it was too
close to an oops for my liking.
A genuine oops
And finally, here's a genuine oops.
The first part of this story occurred
long before I came on the scene but
this is how I pieced it together.
It concerns an up-market stereo
amplifier, a Technics model SU-V460
made by Matsushita. This amplifier is
rated at up to BOW per channel and
has all the usual bells and whistles
dear to the heart of the audiophile at a
price somewhere between $700 and
$800.
It had been sold by my dealer colleague only about 18 months previously and, by all accounts, the customer was very happy with it. Or at
least he was until the night of the
party. The details of the party are a bit
vague - nobody was saying much,
even if they could remember, which
is probably understandable in the circumstances.
However, there was a broad hint
that the party was what one might
term an unofficial one - that the owner
and his wife were away and "while
the cat's away, the mice will play" the "mice" being the teenage offspring,
mates and girlfriends. Anyway, someone managed to overturn a glass of
drink on top of the amplifier. After
which it didn't go.
And so the owner took it back to my
dealer friend, sheepishly confessed
that something had been spilt into it,
and asked what could be done about
it. My friend removed the cover, took
one sniff at the inside, and nominated
Coca Cola.
He then suggested that he should
pass it over to me, as an authorised
Panasonic Service Centre. But no; the
owner insisted that it be returned directly to Panasonic, to ensure that the
job "was done properly". (Bit of a
blow to my ego, that).
And so this was done. But it received short shrift at Panasonic. It
came back with a brief note saying
that the unit was not worth repairing;
it would be cheaper to buy a new one.
Naturally, the owner wasn't very
happy but at least he accepted the
diagnosis. And, in retrospect, I'm
thankful that it was Panasonic that
delivered the verdict, rather than myself. I wonder whether he would have
taken my word as readily as he did
their's. And that softened the blow to
my ego somewhat.
It was at this point that I first learned
of these happenings, mainly because
my dealer friend was stuck with the
relic, and was about to send it to the
tip, when it occurred to him that it
might be of some use to me. He was
TV TEST EQUIPMENT
(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
* * * * * * * * *
DEGAUSSING WAND
Strong magnetic field ,
larger than usual coil with
multicore centre. Double
insulated for your safety.
Also fitted with momentary on/off switch. 240V
AC 2.2 amps.
Just about as important as having a soldering
iron in your toolbox!
$75.00 + $10.00 p&p
* * * * * * * * *
HIGH-VOLTAGE PROBE
Built-in meter
reads positive or negative voltages
from 0-50kV.
For checking EHT and focus as well as any
other hi-tension voltages.
$98.00 + $5.00 p&p
LOW-VOLTAGE PROBE
Ideal for checking microwave ovens & TVs.
The ranges are 0-5kV ·negative & 0-10kV
positive. Double insulated for safety .
$84.00 + $5.00 p&p
* * * * * * * * *
REMOTE
CONTROL
TESTER
(INFRARED OR
ULTRASONIC)
Designed to test inlrared or ultrasonic
control units. With the extension lead you
can also test inlrared units which cannot
be placed in front of the testing unit.
Requires a 9V battery. Output is via a LED
and piezo speaker.
$84.00 ~ $4.00 p&p
. ·.· .·
·,
♦', >;:..,
/
· ~- TUNER REPAIRS ./..r.
~ .
From $19 exchange
~
.
-
plus p&p ...._ ~
- ✓ P,~.~~---
Cheque, Money Order, Visa Card
Bankcard or Mastercard
216 Canterbury Road,
Revesby, NSW 2212.
"'2:r
(02) 774 1154
.·-...!!!!(02) 774 1154
DECEMBER 1991
43
SERVICEMAN'S LOG - CTD
Fig.1: general layout of the
SU-V460 amplifier, showing
the output stage heatsink,
with the amplifier output IC
(SV13205) on the front of it.
It was in this area that the
Coca Cola landed.
aware that I had been thinking about
such a piece of gear and that I might
be able to salvage if I could afford to
put some time into it.
The idea intrigued me too. Without
for one moment questioning Panasonic's diagnosis, I felt there was a
fair chance that it might be possible to
restore the amplifier for a modest
amount, in my own time. And I already had a manual for it.
And so it finished up in the workshop. I pulled the covers off and began assessing the damage. It was a
mixed bag; there was not much to be
seen at first glance but closer inspection soon revealed the details.
This unit has a large main board
(about 300 x 220mm) sitting on the
bottom of the cabinet. Among other
things, it carries the driver stages and
the output IC for the power amplifiers. This output IC is mounted on a
large finned heatsink which in turn is
mounted above a central cut-out area
measuring about 200 x 90mm. And it
was on this heatsink and the surrounding board that most of the Coke appeared to have landed.
The preamplifier and frequency
control stages, along with the switches
and other controls, are on a separate
front panel board. They appeared to
have escaped serious contamination.
Vaporised links
The output IC - IC501, SV13205 connects to 18 terminal pads on the
board, along one edge of the cutout
area. And this was where I first found
obvious - and quite startling - dam44
SILICON CHIP
age. Four of these terminal pads, 3, 4,
5 & 10, are connected to other parts of
the board via short wire links. In fact,
pins 4 & 5 are connected to 45V rails,
while pin 10 connects to a 15V rail.
Or, more correctly, they should have
been so connected. In fact, some links
no longer existed; they had been completely vaporised - a dramatic indication of just how violent the reaction
had been. There were other links
nearby which had also suffered, although not so drastically; these had
simply melted, leaving most the wire
in place.
From here, I turned my attention to
a group of 16 transistors on the main
board - Q451 to Q465 - which make
up the amplifier input and driver
stages. There had been some violent
reactions here also. Some of the transistors had been literally blown to
pieces; all that was left were the legs
with a few scraps of black encapsulation clinging to them.
Of those physically intact, an insitu meter test indicated that most
were faulty, while the others were
doubtful; which meant that they
would all have to be replaced. And,
iri the same area, there were a number
ofresistors which had obviously been
badly overloaded.
At this point I decided to make a
rough estimate as to the cost ofreplacing those components that were obviously faulty. A quick check with
Panasonic revealed that the output IC
(IC501) would cost well over $100.
Add to that the 16 transistors, sundry
resistors known to be damaged, plus
allowance for components not yet
checked, and there wouldn't be much
change out of $200.
Well, OK, an $800 amplifier for $200
(plus the labour ofreplacing the faulty
bits) doesn't sound such a bad proposition. But was that all that was involved? On closer inspection, I real 7
ised that it was not. Between the 18
pads and tracks feeding the output IC,
I discovered considerable carbonising of the board.
Now, it is usually possible to correct this situation by gouging out the
charred material and is always worth
a try where it is only minor and there
is little other expense involved. As I
say, it usually works; but not always.
But this situation was quite different. The charring was extensive and
the. space between tracks quite limited. And it would be impossible to
know whether the work had been successful until I had bought and fitted
some $200 worth of components.
And this was only on the basis of
the known faulty components. There
would almost certainly be more and,
in particular, I was worried about those
on the front panel board. These included three ICs, various switches and
potentiometers, plus a swag of minor
components, any of which could have
been damaged in one way or another.
This would add to the cost but,
even if this was only minor, there
would also be the time needed to track
them down, which could be quite
tricky. And if one of them was intermittent, it could take a considerable
amount of time to track down.
The upshot of all this was that I had
to shake my head and push the thing
aside. What had initially looked like
an attractive idea was simply not a
practical proposition.
I have recounted this story for a
couple of reasons. The most obvious
one, of course, is to emphasise that
electronic devices and beverages do
not mix. Such devices should not be
located where people - and particularly inebriated party guests - are
tempted to use them as a table.
The other point is to vindicate the
original diagnosis by Panasonic. Customers often accuse service organisations of taking the easy, and more
profitable, course of writing off a device, when a repair really is economically possible.
Well, perhaps it does happen. But
not with reputable organisations. SC
|