This is only a preview of the June 1996 issue of Silicon Chip. You can view 23 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 "A High-Performance Stereo Simulator":
Items relevant to "A Low Ohms Tester For Your DMM":
Articles in this series:
Items relevant to "Automatic 10-Amp Battery Charger":
|
SERVICEMAN'S LOG
Chuck it away and buy a new one
No, that’s not my advice but it is the philosophy
from one of this month’s stories – a story from
the USA, where the servicing scene is very
different. Even so, it has a very interesting
connection with the local scene.
My first story concerns a National
Panasonic model TC-1407, 34cm
colour set using an M12H chassis.
This model can be anything up to 10
years old and is a very well made and
reliable set, regarded by many as one
of the best that National ever made.
But of course faults do occur and,
in this case, the complaint was total
lack of colour. This is not an unusual
fault in itself but the actual cause was
unusual, as we shall see.
When tackling colour problems, I
automatically reach for the CRO leads.
And the first thing I checked was
whether a colour signal was coming
into the chroma decoder (IC601) – see
Fig.1. In fact it was, on pin 7. It was
only about 0.8V p-p but this tallied
with the circuit.
Next, I checked the crystal oscillator. Again there was no problem, with
plenty of 4.43MHz signal on pin 16 of
the same IC.
Well, that ruled out the more obvious possibilities. The next thing
to check was the gating pulses (horizontally derived) which control the
burst gate and similar functions. Such
pulses should appear at the burst gate
terminal (pin 14) of IC601 and are
shown on the circuit as waveform 30,
with a p-p amplitude of 4.6V. But not
in this set – pin 14 was dead.
Well, at least I was on the right
track. Unfortunately, the track wasn’t
very clear, which is just another way
of saying that the best circuit diagram
I could find left much to be desired.
It wouldn’t have been so bad if I had
been able find an original circuit but
the best I had was a much copied copy
of a copy – if you follow me.
To be fair, the original circuit was
undoubtedly very good, featuring lots
of information in the form of voltages
and waveforms, but it had been much
reduced and, when copied, a lot of fine
detail was lost.
Anyway, I was faced with the task of
tracing the circuit to find where these
pulses originated and at what point
they were lost. And I thought I had
cracked it in one as soon as I started.
Not far from, and to the left of, pin 14
is a diode, D602, which connects to
chassis. Naturally, I checked it and it
was a dead short. Unfortunately, my
jubilation was short lived because
when I replaced it and switched on
there was still no colour.
A literal drawing board
So, back to the drawing board –almost literally. The line split into two
at this point, both moving parallel
down the page for a short distance.
Then one turned left and one continued down. I turned left and finished up on pin 10 of the video chip
(IC301). There was supposed to be
a waveform here also (designat
ed
waveform 21) which was similar to
the one on pin 14.
But there wasn’t. Fairly obviously,
pin 10 was supposed to receive this
waveform, not supply it. I followed
the other line down until it spilt,
going left and right. I went right, but
drew another blank. It finished up at
the base of transistor Q601, a blanking
pulse generator – which didn’t have
any pulses either.
OK, back to the junction and turn
54 Silicon Chip
Fig.1: the chroma decoding circuitry in the National TC-1407. IC601 is at top right, IC301 at top left, and IC501
at bottom centre. Many of the IC pin numbers are quite difficult to read.
left. This brought me to pin 16 of the
jungle chip, IC501. And this, I felt, had
to be the source of the missing waveform. This chip performs a whole host
of functions, including horizontal AFC
and sync separation, both requiring a
reference to pulses from the horizontal
output transformer. In fact, pin 16 was
connected to the sync separator block
within the IC.
What’s more, this IC appeared to
be performing these, and its other
functions, because we had a perfectly
locked picture but in monochrome.
So it had to be receiving pulses from
the horizontal transformer. In fact,
after much more laborious cir
cuit
tracing, too complex to detail here, I
confirmed that the pulses at pin 2 of
the horizontal transformer were applied to pin 1 of IC501. And the CRO
confirmed that the pulses here were
as they should be.
So why wasn’t it delivering a pulse
at pin 16?
I could only conclude that there was
a fault in the IC. And, since I didn’t
have any on hand, I had to order one.
But while waiting for it, I still had
an urge to confirm that this was the
problem. Referring again to IC301,
I noticed that on pin 11 there was a
waveform (No.24) of similar shape
and amplitude to the waveform that
should have been present on pin 10,
though somewhat smoother.
So what would happen if I connected pins 10 and 11 together. At
best I might get some kind of colour
response. At worst, I could blow up
IC301. After studying the voltages on
the two pins (1.2V on pin 11 and 0V
on pin 10) I decided that the risk to
the IC was minimal and connected the
two pins together.
And it worked – well, partly. It did
produce colour on the screen but it
was not locked, drifting through the
spectrum and producing some weird
coloured scenes. But it was enough to
suggest that my diagnosis was probably correct.
And in fact it was. When the new
IC arrived, I fitted it and everything
came back to normal. Another satisfied
customer.
The American in-laws
And now for a change of scene; quite
a big change in fact, because my next
story comes from the USA. But it also
has a very close relationship with some
of my previous notes and one story in
particular.
By way of background, my regular
readers may recall that, from time
to time, I have featured stories from
a colleague who worked down the
south coast of NSW. And I was always
pleased to feature these stories because they involved factors peculiar
to the area; UHF almost exclusively,
long distances in many cases, and
much hilly terrain. And, of course,
we exchanged technical experiences
and picked each other’s brains from
time to time.
So it came as rather a shock when,
some months ago, my colleague
announced that he had decided to
retire and move to the US where he
June 1996 55
being that muggins, “who knows all
about TV sets”, could probably fix it.
So, it finished up on my doorstep.
I’m afraid I agreed to the idea with
mixed feelings. On the one hand I had
brought all my test instruments and
tools with me, had organised some
workshop space, set up a bench, and
begun to sort things out.
However, it was all very well for
the rest of the family to assume that
muggins “knows all about TV sets”.
The truth was that all I knew about
projection TV sets was secondhand
and did not even involve the same
model. Nor did I have a manual, have
any idea of where to find one, what it
would cost if I did, and whether such
an outlay could be justified.
I short, I would have to fly by the
seat of my pants.
A bright spot
had various in-laws and other family
connections. So, no more stories from
that source.
But do TV servicemen ever retire
completely?
Significantly, my colleague shipped
all his equipment, which was considerable, to the US with him. He had no
intention of setting up in business but,
I imagine, he knew he would feel lost
without the means to look after his
own devices and ap
pliances, along
with those of his relatives and friends.
So that is one part of the background. For the other part I would refer
readers to one of my own stories which
appeared in the May 1995 notes under
the heading “All it needs is a new
fuse”. In greater detail it concerned
a Mitsubishi VS-360A projection TV
set and the difficulty of convincing the
owner that it needed a lot more than
a new fuse.
Linking all this together is the fact
that my colleague was “in” on that
story from the beginning. Neither of us
had tackled projection TV sets before
and he was anxious to learn all he
could from what I had to learn. It was,
therefore, sheer coincidence that one
of the first family jobs he encountered
after settling into his new home involved a Mitsubishi projection TV set.
56 Silicon Chip
Anyway, here is my colleague’s story
as he tells it.
The set involved was a Mitsubishi
VS-405R projection type, featuring a
100cm (40in) screen. It was about 11
years old, roughly the same age as
the one my colleague had dealt with
in Australia. And it came to me by a
somewhat round about route.
It had originally belonged to a
friend of one of my in-laws and had
failed some time before I came on the
scene. The owner had called in a local
serviceman who repaired it for $170
(I’m quoting $US, of course). It was
quite a reasonable charge and the set
performed perfectly.
Unfortunately, it did not perform
for long, failing again after about six
months. And this time the owner did
what so many people do in this country when something fails (particularly
something 11 years old which has
failed for the second time) – he chuck
ed it away and bought a new one.
And, as a matter of interest, the new
set – a larger 125cm (50in) model – cost
about $1200, roughly half the price of
a similar set in Australia.
At the same time the old set was
“chucked” in the direction of my inlaws – not literally I hasten to add, because the thing weighs a ton – the idea
One bright spot was that the device appeared to be fairly easy to get
at; much easier than was apparently
the case with the one my Australian
colleague worked on, due to a slightly differ
ent layout. One important
difference was that the top part of the
cabinet back could be removed, as
well as the lower part, giving much
better access.
In addition, the 3-gun assembly was
mounted on a steel subframe supported on runners on the inside of the
cabinet. This allowed the subframe to
be withdrawn, although not without
some difficulty – more on that later.
The whole system was made up
from a collection of PC boards, each
mounted on a light metal frame which
served as a nominal chassis.
These included a power supply
board, a horizontal and vertical scan
board, a signal processing board, a
stereo sound board, and a convergence control board. The scan and
signal boards were secured to the
floor of the cabinet while the remaining boards were secured to the sides.
They could all be easily unplugged
and withdrawn.
That much established it was
time to apply power and see what
happened. The answer was simple –
nothing. This lead me to a 3A fuse in
the power supply which had failed.
This was replaced and power applied
again.
This time there were signs of life.
The set tried to fire up but then would
shut itself down and try again. In
other words, there was a slow hiccup,
suggesting that a protection circuit
somewhere was taking over. I suspect
that the set had been inadvertently left
on in the hiccup condition which, if
it continues long enough, can blow
a fuse.
I went straight to the horizontal
output transistor, which was readily
accessible, and picked it in one; it was
shot. This was easily fixed. I didn’t
have a direct replacement type but
settled for a 2SD380.
Unfortunately, when I switched on,
the set was still hiccuping. I found a
HT rail check point and monitored it.
It looked as though it was about 125V
but this called for some judgement as
it rose and fell.
I suspected that the fault was either
somewhere in the horizontal scanning
circuit or in the protection circuit
covering this section.
Normally, I would check this second
possibility by momentarily disabling
the protection circuit and noting what
happened. The trouble was, without a
suitable diagram, I had no idea where
to look for this circuit, so I put that
on hold.
I was also curious as to the nature
of the previous fault and what work
had been done. I could see that the
boards had been pulled because the
wiring looms had been released from
their clips and, although most had
been restored, a few had not.
I examined the boards very carefully, in search of a clue, and finally
concluded that no work had been done
on them. One obvious indication was
a fine layer of dust, as normally found
in such situations, which had not been
disturbed.
At the same time, I went over each
board and checked for dry joints, particularly around the horizontal scan
circuitry and the four pin connections
to the horizontal driver transformer.
Dry joints to these pins have been a
common problem with many sets in
the past. In this case, the soldering
quality was very good. I did remake a
couple of joints which were vaguely
suspicious but it was more of a gesture
than anything.
Why the failure?
At this stage, I began to wonder
why the horizontal output transistor
had failed. I also wondered if there
had been a previous failure and if the
faulty transistor I had replaced had
itself been the correct type number?
However, without a circuit I really had
no way of knowing.
Finally, having checked the most
likely possibilities as I far as I could,
there seemed to be only one positive
check left that I could make; a check
for shorted turns in the deflection
coils. Fortunately, I had brought my
trusty shorted turns tester with me
and unearthed it after some searching.
The deflection coils were plugged
in, so it was easy to make the checks
without pulling the metal frame
assembly. I checked the horizontal
windings first and the first two tested
OK. But not so the third one; there was
a clear indication of a short.
So now I had to pull the metal
frame assembly. This wasn’t quite as
easy as it looked. To understand why
it will help if I describe the device in
greater detail.
Imagine a rectangular metal frame
running the width of the cabinet and
sitting horizontally on two runners,
one on each side of the cabinet. This
frame extends from the back of the
cabinet to about two thirds the way
to the front. A second frame is then
attached at right angles to the front of
the first frame and this extends downwards to the floor of the cabinet. This
supports the picture tubes and some
associated circuitry.
OK, having envisaged all that, consider how the frame is constructed. It is
made from box section mild steel –actually two lengths of angle iron welded
together to make the box section. All
of which adds up to a lot of steel. Add
the weight of the three picture tubes
– they may be small, but they’re not
light – plus a few odd pieces and you
have a total weight of around 25kg.
It’s not at all easy to manhandle in an
enclosed space.
Nor was the operation made any
easier by the location of the picture
tubes. These are mounted at about 45
degrees, below the level of the horizontal section so that, if it were simply
removed and placed on a bench, the
whole assembly would be resting on
the tube necks. Ouch!
I tackled the problem by arranging
some suitable blocks on the bench.
I then unclipped all the connecting
leads and carefully manhandled the
frame out and onto the blocks.
It was worth the effort. With
everything out in the open it was
immediately obvious why the set had
failed in the first place – and why it
had failed in the second place. In fact,
one doesn’t often get an explanation
presented as clearly and positively
as this one. The faulty deflection coil
assembly was coated with a brown
varnish, the appearance of which
exactly fitted the description of the
coating on the failed transformer in
the set handled by my Australian
colleague.
Even without any other warning I
would have been highly suspicious
of this mixture. I had encountered a
similar witch’s brew before – sometimes brown, sometimes yellow – in
other makes of sets. It is sometimes
used as a varnish on windings, and
sometimes as a glue to secure an extra
component on the copper side of a PC
board. More particularly, I was well
aware of its corrosive properties. It will
eventually eat away any copper with
which it comes in contact.
In this case, of course, I had been
June 1996 57
Serviceman’s Log – continued
warned. I recalled that the technician
in the Australian Mitsubishi service
department had advised my colleague
to check for this varnish on the trans
former and for any damage it might
have caused. The rest is history; that
transformer was a write-off.
There was one other interesting development. One of the three deflection
coils – presumably fitted during the
previous service –was quite free of the
witch’s brew. Perhaps a message had
finally penetrated.
But that was all rather academic
from my point of view. My more immediate concern was the fate of this set.
Would I be justified in pressing on with
the job or should we cut our losses?
There had been a tentative agreement
before I started that we would put a
limit of around $100 on the cost of
replacement parts. Labour, of course,
would be no more than an extra serving of turkey at next thanksgiving – if
I was lucky!
I had determined that a new deflection assembly would cost about $100
58 Silicon Chip
which, with a new horizontal output
transistor, was already stretching this
limit. To that would need to be added
a second scan coil assembly, because it
would be pointless to simply replace
the one faulty one. The third coil was
another failure just waiting to happen.
On top of that, how much more of this
brown varnish was there elsewhere in
the system?
So we were looking at $200 plus to
repair an 11-year old set, which could
be replaced with a similar size modern
set for around $1000. I am well aware
that, in Australia, many people would
regard such an opportunity as a gift
and be prepared to take the risk.
Not so in this country. Because appliances are so much cheaper, in real
terms, than in countries like Australia,
the concept of service is very much
different. Many people will simply
discard an appliance at the first sign
of trouble, without any attempt to
determine whether the fault is minor
or major, and whether a repair might
be worthwhile.
Against this background, and on
my advice as to what would be involved, the decision was made to cut
the losses, such as they were ( mainly
my time, which nobody regarded as
being very important). Not that this
really worried me very much; it had
been an interesting exercise and an
opportunity to study another version
of this type of set.
The cabinet of the old set was salvaged, however. It was a very useful
piece of furniture and one member of
the family had it earmarked as a useful
storage unit.
I suppose the most interesting
aspect of the story, from a technical
viewpoint, was the coincidence of
finding two sets, of the same make, so
far apart in different countries, suffering from almost identical faults. Also,
it is to be hoped that, at long last, the
havoc caused by these witches’ brews
has been recognised and that they
will be suitably disposed of – with
due regard to the environment, of
course.
Well that’s my colleague’s story; and
a very interesting insight it is into the
TV scene in another country. Thanks
SC
mate.
|