This is only a preview of the February 1997 issue of Silicon Chip. You can view 25 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 "Computer Controlled Dual Power Supply; Pt.2":
Items relevant to "Control Panel For Multiple Smoke Alarms; Pt.2":
Articles in this series:
Articles in this series:
Articles in this series:
Purchase a printed copy of this issue for $10.00. |
SERVICEMAN'S LOG
A tale of two Sharp VCRs
VCRs can fail for all sorts of reasons but I
recently had one that really takes the cake.
Fortunately, not all jobs are like that one,
with most being quite routine.
Just when I’d thought I’d seen every
thing in the servicing game, along
comes something to really set me back
on my heels. Spilt drinks or other
liquids are common reasons for TV
and VCR failures but monkey urine?
–you’ve got to be kidding!
Of course, out of the thousands of
servicemen in Australia, it had to be
yours truly that got saddled with the
job.
The story started out innocently
enough. Some months ago, the local
vet brought in a mid-drive Sharp VCR
with the complaint that it stopped
working after about two seconds on
“play”.
The set was a 1992 VCA34X which
looked to be in good condition – at
least from the outside. Unfortunate
ly, it didn’t smell quite so good and
had a quite distinct “pong” of stale
urine about it. However, seeing that
it had come from the animal surgery,
I assumed that this “pong’ had been
picked up from something in the air.
The fault description turned out to
be quite accurate and when I removed
the top cover, I could see that fast
forward and rewind were OK. The
play mode was a different matter,
however – the arms loaded the tape
properly, the drum motor started and
the capstan motor started but it only
did a revolution or two before stopping
and unloading the tape. It all looked
OK, so why didn’t it work?
Unfortunately, I don’t have the
service manual for this model but the
deck was a very common type; only the
electronics were different. So the first
question was “is this a mechanical or
an electronic problem?” I decided to
inspect the whole machine carefully
30 Silicon Chip
and turned it upside-down to remove
the lower cover plate. Immediately, it
was obvious from the severe corrosion
where the smell was coming from. It
was also obvious that this was where
the problem lay.
Fairly obviously, the machine had
been sitting in a pool of urine, perhaps
up to 12mm deep. And although this
had long been cleaned off by someone
else, the corrosion was abundant to
see up to the high level mark. I was
about to phone the vet and tell him
that all was lost and that he should
get a new VCR but then I had a little
think to myself.
To fix or not to fix
Perhaps this wasn’t a hopeless case
after all. First, all the major motors
were actually turning and secondly,
someone else had already done a pret
ty good job of cleaning up the mess
inside. Added to that, all the bottom
printed circuit boards looked OK and
so I concluded that the job was worth
investigating further to see if the ma
chine could be salvaged.
At this stage, my main suspect was
the drum motor as it seemed hesitant
to start and its speed appeared to be
intermittent when it was strobed under
a neon mains light. I also discovered
that, occasionally, when the video was
switched from EE mode (ie, Tuner) to
the Play mode, a picture could almost
be seen although not always in sync.
My next step was to fire up the CRO
and check the output of the PG (pulse
generator) head. This revealed a pulse
generator signal but it was fluctuating.
Also there was 12V on pin 3 of the plug
to the drum motor.
From this evidence, it seemed likely
that the drum stator board or its com
ponents had corroded and so I decided
to remove it to see if anything could
be done.
The connection to this motor is via
a special 6-lead flat printed circuit
ribbon cable harness whose end is
just pushed into a receptacle. When I
removed this, I found it to be badly cor
roded. I cut about 12mm off, scraped
away the white varnish covering the
tinned strands, cleaned the socket and
reinserted the harness.
When I powered up and pressed
“Play”, the motor spun up quickly
and continued to play correctly. I then
checked the other sockets in this area
but they were all OK. In hindsight, I
should have suspected a cable connec
tion but everything looked fine until
the cable was actually removed.
I returned the video to the vet with
a warning that despite cleaning it, he
would probably have further problems
in the long term as it is very hard to
prevent a chemical reaction of this
type from continuing.
He apologised for the state of the
VCR – apparently a monkey (would
you believe it) had had an “accident”
in that corner of the room and even
though he had got a friend to check
the VCR out, he must have missed that
particular socket.
Just how long the machine will last
is anybody’s guess and I didn’t press
the vet for any further details on the
monkey or what it was doing there.
Let’s just say that it doesn’t pay to
have a monkey monkeying around
near a VCR.
Another Sharp VCR
I thought no more about this repair
until a few months later when another
similar Sharp VCR came in. This time
it was a slightly older model, a VCA105X, and the complaint was that
it wouldn’t play. The unit looked in
good condition throughout and this
time there were no unpleasant odours!
As with the previous unit, this too
would load up, the drum motor would
spin and the capstan motor would
spin a few revolutions before it would
unload and stop. This time, I did have
the service manual and it also had a
troubleshooting guide for the exact
symptoms being experienced.
The first step is to check the head
switching pulse applied to pin 3 of
IC801 (IXO491) and the PG (pulse gen
erator) signal applied to pin 4 of servo
control IC701 (IX04313GE). Well,
the PG pulse was there but there was
definitely no head switching pulse
(HSWP) . In fact, all the voltages and
inputs seemed correct going into IC701
but nothing was coming out of pin
28 and there appeared to be no short
circuit on that line.
In view of these symptoms, I felt that
the problem had to be electronic and,
at this stage, considered IC701 and/
or the system control microprocessor
IC801 as the main suspects. Before
replacing these devices however, I first
checked out all the B+ lines and the
various clock signals but could find
nothing wrong. The HSWP also went
to the Y/C module and onto the head
preamp board. I tried disconnecting
these boards in turn but there was still
no sign of a pulse.
With reluctance, I ordered IC701
first as I knew these ICs would be ex
pensive. It arrived a few days later and
I wasted no time in fitting it. Unfortu
nately, it made no difference, so that
was expensive mistake number one.
I was now faced with the prospect of
having to replace the main micropro
cessor (IC801), which was even more
expensive. Before ordering it however,
I tried replacing the mode select switch
in case it wasn’t engaging quite correct
ly – to no avail. Finally, I went ahead
and ordered IC801. It too arrived after
a couple of days and I went about the
laborious task of removing the old 64pin IC and soldering in the new one.
And that was expensive mistake
number two because it also made ab
solutely no difference – there was still
no head switching pulse. Murphy was
really working overtime on this job!
A new approach
I was discussing my expensive folly
with several colleagues when one said
that he had recently obtained the same
model as a trade-in.
What’s more, he offered to lend it
to me so that I could pinpoint the
location of the faulty parts without
spending any more megabucks on use
less guesses. He was also of the strong
opinion that it was an intermittent
capstan motor that was causing the
problem as this occasionally occurs
February 1997 31
Serviceman’s Log – continued
on this series of decks and gives many
similar symptoms.
Well, this was indeed a stroke of
luck and I started by swapping over
the capstan motor but that too made
no difference. By now, I felt that my
original diagnosis – that the fault
was electronic – must be correct.
As a result, I began swapping all the
electronic circuitry between the two
machines, board by board, but again
this made no difference.
After changing the last board, I was
forced to conclude that I was looking
at a mechanical/motor problem. In
fact, it had to be the drum motor, even
though all the pulses and voltages
from it were correct and it looked as
though it was reaching the correct
speed when viewed under the strobe
light. I couldn’t be certain of this, how
ever, as it was unloading as soon as it
apparently reached the correct speed.
In fact, the machine did not even have
time to switch from EE mode to Play
mode, so no picture was available.
Anyway, I proceeded to replace the
drum motor and try again. Unbelieva
bly, it worked this time but I couldn’t
understand why.
Out of curiosity, I measured the
output from the new motor and it
matched the old one exactly. How
could this be? To solve this mystery,
I replaced the old motor and in the
process noticed that the ribbon con
nector cable had come away from its
32 Silicon Chip
hardened plastic support and that
the tracks were somewhat loose and
frayed. I didn’t pay much attention to
this until I retried the old motor which,
to my amazement, was now working.
So what was the answer? As it is not
possible to prove, I can only speculate
that the ribbon connector cable was
either shorting or not making a proper
connection with the socket. I guess I
should have remembered the symp
toms of the earlier repair and examined
this connection more carefully first.
Bread & butter
The next morning, I faced up to
two TV sets that were awaiting my
attention. The first was a 34cm Toshiba
144R8A made by Samsung (a P54S)
and it was quite dead. Often, a set
of this size is not really worthwhile
repairing as they are so cheap to buy.
It all depends on how difficult the
fault is to fix and sometimes it can be
quite difficult to decide whether to go
ahead or not.
On the plus side, the faults in this
chassis are fairly well known which
does reduce the amount of time spent
in diagnosis.
In the 34cm model, most of the
problems are in the power supply, due
to the electrolytic capacitors drying
out. This causes the chopper power
transistor Q801 (2SC3552/BU508A)
in the switchmode power supply to
blow, which also takes out the fuse
and/or R801 (5.6W 7W).
In this case, the transistor had tak
en out the resistor and so I replaced
these parts and four electros (C808,
C813, C812 & C811) all at once. When
I switched it on again, the set was ob
viously struggling to fire up and was
making funny noises in the power
supply. I immediately switched it off
again to prevent another failure of the
chopper transistor and then started to
check for shorts on the B+ rails.
The line transistor Q404 and capaci
tor C413, a common culprit, were both
OK. In fact there were no shorts and
no, or extremely small, output from
the chopper transformer.
Checking the voltage across C807
confirmed there was 340V out of the
bridge rectifier and I checked and
cleared resistors R806 and R807. The
oscilloscope confirmed that the circuit
was oscillating though the waveforms
were incorrect and Q801 was getting
hot. I changed IC801 which is the main
IC in the power supply but that made
no difference.
And that really only left the trans
former (T801). Removing it, I tested it
with a shorted turns oscillator/tester
which indicated that pins 1 and 3 were
shorted. This test is not always conclu
sive as I don’t know what frequency
the circuit is designed to resonate at.
However, I decided to take a punt – a
new chopper transformer was ordered
and it subsequently proved that my
diagnosis was correct. The repair cost
really made it quite a marginal exer
cise, however.
Sony KV2064
The second set was a Sony KV2064
and the customer thought that it had
to be the on/off switch because it was
intermittently dead. Of course, power
switches can sometimes be faulty but
it never ceases to amaze me that many
people think that a TV set consists of
a tube, a valve, a switch and a fuse.
That’s it – it has to be one these items
which is faulty if the set doesn’t go!
As it turned out, the owner of the set
was a very heavy smoker and there was
a film of nicotine over every surface
and component inside the set. The
switchmode power supply had 240V
AC coming into it, which cleared the
on/off switch, and there were no fuses
blown. This of course left only the tube
and the valve!
A close examination of the power
supply circuitry revealed that the start
up resistors were the likely culprits.
There was over 300V on the collector
the chopper transistor (Q602) and
the four high-value 0.5W resistors
were all discoloured, although only
one (R602, 330kW) was open circuit.
I changed them all for high voltage
types, reworked the solder just in case
and switched on. It was all an anti
climax – everything worked perfectly
and the set now started every time it
was switched on.
All I have to do now is explain to
the customer that it wasn’t the on/off
switch.
The reluctant NEC
Later that morning, a lady dropped
in her NEC video with the symptoms
that it wouldn’t fast forward or rewind
and turned itself off.
When I eventually got the chance
to look at it, I found that there was
virtually little or no take up torque,
which suggested a reel idler problem.
Being an N9083A, this machine was
a later (1991) version of the N9000
series, with the reel drive being a
sub-assembly which is fairly easy to
remove from underneath.
The main problem is due to two
tyres within this assembly which
are supplied in the VBK83 tyre and
belt kit. However, one of the tyres is
an integral part of an idler assembly
and the replacement will not stay on
satisfactorily without gluing.
You can either purchase the entire
sub-chassis assembly for around
$55.00 trade (part no. 016192683 – this
not shown in the service manual) or
forego replacing that one tyre. It’s a
real nuisance that it is not available
as a complete pulley idler on its own.
It is also worth noting that earlier
models also require a modification to
the white CR slider – this part is now
black (part no. 016457582) and costs
around $2.00.
One can only hope that the customer
clearly understood that the cutdown
repair would not last as long as the
original. Still, I spelt it out as clearly
as I could.
This afternoon saw a dead IBM
monitor arrive. It was PS/1, manufac
tured in June 1993, series 028-004. I
have seen quite a few of these units
with blown mains fuses (F601 2.5AT),
usually without any apparent cause.
In one case, however, it turned out to
be the chopper IC itself that intermit
tently blew the mains fuse.
This particular unit also had a dud
mains fuse. I replaced it, checked the
rest of the supply for any obvious
faults and then, just to be on the safe
side, switched the set on with a 200W
globe in series with it. Initially, the
globe went bright, then dimmed and
came on again at half about luminance.
The first current surge was obvi
ously due to the degaussing circuit
kicking in and there was obviously
still quite a lot of current being drawn
at the end of the degaussing cycle – at
least 0.75A. However, the compliance
plate said the full current drawn was
1.4A, so I tried it without the globe
and the fuse held and the screen gave
a good picture.
For the rest of the day, I cyclically
switched the monitor off and on for
periods of 15 minutes, without further
problems. In the end, I could only ad
vise the customer that the cause was
probably either due to a mains surge
or fatigue in the old fuse.
All we can do now is keep our fin
SC
gers crossed.
February 1997 33
|