This is only a preview of the May 1998 issue of Silicon Chip. You can view 35 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:
Items relevant to "Build A 3-LED Logic Probe":
Items relevant to "A Detector For Metal Objects":
Items relevant to "An Automatic Garage Door Opener; Pt.2":
Articles in this series:
Items relevant to "Command Control For Model Railways; Pt.4":
Articles in this series:
Items relevant to "40V 8A Adjustable Power Supply; Pt.2":
Purchase a printed copy of this issue for $10.00. |
SERVICEMAN'S LOG
Lightning didn’t strike this time
If a TV set stops working while there’s a storm
around, many people assume that it has been
struck by lightning. They automatically link the
two phenomena together simply because they
happen at the same time but it ain’t necessarily
so.
My first story this month concerns a
1990 Grundig 68cm TV set. According
to the customer, it had gone off during a storm and so she immediately
jumped to the conclusion that lightning was responsible.
I suppose it’s a natural enough
conclusion under the circumstances
but in this case, it turned out to be
quite wrong.
The set was fitted with a West German 2-pin plug (with an Australian
adaptor added) and the lady had
obviously brought it with her from
Germany. As I very rarely see German
television sets these days, I was intrigued to see what might be involved.
The set modestly proclaimed itself
to be a “Grundig Monolith Blackline
Multisystem”.
A European set of this age could
present problems. For one thing, their
power grid is 220V and the increase in
heater voltage, without a modification
for our 240V, could be a tube killer;
the picture tube could very well be
low in emission.
Another problem concerns model
identification. The number of models
in the Grundig lineup is bewildering.
In fact, identi
fying this model set
requires several numbers: model M70575/90, chassis series CUC5836, and
main chassis part number 29701-057.
Despite all this, when a service
manual which I ordered arrived, it
was only a supplement which covered
the differences between this and the
CUC 5820, 5835, 5860 and 5880 models. So it was very much incomplete
and among other things, lacked all the
information for the plug-in modules.
Later, I acquired the main manual
but even then, this was incomplete.
The PC board component layouts
were not included and there are
many differences and modifications
between part numbers and the component reference numbers, which aren’t
marked on the board.
The back was easily removed but
access to the horizontal main chassis
underside was not as easy as on earlier
models. The service manual suggests
it should be pulled out and lifted onto
its righthand side on the bench. This
didn’t work because the speakers are
directly below the chassis and the
leads weren’t long enough.
There wasn’t much life in the set
except a low level motor-boating,
which suggested the power supply
was functioning – at least in a fashion.
What was more immediately obvious
was rust and corrosion everywhere
– definitely not something caused
May 1998 27
Fig.1: part of the deflection module circuitry in the Grundig M70-575/90. IC7010 (TDA8146) had blown apart and
had to be replaced along with IC7020 (TEA8170A), zener diode D7012 and resistor R7033.
by lightning and storm damage! The
set had obviously had a hard life in
a damp environment or near the sea.
A glance at the tripler revealed
several telltale carbon track deposits
leading from holes in its insulation
where it had been sparking and burning. For the time being, I disconnected
it and went to the horizontal output
transistor T541, a BU508A. It measured short circuit and so I replaced it,
hoping that might be all the damage.
With power applied, the motor
boating disappeared but not much
more was happening. And then suddenly, I heard a slight click and it
started motor boating again. Not only
had transistor T541 become red hot
but so had diode D546, a BY228, and
both had failed. The overall damage
looked pretty severe – two horizontal
output transistors, two diodes (D546
and D547), the tripler, almost certainly the horizontal output transformer,
and probably the horizontal drive IC
(IC500 – TDA8140).
I would have to give an estimate
on these before proceeding; if there
28 Silicon Chip
were any other problems they would
probably be minor. The lady had no
problem with this and so I ordered
the parts which arrived promptly. I
decided to replace them all at once,
to eliminate the possibility of further
damage due to faulty parts.
On switch-on, I was disappointed
to find only limited im
provement.
Granted, the EHT was now working,
the HT rail measured 165V and there
were no signs of distress or overheating in T541 and D546. However,
there was still no sign of a raster or
sound, apart from a few noises in the
speakers.
I checked all seven voltage rails and
this quickly revealed that there was
no voltage on the 29.5V rail. This was
quickly traced to R525 (0.33Ω) which
was open circuit. Replacing this produced some sound but it was garbled,
although this could have been due
to mistuning. And there was still no
picture; just a blurred blob which was
uncontrollable.
The EHT meter confirmed 25kV
on the picture tube final anode and
there was a healthy spark from the
CRT socket focus pin. There was also
plenty of G2 volts but shorting the
tube cathodes to chassis produced
no intelligent raster or picture, other
than the unfocussed blob (the focus
control had no effect).
I prayed the tube wasn’t U/S; shorting any cathode to chas
sis should
give an intense bright raster. The only
thing left to check was the deflection
circuit. The CRO showed that there
was no vertical deflection on the yoke,
while there was too much signal at the
chassis end of the horizontal coils. It
was time to unplug and examine the
deflection module.
Once the deflection module was
on the bench, it was obvious it had
sustained major damage – IC7010
(TDA8146) had literally blown apart,
leaving a blackened hole in its lower
half. A lot of power would have been
required to do this, which meant that
there was bound to be collateral damage. And sure enough, a quick check
soon revealed that D7012, a 36V zener
diode, was short circuit, while R7033
(100Ω) had gone high. In addition,
R546 (4.7Ω) was open circuit.
Until now, I had managed to obtain
all the parts easily and cheaply from
my local supplier and so I ordered a
replacement TDA8146 (IC7010) from
them. But IC7020 (TEA8170A), which
I also strongly suspected, had to be
ordered from the Grundig agents.
While waiting for these parts, I had the
opportunity to clean up the corrosion
and to check the other components
around these circuits. Nothing else
appeared to be amiss.
The ICs duly arrived and, when
fitted, restored the pic
ture. After
some tuning and setting the CCIR B/G
system standard (which is actually
a story in itself), I had a watchable
picture with good colour, sound and
focus. The only problem remaining
was severe pincushion distortion.
Adjusting the pincushion control
(R7011) made no dif
ference. Ditto
for the width control (R7002) and the
trapezoid control (R7007). In fact, no
east-west correction controls were
working around IC7010 (TDA8146). I
tried two more TDA8146s for IC7010
but to no avail – a matter of some
significance, as it turned out.
After a lot of CRO measurements,
the only two waveforms I could fault
were waveform 10, which is the vertical input to pin 2 of IC7010, and waveform 9, which was slightly different
on output pin J7 to the deflection coils.
I was really stuck now, as I couldn’t
determine whether the pincushion
fault lay in the motherboard or in the
deflection module.
panied by a whistling noise. Close
inspection revealed that though both
sets were similar, the Blaupunkt mod
ules had major differences. Because
the Grundig was a multi-system,
with picture-in-picture, the boards
were larger and contained a lot more
components.
I was mainly interested in the deflection module and the only major
difference between the two was that
IC7010 in the Blaupunkt deflection
module was a smaller 8-pin TDA8145
device instead of the 14-pin TDA8146
in the Grundig. Despite this, I decided
it was worth the risk and swapped the
modules over.
Surprisingly, this fixed the Grundig’s pincushion distortion problem
completely. I was somewhat taken
aback at this because I was sure I
had checked every component in the
module. Conversely, the Grundig’s
deflection module (29504-107.80) did
down to the IF module, of all things.
This was established by temporarily
fitting the IF module from the Grundig
set. Unfortunately, I ground to a com
plete stop here after replacing IC2340
(TDA2579) in the IF module, which
made no difference.
I now had two problem sets: the
Grundig with pincushion distortion
and the Blaupunkt with lack of
height. It was then that my friendly
leprechaun came to my rescue for the
second time. By chance, I was talking
to a colleague and happened to mention my problems with these two sets.
And as luck would have it, he knew
the answer to the lack of height in the
Blaupunkt, as he had once spent a lot
of time tracking down this very fault.
In his case, it turned out to be diode
D2334 (TD190) in the IF module.
And he was spot on. I subsequently
discovered that the diode in my set
was slightly leaky and replacing it
A friendly leprechaun
not fix the height or width problems
in the Blaupunkt, nor did it correct its
east-west foldover problems.
Well, at least I had localised the
Grundig’s problem. However, I decided to leave this set for the time being
and concentrate on the Blaupunkt
instead. The first step, of course, was
to reinstall its deflection module.
The lack of width and the horizontal foldover turned out to be due to
IC526 (TDA8140), C527 and C574,
which were causing transistor T572
(BU508A) and resistor R574 (18Ω)
to overheat. The exact sequence of
events involved here is not clear; I
could not determine which had failed
first and what was damaged as a result.
All I can tell you is that all these parts
had failed and had to be replaced.
The lack of height was tracked
completely cured the problem.
It was at this time that I was blessed
with an surprising coincidence (and
for this, I imagine I should thank some
friendly leprechaun – I’m writing this
on March 17 which is St Patrick’s
Day). As stated before, I very rarely see
German TV sets and yet, incredibly, it
was just then that another customer
brought in his Blaupunkt IS 70-39
VT (FM 500.40 chassis 7669 800)
which is manufactured by – yes, you
guessed it – Grundig. And the main
chassis Grundig part number was
29701-056 – only one digit different
from the Monolith. Luck doesn’t often
come like this.
The problem with the Blaupunkt
was a shrunken picture (both vertically and horizontally) which became
folded after about 10 minutes, accom-
Back to the Grundig
My next problem was what to do
about the faulty Grundig deflection
module. One option was to send it
back to the Grundig agents for servicing, a process that would take two
or three weeks. Another option was
to buy a new one but this was not
readily available and, in any case, is
rather expensive.
In the end, I played a hunch. I had
begun to suspect that the TDA8146s
(IC7010) which I had purchased from
my supplier might be the problem.
And I became suspicious because
of their price. They were only about
$7.00 each from my supplier whereas
they were closer to $20.00 each from
the Grundig agents.
May 1998 29
Fig.2: part of the IF module in the Grundig M70-575/90. Diode D2334 (TD190) at lower right had gone leaky.
OK, that’s fine; I’ll buy in the cheapest market, all else being equal. But
were these parts really equal? To test
my theory, I ordered a TDA8146 (part
no 8305-358-146) directly from the
Grundig agents and when it arrived, I
noticed one obvious difference in the
batch number. The original Grundig
number was W994A9409 but the
ones I had fitted during testing were
marked W994A9422.
And that was the answer; it was all
that was required to cure the problem
completely and the east-west correction circuits now worked perfectly.
Both customers were pleased that
their sets were now working properly
but neither was nearly as pleased as I
was. And what was the final verdict?
Was the set struck by lightning? I
dunno, please. I have no doubt that
the Grundig set failed while a storm
was in progress but that doesn’t necessarily mean that the two events are
directly related.
From the evidence before me, I
would have to say that lightning was
not the probable cause. However, the
customer is firmly convinced that
lightning caused the problem and the
matter is still before her insurance
company. Well, if she’s happy, who
30 Silicon Chip
am I to disillusion her?
For my part, I returned the three
cheaper ICs to my local supplier with
a please explain note and am awaiting
their re
sponse. Unfortunately, this
wouldn’t be the first time that bodgie,
off-tolerance components had found
their way onto the local scene and
been marketed quite innocently by
local dealers.
Yes, it’s a rough world out there.
The JVC video
JVC videos don’t seem to like me.
This week, a lady brought in her JVC
HR-D600 VCR and complained that
the picture was “sort of distorted but
only with some tapes some of the
time”. Obvious
ly, she imagined, I
must know what the trouble was, so
could I please tell her what it was,
how much it would cost and how long
would it take to get it fixed.
Ironically, I had a sneaking suspicion as to what the trouble might be
but I certainly wasn’t going to commit
myself to a guess. Instead, I suggested
she leave it with me and I would make
an assessment after I had seen the
fault. I put it on the soak bench with
an E240 cassette in standard play and
left it producing an excellent picture.
Every so often, I would check that
it was going OK and restart the tape.
After three days of this, I was beginning to despair and so, when she
phoned to enquire about progress, I
had to confess that there was none;
it hadn’t missed a beat.
Perhaps it was her tapes? Her answer to this question was an emphatic
“no”, because it also happened with
hire tapes. Not wishing to start an
argument on that basis, I told her she
might as well pick it up and try it
again later.
She didn’t call for two more days
and when I took her to meet the
monster, the tape had stopped. So I
pressed play again to show her it was
still going OK and would you believe
it, the wretched machine started to do
its trick (I told you they don’t like me).
The picture had three or four sets of
noise bars across the screen – permanently. Somewhat embarrassed, I did
my best to assure her that this was the
first time that the fault had appeared. I
don’t know whether she believed this
or not but, in any case, I could only
mentally shrug my shoulders; after all,
that was the true situation.
“Leave it with me”, I said. “Now
that the fault has shown up, I should
be able to get at the problem”. The
customer readily agreed to this, so my
explanation must have been at least
partially accepted.
Despite being busy, I decided I
would tackle it straight away while
the fault persisted. The symptoms
were typical of a misaligned tape
path and the tape guides used in this
deck do give a lot of trouble. The top
cover comes off conveniently but the
bottom is not only screwed in via the
fancy feet but also clips in on the sides
and centre. And it was while I was
removing this cover that the so-and-so
bit me, a sharp edge cutting one of my
fingers and causing it to bleed.
The next step is to move the top
PC board into its service position.
This involves removing a screw at
the rear, then using a smaller Phillips
screwdriver to remove five self-tapping screws that hold it. The board
was then parked in a vertical position
along the rear of the video.
Next, three more screws had to be
removed from the metal cover over the
heads. With the tape in the play mode,
everything looked OK except that
the entry and exit tape guides hadn’t
engaged properly at the end stops.
By wiggling them, I could restore the
picture. It was while I was wiggling
these guides that the machine bit me
again; it cut another finger, this time
on the sharp edge of the cassette ejector housing.
Things were bad and getting worse.
I removed the tape and put in a dummy cassette (one with no insides,
just the outside framework). I then
pressed the play button and pulled
the mains plug when the two arms
were only half-way along their action.
Now I could see how loose things
really were.
On the underside of the deck, the
guides are held in place with a brass
plug/collar assembly and a plastic
pin. Both of these are just a snug push
fit and the brass ones were not all the
way home. I removed them and applied some superglue before pushing
them all the way home while holding
the guides in on the other side.
This done, I cleaned the heads and
tape path and checked that everything
was shipshape before trying another
tape in the play mode. Everything now
worked OK, so I refitted everything
and gave it one more final test.
I couldn’t believe it – not only was
the original problem still in evidence
but there was now another even heavier noise bar at the top of the picture.
I stopped and started it several times
but it wouldn’t go away. There was
nothing for it but to go back in again
and access the tape guides.
Well, the cause of the problem was
immediately obvious – the pin in the
input guide had fallen out, presumably when I turned the machine upside
down to fit the bottom screws. The
annoying thing was that I had originally checked it and it had appeared
to be tight.
I decided to superglue it in and
put a few drops on the end of the
pin before pushing it into its hole.
Big mistake – the pin set hard in the
socket but wouldn’t go all the way in.
There was nothing for it but to remove
it and start again. It took a lot of aggro
to remove and clean it with acetone.
Before starting again, I made sure the
pin could slide all the way home in its
socket and that it did so easily.
Another drop of glue and once
again it seized half way in. I was on
the verge of chucking the whole thing
out the window when common sense
told me to try some lateral thinking.
After removing and cleaning the pin,
I refitted it in the hole and applied
the glue around the edges afterwards.
Sure enough, it finally locked in place
and capillary action made the glue
sink in.
I did the other guide too and then
after cleaning every
thing, tested it
again. I was dismayed to find that the
original symptoms were still there,
although they were now quite inter
mittent.
In desperation, I went in for the
third time and after a very careful
examination of the guide assembly, I
found I had inadvertently pulled the
plastic pin connecting it to the loading
arm. Consequently, this arm was also
loose. I glued this and its partner on
the other side, and anything else that
might possibly come loose.
Fortunately, that was the end of
the story because it now worked like
a steam train and was still working
when the lady called to pick it up. I
just hope that I managed to wipe off all
those bloodstains and that she won’t
notice the dents in the sides!
SC
May 1998 31
|