This is only a preview of the January 1992 issue of Silicon Chip. You can view 43 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 4-Channel Guitar Mixer":
Articles in this series:
Items relevant to "Experiments For Your Games Card":
Articles in this series:
Articles in this series:
Articles in this series:
|
SERVICEMAN'S LOG
Same set, same symptoms, but. ..
Usually, we welcome a set of symptoms we've
seen before; it can mean a quick cure for what,
originally, might have been a sticky one. It's what
we call experience and it's what the customer
pays us for. But sometimes two different faults
can produce identical symptoms.
My stories this month demonstrate
the confusion and red herrings this
possibility can sometimes cause. The
first one involved an AWA Mitsubishi
34cm colour TV set, model C-3423. It
was a new set, only about nine months
old and, of course, still under warranty.
I was .making a house call on a
different matter when the teenage
daughter asked me if I would look at
her TV set. According to her, it had
simply failed in the middle of a program and was completely dead. Could
I fix it?
I assured her that I should be able
to and, if she had the necessary documents to cover the warranty, it should
noi cost her anything. In fact all these
were in order, so I loaded the set into
105
the van and headed for the shop.
When I set it up on the bench and
switched it on, the result, superficially, was as she had described it: no
picture, no raster and no sound. But it
was not quite dead. At the moment of
switch-on, I had heard a faint "boing"
as the de-gauss circuit activated.
This meant that the mains fuse was
intact, eliminating one initial check. I
pulled the back off and switched on
again, looking for any obvious damage. I found none but I was able to
observe that the picture tube heaters
were up and running, which ruled
out a lot more possibilities.
Delving a little deeper, I established
that the main HT rail was spot on at
103V. I pushed the EHT probe in under the ultor flap and came up with a
1N•9'&
T 40 2
F BT
C421
IOOOfliF
2~V
Fig.1: the horizontal
output transformer
circuitry in the
Mitsubishi Model C3423, The 16.5V rail
is derived from pin
5 ofT402 via
resistor R435, diode
D406 and capacitors
C422 & C421.
similar answer - EHT spot on. So
where to now? I began imagining all
kinds of weird and subtle faults, involving a faulty IC or something similar. Fortunately, calmer thoughts prevailed. There are always other voltage
rails and it is essential to check all of
these before suspecting individual
components.
And so I very soon discovered that
what should have been a 16.5V rail
was actually sitting at only 3V. I'm not
sure of all its functions but one is to
provide another rail - a 12V rail via a
voltage regulator, 1502. In any case, it
was the most likely cause of the failure and had to be fixed.
There is very little to this circuit
(Fig. l). It is derived from a tapping,
pin 5, on the horizontal output transformer (T402) and consists of a lQ
fusible safety resistor (R435), a diode
(D406), a 1500pF capacitor (C422)
across the diode, and a lO00µF 25VW
electrolytic capacitor (C421) as the
filter.
Well, . with only four components
involved, it didn't take long to find
the culprit. It was C421, the lO00µF
filter capacitor, which was virtually
open circuit. In fact, it couldn't even
struggle up to lµF on the capacitance
meter.
Anyway, it was easy enough to fix;
a new electro and the set was up and
running like new. I rang the house,
spoke to the girl's mother and told her
that the set was ready. She said she
would be in to pick it up the next day.
In the meantime, I left it running.
And so the next day the lady collected the set and I completed the
necessary warranty claim forms and
sent them off.
·
All of which added up to a perfectly routine exercise involving a relatively simple fault. In fact, the reader
is probably wondering why I bothered to relate the story.
Mum returns
There is a very good reason; the set
34
SILICON CHIP
bounced - and bounced in a particularly nasty way. Mum was back the
next day with the set and rather testily complained that it had run for
only 15 minutes, than failed with exactly the same fault as before. (It's
always the "same" fault when a set
bounces).
Naturally, I apologised, tried to explain that these things do happen occasionally, and assured her that the
warranty conditions would still apply. This mollified her somewhat but
she wasn't laughing when she left.
Why did that crucial 15 minutes have
to happen in the customer's home
rather than on my bench?
And so I started all over again. I
switched the set on and there was the
reassuring "boing" from the degauss
circuit. And with the back off I could
see the picture tube heaters were
alight. And the main HT rail was spot
on, as was the EHT.
Of course, it couldn't be the 16.5V
rail again but it was. It was down to
3V, exactly as before. I was starting to
pinch myself now, expecting to wake
up and find it all a horrible dream. (I
do have such dreams!)
But it was real enough. And, with
Fig.2: power supply circuitry in the Mitsubishi model CT-1447AM. The bridge
rectifier is at centre and the voltage regulator, IC901, at lower right. Note the
voltage on pin 4 and the incorrect voltages shown for Q902.
identical symptoms, I was totally confused. On the one hand, it appeared
that the brand new capacitor I had
fitted had lasted only a few hours. On
the other hand, the odds on such a
coincidence seemed too long to consider.
Nevertheless, I felt that it had to be
the capacitor, so I pulled it out and
checked it on the tester. But I felt a bit
foolish when the tester showed that it
had the full lOOOµF plus; there was
nothing wrong with it.
So what was wrong? With the capacitor cleared, the wasn't much left.
The lQ safety resistor? A likely culprit - only it wasn't. That really left
only the diode. The only snag was
that an in-situ check with the meter
showed it also was OK.
I was getting the horrors now. Was
it a faulty tapping on the transformer?
I put that idea to one side and decided
to try a long shot. In spite of the test, I
was having second thoughts about the
diode. I grabbed another diode, tacked
it across the existing one and tried
again.
Bingo! That was it. The 16.5V reappeared, and the set began to play. I
pulled the original diode out and
checked it again, but couldn't fault it.
Yet it wouldn't work; it obviously
could not handle the voltage and/or
-~---·· ij~t<t
)I.
.
.
<at>?)o~=--·
cs~-r :BOUNC..E:..D 11-\e:. NE:.X, PAY
WHF;.N ~UM "RA~E.~ 1"E.S1iLY COM?LA\Nl::P.u ..
11-\~
JANUARY 1992
35
SERVICEMAN'S.LOG - CTD
current conditions off the circuit.
I fitted the new diode properly, ran
the set for a couple of days, crossed
my fingers, and advised the owner
that it was ready. That was several
weeks ago and all is well so far.
But why did those two components
fail in quick succession? I've tried to
think of some reason why the capacitor failure may have contributed to
the diode failure but without success.
I have finally concluded that it was
sheer coincidence - but I wish that
the diode had chosen -to fail in the
workshop rather than in the customer's home.
So that was the first example of two
faults with identical symptoms.
It was another case of a totally dead
set. As before, preliminary observa1 .---t-lJol-,
tions, indicated that the mains fuse
.,.bi
~
was intact but there the similarity
~--~ I ®
!l!l!il
ended. There was no HT rail voltage
and, not surprisingly, no EHT.
I went back to the bridge rectifier
(Fig.2), which is directly across the
mains, and measured its output across
the main filter capacitor, C906 (l00µF).
The reading was around 300V, which
was a little on the high side and suggested a lack of load. From there, I
went to pin 4 ofIC901, a voltage regulator package which provides the main
HT rail at 115V.
(Incidentally, some of the voltages
quoted around this part of the circuit
are quite strange. Pin 4 of IC901, .
More Mitsubishis
c________ , j
shown as 115V, connects directly to
The second example involved two the collector of Q902 where, miracuQ
more 34cm Mitsubishi colour sets, lously, it becomes 165V. The other
models CT-1447AM and CT-1440A. voltages on this transistor are equally
Fig.3: horizontal output transformer
And perhaps I should add at this point strange.)
But there was no voltage at pin 4. circuitry in the Mitsubishi Model CTthat any apparent emphasis on
Mitsubishi by these stories is purely So, was it a regulator failure or was 1447 AM. The collector of the
horizontal output transistor, Q503,
coincidental. In no sense do I intend something loading the circuit? In fact,
connects to pin 1 of the primary
to imply that they are more prone to it appeared to be the latter because a winding, while the 115V rail goes to
faults than any other set. It is just that resistance check from pin 4 to chassis pin 2.
· the need to present these stories in showed only 3.5Q. Aha!, I thought,
sequence may give a wrong impres- I'll bet that's the horizontal output
sion.
transistor broken down - this being mary, pin 1, while the 115V rail conThe first set appeared a couple of one of the more common causes ofHT nects to the other end, pin 2.
weeks after the incident described rail loading.
So it was off down to the other end A wobbly diagnosis
above. The owner was going on holidays for several weeks and the set had of the circuit (Fig.3) to Q551, the tranAnd this was where my snap diagnosis began to look a bit wobbly. A
failed a couple of days earlier, so they sistor in question. In typical fashion,
decided to leave it with me while the collector connects to one end of resistance measurement from the colthey were away. That suited me fine.
the horizontal output transformer pri- lector ofQ551 to chassis gave a virtually identical reading to that at pin 4
of the regulator IC. I tended to over11ao•
look
the full significance of this iniHK(lll
112W
tially, regarding it as no more than a
confirmation of the load on the HT
rail.
But it suddenly struck me that there
was the resistance of the transformer
C802
primary winding in between these two
oaoe
Rf-I
6
points. Granted, it is quite low - only
AIOI$
11011
about JQ - but it is measurable and
HO
should have shown up as a difference
between the short at one end of the
winding and that at the other. So how
M:.2,r.'..,
JVOI
come they both read the same? How
come both ends of the transformer
PI04
OICIIUI- coi.
primary were showing the same resistance to chassis?
Fig.4: power supply circuitry in the Mitsubishi Model C-3423. The power supply
If the fault was in Q551, it was a
bridge rectifier is at left, the chopper transformer (T801) at centre, and the
voltage regulator (Q801) at right. The 103V HT rail is derived from pin 4.
very funny one. Either that or we had
36
SILICON CHIP
OS!5
more than one fault; or the fault was
in the horizontal output transformer.
That was a nasty thought because there
are some 10 pins connecting the transformer into the PC board and, together
with the physical layout, this could
make getting it out rather tedious.
On the other hand, it was a simple
job to lift Q551 to clarify the position.
And as far as I could determine from a
rough check, the transistor was OK.
However, it is important to realise
that these transistors have an internal
resistance of about 400 between base
and emitter, plus a diode between
collector and emitter. This can sometimes confuse simple tests.
But the more important check was
from pin 1 of the output transformer
to chassis and this still showed 3.50.
So it wasn't Q551. It had to be the
transformer- or was it a "funny" somewhere else in the set? I felt that the
latter possibility was a long shot and,
in any case, tracking it down (or disproving it) could be a long and involved business.
Among other things, it would probably involve cutting various copper
tracks. This is not always as easy as it
sounds if the tracks are crowded to-
gether and is something I try to avoid
if possible.
So before pulling the transformer, I
decided on one more check. One of
my more valuable pieces of test gear
is a shorted turns tester. I wasn't sure
how well it would work in this situation because the transformer second.ary winding was pretty well loaded
with, among other things, the picture
tube heater winding.
Nevertheless, I connected it across
pins 1 and 2 and made the test. And
the indication was that there were no
shorted turns in the transformer. Well,
that was a useful piece of information
but not really conclusive.
So where to now? After some cogitation, I decided that in spite of the
work involved, the quickest and most
satisfactory approach was to remove
the transformer. Granted, if it wasn't
the culprit, the work of pulling it out
and putting it back might appear to be
wasted but at least it would settle the
question once and for all.
I set to work with the iron and
solderwick and, after about 15 minutes, was able to lift the transformer
clear of the board. This finally gave
me a clear go. A quick check at the pin
AUSTRALIAN MADE TV TEST EQUIPMENT
Happy New Year Specials. 10% discount on all test
equipment. 15% discount on two or more items
SHORTED TURNS TESTER
Built-in meter to check EHT transformers including
split diode type, yokes and drive transformers .
$78.00 + $4.00 p&p
TETIA TV TIP
Princess 14CT8 (PC-02X
chassis)
Symptom: weak vertical hold.
Video input to the sync separator
(IC301) is normal but the vertical
oscillator is inside the same chip
so it is not practical to see if the
sync pulses are normal.
Cure: the clue is that the voltage
at the input to the sync separator
is lower than specified. The cause
is that the first video amplifier,
(0201 2SC1815) is slightly leaky
collector to emitter. This does not
seem to affect its gain but does
put a wrong bias on the sync separator.
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.
1 and pin 2 positions on the board
showed that there was no short to
chassis, which meant that it had to be
PRESTON ELECTRONICS
217 High Street, Preston, Melbourne
Victoria 3072
(Cnr of Bell & High Street)
TEEPROM PC Eprom Programmer Card,
Cable + Software
$250.00
HIGH-VOLTAGE PROBE
Built-in meter reads positive or negative voltages from 0-50kV. For checking EHT and other HT voltages.
$98.00 + $5.00 p&p
1991 European/Jap Transistor Equivalent
Book, 790 Pages
$39.95
DEGAUSSING WAND
Strong magnetic field. Double insulated for your
safety with momentary switch operation. 240V
AC 2.2 amps. As important as a soldering iron!
$75.00 + $10.00 p&p
REMOTE CONTROL TESTER
Designed to test infrared or ultrasonic
control units. Supplied with extension
infrared detector lead. Output is via a LED
and piezo speaker. $84.00 + $4.00 p&p
~~
~
From$19exchange
plus post & pack
Cheque, Money Order, Visa, Bankcard or Mastercard
216 Canterbury Rd,
Revesby, NSW 2212,
Australia.
Phone (02) 774 1154
Fax (02) 7741154
Economy Soldering Iron - 240V Fine Point Tip
$19.95
Video Recorder Replacement Belt Kits (Specify
Make & Model}
$12.95
CB Radio Power Amp, 35W AM, 70W SSB
Output
$199
Fujitsu 12V 1A PC Relays
$1.00 ea.
Mail Order Hot-Line
8 /Card-Mast/Card-Visa-Cheque
Phone (03) 484 0191
JANUARY 1992
37
SERVICEMAN'S LOG - CTD
the transformer. Fairly obviously, there
had to be a short between the primary
winding and one of the secondary
tappings.
A resistance measurement between
pins 1 & 2 and pins 9 & 10 (the picture
tube heater winding) gave the all clear.
But from primary pin 2 to pins 3, 4, 5
& 7 (which are all part of one winding), there were very low resistance
readings. And the lowest was 3.5Q to
pin 7, which connects to chassis. Also
significant was the fact that the readings from pin 1 were almost identical.
Conclusion: the breakdown between primary and secondary must
have been at, or very close to, the
centre of the primary winding. And
that makes it one of the dirtiest tricks
that Murphy has pulled for a long
time.
And the shorted turns test? No, it
didn't fail; the fault was not a shorted
turn. On the other hand, it was areasonable try as faults in output transformers commonly result in shorted
turns.
Anyway, these thoughts aside, I had
finally tracked down the fault. A call
to AWA for a replacement transformer was handled with commendable
speed and it was on its way in few
hours.
And that, apart from the routine
fitting, was the end of the story. I ran
it regularly each day until the customer returned from holidays and
have heard nothing from him since.
The twist to the story
But now comes the twist to the
story. While that set was still in the
workshop waiting to be collected, another 34cm Mitsubishi arrived, this
time a model CT-1440A. Although a
different model, with some superficial differences, it uses essentially the
same circuit and chassis as the CT1447 AM.
The complaint was the same; completely dead. I started at the bridge
rectifier as before and found a similar
situation; a slightly higher than normal voltage of about 300V. I then
moved to pin 4 of the regulator, IC901,
only to find no voltage. Almost automatically I made a resistance check to
chassis. And what did I find? Yes,
you've guessed it; 3.5Q
I could hardly believe it. More to
the point, I hardly knew what to do.
Remembering how I had been caught
in the first story, I had a horrible feel-
~MONG 011-1~ 11-lt~G'E> rf WOUl.--D PROBABL-V
\NVOL-VE'. CV"'\I\NG VA'R\OUS COP'P6:.'R -.1RAC.KSon
38
SILICON CHIP
ing that Murphy was having a lend of
me. On the one hand, the symptoms
seemed so obvious - perhaps too obvious - while, on the other hand, the
odds against two such identical faults
landing on my bench within a week
or so of each other seemed pretty long.
I couldn't escape the feeling that
which ever way I jumped, it would be
wrong. IfI pulled the transformer out,
there would be nothing wrong with it.
If I didn't, I would spend hours looking for an elusive fault with the same
symptoms, while it was the transformer all the time. Either way, I couldn't
win.
But, of course, I had to start somewhere. I went over the circuit in fine
detail, repeating every measurement I
had made on the first set, looking for
something that didn't match and
which might provide a clue.
It was a vain effort; as far as I could
determine everything was exactly as
it had been in the first set. So, finally,
there was nothing for it but to pull the
transformer.
And it really was the same fault. So
I'd been caught out by my own caution. But I still reckon it was long
odds for those two faults to turn up in
such quick succession. Or have I
latched on to an inherent fault in a
batch of these transformers. Well, only
time will tell but, if I encounter these
symptoms again, I reckon I'll take a
punt on the transformer.
Postscript
Well, that was supposed to be the
end of this month's contribution. However, shortly after I had finished bashing it out on the computer, another
dead Mitsubishi turned up. It was a
model C-3423, the same as the first
one with the failed 16.5V rail. And,
like that one, it was a new set, only
about one month old and still under
warranty.
In view of the previous stories, my
first reaction was to suspect that 16.5V
rail. But then another thought crossed
my mind; I realised that this set uses
the same horizontal output transformer as the models CT-144 7AM and
CT-1440A. So I couldn't overlook the
possibility that this was going to be
number three with the primary/secondary breakdown.
I checked the 16.5V rail first. It was
completely dead whereas previously
there had still been about 3V. Next, I
checked the 103V HT rail and this
You can now
afford a sate II ite
TV system
For many years you have probably
looked at satellite TV systems and
thought "one day".
You can now purchase the fol lowing K-band system for only:
$995.00
Here's what you get:
* A 1.8-metre prime focus dish
antenna, complete with all the
mounting hardware.
*or better).
One super low-noise LNB (1.4dB
*magnetic
One Ku-band feedhorn and a
signal polariser.
* 30 metres of low-loss coaxial
cable with a single pair control line.
* A 99-channel infrared control
This horizontal output transformer is from the CT-1447AM (second story),
although it is also used in the C-3423. It had shorted from the centre of the
primary winding to one of the secondary windings.
was also dead, a symptom more like
that of the output transformer fault.
It looked even more like it when I
checked around the voltage regulator
package, Q801 (Fig.4). Voltage is applied to pin 1 of this from a chopper
transformer, T801, and was sitting at
around 300V. But at pin 4, which delivers the 103V rail, there was nothing which indicated that the voltage
regulator wasn't working.
A resistance check from pin 4 to
chassis showed only 2.6Q. And hard
alongside the voltage regulator on the
circuit were two likely culprits: a
lO0µF electrolytic capacitor (C807)
and a 160V zener diode (D807) which
protects against excessive HT voltage. It was easy enough to disconnect
both of these but I drew a blank; the
2.6Q remained.
Another possibility was a breakdown in the voltage regulator. But
while not ruling it out, I felt it was
most unlikely considering the symptoms. A breakdown of the transistor
in the regulator would most likely put
300V on pin 4.
So was it the previous transformer
fault? This still seemed like a possibility until I took a closer look at the
circuit. Between the HT rail and pin 2
of the output transformer primary
(Fig.1) is a diode, D408. It seemed to
me that it would be most unlikely that
I would be able to get a reading of
2.6Q through this diode, considering
the low voltage of the meter.
Nevertheless, I made some readings at the transformer and this
quickly ruled it out. So it was back to
the circuitry around pin 4 of Q801.
The HT rail also connects to a tap on
the lower winding of the chopper
transformer. So was it a fault in the
chopper transformer?
I had a lucky break here; using
solderwick, I was able to soak up the
solder around the tapping pin until it
was isolated from the copper pattern.
This cleared the fault from the HT
rail, which was my first real breakthrough.
So it could be the chopper transformer. But then I spotted something
else; the left hand end of this winding
connects to chassis via diode D805
and choke L802 in series, and also via
a parallel 2200pF capacitor, C808. So
it could be either the diode or the
capacitor.
I pulled the capacitor out first and,
of course, drew a blank; nothing wrong
with it. But the diode was a different
story; it had broken down and the
2.6Q was mostly from the choke (L802)
and the transformer winding.
So I finally cracked it, in spite of
the red herrings. But that's all part of
the game.
SC
satellite receiver with adjustable IF
and audio bandwidth, polarity and
dual digital readout.
Your receiver is pre-programmed
to the popular AUSSAT transponders via the internal EEPROM
memory. This unit is also suitable
for C-band applications.
Call, fax or write to:
AV-COMM PTY LTD
PO BOX 386, NORTHBRIDGE
NSW 2063.
Phone (02) 949 7417
Fax (02) 949 7095
All items are available separately.
Ask about our C-band LNBs, NTSCto-PAL converters, video time date
generators, FM 2 and EPAL.
I
I
----------YES GARRY, please send me more information on K-band satellite systems.
I
I
: Name ... ... .... .......... .. .. ................ ......
:
I Address.. ... ... .. ... .. .. ....... .. ............. ....
I
I ...........................P/code ..... .. .. ..... ..
II Phone ..... ...... .. ............. ............. ..... ..
I
I
I
II
. ___________ .I
I
AC N 002174478
01 /92
JANUARY 1992
39
|