This is only a preview of the July 1990 issue of Silicon Chip. You can view 49 of the 104 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 "Digital Sine/Square Wave Generator; Pt.1":
Articles in this series:
Articles in this series:
Articles in this series:
Articles in this series:
|
SERVICEMAN'S LOG
The dangers of a quick fix
A quick turnaround is something every serviceman
aims for. It minimises the number of devices
cluttering up the shop, it helps cash flow, and .it
provides a certain sense of satisfaction. What's
more, the customer expects it. But it isn't always
the best approach, particularly with older sets.
IC provides a number of functions,
including the horizontal oscillator,
sync separator, etc. I quickly found
that R458 needed only a minor adjustment for the system to jump
back into lock, after which the set
produced a good picture. In fact, on
a quick turnaround basis, I could
probably have regarded the set as
fixed.
But a couple of things made me
suspicious. One was the owner's attempt to describe the fault which,
garbled though it was, hinted at
more than one symptom. The other
was that, now that the picture was
locked, it was obviously overscanning horizontally. Not grossly - and
the customer might well have accepted it - but enough to warrant
investigation.
That little bit of philosophy was
prompted by a recent experience
involving a Kriesler 59-4 colour set.
This model is now about 12 years
old and is about the last of the
"true" Kriesler sets; ie, before they
became a Philips chassis in a
Kriesler cabinet. And of course, at
that age there is always the question as to whether a lot of expense
can be justified.
In this case, the Kriesler was a
"second" set, having been replaced
by a more recent model a few years
ago. But it still had its place in the
household and the owner was keen
to keep it going. Unfortunately, he
was unable to provide more than a
very garbled description of the
trouble which, boiled down, was
best summed up in his own
words, "It doesn't go properly."
Fortunately, he and his family
were going on holidays for several
weeks, so there was no rush for at
least that period, or even longer if
necessary. He just wanted to be
sure it was fixed.
TETIA TV TIP
Philips 14CT2006 etc (CTXA-1 S
chassis)
Symptom: No sound . Voltages
around output IC 7181 (TDA
2611 AO) are all wrong but some
No sync
When I eventually switched it on,
one fault was immediately obvious;
the horizontal system was out of
lock. This could mean either a loss
of horizontal sync pulses or serious
horizontal oscillator drift, such that
the sync system could no longer
control it.
Removing the back of the set, I
sought out the horizontal hold
preset pot (R458), a 47k0 trimpot
connected via R459 to pin 15 of
IC450 (TDA2590) - see Fig.1. This
40
SILICON CHIP
are more wrong than others. Audio
output on pin 8 of SIF chip 7164
seems to be OK but nothing
reaches the power chip.
Cure:
C2177,
a
22nF
50V
ceramic capacitor, has gone leaky
and shorts the signal to ground.
i
TETIA TV Tip is supplied by the
Tasmanian branch of The Electronic Technicians' Institute of
Australia. Contact Jim Lawler, 16
Adina St, Geilston Bay, Tasmania
7015.
The width control, R687, 10k0
(lower right of Fig.1 ), had no effect,
suggesting that the overscan was
more than a maladjustment. In fact,
I decided to bypass that problem for
the moment and simply let the set
run while I attended to more urgent
jobs. In particular, I was anxious to
make sure that the horizontal hold
problem had really been solved and
was not intermittent. In fact, the set
ran perfectly for the rest of the day.
But next morning there was a
new fault. As the set warmed up, all
it presented was a 50mm high band
of picture across the centre of the
screen; ie, partial frame collapse.
And I had a feeling that this was
only the beginning.
On the off chance that this might
be a faulty height control pot, R637,
lkO (lower left of Fig.1), I tried adjusting this but it had only a minor
effect. There is a waveform (64)
given for the signal feeding this pot,
indicating a p-p value of 3V. But a
check with the CRO showed only a
fraction of this value.
From there it was a fairly simple
step to transistor TR636 which
feeds the pot. This transistor had
dropped its gain and replacing it
brought the scan back to normal. .
So back to the width problem.
This turned out to be due to R688
(47k0) which connects to the moving arm of the width pot. The
resistor was found to be open circuit, so that problem was easily
fixed.
More trouble
By now the set was beginning to
look pretty good. I tidied it up a bit,
then put it to one side where I could
let it run in case anything else
decided to show up.
For the next three days it went
like a bought one but then the picture started to tear, as though it
was about to lose horizontal lock. It
didn't actually lose lock but the
1sae1 ..
u
I
~£
~~
To R-425. R.416.
L501 . UU
:
IT.
R47S .
SK2:42 Pin 6,
I
SK211 Pin I .
I
, :::, I
: Lrn
, C4U
~
'
-
-
-
-
-
-
-
-
-
-1
-
lDo_ -
.3;- - - -
:~v
FRAME S
FRAME FLVBACK PULS 46
J;
-~:::1
JOV
DEFLECTION SOARD
to-1U55
CIS1
'r-{Il]---.--
-,-~22~~v
R&St
;;J;
35V
7TR65i
80213
cna
1::
RH5
TCH3
n'n 1Hn
rlHF'-tl~l1?,'z51.
fRAMf
OUTPUT
RH4
OEFUCT10N YoKE - -
FRAME
OSCII LATOH , J
Fig. 1: section of the Kriesler 59-4 circuit. IC450 (TDA2590) is at the top, the vertical oscillator stage at
bottom left, and the width control at bottom right.
tearing action was accompanied by
loss of colour.
My first reaction was to suspect
a fault in either the sync circuitry
or the horizontal oscillator. And I
assumed that the loss of colour was
probably due to misplacement of
the burst gate and its effect on the
colour killer. In any case, either
fault would be in the TDA2590 IC.
So the only real test was to replace
the IC.
I was lucky in this regard. I had
such an IC on hand and the IC in the
set was mounted in a socket. So it
was only a few moments work to
make the change. And that was it;
the picture came up perfectly
stable, with steady colour.
So, again, the set went back into
"soak corner" where it ran perfectly for several more days. Then, one
morning, it came up with the picture rolling. I let it run in this mode
just to see what would happen and
after about half an hour, when the
roll rate had slowed appeciably, it
suddenly locked. And it remained
that way for the rest of the day.
It rolled again the next morning
at switch-on and I immediately
reset the vertical preset pot (R622)
until the system locked. Of course, I
knew what would happen; after
about half an hour the picture
started rolling. So it seemed like a
thermally sensitive fault in the vertical oscillator circuit.
I let it cool and tried again. Again
the picture was stable at first, then
began rolling as it warmed up. I
reached for the can of freezer. The
vertical oscillator consists of two
transistors: TR621 (BC548) and
TR622 (BC558). I froze each of
them, one at a time, but this had no
effect. I sprayed the surrounding
components and even the sync
amplifier transistor, TR615 (a
BC548), all to no effect.
But I still suspected one or both
of those two oscillator transistors.
So out they came and in went two
new ones. And I was right; that fixed it. After that, the set ran for
several more weeks before the
customer called for it.
In the meantime, I checked. the
two transistors. Both seemed OK on
a rough "continuity" test, even
when frozen or heated. But it was a
different story when I tried them on
a curve tracer. I could find nothing
wrong with the BC558 but the gain
of the BC548 was all over the place.
But strangely, heating or freezing
had no effect, one way or the other.
Naturally, all this had chalked up
a fairly substantial account and I
was a little apprehensive as to how
this would be received. After all, it
was an old set. But the owner didn't
quibble; in fact, I think the list of
faults convinced him that I had
done a thorough job.
This all happened several months
ago, and a recent check confirmed
that all was well. So everyboby's
happy.
But it was lucky that I had plenty
of time to run the set. Had it been a
rush job, the set would have bounced more times than a rubber ball.
A thorny Thorn
And still on the theme of not-soyoung sets but from a slightly difJUL Y 1990
41
apart from the condition of the picture tube.
So I wasn't too happy when this
set turned up. The owner said that
the picture had gone but the sound
was OK and the screen lit up. That
absolved the power supply and
made the job a little less discouraging. So I agreed to take it on.
The first thing was to test the picture tube emission. This was a real
surprise. Most of these tubes now
test in the 10% to 30% range well into the red on the meter. This
one read 105%, only a little below
full scale, and the same on all three
guns.
It would be a shame to junk such
a set, so it seemed to be worth a
couple of hours of labour and a
moderate figure for parts. The
owner set $100 on the repair and I
imagined that I could do it for that
much. So I set to work.
Blue screen
ferent angle, here is a story from
my Tasmanian colleague, J.L.
Just how much is an old TV set
worth? It's one of the more unpleasant duties of a television serviceman to have to tell a customer
that his TV set is too old to be
economically repaired.
He might have paid nearly $ 1000
for it 15 years ago and finds it hard
to accept that his investment has
gone sour. Sets of that age are
usually worthless but there are
exceptions.
A Thorn 9104 came in recently
and at first glance I didn't want to
have anything to do with it.
Although it was still in excellent
mechanical condition, I had no
doubts about what it would be like
to service.
Over the years I have repaired
many of these sets and have also
42
SILICON CHIP
written off just as many with worn
out picture tubes. Most of the
repairs involved the switchmode
power supply and so long as this
could be fixed easily, the repair
was worth doing.
Unfortunately, this chassis was
built in the UK, to the usual UK
standards of complexity and inconvenience. The power supply is
extremely awkward to work on, being a three layer assembly with
very short leads linking the various
parts.
The supply cannot be operated
while it is open for service so one
is continually unsoldering and
resoldering leads if chasing a difficult fault. This alone is enough to
turn me off the model but other
parts of the set are just as confusing and any obscure fault is all it
takes to write off the set, quite
When I switched the set on, it
came up with a blue screen, retrace
lines and a faint trace of picture. It
seemed at first like a screen (G2)
supply problem and I checked the
screen voltages on the tube base.
This didn't help because they were
all about normal and the same on
each gun.
Next, I tried the grid voltages and
again, all were normal and equal.
Finally, I measured the cathode
voltages and found only the red
cathode to be normal, with the
other two only about half of what
they should have been.
I removed the video output board
and began a systematic check of the
transistors in the blue and green
channels. I could find no fault there
so I began a check of the diodes, inductors and finally resistors in both
channels. In fact, I found the fault
among the latter components
without having to measure them.
I happened to touch the heatsinks
on all three output transistors and
only the red one was warm. The
blue and green heatsinks were
stone cold. This could only mean
that the transistors had not been
turned on, or were only just turned
on. And, if so, then it could only be
due to a high resistance collector
load.
And so it turned out to be. The
special thick-film network carrying
the load resistors for all three transistors had gone funny and the two
transistors were trying to work
with loads six times higher than
normal.
Suddenly, something occurred to
me. " Whoa there! Back up a moment!" I said. " If the green and
blue channels are turned off, how
come we have a blue screen?"
I went back to the screen controls and gun switches on the convergence board. Switching off the
green and blue guns made not the
slightest difference, while switching off the red gun removed the
blue screen!
So the main problem seemed to
be gross impurity and I'd have to
get that right before I went any further. And this is where my problems really began.
I started the process by manually
degaussing the tube. This removed
the nice even blue screen and produced a varied mass of uneven colour similar to that which usually
follows the next procedure - moving the yoke to the end of its travel.
At this point I wondered if that
was all that the trouble was - just
the yoke out of position. But try as I
might, I could not find any combination of yoke and purity ring positioning that would give an all over red
screen. The best I could get was
reasonable red down the centre
half of the screen, with a wide band
of blue on one side, and the same of
green on the other.
For the moment, I returned to the
video output board. I patched a couple of 12k0 1W resistors across the
faulty thick-films, replaced the
board, and restored the three gun
switches.
When I turned the set back on,
the nice, pure centre screen that I'd
had earlier had become all blotchy
again. I degaussed the screen
again and this restored a monochrome picture down the centre,
with the green and blue bands
down the sides.
In fact, this procedure was
repeated three more times and on
each occasion the tube somehow
altered its purity each time it was
allowed to cool down. It was very
puzzling, and I had almost come to
the conclusion that the tube had an
internal fault. A loose shadow mask
was about the only thing I could
think of and I began to wonder if we
would have to write it off after all.
Degaussing circuit
Then I realised that whenever I
degaussed the tube, the central red
band became a little wider. If I kept
this up, then I might have a good
red screen by tomorrow morning!
And this led to the realisation that
in all probability, the automatic
degaussing circuit was not working.
One of the idiosyncrasies of
English design is that nothing is
where you'd expect it to be. The
auto degaussing circuit is not on the
power supply board, or the vertical
output board, or even the convergence board. It's on the video IF
board!
Fortunately, it is easy to get at
and I soon determined that the coils
were OK. I wasn't so sure about the
rest of the circuitry because they
gave some funny readings. The cir-
TV TEST EQUIPMENT
PCB POWER TRANSFORMERS
(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
HI-VOLT AGE PROBE
Built-in meter reads positive or negative 0-50kV. Can
also be switched to check negative voltages. For
checking EHT and Focus voltages as well as TVs and
Microwaves. Hi-tension voltages.
$89.00 + $5.00 p&p
LOW-VOLTAGE PROBE
Ideal for checking microwave ovens and TVs. The ~
ranges are from Oto 5kV and from Oto 1 OkV. Double
insulated for safety.
$79.00 + $5.00 p&p
DEGAUSSING WAND
•
Proudly manufactured in Australia by
HARBUCH ELECTRONICS Pty Ltd
90 George St. , HORNSBY, N.S.W. , 2077
Phone (02)476-5854, Fax (02)476-3231
Strong magnetic field, larger than usual coil with
multicore centre. Double insulated for safety with
momentary switch operation. 240VA/C 2.2
amps. As important as having a soldering iron!
$75.00 + $10.00 -,&p
TUNER REPAIRS
.
We repair most tuners on an exchange basis.
From only $17 .00 + p&p
. ...
,
.
.
. .
Cheque, Money Order, Bankcard or MasterCard
Y.V.TuNERs)
216 Canterbury Road,
Revesby, NSW 2212
Phone (02) 77 4 1154
JULY
1990
43
WOOD FOR CHIPS ... WOOD FOR CH IPS .. WOOD FOR CH IPS ... WOOD FOR CHIPS . . WOOD FOR CHIPS .. WOOD FOR CHIPS ... WOOD FOF ·
Emtek E-301
Logic Probe
•
•
•
•
•
•
.
Level Detector
Pulse Detector
Pulse Memory
HVLow/Pulse Leds
Suits TTUCMOS/HTL etc
Operates from 4.5 to 20Vdc
Specifications
Max Input Frequency
Min Pulse Width
Input Impedance
,3!> - . Over-Voltage Protection
~
. ·:, Pulse LED flash time
17MHz
30nS
1Mn
220Vac
500ms
": . •
$31.50
~,,~ :~~si=~~~:!~s
NEW Digital Engine
Analyser
~
::r:
(.J
cc
a
lJ..
Cl
a
~
~
::r:
(.J
cc
The APPA23 gives you 8 functions dwell, tacho, duty cycle, continuity,
volts, amps, ohms and diode
testing. Huge 20mm high contrast 3
1/2 digit LCD readout. Auto-poweroff to save battery life. Large rotary
switch to simplify function
selection . Case is drop-proof and
water resistant.
Brief Spees
DC Voltage. 200mVto 200V
Tacho: Oto 10000rpm and
Oto2000rpm
Duty Cycle: Oto 100%
Dwell : Oto 90" (4cyl)
Resistance : 200 to 20Mo.
DC Current: 15A
Great value at $119.00 inc tax
($99 .00 ex tax)
-::,
L: .
~
~
i(.J
a
lJ..
Cl
a
2
.,.
~
i
(.J
cc
a
lJ..
Cl
Cl
~
~
~
cc
Cl
lJ..
Cl
a
~
Fluke 80
Series ...
Top of
the line
Multimeters
The Fluke 80 series offers a huge
range of extra features like a 200kHz
frequency counter, capacitance
meter, combined analog and digital
scales plus the ability to
automatically record max/min and
average readings and store them for
u2 to 36 hoursf
c:.
a
cc
Proven the best under all conditions
from -100°C to +300°C. Laminated
cage springs on solid brass pin to
ensure virtually non-resistive
contact. Sliding action is self
cleaning. Soldering irons won't burn
them. Highest quality and flexibility.
Another must for professionals. H,A
rating. Available in black, red , blue,
yellow, green and violet and in
lengths of 500mm $6.85, 1000mm
$8.70, 1500mm $10.45.
Sa Series Features
NEW
APPA76
Meterplus
•
•
•
•
You can 't really call this one a
multimeter1 Does everything your
multimeter doesn't - tests resistors,
capacitors, trimmers, VRs, diodes,
LEDs, transistors, SCRs, batteries.
We call it a component tester
because it has none of the
conventional multimeter functions.
Brief Spees
Capacitance : 200pF to 2M
Resistance : 2000 to 20Mn
Transistor hfe and lceo Test:
0 to 1000hfe (PNP or NPN)
lceo from /OnA to 20.0£¥4
Diode: displays Vf at 1mA
LED : displays Vf at 2mA and /OmA
Battery : Tests 9V and 1.5V cells
under load
SCR: Pass/Fail
Another great instrument at
$138.00 (inc tax), $115.00 (ex tax)
•
•
•
•
•
4 3/4 digit 4000 count resolution
Digital Display updates 4 times/s
87 has analog pointer
83/85 have analog bar~raph with
zoom mode. Analog display
updates 40 times/sec
0.5Hz to 200kHz Counter with
duty cycle from 0.1 % to 99.9%
1OpF to 5µf Capacitance range
Input alert warns of incorrect
connections
87 is a true rms 1ms Peak Min
Max hold with backlit display
Touch Hold and Relative modes
83 $478.80 ($399 ex
85 $576.00 ($480 ex
87 $696.00 ($580 ex
Also ...
Fluke 23 Multimeter
tax)
tax)
tax)
Protected 1OA range plus GkV line
surge protection to IEEE-587 Cat B.
Touch Hold. Analog Bar Graph.
Yellow Industrial ABS Case.
$393.60 ($328 .00 ex tax)
'II
Weller ,~ ,
WTCPS · -~
Soldering
Iron Station
Transformer powered low voltage
features Weller closed-loop method
of controlling maximum tip
temperature to protect temperature
sensitive components.
• 48W stainless steel heater
• Non-burning silicon rubber cord
e 15 tip sizes - 0.8mm to 6.00mm
e 11secs to recover for 37°C drop
• Choice of tip temperatures 315°C/
600°F 370°C/70°F 430°C/800°F
'Weller WTCPS $137 .00
Weller EC3001 D Electronic
Controlled Solder Station $325 .00
Weller EC1001D Electronic
Controlled Solder Station $195.00
Replacement Tips
$8.95
We have a huge range of irons lo
suit every requirement
Witte Screwdriver
Set
This set is for the professionals.
Rated to 1000V. Fully insulated right
through to the tip in a single hard
moulding. Contains 4 flat blades,
6.5 5.5, 4, 2.5mm and No 1 and No
2 Philips head plus a pocket mains
testing screwdriver. Made in
Germany with a 5 year guarantee
and VDE approved.
A must for electricians .
$49.50
Individual screwdrivers available
Flat• 2.5mm $4.20, 4mm $5.40,
5.5mm $7.80, 6.5mm $10.50
Philips Head • No 1 $6.40, No 2
$8.40
Wire Wrap
We carry exte_nsive stocks of wire
and tools.
Wire Wrap comes in two pre-cut
lengths - 10.16cm and 12.7cm
30AWG at $4.80 and $5 .60 per 100
lengths. Also on spools of 30m for
$9.95 .
We also have a wire kit consisting of
200x 6.25cm, 250x7.62cm,
100x8.89cm, 100x10.16cm
100x11.43cm and 100x13.97cm. All
30AWG and colour coded for$34 .80
Tools
The OK MST hand wrap tool is just
$19 .50
The OK Cutter and Stripper is
$28.50
Enquire about wire wrap sockets.
Our range is too extensive to list
here.
.,.... """.,."'
""'"' ' l. .,. .,..,,.,~"'
.~~
AilJ!!!'i;ill""'
AT&M KITS
We carry the full range of the ureat
kits. Why develop your own circuits
when these building blocks are
already available?
Kit
Description
Price
ATM01 3 digit Counter
$33.60
ATM02 4 digit Counter
5D.40
ATM03 31 /2 dig it DVM
59.40
ATM04 1Hz Timebase
22.40
ATM05 Fibre Optic Modem 75 .00
ATM06 Reciprocal Calculator 29.40
ATM07 10MHz Timebase
45.60
ATMOB V/F Converter
42.60
ATM09 FN converter
42 .60
ATM/0 Quad Power Supply 60.60
ATMl 110 Minute Controller 46.20
ATM/2 6 dig Batch Counter 75.00
ATM13 3 Digit Counter
65.40
ATM/4 3 1/2 digit LCD DVM 42.60
ATM16 10kV Isolator
70.80
ATM/ 7 Storage Adaptor 193 .80
ATM/8 8 bit UART
66.00
ATM/9 6 dig Up/On Counter 69.60
ATM20 IR Transceiver
39.60
ATM21 Multi-Timebase
42.60
ATM2216Ch Tranmitter
77.40
ATM2316Ch Receiver
77 .40
ATM24 22/3 Com ms PCB
34.80
ATM25 22/3 Radio Modem 66.60
All prices INCLUDE sales tax.
~
Cl
Cl
.,.,
a
::0
"::r:
~
Cables
Al/2m long
D9 plug to D915 socket (that's the
triple row 15 pin in a 9 shell) $29.95
D9 socket to D25 Plug
$19.95
D9 plug to 8 pin Mini-DIN $15 .00
D9 Plug to D9 Plug
$19 .95
D9 Socket to D9 Socket
$19.95
D9 Plug to D9 Socket
$19 .95
AM/GA Users
We have the D23 Plug to D23
Socket you're looking forl $29.95
Another Problem Solver
Our 4 way, 2metre adaptor.
Consists of Male and Female D25 to
Male and Female D25. Ideal for
service technicians and installation
people. A very versatile piece of
cable!
$34.00
NEW 2-Way' /,""
Automatic !/
Data
II'
Switch
............
Now two PC's
can share one
printer automatically.
it
The printer (or other peripheral) is
allocated on a first-come-first-served
basis. LED indication tells you which
PC is connected. No external power
and can be easily attached to any
surface using velcro pads supplied.
Suits any PC and Printer with parallel
(Centronics) connections using
DB25 connectors. $99 .50
la
.,.,Cl
a
::0
"::r:
~
la
a
.,.,
a
::0
"::r:
~
lCl
Cl
.,.,
a
::0
":i::
~
lCl
Cl
.,.,
Manual Data Switches
a
Total flexibility at very modest cost.
2 Way Seriaf (RS232)
$45.00
Parallel (Centronics)$55 .00
4 Way Serial (RS232) . · $70.00
Parallel (Centronics)$80 .00
"'ii::r:
::0
V)
lCl
GEOFF WOOD ELECTRONICS PTY LTD
229 Burns Bay Road, (Corner Beatrice St.) INC IN Nsw
Lane Cove West, N.S.W. P.O. Box 671, Lane Cove N.S.W. 2066
Telephone: (02) 428 4111 Fax: (02) 428 5198.
8.30am to 5.00pm Monday to Friday, 8.30am to 12 noon Saturday.
Man Orders add S_5 00 to cover postal charges.
Next day delivery in Sydney add $5.00.
c
Cl
.,.,
a
::0
~
"":"'
1
.
ro
GLADESVU.E
Tax exemption certificates accepted if line value exceeds $10.00.
BANKCARD, MASTERCARD, VISA, CHEQUES OR CASH CHEERFULLY ACCEPTED
"'ii::r:
V)
SERVICEMAN'S LOG -CTD
cuit consists of just three components. Only one has a logical part
number - R199, a 6800 5W wirewound resistor. The other parts are
X199, the usual PTC thermistor,
and Z199, a device which looks like
a 40mm diameter disc ceramic
capacitor.
This latter device was in series
with the degaussing coils, and had
a resistance of several megohms.
This seemed to be all wrong and for
the moment, I couldn't identify its
symbol on the circuit diagram. So I
had to find out what it was, and if it
was indeed faulty.
In some respects, I'm glad I'm a
hoarder. Although it is hard to live
among all the junk, there are times
when it becomes invaluable. Like
when I want an old Thorn IF board
complete with degaussing components.
In fact, I found three such boards
and all three gave readings similar
to the suspect one, which seemed to
clear Z199. I still wasn't sure what
Zl 99 was but was beginning to
recall that the symbol could mean a
voltage dependent resistor, which
would have a high resistance at low
voltage - including the voltage ap-
plied by the ohmmeter. On this
basis, a theory as to the circuit's
purpose and operation began to
evolve.
It seemed most likely that its purpose was to reduce the residual
current in the degaussing coils to a
minimum, following the degauss
function. This theory was eventually proved correct but more on that
later.
Anyway, since the circuitry
seemed to be OK, the next test was
X199 PTC
THERMISTOR
200V'--.\oliliill.J.-----,
Fig.2: the degaussing circuit used in
the Thorn 1904. The VDR (Z199) and
the 6800 resistor (R199) network was
an early refinement to the
conventional degaussing system.
to see if it actually passed any current. For this I used an AC clamp
meter. I hooked the clamp around
one of the AC leads and switched
on. The needle didn't move.
This meant that there was either
a break in the circuit that I hadn't
found, or no voltage was being applied to it. A check across the
degauss network showed only BOV,
whereas the circuit diagram indicated there should be at least
200V.
There were no less than three
plug and socket connections between the degaussing circuit and the
power transformer. None of these
was easy to test but I soon found
myself at the relevant terminals on
the transformer. And there was the
trouble - the winding was open
circuit.
This was quite a puzzle because,
according to my circuit , the
degaussing system is fed from the
same winding that supplies the HT
to the chopper transistor. The chopper was working quite normally, so
I had to assume that there was a second secondary, so to speak.
Then my magpie tendencies came
to my salvation - again! Just outside the workshop was the latest of
these Thorn sets to be junked. It
was only a few minutes work to
remove its power transformer and
fit it in place of the faulty one. At
the next switch-on, the degaussing
coils gave out a good solid "boing"
and the last of the impurity
disappeared.
The next thing this set needed
was a thorough convergence setup.
These old delta gun sets never had
enough convergence adjustments; it
always seemed that the one you
wanted was on a different brand of
set. This one needed at least four
controls which were not provided.
But I perservered and eventually
produced a reasonably well converged picture.
But it was still only in monochrome.
By this time I had spent much
more than the two hours I had
budgeted for the job. I had replaced
the load resistors on the video
board and the power transformer.
I had no idea what would be involvcontinued on page 99
JULY1990
45
Dual power supply:
ctd from page 83
hence reliability. A larger heatsink
may also be necessary.
Performance
This unit can be thought of as two
power supplies in one. As a dual
voltage variable supply it is capable
of delivering up to 150mA at ± 18V
and in excess of 250mA at ± 15V
or less at very low ripple, and with
excellent line and load regulation.
The Serviceman's Log:
ctd from page 45
ranges from 4k0 to 400k0, ± 1.2%
for the 4MO range and ± 3.0% for
the 20MO range.
The accuracy for frequency
measurements is ± 2 % on the three
ranges from 4kHz to 400kHz.
The unit also has a continuity
and diode test range with audible
alarm and a data-hold feature
which stores the last measurement.
Priced at $169.50, the Metex
M-818 is available from all Jaycar
Electronics stores.
48V switchmode
supplies from Amtex
ed in restoring the colour, and was
beginning to regret ever having
started the job.
Then I thought, "Damn it! I'm
never likely to need the chroma
board out of this junked set. Why
not swap it for the one in this set?"
And it was no sooner said than
done.
That fixed it and the old Thorn
went home with a new lease of life.
Its picture tube is likely to last
another 15 years but whether the
circuit boards will last that long is
another matter. Then again, there
are likely to be more of these old
Thorn sets coming in with dud
tubes, so perhaps I can keep this
one going beyond my retiring age!
How it works
This new range of switchmode
supplies provide an output of 48
volts DC with output currents ranging from 1A to 10A.
Each unit has short circuit and
output overload protection as well
as switchable 110/230 VAC inputs.
Also included is soft start power up
circuitry and input surge protection.
For further information, contact
Amtex Electronics Pty Ltd, 13 Avon
Road, North Ryde, NSW 2113.
Phone (02) 805 0844.
With time to think, I evolved the
following theory about the workings
of the degaussing circuit. At the moment of switch-on, the PTC thermistor, X199, is cold and its
resistance is low. This allows
almost the full voltage to be applied
to the degaussing coils and Z199
which, at the high voltage, offers
minimum resistance.
In its fixed voltage guise, it can
provide in excess of 550mA at 5V
with short circuit protection provided by R7 and internal foldback current limiting.
As a combination unit, with both
supplies in use, the 5V specifications still stand but, as you might
suspect, the 18V output may not be
usable depending on how much current is drawn from the fixed supply.
The lower voltage supplies are
unlikely to be affected if current
demands a re reasonable.
~
As Xl 99 heats up and its
resistance rises, the voltage applied
to Z199 and the coils will fall and
the reduced voltage across Zl 99
causes its resistance to rise. But
this would allow X199 to cool and
increase the current through the
coils.
So the function of R199 is to
maintain sufficient current through
Xl 99 to keep it warm and its
resistance relatively high. At the
same time, most of the current it
does pass will flow through R199,
rather than the coils.
Subsequently, a colleague came
up with two text books which confirmed this theory: "Colour Television;' by R. N. Patchett, and "PAL
Colour Television for Servicemen"
by W. C. Cook.
That leaves one point still unexplained; the 80V I measured across
the degaussing circuit, connected to
an open circuit transformer winding. How come?
It was another of those (expletive) digital multimeter misreadings. Apparently there was enough
capacitance, or leakage, across the
break in the transformer winding to
develop 80V at the picoamps needed to drive the DMM. Later, using a
100,0000/V analog meter, I could
not detect any voltage.
~
Is Your Product Getting
The Exposure It Deserves?
Consumers need to see your product
if you want them to buy it
Contact Paul To Reserve
This Space - (02) 979 5644
]UL Y 1990
99
|