This is only a preview of the February 1992 issue of Silicon Chip. You can view 45 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 "Build A Stereo Power Amplifier":
Articles in this series:
Articles in this series:
Articles in this series:
Articles in this series:
|
SERVICEMAN'S LOG
Some sets go on forever
There is always a small percentage of older sets
being presented for service and they usually call
for a special approach. Are they worth fixing?
Are any hard-to-get components involved? How
valuable is the set to the customer? In practice,
each case has to be treated on its merits.
This story concerns one of those
old timers - a Rank Arena model C2053. This was one of several Rank
models which appeared shortly after
the introduction of colour TV in Australia. These included the C-2601, 02,
03, 04, 05 and 06 (all 63cm models),
and a similar range of 51cm models
starting at the C-2201.
The set in question comes from a
third range with the 20 prefix - 48cm
(20in) - but the suffix 53 suggests that
it is a later model than those listed
above. However, there is some doubt
about this and I am not sure exactly
where it fits into the manufacturing
sequence. Significantly, it has a VHF
only front end.
However, the point about mentioning these ranges is that all the sets
from them are very similar. Many of
the boards are used in all models, the
main differences generally being confined to just one or two boards. Typically, this includes the output deflection board which is changed to suit
different size picture tubes.
Something old
In this case, the set was almost the
same as the 2601/2201 series, the most
obvious difference being the "Deflection Out" board which is designated
PWC-497 (PWC-315 in the 2201).
The customer brought it in together
with a Sharp video recorder which, in
addition to its normal role, was being
used with the Rank as a UHF downconverter. Both units needed attention but the video recorder was a fairly
routine problem. The
Rank TV set was not
quite so straightforward.
These early Rank
sets must have set
some kind of record
for longevity; some
owners have likened
them to the old DC3
aircraft - they just
seem to go on forever.
One 2601 that I know
of started life just
after colour was introduced (in 1975),
probably chalks up
three
hours or more a
SON\~ OWN~'R'S 'r\A.\J6:
day, and is still going
L-\ Ke:.N E.t> 1l-lE.SE::'. S:.A"R\..-'(
strong with plenty of
RANK St;."'t'"S , D OL-D
brightness and con-OC. 3 A\~Ct<.AF~, ..
trast in the tube.
56
SILICON CHIP
And that raises the question as to
how long is it economical to keep any
particular set going. Much depends
on the condition of the tube. If it is in
good nick, and the fault is unlikely to
involve other specialised components,
then repairs may well be justified. It
is really up to the owner.
On the other hand, the need for a
new tube is another matter. Normally
a reconditioned tube would be available and, for what it is, the price is
very reasonable. But by the time fitting and other costs are added, the
proposition begins to look a bit dubious.
To justify it, one must assume that
the set is going to run for several more
years without developing any serious
faults. Again it is up to the owner but,
when I point out these factors, most
decide to cut their losses and put the
money towards a new set; one which
will normally offer more features than
the old one.
And so the old set finishes up on
the tip. But not before yours truly has
picked it over. The various boards, in
particular, are well worth salvaging,
especially in view of their universal
nature, as mentioned above.
Something wrong
And that brings us back to the set in
question. A quick check suggested that
it was well worth saving. It was working (almost) and the picture tube appeared to have plenty of life left in it.
The main problem was that it had no
vertical or horizontal sync.
1\viddling the vertical hold control
brought this back into sync and with
a good.normal locking range. No problem there. But the horizontal system
was a different matter. I could lock it
but it was extremely critical. Something was obviously very wrong.
Nevertheless, I felt that the set was
well worth saving and advised the
customer accordingly. I was banking
on the fact that the fault would almost
certainly be in the "DefBoard", PWC-
PWC - 367
DEF BOARD
RS6C
6. 2 K
I
1
RS':> 5
~
- ~==
~.5'l.
1
..''c,'
fi~J.~
SL:' :;,
_: R.:."~ 1$C~!lJ..~.L
0501
s~if~1M
~-~l;
J;;: ◄ 7 l ; ii~R
t~3~
~~s~
,.,.. -,.
~~i~
r ~~J
...
1n~:~
"-,~,:',::::~,·";::_l~.-c,-_s,,_: _-,;;-_-~.-':S
!____ , f/;_'..:,_. ,. ~,,
_,_. - - - - - - c""'-o~ 5=,-
' ...
I ~•;~·/':/ _j
"'
SJ ;
':'CF
-
,-.
. f :>??. ,I
I
l
3..
7
l- ~·!
1
~:-:·f'
i ~}--:c .,
• ;i:' i l
; ~ --
f'-+ -
I
r!!io
,J,
--'-'-0
~~;;::'-_-·. ·-~;-_
,:~_-~\-._ ______,I
n. I 'f',_,,,
,I Il
'
I
~-LJ
9:~-- - - - t.___. ,,. :- : ,,
Fig.1: part of the horizontal deflection circuitry in the Rank Arena C-2053. The
horizontal output transformer, T552, is at extreme right, the connection between
PWC-497 and PWC-367 at centre, and diodes D501 & D502 at left.
367, which carries the vertical and
horizontal oscillators, plus the sync
circuitry.
And, with the stock of spare boards
I mentioned earlier, it should be a
snack. In fact , this is a technique I
have been using extensively on old
sets like these. A board replacement
is usually all that is needed, the old
board can be tossed,, and there is no
time wasted searching for elusive
faults. Nothing to it!
So it was out with the box of discarded boards which quickly yielded
a PWC-367. I then tackled the fiddly
job of unplugging the umpteen connectors on the suspect board, fitting
the replacement board, and replacing
the connectors.
The only snag was that it didn't
achieve anything; the fault was just as
evident as before which was a bit of a
blow. I had been so confident that the
fault would be on this board that I
hadn't considered any other possibilities. But now I had to.
The next likely suspect was the
Video IF board, PWC-312 , and the
possibility that it was not delivering
proper sync pulses. The video signal
comes out of PWC-312 on connection
B1, enters PWC-367 on B1, and goes
to the sync separator (TR401).
I took a look at this waveform on
the CRO and, as far as I could judge,
all was normal. But I was at something of a disadvantage here because
the circuit diagram carried no waveforms - a common shortcoming of the
Rank manuals. The closest data I had
was from the 2201 manual; one of the
few that did carry waveforms. In this
case, the relevant waveform, from
TP17, looked similar to the one I was
viewing. However, the manual's reproduction wasn't particularly good
and I couldn't be certain that there
were no subtle differences.
I decided that the easiest way to be
sure was to make another board switch
- except that when I delved into the
box for a replacement PWC-312, there
was none. So that brought the whole
exercise to a halt for the moment.
In any case, pressure of other work
dictated that I put the set aside for the
TETIA TV TIP
Sony KV2764EC
Symptom: the set had been cutting out but finally died altogether.
The fault was on the 20V rail and
turned out to be a shorted sound
output chip IC252. Within a few
days, the set was cutting out
again and the serviceman attended to it in the owner's home.
Cure: it turned out that the owner
had connected two large external speakers to the set. The
sound output chip wasn't intended to drive this sort of load
and is fitted with a very small
heatsink. Removing the speakers solved the problem, although
a larger heatsink might have
been an alternative.
TETIA TV Tip is supplied by the
Tasmanian branch of the Electronic Technician's Institute of
Australia. Contact Jim Lawler, 16
Adina St, Geilston Bay, 7015.
time being, which would also give me
time to think. It was a couple of days
later before I thought much about it
again and then only because I was
looking for something else in the storeroom. Among other things, this storeroom also houses scrapped sets and
components and I realised that there
was a complete Rank chassis waiting
to be stripped.
So there was the PWC-312 video IF
board I needed. I fished it out, dusted
it off and fitted it into the set. And
that was the second setback; the fault
w as still there, exactly as before. That
meant that the time for short-cuts was
over; I now had to get the brain _into
gear and do some real troubleshooting.
Something on my mind
However, right now, there were
more important jobs to be attended to
and so, once again, the set was put to
one side. But it was never too far out
of mind; I once knew these sets inside
out and it was shock to realise that
time had erased this familiarity. If the
fault wasn't on either of those two
boards, where was it? I was obviously
missing something but I couldn't figure out what.
The set was apparently more on my
mind than I realised. In the wee small
hours one morning several days later,
I awoke with a start and could visualise the circuit quite clearly. More than
that, I knew where to look for the fault
(the brain sure works in a mysterious
way).
Cursing myself for being all kinds
of a stupid clot, I realised that the
third board in this group, the "Def
Out" board (PWC-497), had to be the
culprit. Why? Because the discriminator section of the flywheel sync cirFEBRUARY 1992
57
SERVICEMAN'S LOG - CTD
cuit- diodes D501 and D502 on PWC367 - needs a reference pulse from
the horizontal output transformer before it can function.
I could hardly wait to get to the
bench that morning, fish out the circuit, crank the set up and delve into it
w ith the CRO. But there was the major snag I mentioned earlier; the
manual has no waveforms and very
few voltages. The best I could do was
fall back on the 2201 manual but this
was of limited value in this case because the Deflection Out board for the
set in question uses a different circuit
arrangement.
However, the 2201 manual was of
some help. It shows the pulse line
coming off pin 3 of the horizontal
output transformer and designates this
as waveform TP32. This shows sharp
spikes, at horizontal frequency, with
a value of 250Vpp.
The 2053 shows the pulse line coming off pin 9 and the CRO indicated a
virtually identical waveform. But
58
SILICON CHIP
whereas the 2201 pulse line uses a
passive arrangement, the 2053 adds a
transistor (TR505) as an AFC amplifier.
In both circuits, the pulse line leaves
the Def Out Board and enters the Def
Board via the H4 pins. The 2201
manual gives no waveforms along this
line - which features various wave
shaping comp£?nents - but it does
show the waveform (TP26) at the junction of the discriminator diodes as a
sawtooth pattern with a value of 4 Vpp.
From this , I felt that it was safe to
assume that a similar waveform would
be required at this point in the 2053 .
Well that was something, even ifI had
no indication of what to expect at
various points along the line.
In fact, a check at H4 on the Def Out
Board was most revealing. The best
description of the waveform is that it
was similar to that from a half-wave
rectifier, consisting of a series of
roughly sinusoidal pulses separated
by spaces of approximately equal
length, and with an amplitude of about
2Vpp.
Well, even if I didn't know exactly
what the waveform should be, I was
fairly certain that it shouldn't look
like that. And because there was no
problem with PWC-367, the fault had
to be on PWC-497.
Something suspect
My first suspects were two electrolytic capacitors in the pulse line: a
lµF type, C558, ahead of the AFC
amplifier transistor (TR505); and a
10µF type, C556, following it. Low
value electrolytics are notorious for
drying out and these were quite old.
The lOµF capacitor was marginally
easier to get at, so I lifted this first. It
checked OK. Not surprisingly, lifting
this capacitor completely killed what
little horizontal sync there had been. I
could float the picture but that was
all. But the real surprise was that this
also killed the vertical sync.
I have no explanation for this. I
have poured over the circuit trying to
find a link - even an accidental one but without success. No doubt, given
the time and the right facilities, the
reason could be found
but there is a practical
limit to the time one can
spend on such problems. Perhaps a reader
has the answer, in
which case I'd be happy
to hear it.
Getting back to the
main problem, I next
tried the lµF capacitor,
feeling certain that it
would be the culprit.
Again I drew a blank. I
also drew a blank with
all the resistors in the
chain, which left only
one thing - the transistor. It tested perfectly
but I've been caught that
way before. The only
real test is to replace it.
I didn't have a 2SA539
on hand, so I substituted
a BC556.
And that was it - the
system locked rock solid
with a good wide locking range. Problem
solved. And when I
checked the waveform
at H4, there was a sawtooth similar to the one
shown at the discriminator diodes in
the 2201. Its amplitude might have
been a fraction less but, if anything, it
was a better sawtooth than the one
shown.
But what was wrong with the transistor? It tested as having normal gain
and there was no indication of leakage. I can only speculate that, for some
reason, it had lost its high-frequency
response . As a result, it could no
longer handle the higher frequencies
which make up a sawtooth waveform
and reduced it to something like a
sine wave.
So that was the end of the exercise
and the set went back to a very satisfied customer. However, one of the
points arising out of the exercise is
the difficulty of working without
waveform and voltage information in
service manuals .
And I'm not just harping on about
the Rank manuals; they are history.
What I'm really concerned about is a
similar trend in the latest manuals; a
complete lack of waveform data and
only limited voltage data. Samsung is
one of the most serious offenders at
present.
So serious is this that I am looking
for some convenient way of building
up my own waveform references as
sets pass through the workshop.
Photography is one approach but it
has serious practical limitations. For
starters, this would require a suitable
photographic jig attached to the CRO,
to reduce the process to simply pressing the button. And there is a real
need to record the test point details
on the film, along with the waveform.
Any other arrangement is prone to
mistakes .
A more recent approach is a system
which can couple the CRO to a computer and, ultimately, reproduce the
waveform via the computer's printer.
I'm looking into this one. But it would
be a lot easier if the makers did their
job properly.
FRACARRO
SAVES...
... more than $10. from every UHF
television antenna you buy.
Fracarro's
4PUN
for better
performance
from weak
UHF
signals.
Peter C. Lacey Services Pty. Ltd.
P.O. Box 678 (74Fulton Rd.) Mount Eliza 3930
Tel:03 787 2077 Fax:(03) 787 3460 ACN006893438
Something new
For a change of theme, we now go
from one of the oldest sets around to
one of the latest. This is a 37cm colour TV set which I recently acquired
for use as a workshop monitor. Apart
from a minor technical problem, there
are a number of interesting points
about this set. It carries the Hitachi
brand - or Hitachi Fujian Model HFC-
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
Bu ilt-i n meter to check EHT transformers including
sp lit diode type, yokes and drive transformers .
$78.00 + $4.00 p&p
1425B to be precise - but it is not
marketed by Hitachi. It is made in
China and marketed in Australia
through the China Trade and Exhibition Centre, 26-34 Dunning Avenue,
Roseberry, NSW 2018.
I have dealt with Chinese sets before in these notes and, not to put too
ANTRI
TOROIDAL TRANSFORMERS
HIGH-VOLTAGE PROBE
Built-i n meter reads positive or negative vo ltages from 0-50kV. For checking EHT and other HT vo ltages.
$98.00 + $5.00 p&p
DEGAUSSING WAND
Strong magnetic fie ld. 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 ultrason ic
contro l un its. Supplied with extension
infrared detector lead . Output is via a LED
and piezo speaker. $84.00 + $4.00 p&p
:s.. _, "S;'--
-.J\+~// t-~
~~ -- TUNER REPAIRS -~
~~~
From $19 exchange
plus post & pack
Cheque, Money Order, Visa, Bankcard or Mastercard
.,;v! !,U~ERS} ~~~e~Jl~:i~.~~1d2,
Phone (02) 774 1154
Fax (02) 774 1154
•
•
•
•
Now made in Australia
Standard range 1SVA to 2kVA
Large ex-stock quantities available
Competitive pricing available for OE Ms
& resellers
• Simply the best transformers all round
HARBUCH ELECTRONICS PTY LTD
90 George St, Hornsby, NSW 2077.
Phone (02) 476 5854 Fax (02) 476 3231
FEBRUARY1992
59
spare parts backup? This a grey area
at the moment. The set is sold with a
12-month replacement warranty; that
is, if the set fails within 12 months, it
is replaced with a new set. After that,
it appears that you are on your own.
This may not be too bad provided
there is a guaranteed supply of major
spares but I have not been able to get a
clear ruling on this to date. Frankly, I
consider that this point could make
or break the commercial operation in
Australia.
SERVICEMAN'S LOG - CTD
fine a point on it, I was not particularly impressed by the workmanship
involved. So to be fair, I must say that
this effort appears to be a better proposition.
Time alone will tell, of course, but
my first impression is of a very wellmade unit offering a large number of
features at a very competitive price
(around $350). Recently, I was interested in acquiring another VCR monitor for the workshop and so, when my
dealer colleague put some on display,
I had a closer look. In fact, he was
happy to lend me one to evaluate, as
much for his benefit as mine.
I was particularly interested in the
remote control features and on-screen
displays, which are quite extensive.
This can be very valuable in practice,
as when soak-testing VCRs for intermittent faults. It also allows the monitor to be located in the best viewing
position, regardless of accessibility.
The remote facilities include: on/
off control; off-timer, selectable for
30, 60, 90, or 120 minutes; program
Something wrong
mode providing brightness, colour and
volume control (with on-screen value
bars); on-screen clock; and step-bystep channel selection or specific
channel selection via a set of 12
pushbuttons. In the end, the features
were impressive enough for me to
clinch a deal with my colleague.
But what about the warranty and
Getting back to my own set, I soon
discovered a problem with the onscreen clock, which seemed to work
OK initially but, after about a day,
appeared to be running fast. Sure
enough, a more careful check revealed
that it gained nearly five minutes over
24 hours. That puts it on a par with
the old alarm clock that used to tick
away on our kitchen mantlepiece
when I was a kid but I expected something better from a crystal.
The crystal in question (Z101) is
the same crystal that serves the microprocessor and is connected between
pins 28 & 29 of this IC. It is marked as
Introducing the Protel advantage.
Protet®for Windows':'
Prate!for Windows is an entirely new Printed Circuit Board design system created for Windows 3 . Prate! for Windows
brings the professionalism of Protel's proven PCB design software to the highly productive Windows 3 environment.
New Prate! users will start designing sooner and experienced users will work more efficiently than ever. In 386
advanced mode, Windows 3 provides vi11ual memory capability, which allows advanced PCB users to design without
restrictions on the total number of components, nets, tracks, etc. Prate! for Windows supports the multiple document
interface (MDI) standard. Any number offiles can be loaded
Special-------~
Copper Trace Layers
□ Midl
- □ Mid8
- □ Mid9
□ Mid 3
!ll □ MidlO
l!ffll □ Mid 11
iZITop
□ Mld4 - □ Mid12
□ Mid 5
OMid6
□ Mid 7
liifil □ Mid 13
□ Drill Guide □ Drill Draw -
fiiil
-
181 Keep Outs Ill
IZl Multi Layer !iifil
-
M
-
-
Rats Nest Selections
CJ
Background
ech 1
D Mech 4
IPoslScript Printer on LPT1:
D Bottom
D
Batch Mode
I [Silks
Bottom -
.
IZl Tc
0
IZl Grid 211tiJ
Seperate Page For Each Print
0
Composite Print
®
Panels (Multiple Layers Per File}
Border Between Prints
~
at the same time, using standard Windows routines, to move
iriformation between different files or different applications.
The Prate! for Windows family consists of Advanced PCB -
-
~--~
IZ] Transparent
Ill D Mech 3 lll.ll D Dithered Colors
ech 2 -
IZ] Bottom -
rsMT Paste Mask
LD Top
D
hanical Layers
! Solder Mask
I D Top
□ DRC Errorsfiiil IZ] Grid 1-
CJ
fll
CJ
- □ Mid14 (Ill
Visible-
advanced level PCB layout - $1990. Options: Advanced Place
$995 , Advanced Route $995 . Professional Schematic' -
schematic capture -$129,.5. Option: Digital Simulator• $795.
■
Protel Technology Pty Ltd
PO Box 204, Hobart Tasmania 7001
Phone (002) 730100 Fax (002) 730944
Tollfree 008 030 949
T?! chnology
~Available April 1992. Prate/ is a registered trademark. Admnced PCB and Professional
Schematic are registered trademarks of Protei Technology Pty Ltd. V1indows is a registered
trademark of Microsoft Corporation Inc.
60
SILICON CHIP
4MHz, although I considered the possibility that this may
be only a nominal value.
Initially, I had some ideas about trying to find out
exactly what had gone wrong but the more I delved into
the problem, the more complex it became. For one thing,
what was the precise intended frequency of the crystal?
Was it exactly 4MHz or merely something very close to it?
On the basis of a simple sequential 2:1 division chain,
a 4MHz crystal is not practical. But 4.194304 will divide
exactly to lHz, or 3.932160 will divide exactly to 60Hz,
either of which can form the basis for a clock display. But
the simple 2:1 sequence is not the only option. Modern
chips can provide virtually any order of division and it is
quite conceivable that the makers used a more complex
network in order to accommodate a standard 4MHz crystal. These are used extensively in computers and are quite
cheap.
The situation was further complicated by the virtual
impossibility of making an accurate crystal frequency
measurement directly across it. Any connection here
would load the crystal and make the reading meaningless.
Significantly, several computer buff acquaintances have
told me that they have similar problems with the realtime clocks in their personal computers. In all cases, their
clocks run fast. One enthusiast, who uses the clock mainly
to maintain the calendar, realised recently that, since he
had last set it, it had gained 17 minutes!
I finished up using the more or less brute force approach of loading the crystal circuit with additional capacitance until an acceptable result was achieved.
The circuit shows two 33pF capacitors, C152 and C153,
connected from each side of the crystal to chassis. I
decided to add some capacitance in parallel with each
and started with two 13pF units, tacked on under the
board.
This came out better than I had hoped. Not only did the
crystal accept the idea - they can sometimes drop out or
behave erratically with excessive loading - but the error
was significantly reduced. The clock went from gaining
five minutes in 24 hours to losing about 25 seconds over
the same period.
Electronics
Workbench®
66660660
Build and simulate both analog and
digital circuits, complete with simulated
instruments on your computer!
FEATURES
• Quick and simple circuit entry • Digital and Analog
Modules included, complete with all components • Simulated
instruments: dual trace scope, spectrum analyser, function
generator, multimeter, digital word generator and logic
analyser • Complete control over all component values and
parameters • Print: circuit schematics, parts list, instrument
readings, macros • Logic conversion -truth table to Boolean
formula to logic gates • Customisable hypertext help system
THREE VERSIONS
• Professional Version: EGA/VGA colour display;
unlimited components
• Personal Plus Version: Monochrome display; unlimited
components
• Personal Version: Monochrome display; limited to 20
----------------I
YES
■
Something logical
OK, so I had overshot slightly and needed to decrease
the additional capacitance. A small trimmer seemed to be
a logical alternative at this stage. I didn't have anything
suitable in stock but, while looking for something else
later on, I found an old VCR board with a miniature
trimmer across a crystal.
It was close to ideal. It measured from 7-20pF and
needed only minor doctoring of the leads to make it
suitable. I removed the 13pF fixed unit from C152 and
fitted an 11 pF capacitor in its place. I then fitted the
trimmer across C153.
Over the next few days, I juggled the trimmer on a trialand-error basis and eventually reduced the error to a
couple of seconds a day. And that's probably as good as I
will do, because there does seem to be some drift which is
probably due to small variations in the trimmer's value. A
better quality unit ::nay help but would be hard to justify.
Well, that's it for this month. If I learn anything more
about this set, I will pass it on.
SC
~ want ELECTRONICS WORKBENCH
m my computer
$429
$299
$149
Professional Version
Personal Plus Version
Personal Version
Plus $7
Delivery
Enclosed is my cheque/money order for $._ _ _ _ _ or
Please debit my
Ca
rd
No.I
J
J
O
J
J
O
Bankcard
J
J
J
J
J
J
Mastercard
J
J
J
J
J
I
Signature _ _ _ _ _ _ _ _ Expiry Date _ _/_ _
Name _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ __
Street _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ __
Suburb/Town _ _ _ _ _ _ _ _ Postcode _ __
Send to: Emona Instruments,
PO Box 15, Camperdown, NSW 2050.
~-----------------·
Tel: (02) 519 3933
Fax: (02) 550 1378
FEBRUARY1992
61
|