This is only a preview of the September 1988 issue of Silicon Chip. You can view 43 of the 104 pages in the full issue, including the advertisments. For full access, purchase the issue for $10.00 or subscribe for access to the latest issues. Articles in this series:
Items relevant to "High Performance AC Millivoltmeter":
Articles in this series:
Articles in this series:
Articles in this series:
Articles in this series:
|
Helping the old folk at home
Pride of place this month goes to a story from my
Tasmanian colleague, J.L. Apart from technicalities,
it has a high human interest, serving to remind us
that, TV program knockers aside, there are many
people for whom this is the only form of
entertainment available.
Technically, J.L. 's story emphasises the problems created by
the need to expand an antenna
distribution system as the demand
for outlets grows. Anyway, this how
J.L. tells it.
"Got a light mate?" The sepulchral voice boomed out of the
darkness in the ceiling space over
the old people's home at Taroona,
south of Hobart. I had been called
in to add a new outlet to the 60-odd
outlets already connected to the
home's TV antenna system. The last
thing I expected was to be asked for
a light amid the dust and fluff between the rafters.
The enquiry came from a
Telecom technician who was looking for a phone line terminal board.
I had often come across phone lines
among the power and antenna
NURSE'S
LOUNGE
KITCHEN
CHAPEL
➔-~
o-4 ·nl
➔
OINING
ROOM
Basic layout
RESIDENT'S
LOUNGE
2 -WAY
SPLITTER
--<>
TV OUTLETS
4- AY
SPLITTER
I
<$>-o
~
I
I
D
SPLITTER
[>
AMPLIFIER
◊
OROPTAPS
ADMINISTRATION BLOCK
HOSPITAL ANO RESIDENTIAL
WEST WING
Fig.1: general layout of the retirement home installation. It is not possible to
show each individual T and outlet, but the order and value of T's shown is
typical. Note the splitter feeding the nurses' lounge.
38
STLICON CHIP
wires but I couldn't help regarding
the terminal panels. The best I
could do was lend him my lantern
so that he could continue his search
in greater comfort.
He surprised me by revealing
that Telecom employees don't rate
a torch or lantern as part of their
normal kit. They have to draw one
from store and he had not done so
because he didn't realise that he
would be working in the ceiling.
This is probably the most surprising encounter I have had since I inherited the installation. I have had
many other encounters over the
years, some uplifting and some
tragic but all of them interesting.
I was called to the home for the
first time when the system consisted of about 40 outlets. Since
then I have added another 20 and
another 20 need to be added before
every room has an outlet. It is quite
an installation.
The home is made up of two
residential wings which form two
parallel sides of a rectangle. What
is called the west wing is a single
storey building which houses the
hospital and some 25 residential
rooms. The east wing is a double
storey building of about 46 rooms.
An administration block forms the
third side of the rectangle and a
chapel, residents' lounge, dining
room and kitchen, the fourth side.
The system began soon after the
home was completed, not long after
TV started in southern Tasmania.
But as with many new buildings, little thought was given to the provision of either telephone or TV services, and both facilities have had
to be added piecemeal over the
years.
It seems that the story began
when a local doctor's mother moved
into the home, and found it impossible to receive satisfactory TV
About this time the firm that had
installed and maintained the
system was sold. The new firm cut
out all domestic service and the
home authorities had to find a new
firm to serve their needs. I was the
third or fourth technician offered
the contract but it appears that I
was the only one game enough to
take it on. (I sometimes wonder if
"thick" should be substituted for
"game").
Although I have cursed the job at
times, particularly on hot summer
days in the stifling atmosphere
under the roof, I still think it is one
of my more interesting occupations
and I'll be very sorry when age
makes me give it up. Anyway, that's
the background to the story; now
for some technical and physical
details.
~
~
<G>~~~
SOME OF -rH~ -RE.SIDE:"1\S
WeRE. GE1\\NG
VJORM'-t' 'P\G,u~e:.s •....
signals with an indoor antenna in
her room. So her son had an antenna installed for her.
When her fellow residents saw
the lovely picture she was getting
they all wanted an antenna. But the
home authorities, quite understandably, didn't want a forest of antennas on the roof. So a well known
and respected TV service firm was
called in to install a system that
would accommodate any resident
prepared to pay for an outlet.
The heart of the system was a
distribution amplifier near the
antenna, of about 25dB gain. From
this a long trunk was run the length
of the east wing with "T" taps for
each subscriber.
A "T" tap (or simply a "T") is
designed to isolate each set from
the line and minimise local
oscillator and similar interference
to other sets. It has a relatively high
loss between the line and the set
(side loss) and a minimum through
loss to the remainder of the line.
Typical figures would range from
30dB side loss with 0.2dB through
loss to 12dB side loss for 1.5dB
through loss. Knowing what value
of T to use in each part of the
system is a vital aspect of distribution engineering.
The original system worked well
for the first 12 to 15 residents in the
east wing but more outlets were
soon needed and the hospital and
west wing residents also wanted
outlets. The much longer run was
clearly beyond the capacity of the
existing amplifier.
The solution was to fit a larger,
30dB, amplifier at the antenna end
and move the existing amplifier to
the other end of the east wing
trunk. From here another trunk was
taken through the administration
block to the west wing and a new
network begun.
Wormy pictures
The 30dB amplifier had been fitted at the antenna end not long
before I took over the system, and
some of the residents close to it
were getting wormy pictures. This
was not surprising, since some of
them had 80dBu or more at their
outlets.
At this point it might be wise to
digress briefly to discuss distribution network terms and standards.
TV signal strengths can be most
conveniently expressed directly in
decibels (dB), provided this is done
relative to some accepted reference. By using decibels, gains
and losses can be easily added and
subtracted, rather than resorting to
the complicated multiplication and
division needed if other values are
used.
The reference used in this application is 1 microvolt (lµ V), expressed as dBu. Most modern TV
sets require 1 millivolt (lmV) or
1000µ V for a good stable, snow free
picture. On the basis of the lµ V
reference, lmV is + 60dBu. Most
sets would work, after a fashion,
down to 50dBu, but 40dBu would
result in a very snowy picture. At
30dBu there would be more snow
than picture and a 20dBu signal
would probably be unrecognisable
as a TV picture.
In the other direction, most sets
can cope with 70 or 75dBu. Their
AGC (automatic gain control)
SEPTEMBER
1988
39
SERVICEMAN'S LOG
should be able to take care of such
levels. But at 80dBu and above the
AGC cannot always cope and wriggly patterns on the screen are the
result.
My aim was to provide 60dBu at
each outlet. This was most conveniently done by starting with a high
level signal and reducing it as required at each outlet with an appropriately rated T.
In this installation the antenna
was delivering about 65dBu to the
amplifier, which was raising the
level by 30dB to around 95dBu.
Thus, sets close to the amplifier
were being overloaded, even after
the losses in the first string of T's.
The original system had used
16dB and 12dB side loss T's, which
had been OK with the old 25dB
amplifier, but were not good enough
with the new, more powerful
amplifier. My problem was how to
reduce the level to the first 20
rooms without the expense of
replacing all the T's down the line
with ones having higher side loss
values.
The first step was to split the
amplifier output into two trunks, using a four way splitter with the two
unused splits terminated. A twoway splitter causes a 3.5dB loss in
each leg and a four-way splitter
about 7dB. Thus I was able to in-
troduce a 7dB loss into each trunk
before I had to worry about changing any of the T's. In addition, I
have 4dB in reserve which I can
reclaim at any time by changing to
a two-way splitter.
Then I took about 10 rooms
closest to the amplifier and connected them through new 30dB and
26dB T's. This solved the overload
problem for this group and left only
a few troublesome sets on the other
trunk running to the second
amplifier. These were cured on a
temporary basis by fitting attenuators at the outlets. Later, a
more permanent cure was made .
when new outlets nearby justified
the installation of more appropriately rated T's.
Nurses' lounge
I was handed a sticky problem at
one stage when I was asked to extend the system out of the west
wing, along a covered walkway and
into a nurses' lounge some 40
metres beyond the end of the west
wing trunk.
I removed the termination from
the last T (the last T must always be
terminated), and ran an extension
to the lounge, feeding the new outlet
from a terminated 12dB T. But
when I connected a TV set to this
outlet there wasn't enough signal
for a satisfactory picture. It was too
long a run from the last amplifier,
there were too many T's in the
cable, and the 12dB in the final T
was the last straw.
To solve this problem I tried a different approach. I went back to the
start of the new run and replaced
the last T with a 2-way splitter, and
removed the Tat the end of the new
run; ie, in the nurses' lounge. This
meant that there was now a 3dB
loss to the new run, but no 12dB
loss at its far end, since the T had
been removed from this position.
This meant a gain of at least 9dB,
but actually a bit more because the
T at the end of the old run would
have introduced at least ldB
through loss into the new run.
Thus the nurses' set should have
been at least 10dB better off and
the end result, a first class picture,
would seem to confirm this. But was
I l
11
40
SILICON CHIP
I tempting fate in terms of isolation,
particularly from either of the sets
on the splitter back into previous
sets on the line?
Yes, I was breaking the rules. A
splitter typically provides around
22dB of isolation between the two
sets it is feeding, but only about
3.5dB between either of these sets
and the line into the splitter. So,
while there was little chance of interference between the two sets,
there was a risk that either one
could interfere with other sets on
the main line. But I was lucky; I
have had no problems or complaints.
Putting new outlets into an old
building often presents real difficulties and taxes both my ingenuity and physical endurance. I often
need to be in the roof and on an outside wall at the same time. And you
can't nail a cable to an aluminium
window frame.
Most of the cables are brought
out under the eaves and down the
outside walls. This isn't an ideal
situation - I would prefer to run
them inside the wall cavity. Unfortunately this is impractical in most
old buildings and impossible in twostorey ones. So I run them as inconspicuously as possible on the
outside walls.
Civil engineering
One particular outlet will stay
long in my memory. It was into a
downstairs room of the two-storey
block, with steeply sloping ground
outside the window. The first problem was to provide a stable, level
base for the ladder. This took over
half an hour of civil engineering, involving quite a bit earth moving and
the use of bricks and planks to
create a retaining wall. (This ladder base is now a garden bed;
another feather in my multitalented cap!)
When I put my ladder against the
wall I found that, fully extended to
eight metres, it only just reached
the top of the upstairs window. If it
slipped, I would go through the
glass. So I had to modify the top of
the ladder with a plank held in
place with G-clamps. This worked,
but I felt decidedly unsafe perched
so far above ground.
After all this preparation it was
up the ladder, drill a hole, down the
ladder, grab the coax, up the ladder, push the coax through the hole,
down the ladder, into the building,
up the stairs, into the ceiling, and
search for the end of the cable.
Alas, the low pitch of the roof
prevented me from getting closer
than three metres from the coil of
cable, clearly visible under the
eaves.
I had the idea that moving a few
tiles might give me enough freedom
to push the cable into a more accessible position. But my ladder
wasn't long enough and I could only
stand on the top rung, with my chin
on the guttering, wondering how I
might move those tiles only inches
from my nose.
Then I had the first really good
idea I'd had all day. I came down
the ladder (again), got into the van
and drove into town to buy a
4-metre length of 3mm mild steel
rod. With the end of the rod bent into a small hook I effectively had an
arm four metres long and could
easily reach the cable - without
having to move tiles or buy a longer
ladder!
The rest of that job was an anticlimax, although there was still a
lot of up and down the ladder while
I fixed the cable to the wall. But my
elderly client was ecstatic when
she saw the "lovely pictures" and
her delight made all the hassles
seem worthwhile.
As I mentioned at the beginning
I've had both happy and sad contacts on this job. One of the sad
ones was a dear old lady who was
so crippled with arthritis that she
couldn't press the buttons on her
TV remote control unit. And I complain because it sometimes hurts
me to hold a screwdriver!
Other clients have died before
they got full value from my work.
Then I have had others (metaphorically) leaping over chairs in
delight at the birth of a great-greatgrandchild. Although most of my
clients at the home are, in a sense,
really there just waiting to die, they
are almost all a happy and nice-tobe-with crowd. One old chap likes
to help me, even though he has trouble picking up cable clips, one at a
time.
Well, that's J.L.'s story, and I
found it most interesting. The only
comment I would offer concerns the
lack of problems with the splitter.
The most likely cause of interference between sets on the
same antenna system is where the
local oscillator frequency of a set
on a lower frequency channel falls
within the video band of a higher
channel. Granted, there are other
possibilities, involving harmonics
etc, but this is the major problem.
I'm not sure about all the channels in this area. The best reference
I can find lists channels 2 and 6 in
Hobart, and a channel 8 translator
in Taroona. By my calculations,
SEPTEMBER1988
41
SERVICEMAN'S LOG
~
~
-- ·
~7
MRS '-Nl>l'Altf'
( DOt.S~'i ~NOIN
WHA1 Mt.iAfftORIG
more likely to be a component fault,
probably the dew sensor itself. I
suggested he bring it in for me to
check.
So he turned up a couple of days
later with the machine and I connected it to a monitor and turned it
on while he was there. The dew
light came on as predicted and,
since the weather had been fine
and dry during those few days,
there seemed little doubt that it was
a false alarm. I warned him that if
it was a faulty dew sensor there
might be some delay, since I had no
replacements in stock. He indicated
that there was no great hurry.
At the first opportunity I pulled
the covers off to have a closer look.
And it was then that I observed the
second fault; as soon as power was
applied the video head drum commenced turning, which was
definitely not normal. I pushed a
cassette into the carrier and it accepted it and put it down on the
deck. But that was as far as it
would go; pressing the play button
.produced no response - which
wasn't really surprising.
Two faults or one?
none of these would qualify as likely to either cause or suffer from
such interference. But other
localities may not be so fortunate ,
particularly those within range of
two different TV transmission
centres.
For example, a combination of
channels 2 and 5 would be suspect,
while channels 6 and 10 clashed
badly on the NSW north coast some
years ago, to the point where channel 10 had to be changed to channel
11. And these are only two
possibilities.
Faulty dew sensor
From my own bench this month I
have a story about a puzzling dew
sensor fault in a National NV-370
video recorder. In fact, there were
two faults, seemingly unrelated, but
42
SILICON CHIP
the dew sensor fault was the one
which alerted the owner, and the
only one that he knew about.
The story started with a phone
call from the owner - a new
customer - who explained that he
had been away on holidays for a
few weeks and that, on switching
on the recorder when he returned
home, the dew light warning appeared. Since there had been a
bout of wet weather while he was
away, he simply left the machine
turned on for a few hours, expecting that the condition would correct itself.
When this did not happen, he
rang me to ask whether it was
reasonable to expect that there
would be any moisture left in the
machine after this treatment. I said
it seemed unlikely and that it was
So, did I have two separate faults
or was it one fault producing two
apparently unrelated symptoms? I
decided to pursue the dew sensor
fault first, clarify that situation,
then take it from there.
The dew sensor in this machine is
mounted on a small sub-assembly
bracket mounted on the rear right
hand corner of the deck. The accompanying exploded view of this
corner shows the bracket (item 69),
the dew sensor (67), the loading
motor connection board, VJB00A54,
to which the dew sensor is connected, and sundry minor components.
Dew sensors are strange beasts.
As I have found on previous occasions, they are basically resistors
which increase in value in humid
conditions. (Don't ask me how they
work!)
On this basis, an open circuit or
high resistance dew sensor would
simulate a high humidity situation.
And the logical way to test for such
a condition would be to short out
the dew sensor, at least in theory.
In practice, in this machine, it
y-/
_/ ✓
68
ELECTRONIC
COM ONENTS
We stock a wide range of
electronic parts
'7
.. •
1
~·
11
• For service • For Hobby
• For Transmitters
• For Receivers
Also in stock:
~D/:}__·1!
Valves for Transmitters - 6146,
8950, 4X150, 6JS6, 811 and
many others.
Valves for receivers, made by
Rhode & Schwarz, Siemens &
Collins. R-388, R-390(A}, R392
and more .
Fig.2: exploded view showing the dew sensor bracket (69), the dew
sensor (67), and the loading motor connection board, VJB00A54. The
relevant corner of the main chassis is also shown.
wasn't all that simple. While the exploded view (Fig.2) may give the impression that it should be easy to
get at, this is not so when
everything is in its proper place.
The best I could do was release
the bracket and tilt it at an angle
which allowed me to get a pair of
sharp pointed test prods onto the
sensor terminals to provide a patch.
This had no effect, which seemed to
rule out the sensor as the faulty
component and to support the idea
that it was a common fault creating
both symptoms.
At this stage, pressure of more
urgent jobs and the need to think
about the problem prompted me to
put the job aside for a few hours. In
fact, it was the next day before I
turned it on again, only to find that
the fault had vanished; the dew
Mail order welcome
D. Dauner
Electronics
51 Georges Crescent,
Georges Hall 2198, NSW
Phone: (02) 724 6982
Telex 178 401
TV TECHNICIANS &
SERVICE COMPANYS
Have your
PHILIPS VARICAP
TUNERS (ELC2060)
Repaired or
Exchanged
ONLY $17.oo ea
TEN or MORE $15.oo
Quick service
3 Months Warranty
Send Faulty Tuners
Cheques and money orders
+ P&P
8
~
cs-~Awo,, - ·
WC\RN\NG AP~RE.D.
$2.00 to:
-:r.v.TUNERs)
216 Canterbury Road,
Revesby 2212, Sydney Australia
Phone: (02) 77 4 1154
SEPTEMBER 1988
43
SERVICEMAN'S LOG
P1512
1
2
3
4
5
6
CAPSTAN FG
LED GND
LOADING <at> t- 1
LOADING (f;l) l+I
SENSOR LED
DEW SENSOR
BP1512
BP1512
BP1512
BP1512
BP1512
BP1512
-
I
2
3
4
5
6
Pl515 I TO LOADING MOTDRI
I
2
LOADING
LOADING<at>
Fig.3: interconnection board, VJB00A54, and the dew sensor. Note the panels
indicating the various plug connections.
light was out and the drum had
stopped turning. I loaded a tape
and it played without hesitation;
everything was back to normal.
So there was the further complication that the fault was intermittent. And, as if to prove the
point, the next time I turned it on,
the fault had returned.
Racking my brains for an inspiration I remembered that the mode
select switch is something of a problem child on the National NV-450
and, while I had never had any such
problems with the NV-370, it uses
the same switch. I had such a
switch on hand and, since there are
only four leads and a minimum of
mechanical work involved, it was
a fairly simple job to swap them.
Unfortunately it proved to be a
false trail; it made no difference.
So where to now? Delving back
into the manual I began concentrating on the previously mentioned
loading motor connection board to
which the dew sensor is connected.
The pattern of it is reproduced here
and, as can be seen, it is little more
than an interconnection device between various sections of the
machine, which are fed into it via
four plugs, P1512, 1513, 1514 and
1515.
The sections connected to three
of the plugs are shown in the panels
alongside the board pattern; P1513
goes to the sensor LED, Pl514 to the
44
SILICON CHIP
capstan motor, and P1515 to the
loading motor, all three then being
re-routed via P1512. (Incidentally,
notice how the common Asian confusion between the letters "L" and
"R" has crept into the designation
of pin 1 in the P1513 panel).
It was while I was studying this
board pattern and the various interconnections, seeking inspiration,
that I noticed a possible common
factor. The wide copper track, surrounding a blank circle near the
centre of the board and branching
in two directions, is a chassis connection using the mounting screw
(413 on the exploded diagram Fig.2) to complete the c.onnection.
This pattern provides two
chassis connections; one for the
TETIA TV TIP
General Electric TC53L2.
Symptoms: Horizontal shakes of a
few lines at a time . Occasionally
shuts down with a loud screech . If
kept running with an external
power supply, the screech can be
modified by tapping the line output
board.
Cure: TR704 (2SC1 722) line
driver transistor intermittent internal open circuit. The transistor
seems to be insensitive to direct
tapping but responds to vibration
through its connecting pins.
dew sensor and one for the capstan
motor. Could this be the common
factor, assuming that a malfunction
in the capstan motor circuit could
ultimately be reflected in the false
operation of the drum motor? While
such interaction between the two
motor circuits seemed somewhat
tenuous, I was clutching at straws.
But there was one other factor.
Somewhere in my memory chip
something stirred; a vague recollection of a "shop talk" session with a
group of colleagues during which
someone had mentioned a set of
weird symptoms created by a faulty
chassis connection in a National
machine. I hadn't paid much attention at the time - not enough, obviously - but then, someone else
was telling me about another equally interesting fault.
All these thoughts went through
my mind in much less time than it
takes to tell - and they all pointed
in the one direction; the mounting
screw, 413. I reached over and applied gentle pressure to the board.
And that was it; the dew light went
out and the drum stopped turning.
I reached for a Phillips screwdriver and tried tightening the offending screw. It was very tight and
moved only a fraction of a turn, but
that was enough to effect a more
permanent cure; enough to let me
play a tape and generally confirm
that everything was working as it
should.
But I wasn't prepared to trust
that screw. I pulled the board out
and soldered a length of lead to the
copper pattern near the mounting
screw. Then I replaced the board
and anchored the lead securely
under a nearby screw.
Since then I have learned that a
mod sheet has recently been issued
concerning this problem, and
recommending that a separate lead
be fitted along the lines I have just
described.
Now they tell me!
One other point. I realise now
that all my effort to short out the
dew sensor was wasted; it was
already disconnected at the chassis
end, so the test was meaningless.
Had I shorted it to chassis I would
probably have solved the problem a
lot quicker.
Now I tell me!
•~
|