This is only a preview of the April 1988 issue of Silicon Chip. You can view 42 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:
Articles in this series:
Articles in this series:
Items relevant to "Walkaround Throttle for Model Railroads":
Articles in this series:
Articles in this series:
Articles in this series:
|
Beating an intermittent
It is not often that an intermittent fault can be
regarded as predictable. Which is one of the
reasons we hate them; they waste a lot of time but
seldom teach anything. Well this month's story is
different - I've had four so far and, after the first
one, they were a snack.
Perhaps I'm stretching things a
little to imply that these were true
intermittents, but they did exhibit
the characteristic that sometimes
they would work and sometimes
they wouldn't. So as far as the
customer was concerned, they
were just that. What's more, it was
serious enough to fool at least one
professional service organisation.
The story started with a call from
the local video store proprietor. As
well as video cassettes, they stock a
range of video recorders and
players for rental, and I service
these units. In this case the ·device
was a Rank Arena VCR, model
RV340, and the complaint was that
it would not rewind or fast forward.
No mention was made of any other
fault and I suspect that this was
because it was seldom used in a
situation where the fault would
show. In fact, it was used mainly in
the shop to check or display cassettes, rather than for rental.
(It ultimately transpired that this
machine was marketed under at
least four labels. As well as the
Rank label it appeared as the JVC
model HR7200EA, which is the
origin of the beast, as General Electric GE V6900, and as Ferguson
3V29A Ferguson being
associated with EMI-Thorn in the
UK.)
I was at a disadvantage with this
machine in that I had not encountered it before, and had no
manual for it. On the other hand
most machines are broadly similar
and I felt reasonably confident
about tackling it on the basis that it
was probably a fairly straightforward fault. Time enough to worry
about a manual if the fault turned
nasty.
I'm not sure how old this particular machine was, but I would
suspect at least five years. It was a
top loading type and probably
among the last of these before the
front loading types appeared.
Initial checks
I pulled the cover off, checked for
any obvious faults, like foreign objects or mechanical damage, then
loaded a tape into it to see what
would happen. The machine accepted the tape and when I pressed
the play button, it began to play.
The only snag was that the takeup spool was not being driven and
tape began to build up as it came
away from the drum. I hit the stop
button before the automatic shut-
J;IGH~1
54
SILICON CHIP
down functioned, then tried to make
the machine wind the surplus tape
back into the cassette. It wouldn't,
which more or less tallied with the
owner's complaint.
My first reaction was that this
machine looked very like a Sharp
and, based on experience with the
Sharp, I suspected the tyre on the
idler wheel; the wheel which is
driven by the reel motor and which
toggles between driving the supply
reel disc and the take-up reel disc.
When the tyres deteriorate they no
longer provide the necessary drive
to these discs.
I had some spare tyres on hand
and it wasn't a big job to fit a new
one. Then I tried the tape again, only this time I simply checked for
forward and rewind functions.
Neither operated and I realised
that I had a more serious fault.
The most likely explanation now
was that the reel motor was not
running, but confirming this was
not as easy as might be imagined.
The idler wheel is driven directly
from the motor shaft and this is
quite small, precisely ground, and
revolves at high speed. It is also difficult to see. I tested it by touching
the blade of a screwdriver against
it and feeling for vibration. There
was none.
Dead reel motor
So was it a faulty reel motor, or a
faulty circuit robbing it of drive
voltage? Clarifying this involved inverting the machine and removing
the approriate covers. Then it was
a short step to establishing that,
yes, the reel motor had "had it" .
Reel motors aren't cheap; they
can range from $40 to $100 according to make, so I conferred with
the owner. He didn't hesitate and
said go ahead. From there on it was
routine: order the motor, give the
machine a routine clean, drum,
heads, etc while waiting, then fit
the new motor and give it a test run.
It came through with flying colours and produced a first class
picture.
But rather than accept it on the
basis of a few minutes performance
I decided to give it a long test run. I
had no particular reason to do this,
other than the fact that the owner
wasn't in a hurry and I had a spare
monitor available. And so it ran for
several hours, and it performed
perfectly.
Then I turned it off, mainly
because I needed the power point.
And this is most important: the
machine was turned off at the
power point rather than by its own
on-off switch, which would have
left it in the standby condition, with
clock running.
Some time later I gave the
machine another run, a shorter one
this time, and again it performed
perfectly. But when I switched it off
this time, I left it in the power point
with power on, switched off at the
machine, the clock running and
everything ready to go.
TETIA CORNER
Sony KV1800-AS
Symptom: No sound or picture.
11 0V rail up to 135V. There is no
line drive out of the "D" board but
the set will run normally if the line
oscillator is powered momentarily
from an auxiliary supply.
Cure: C531, a 1 0µ,F 160V electrolytic capacitor , open circuit or
dried out. At switch on, this
capacitor charges from the 11 0V
rail and feeds a momentary burst
of power into the 18V rail. This
starts the line oscillator after which
the line output stage supplies the
normal 18V rail.
It sat like this for a couple of
hours, then I decided to play it
again, not so much as a test but
because I had a tape I genuinely
wanted to play and check. And this
was where things went wrong. The
machine went through the loading
procedure - ie, wrapped the tape
around the drum - but after a moment, shut itself down.
I went through the exercise a
couple more times but with the
same result. So there was nothing
for it but to pull the top cover off,
load the tape in once again, and try
to see what was malfunctioning.
It wasn't hard to find. The
machine loaded the tape around the
drum correctly, started the drum
motor, and was all ready to play except for one thing: the pinch roller
had not closed on the capstan. As a
result. the reel motor did not start
and the machine, sensing that
something was wrong, shut down.
Two step pinch roller
I went through the exercise again
and observed the pinch roller action more closely. It actually moves
in two steps, starting some distance
away from the capstan and making
its first move as the tape is being
loaded but stopping slightly short of
engaging the tape and capstan.
Then, when the tape is fully loaded,
A l'llll. "1988
55
it moves into contact with the
capstan.
Or that was what was supposed
to happen. In reality, it baulked at
that last movement. And, incidentally, that final movement signals
other functions, such as the reel
motor operation, to commence.
This was a serious setback. I had
seen enough of the mechanical
setup, plus the electrical system
as represented by innumerable
cables, plugs and sockets and
solenoids to realise that this was no
job to tackle without a manual.
I rang the owner, explained that
we had a second fault, and asked a
few discreet questions to try to confirm a theory I was nurturing. The
result was inconclusive, although
he recalled that it had failed to
operate on odd occasions and, yes,
this could have been in the circumstances I suggested.
Hot theory
Well, at least he hadn't shot
down my theory. Whatever it was, I
suspected that the fault was
temperature conscious. If the
machine was plugged in cold and a
tape loaded immediately it would
play it for as long as it lasted - up
to four hours. But remove the tape
and try to play another one and the
machine's internally generated
heat would prevent it.
Similarly, if the machine was
simply left for a time in standby
mode, its own internal heat would
then stop it playing a tape.
As used in the video shop it was
seldom plugged in until needed, so
the fault had gone unnoticed except
for a few occasions when it had
been left on. And since it came good
the next time, it was not of much
concern.
I went on to explain to the owner
that I feared it was going to be a difficult job, aggravated by the fact
that I didn't have a manual. This
latter now appeared to be essential
and I wasn't sure I could get one.
Obtaining a manual was complicated by the fact that I was still
under the impression that it was a
Rank machine. It was only during a
visit by a rep. from a parts distribution firm that I learned its real
origin - JVC. More to the point he
said he was sure he could get one,
but warned that it could take time
56
SILICON CHIP
I F\N\~~EP U? Wrn-\ ~E T\>JO
IV\~C. \·H ~E.S 'S'rR \'?~ED l'l P...~ ED;
CS\~~'<- S\DE. ON \r\~ "BE~C..~.. ·
and would be fairly pricey, which is
par for the course for these
manuals. I told him to go ahead.
And he was as good as his word
- in more ways than one. He obtained a manual, but it took over
three months and it cost $65. I
didn't qibble over the price. As I
said before, that kind of figure is
typical for such manuals, but I do
wonder if the general public realise
the kind of "incidental" costs like
these that a serviceman has to
meet. I'd hate to speculate on the
capital cost represented by the
stack of manuals in my workshop;
costs which must ultimately be
passed on to the customer.
On the bright side was the fact
that it turned out to be a first class
manual; very comprehensive and
virtually devoid of Japanese
English. And that's a plus in
anybody's book.
Back to the fray
So back to the problem, which
was pretty confusing. The fact is
that the inside any video recorder is
the most complex arrangement of
levers, cams, gears, belts, pulleys,
clutches, springs, shafts and
pushrods that anyone can imagine.
Add to this sundry switches, plus
the occasional photocell and exciter lamp, and you have a combination against which Heath
Robinson's collections of strings,
springs, weights, and lighted
candles pales into insignificance.
More to the point, one can never
be sure, initially, whether a particular pushrod is pushing an adjacent lever, or being pushed by it.
It's all very complicated.
The obvious place to start was
with the malfunctioning pinch
roller. This is operated by the pinch
solenoid which is controlled by two
transistors, Q4 and Q5. It is an
unusual arrangement in that Q4
feeds the whole of the solenoid
winding, while Q5 feeds a tapping.
My initial reaction was that this
was a way of providing the dual action of the pinch roller, as previously noted. In fact I was wrong. The
first movement of the pinch roller is
purely mechanical and only the
final, shorter movement is enabled
by the solenoid. (I will deal with the
reason for the two transistor arrangement later.)
But this was of little consequence. The important point was
that the roller was not operating for
the simple reason that the solenoid
was not being activated. There
followed a long and involved chase
through both the circuit and the
machine to find out why. This was
prolonged by the need to set up the
hot and cold conditions needed for
"fail" or "perform" behaviour.
I won't bore you with the details.
Suffice it to say that I cleared the
,,,i..,
,"r1, I
/ I
1/ f
I I r1
I
I
I
'7"1
®
/
/
I
f'\1, I
I I t:.J
L--v /
/
/
Loading motor-++--~'
\
Capstan motor
Capstan belt
------
Capstan flywheel
'--
Capstan
Take-up clutch belt
Take-up clutch
Take -up idler
Take-up reel disk
/
Supply reel disk
Reel idler
Reel motor
Reel idler
Fig.1: general layout of the video recorder, with the loading motor and associated gearing shown dotted in the
top left corner. The pinch roller (not shown) is adjacent to the capstan.
transistors and finally established
that they were supposed to get a
signal from what is called the
"after-loading switch"; (AL) SW
S002 on the circuit. So I had to find
this device and more importantly,
what was supposed to activate it.
To tell the truth, I never did lay
eyes on it. Its location is indicated
on some of the drawings, but it is
totally hidden by other components.
To replace it would probably mean
stripping down most of the deck.
Faulty twins
At this point I put the machine
aside, partly to give myself time to
think, and partly to attend to more
urgent jobs, the owner having indicated he was not in a hurry. So it
sat on the bench for a couple of
weeks, until, in fact, another
customer appeared with the GE version of this machine. His was a long
tale of woe but, in essence, it boiled
down to exactly the same symptoms
as the Rank.
So now I had two nasties to deal
with. Which meant two failures if I
didn't crack it, or two service fees
for the one effort if I did. (That's
how servicemen think!)
I finished up with the two
machines stripped naked, side by
side on the bench. After a number
of test runs I confirmed that both
not only had the same symptoms but
that behaved in exactly the same
way in regard to the pinch roller. So
it really was the same fault.
So what now'? I was still trying to
work out what activated the after
loading switch and, after numerous
tests in both the fault and no-fault
condition, I latched onto a particular lever. I think it is called the
pinch roller arm push plate and is
shown as ·having one end in the
area of the after loading switch,
with the other end attached to a
spring in close proximity to the
pinch roller arm.
It is made with two slotted holes
which fit over two studs, and it
moves back and forth for the length
of these openings during the loading
and unloading sequences. (I later
concluded that it provides the initial movement of the pinch roller.)
I picked on it partly because of its
proximity to the after loading
switch and partly because I noted
that, in the no-fault situation, it
moved over the full length of the
slotted holes but, in the fault condition, only part of the way. In an effort to clarify things I set up a fault
condition, then pushed it gently the
rest of the way when it baulked.
The pinch roller snapped home
immediately and the machine
played, but the picture was poor
/ \I'll/I.
Hlll8
57
loading ring was obviously what activated the after loading switch, it
was clear why the system failed. I
was able to prove this by gently
prodding the loading ring the remaining distance when it baulked.
Everything came good immediately
and the machine produced a
perfect picture.
. . : Unloading
c>: Loading
Slack belts
Wormwheel gear
Worm gear
Take-up loading ring
Loading motor
Loading belt
Fig.2: exploded drawing of the loading rings, driving gears, and loading
motor. Note the direction of rotation.
due to an apparent tracking problem.
VCR revision
To follow the story from this
point the reader will need to have
some idea of how the loading
mechanism works in a typical VCR.
When the cassette is inserted it is
located over the two reel discs (supply and take-up) with the tape in
front of two guide rollers. The protective cover of the cassette is then
opened and the two guide rollers
then move away from the cassette,
one each side of the head drum,
drawing the tape with them and
wrapping it around the drum for a
little over half its circumference.
It was the mechanism associated
with these guide rollers which took
my attention next, and their operation is most easily explained by
Figs.l, 2 and 3 taken from the
manual. The guide rollers are moved by two loading concentric rings
which rotate in opposite directions.
These are driven by the loading
motor, via a wormwheel and a gear
train which drives the upper ring
directly, and the lower one via a
relay gear to provide the reverse
direction.
58
SILICON CHIP
As shown in Fig.3 the guide roller
assembly sits above the slide ring
which, in turn, sits over the loading
ring, being coupled to it via a spring. Now notice particularly the
slotted holes in the slide ring, and
the pins on the loading ring which
sit inside them. During the loading
operation, with the upper ring
rotating clockwise, these pins
would occupy the positions shown,
until the guide roller assembly
reached the end of its travel as dictated by stops called pole guides.
At this point the loading ring(s)
would continue to rotate until the
pins reached the other end of the
slotted holes, thus putting the spring under extra tension and driving
the guide roller assembly hard
against the pole guides. This is to
ensure that the guide rollers remain
rigid and hold the tape accurately
against the drum.
Failure to do so can cause tracking errors - as I observed when
trying to brute-force the system. At
least that was what was supposed
to happen. What was really happening, under fault conditions, was
that the pins moved only about two
thirds the way along the slots. And
since this final movement of the
But why was the system baulking? Putting it through the loading
process a couple more times, in
fault condition, produced the
answer. The loading motor . continued to run briefly after the
system baulked, with the belt simply slipping on the pulley. A little extra tension on the belt, artificially
applied, cured the problem, albeit
temporarily.
I suspected the belt was the
culprit, either worn or aged. Starting with the Rank, I pulled th~ belt
out, along with the other two; the
capstan belt and the take-up clutch
belt. All three were in poor condition and needed replacing, even
though the other two had not given
trouble. Since I had none in stock I
put several sets on order.
Incidentally, the loading belt is
the hardest one to get out. The only
way to get it out is to remove the
complete loading motor assembly,
which is relatively simple, involving
only two screws. Then a circlip is
removed from the end of the worm
gear to allow the shaft to slide out
and release the belt.
It is easy enough to fit the new
belt to the motor assembly while it
is out of the machine, and easy
enough to slip the assembly back into place. But unless the correct
associated procedure is followed
the machine will not function.
There is not much in the manual
covering this operation and I finished up doing it on a suck-it-and-see
basis. In fact the procedure is simple enough in theory but, in practice, needs about four hands. All
that is needed is to pull both guide
roller assemblies hard back against
their stops in the unloaded position,
and hold them there while the
loading motor assembly is slipped
into place and the two securing
screws fitted.
I won't try to explain how I did it
with only two hands!
Slant pole
Guide ro ller
Supply pole base
€,;
0
Supply loading ring
Pin
I
'
slide ring
Fig.3: detailed drawing showing the supply slide ring (note tension spring)
and the guide roller assembly driven by it.
Next I tackled the GE machine.
The situation here was the same; a
faulty loading belt. But the interesting thing was that the other
two belts were brand new, having
obviously been replaced during a
recent service. But why had the
real culprit not been replaced? Too
hard I suppose.
Naturally, after the battle with
the Rank, the GE machine was all
plain sailing, and I had it up and
running in short order. Which
helped make up for some of the time
expended on the Rank. Since then I
have encountered two more
machines with the same fault,
which has put the balance sheet for
this episode back in the black.
Post mortem
But there are some points still to
be clarified. The most important
one is the reason for the thermal
sensitivity of these machines when
the loading belts deteriorate. My
first reaction was to assume that
the elasticity or size of the belt
changed with the temperature. On
reflection, however, I'm inclined to
reject that theory. I don't believe
that the temperature change was
great enough to affect the belt
material in this way.
I'm more inclined to the theory
that it is metal expansion that is involved; that somewhere the
tolerances on some metal parts are
just too fine. This is not evident
while the loading belt can provide
adequate drive, but shows up when
the latter begins to wear. Unfortunately, I can't be sure either way.
Next point: why does the pinch
roller solenoid employ two transistors and a tapped winding? It ap-
pears that the solenoid needs more
pull to snap the pinch roller into
place than is desirable when it is in
operation. So Q4, which feeds the
full winding, is turned on the whole
time the machine is playing. Q5,
which feeds the tap, receives only a
330ms pulse at the moment the
solenoid is turned on, to boost the
action.
And that about sums it all up. It
has been a long story, but I think a
detailed explanation was justified.
So keep an eye open for these
machines and these symptoms. My
experience could save you a lot of
time and headaches.
(Since this story was written the
Serviceman has reported that
another three similiar jobs
have come in for repair so
it looks like it will be a
common fault. Ed).
that this MHA was installed when
television was introduced in
southern Tasmania in 1960. Since
then it has worked - and worked
- and worked. I was able to help in
regard to locating a new 6ES8 and
this put the MHA back in working
order again, hopefully for another
27 years.
Out of curiosity I dug out my old
valve manual from under a pile of
forgotten literature and looked up
the 6ES8. It was a variable-mu twin
triode RF amplifier designed for
series operation in TV tuners. In
this mode it required a supply of
180V, together with 6.3V for the
heater, and a variable negative
AGC voltage.
I did not see this antique MHA, so
I can't be sure how these voltages
were supplied. My guess is that
there was a 240V feed up the mast
to a transformer and solid state rectifier. (Yes, power diodes were
available in the 60s.) The RF output
was on 3000 open wire feeder
continued on page 90
j)
Back to valves
To finish off, here is something in
lighter vein from our colleague, J.L.,
of Tasmania. He writes:
I had a call recently from a colleague in a bush town about 80km
south of here. He asked if, by
chance, I had a 6ES8. Now I have to
admit that I asked him to say that
again while I put my brain into
gear. I haven't heard type mumbers
like 6ESB for so long that I have
almost forgotten what they mean.
Then I recalled that these were
common RF amplifier valves in
1960 model TV sets.
I quizzed my friend as to why he
was messing around with these old
sets. It transpired that he was not
servicing a TV set at all, but a mast
head amplifier. He went on to say
,~CHNlC.lAN
TR't'\NG
ro
GE", :Be.L,.°\"S ON o..
1\i'HIL UJ/l/l
59
INPUTS
X = FUSIBLE LINK
CONNECTED
OUTPUTS
Fig.24: a generalised circuit for a programmable logic array.
MSI circuit form. Most of those circuits are extremely
sophisticated and perform not only addition but also
subtraction, as well as many other logic functions.
Such circuits are used as the basis for an arithmetic
logic unit (ALU) in digital computers.
Programmable Logic Arrays
While a high percentage of digital applications can
be implemented with the combinational logic circuits
just discussed, there are also many applications that
require special circuits. Those special circuits can
often be made from the available combinational circuits, plus random gates and inverters as required.
While the resulting circuit usually performs the
desired function, a good number of chips must be U8-
Serviceman' S Log
continued from pag-e 90
which is, I believe, still the original
material.
When I think of the times a solid
state MHA gets blown out by lightning, or zapped by some illegal CB
afterburner, I wonder why they
ever stopped making valve type
MHAs.
My friend was lucky to have asked me for an old valve. I did not
throw away my valves when I stopped using them. They are stowed
under the house, out of the way and
almost forgotten. I still have some
6J8s, 6U7s, 6B8s, 6V6s, and 5Y3s.
Can anyone remember what they
were used in'?
90
SILI CON CI-111'
ed. These chips take up a lot of space, consume power,
require larger circuit boards and occasionally are not
fast enough.
All those problems can be overcome by using a programmable logic array (PLA). A PLA is an LSI or VLSI
circuit consisting of multiple gates and inverters arranged on a chip in such a way that they may be randomly connected to perform almost any logic function.
Semiconductor technology now permits manufacturers to quickly, easily and inexpensively manufacture custom circuits using PLAs.
Other PLAs are field programmable. That is, the
designer may specify his own circuit, then implement
it himself with a PLA. Such circuits make it possible to
replace MSI functional combinational circuits and all
And another query: the twin
triodes in the 6ES8 were described
as being run in "cascade", meaning
one under the other, like a waterfall. In my dictionary, a waterfall is
described as a cascade. I have
never found anyone who can explain why an electronic cascade is
spelled "cascade" . Do you know'?
Thank you J.L., for that little
piece of nostalgia. I can't offer any
explanation as the origin of the
word "cascade", but I doubt
whether it has any particular
linguistic significance. I imagine it
was nothing more than a sudden inspiration by someone groping for a
term to describe the new circuit
concept.
Regarding the supply of power to
the old MHAs. I cannot recall any
systems where 240V was run up the
mast, and it would have been a
rather complex and expensive
setup. As I recall, the most popular
arrangement used the feeder as a
supply line, power being fed up it at
some convenient low voltage typically 32V if I remember correctly - to a transformer in the
amplifier which supplied the required voltages.
As to whether anyone can
remember the valve types you mention: yep, I sure can; they are part
of our history!
~
|