Sometimes I wonder why I bother! This morning I came into the
workshop to discover that the receiver I had spent an hour in
"repairing" last night now had "rolling picture" and absolutely no
sound. It's lucky that I had left it on soak test.
The original report fault had been "wavy pictures". When I first
tested it, there was sound but just a blank screen. Someone had
already been inside and replaced C54 - a common cause of picture
faults. They had also lost three screws. I should have given up right
then but I'm a glutton for punishment. I traced the initial lack of
signal to R9 (470R) which had suffered from the corrosive effects of
black glue. Now there was sufficient signal to give weak inverted
video but this receiver had been well "cooked" during its service -
another reason why I should have jumped up and down on it!
I carried on testing and measuring. More components were replaced but
the video level remained stubbornly low. It was only when I got a
brief flash of good, strong picture as I touched C54 that the reason
became clear. Whoever had replaced this capacitor had broken a track
nearby.
So, this morning I'm left with a choice of still more work or of
cutting my losses. Do you hear that satisfying crunching sound?
That's size ten Doc Martins improving the shape of a zero-value
SRD510. I can use the power supply for another repair.
My first thought was "Oh, no! Another."
This receiver had severe "lines on picture". In fact it was rather
like the Pace SS9000 tuner capacitor fault symptoms. My first move
was to replace the "Relkit 3" capacitors but that didn't help at all.
In fact now I had a blank screen! I got a weak picture back (still
with lines) by removing the "C-Band" switch and fitting a wire link.
TR7 had glue around it so I replaced that. Now I was back where I
started half an hour previously - a picture heavily obscured by
lines. Just for good luck, I replaced ALL the power supply capacitors
with low ESR types. No good. Working blind is not a good idea so I
decided to do the job properly. A quick prod with the oscilloscope
showed a serious high frequency ripple on the 12 volt supply from
TR304. This transistor, in combination with the IC300 Op-Amp
circuitry, is supposed to give a smooth output. Measurement of all
associated resistors showed R8 (1k) to be open-circuit. Fitting a new
one didn't cure the ripple fault. Nor did the replacement of TR304.
In desperation, I replace IC300 (LM392) and achieved success at last.
The picture was clean, stable and the audio was good. Dare I "soak
test" it, I wonder? Nah, let the customer find the next fault!
(applies equally to MSS200/300/500 etc.)
You may recall Frank at the "Lion and Swan" whose PRD800 I repaired
previously. Well, no more than a week after I fixed it and made it
reliable, he had the urge to "trade it in for a better one" from a
customer. This turned out to be a Pace Apollo which apparently
suffered from a sharply defined vertical white line on "Sky
Sports".
It took a long time for the fault to appear and, when it did, I
discovered that it affected all decoded channels (including Sky News
and Ch.5) but no unencrypted channels.
The line was 2mm wide and approx 100mm high on a 14" screen.
It was located about 1/3 from left edge of picture.
It was very intermittent, occurring only at a specific temperature
but it could be forced to appear by using hair dryer and freezer in
the area of the PTV decoder I.C.s
By replacing each I.C. in turn, I traced it to U9, the PTV114 decoder
graphics inserter. This was, of course, the very LAST PTV I.C. that I
replaced!
It was not possible to narrow down the cause specifically to the
PTV114 by heating/freezing because changing the temperature of *any*
PTV I.C. would cause the line to appear or disappear.
Anyway, I get free beer for the next week. (Don't expect me to get
much work done!)
This receiver arrived in the van of a local installer. His customer
had complained that the screen flashed "LNB Short" but, when I
connected the receiver up, it would not come out of standby. The
fault report was useful, however, since it prompted me to check the
LNB current-sensing resistors. On this model there are two 1R8
surface mount resistors connected in parallel, just below the front
edge of the board. They were open-circuit.
The installer was in a hurry so I confess to a "bodge". I soldered a
1R fusible resistor across the burned out resistors and the receiver
worked perfectly. The installer paid up with good grace and I warned
him to find the short-circuit in the cable or LNB before reconnecting
the receiver!
I don't do many customer call-outs nowadays. Since I started quoting
"a pound a minute" the time-wasters stay away in droves. However, a
little old lady just round the corner tempted me with a cup of tea
and a bun so I trotted round immediately.
The Amstrad SRD510 worked perfectly but, after warm-up, the voltage
on pin 8, TV scart, would not drop below 1 volt when switched off.
This caused the TV to stay in AV mode, with the receiver switched
off. Voltage was 12.2 volts when switched on so I felt that the TV
was being very uncondescending in treating a 1 volt level as "high".
I soldered a 10k resistor from the Scart pin 8 to ground. This extra
load pulled the "off" voltage down to 0.85v and the problem went
away. I went away when the teapot was empty.
My wife was complaining again. (She likes to keep in practice).
"It won't record. The timer is stupid." I agreed to take our PRD900
to the workshop, later, to test it but she wanted it done NOW.
"And if it DOES record it plays the wrong program with a flashing
clock or else just a blue screen."
Well, I hate to admit it, but she was right. I tried setting a number
of timed events and the results were somewhat random to say the
least. Sometimes it remained on the wrong channel but flashed a clock
symbol on the screen and a "t" on the LED display indefinitely until
manually cancelled by pressing "F" then "Standby". Sometimes it
seemed to change channel but left a blue screen as if the desired
frequency had been overshot and lay outside the "capture range" of
the phase lock loop system associated with the "Nicky" I.C., U9.
I fixed this particular fault by pressing "F" with the LNB menu
displayed so the AFC was disabled. Then it was simply a matter of
retuning a few channels manually to remove "sparklies". I measured
all the power supply voltages and checked the ESR of all the
capacitors around the power supply. Everything was perfect. However,
the timer function was still erratic so I replaced the
microcontroller, U2, and that cured the problem. I have never known a
micro to develop a fault like this and I doubt if I ever will again.
It's extremely unusual, but my wife thinks it's typical.
Nokia ACU8152 dish positioner
I love decorators! Mr Winstanton is a typical satellite enthusiast.
He has an enormous motorised dish, a Nokia receiver and an Antenna
Control Unit (or 'positioner' as we call it). Nokia equipment is
generally very reliable but Mr Winstanton had "had the decorators in"
while he was at work. He hadn't considered labelling his cable
connections so, upon his return, he found that the decorators had
disconnected everything and left him to sort out the mess.
Apparently, he stuck wires into connectors in a random fashion and
the result was a non-working positioner. It would send the dish East
or West with a few jerks then stop. Normally I turn away anything
associated with motorised systems because the customer always wants
me to spend four hours reinstalling the system for no wages. On this
occasion, however, Mr Winstanton had brought a local installer along
so I knew I wouldn't be lumbered.
There was a screw missing from the ACU8152 and the customer saw my
worried look.
"Oh, Wossname up Church street had a quick look but he couldn't find
anything."
I agreed. He clearly couldn't find all of the screws either! Luckily,
the printed circuit board did not appear to have been touched. I
inspected the pulse input protection circuit which comprises two 6v8
zeners, DA27 and DA27, and two low-resistance 10uH inductors, LA05
and LA05. Of these, LA05 was clearly burned and DA28 was discoloured.
My ESR meter indicated that they were both short circuit. I confirmed
this with a multimeter after removing both components. It seemed that
somebody (I glanced at my customer) had connected the wires
wrongly!
Replacement of the faulty parts appeared to be all that was required.
The installer confirmed that the repair had been successful when we
met in the "Lion and Swan" that evening. I charged him two pints and
a tenner. That must have been too cheap because he grinned.