|
For some reason best known to
the Quad designers their Quad 44 uses an electronic assembly
to select a particular input, rather than a rotary switch.
If you compare your equipment with the Issue 5 of the circuit
diagrams you may notice that Quad originally used two different
chips, an SN16862 and a pair of SN16861 to handle the switching.
The latter is a 4-input switch and the former, a 2-input switch.
There are five possible inputs but an added option is to select
either of two monitor positions.
The requirement therefore is to switch one of five inputs plus
one of two monitor positions, making a requirement of seven in
all.
A later modification dispensed with the SN16862 substituting
for this a third SN16861 chip.
I understand from reading various reports that there are several
different problems encountered with the selector unit, and more
than one explanation, such as switching on account of noise present
on the power rails etc.
The problem I encountered was intermittent
selection of the top monitor position. This has the effect of
significantly attenuating the desired input.
I see from irate users that often an equipment is repaired by
a dealer or even by Quad themselves, only to misbehave later,
in the same way.
The problem seems to be a failure in an SN16861 chip.
As the fault worsens indeed noise may trip a switch, but the
latter is not the root cause.
Each chip has four inputs and if one of these is connected to
a positive voltage for a short period it will cause that input
to latch its related output. The latch will remain set until
another input is selected, when this will latch and the other
to unlatch.
Daisy-chaining is possible via pin 16 of the chip, so that in
a two chip system selecting any of eight inputs will work just
like it did in a single chip.
Quad use another feature. By connecting pin 6 of IC101 to an
input of the IC102 chip, one can select one of two monitor positions
in conjunction with a selected external equipment such as a radio.
Selecting one of two monitor positions will not reset the selected
input. The latter will remain latched. If a different monitor
position is selected this resets any existing monitor position
when a different input is selected.
My fault was due to internal leakage
inside the SN16861 spuriously selecting an input. This generally
cannot be unlatched as long as the leakage persists.
I also found that when a chip is cold the leakage is worst; in
fact after say 30 minutes leakage can drop to the extent the
frozen input will be re-latchable, and everything will work normally.
This, I suspect, may lead a repairer to believe he has fixed
the problem.
To tackle a selector problem. First
one should identify exactly what's going on. To simplify matters
you can unplug the module from the mainboard and apply 12 volts
across the blue and green wires (blue positive and green negative).
The module can then be operated independently of the equipment
which can be left unpowered.
In the unlikely event there's a power supply problem, this will
be readily determined as one can use a variable power supply
to test the selector module. Typically, when you apply 12 volts
to the module you will see around 100mA of current drawn.
If you reduce the voltage you'll find around 9 volts or less
will result in the buttons not working. Don't increase the applied
voltage beyond about 16 volts.
If a problem is apparent the module should then be detached from
the front panel, after determining which input or inputs is faulty.
This will be the one whose LED is permanently on despite which
button is pressed.
Next you will need to examine the printed
circuit in conjunction with the circuit diagram and determine
which pins of the relevant chip need to be modified.
From the circuit diagram you can see the associated inputs and
outputs corresponding to the four elements of the switch.
These are (for the SN16861) Pin 4 and 11; Pin 3 and 12; Pin 2
and 13; Pin 1 and 14.
The plan is to cut the connections to
the faulty input and its associated output then connect an unused
input and output pin in their place.
In my example the fault was a leak at Pin 2 of IC102 causing
its output at Pin13 to permanently latch.
I cut the connections to Pin 2 and reconnected the track to Pin4,
leaving Pin 2 unconnected also cutting the connection to Pin13
and strapping this to Pin11, leaving Pin13 left unconnected.
You should find that each SN16861 has
an unused input/output.
IC100 if it's an SN16861 will have two spares and IC101 will
have one spare.
If you have an SN16862 at IC100 you can use the spare input/output
at IC101.
If you have two or even three bad inputs
for Radio/CD/Disk/Tape1/Tape2, you may still be able to make
a modification if you have three SN16861 chips as there are three
spare input/outputs available.
Unfortunately there is only a single spare input/output in the
Monitor selector.
Please don't contact me for more detail
as I am assuming that you have enough technical know-how to determine
any modifications yourself. If not then do not attempt it.
I assume also you can find the relevant
circuit diagram on the Net... several are available free of charge.
Post Script:
Alas, after successfully testing the
repair, a few days later I tested it again and found the chip
had degraded even more and the internal fault had rendered another
section unusable.
I checked in my box of old TV semiconductors
(I never throw anything away) and was very surprised to find
an SN16861 chip. I seem to recall buying this to repair one of
those small portable TV sets from the 1980s. The chip was possibly
used on the front panel or conceivably in an early remote control.
The old chip worked like a dream after
putting back the original tracking.... |
