a reply to: Shema
OK, but a fair warning, this may get technical:
I'm also assuming that when you speak of a radio plugged in near your bed, you are taking about a clock-radio.
Radios signals are not very powerful, so to be audible, the audio signal has to pass through an amplifier. Within that amplifier are usually a few
stages that pump the signal up, incrementally. In the case of a small transistor amp of the type used in a clock-radio, it would probably consist of a
pre-amp stage and a power amp stage.
Normally both these amplifier stages sit across power supply 'rails' and do not have a switch to turn themselves off independently and because you
usually want the clock to be powered on, the power supply itself is not usually switched, either. It eliminates a switch which would add cost to the
appliance and wouldn't be that useful (because the clock would loose time).
To control the volume at the output of the power amp requires fairly heavy duty components to carry the current generated, so the preferred method is
to put the volume control between the pre-amp and the power amp. This means that even though the volume may be off, both the pre and power amps are
actually fully operating.
So that explains how the clock radio could pick up and amplify the radio "dit dits".
The next thing we'll look at is the 'phone.
You said that it went faulty previously. If the fault is a software fault, it usually is a hard fault that always occurs under the same particular
conditions (like the execution of a particular module of code). However, hardware faults can be very intermittent and can be unpredictable.
In the case of modern electronics (like smart phones) they usually consist of miniaturized surface mounted components. In the case of the processor of
the 'phone, they are usually of a type known as a ball-grid array mounting where hundreds of contact points on the underside of the processor are
matched against contact points on the circuit board and the whole assembly is then heated until the solder on the CPU bridges across to the circuit
board. This allows a great contact density but does have its problems.
The main problem with BGA mounting is differential expansion. The CPU is made of different stuff than the circuit board so thermal changes cause
pressures on those contact points. This constant flexing causes the solder to become crystalline and brittle (especially with ROHS compliant low lead
solder that would be used in a modern phone). Eventually the connection cracks and then the CPU no longer works.
Considering that it is likely that the 'phone was colder at 5 in the morning, it could be that the board/CPU assembly just flexed in the right way to
make contact again and the CPU sprang to life, and started to do what it was designed to do and run the 'phone.
So the radio in the 'phone kicks in and tries to contact the cell towers and log into the network (causing the induced dit dits) but the processor is
now generating heat and the board again flexes out of contact and the CPU dies.
When you examine the phone, the heat from the day or the heat from your hand causes the fault condition, so it seems to stay faulty afterward.
So, that explains the 'phone part.
Generally, I'd advise you that issues with BGA mounting are manufacturing defects and should be covered by warranty. You could recount your story to
the techies/phone people as it identifies particularly that the issue relates to BGA manufacture. You could get a shiny new one out of it!