Quite so Abz

You could put the MAF sensor on the roof or in the boot (with the right tubing) and it would still read correctly - you'd just have crap induction

.
However, removing the MAF from it's factory designed housing is bound to have quite significant effect. The error will be proportional to the square of the difference in diameter of the respective housing. Obviously this assumes the sensor is orientated correctly with respect to the air-flow.
For example if the tubing is 41% larger (than size of factory housing) then this will produce a factor of 2 error. In other words, twice as much air is entering the turbo as the MAF is reporting to the ECU.
What happens now is the puzzle?
You'd expect the ECU to mark the MAF as defective, resort to default values and store an error code. Your tests suggest it doesn't do this :chin:
If it accepts the readings from the MAF then I'd expect a very lean mixture to be produced together with boost errors (remember it would be compressing a lot more air than it thinks it is). Once, again your tests seem to indicate that no errors are stored which I find very puzzling :chin: I suppose the ECU might make an 'on-the-fly' correction to mixture settings using the MAP sensor (remember this is reading the correct pressure within the manifold) but this is pure speculation.
The biggest puzzle to me is how you can generate an engine fault and yet an error code is never stored