Seems likely that this aircraft was being operated with cabin pressurization selected off. This leaves all cabin air entering through a NACA inlet on the lower engine cowling. If this was partially blocked by ice it would be likely that the windshield demisting fan had dropped the pressure around the exhaust such that a leak would be very likely to enter the cabin.
Looking at the system diagram at 7-25 of the Pilots Manual you realise that the design is fine for all conditions apart from unpressurized icing when your relying on the integrity of your exhaust pipe to keep you safe!
NACA inlets don’t ice
relying on the integrity of your exhaust pipe to keep you safe!
I don’t know much about other types, but is that not the case, in general ?
Yes indeed it is.
It sounds like the PA46 pressurisation gives you an extra protection layer.
Mostly you will have ram air pressure around the exhaust muff which minimises the problem of exhaust leaks entering the ventilation duct. I’m not so sure of the immunity of NACA inlets to icing when they are on a surface inclined forwards as this one looks to be.
I saw few CO recordings (up to 20ppm) when taxying on tailwinds or flying with full rich mixture, all of it gets vanished when opening doors on ground or vents while flying, I also never saw much of it when running peak EGT or LOP, does one expect to get some CO exhausts when you save fuel, I bet no?
Obviously, big leaks in the exhausts has problematic smell but that is not the topic when dealing with odorless gases?
FYI, I plugged that CO detector when driving car in London or near an open air barbecue, I saw even bigger figures than 30ppm but I think one does get fresh air and the task in hand requires zero mental capacity 
What do you actually do in a pressurised aircraft if you start to smell smoke? You can’t really open the window.
kwlf wrote:
What do you actually do in a pressurised aircraft if you start to smell smoke? You can’t really open the window.
You dump the cabin. (more complicated than that but that is the gist)
jbees44 wrote:
I’m not so sure of the immunity of NACA inlets to icing when they are on a surface inclined forwards as this one looks to be.
I thought the positioning was pretty standard, for example on jet engines they look like that:
Peter wrote
Just because the inlet itself doesn’t block with ice doesn’t mean ice won’t accrete further down. See this as one example.
True, although unlike an engine inlet, there’s no filter to potentially soak in moisture, and also no low pressure zone like a fuel injector or a carburettor.
I just really think that particular air source cannot be blocked by ice, basically
JasonC wrote:
You dump the cabin
But just the cabin air, not the actual cabin like Thunderbirds 2, unfortunately (I know you know that Jason 
)
kwlf wrote:
What do you actually do in a pressurised aircraft if you start to smell smoke? You can’t really open the window
I actually had this once. We were in a King Air at about FL200 when the smoke started. To cut a long story short, it was descent, depressurise, divert, land with the fire service in attendance. It was the most concerning incident I’ve ever had, a sentiment shared by the pilot in the right seat for that flight, a somewhat seasoned aviator with something over 28,000 hours
