Peter wrote:
That’s what most people do, and it is safest, but it may not be the “right way”.
It depends on your SOP. Both ways are permitted.
Peter wrote:
I still don’t understand the relevance of setting 1024 or 1013, with a descent from FL080 to 3000ft, and a TA of say 6000ft. It means that it won’t be a straight line, but the “bend” at 6000ft will never be visible because the delta is 11mb which is worth about 300ft.
The relevance is that you may get an unexpectedly high rate of descent after passing the bend. When you choose your desired descent rate in the GPS box, it will compute a Flight Path Angle (FPA) based on a constant QNH. If the altimeter setting is increased during the descent, the aircraft will suddenly be above the profile and the AP will increase the descent rate to rejoin the descent profile.
If you have chosen a 500 fpm descent (say), the target altitude is 5000 ft and you reset the altimeter from 1013 to 1024 when passing 6000 ft, the GPS suddenly finds that it is 300 ft above the profile and has to increase the descent rate to 650 fpm. (Possibly more depending on how fast it wants to rejoin the profile.) If you don’t react by pulling back on power, the airspeed will increase which may be a bad thing.
The relevance is that you may get an unexpectedly high rate of descent after passing the bend.
I think we are splitting hairs. Look at some numbers. The 11mb diff in QNH, taking the change in the slope from 6000ft to 3000ft, is 10% of the altitude delta, which taking the small angle approximation (or calculus of small changes) is a 10% change in the descent angle, and most will struggle to notice that. My KFC225, if you set say – 500fpm, might do 500, or 400, or 600. That is +/- 20%. Unless you do a descent in PIT mode (which uses the KI256 pitch feedback, AFAIK – I don’t think there is an “angle sensor” in the KC225), it will be done barometrically and will never be that accurate. Flying is just not that precise… The things an autopilot can do precisely is e.g. altitude hold but that’s because it has two loops: the internal baro, and the KEA130A encoding altimeter gray code input for the long term definitive feedback. For VS you don’t have a precise reference; you would need precision VS feedback. That will never be right either, due to temp and density changes…
A descent will never be a straight line in the vertical plane, with any GA avionics. You could do it with GPS altitude.
Anyway, no matter how you shake this, setting the final QNH at the start of the descent, avoids all this.
Just got this by email:
