Chaders will correct me if I'm wrong but I don't see the length of final being an issue here given that the A/P isn't trying to capture the G/S as he is only in VOR LOC mode. I landed somewhere the other day and the published approach had a final of around 7 miles. The waypoint EH630 expects 2000ft at 6.2 ZWA which suggests that capturing loc/gs shouldn't be a problem.
I think in real world you would always expect to get loc established before g/s but in this case I don't think there is an issue with loc capture.
Hi Ringo,
You're bang on mate. Although his 80 degree intercept is never going to end well that close in.
If you look at Jason's third shot you'll see he did infact activate APP mode as he's captured the GS (shown on his FMA) I think he was just confused by the fact the APP light on the MCU goes out at G/S capture. He also seemed to think it was turning early to capture the LOC, to me it looks like its turn is perfectly timed given his speed/intercept angle.
What I don't in Jason's or thingamjib's shots is why the ILS is identing SP and not ZWA. We're you guys going into a default AMS? If so it may just be old AFCAD data causing that.
Ok, I've reread all relevant posts a bit more thoroughly. However I must disagree and stick to my guns about the final not being long enough. Unless you're doing a GPS approach (which I know absolutely nothing about), you shouldn't really be following a flight plan for that long, which to me it looks like this is what Jason is trying to do. As I said before at some point in your descent ATC will either vector you to the localiser or tell you to track straight to a beacon/way point to begin a procedural approach. Both of these will mean you leave your preplanned route and you should forget about the FMS/LNAV/VNAV. you should also take the ND off map mode and put it on APP mode.
I have Aerosofts Schiphol which provides me with a complete set of fictional airport plates. For the sake of this topic I'm assuming they're accurate representations of their real life counterparts. Looking at the plate for the 18C ILS, if you were flying a procedural (to summarize extremely briefly that means ATC don't say anything and you follow the instructions on the plate instead) you would expect to intercept the localizer at 11.3D or EH629. It would be safe to assume that if ATC were vectoring you then they would also aim to have you intercept around this figure as well. EH630 is the point of the GS intercept, so at 2000ft and 6.2D out you will be bang on the centre of the GS and if not then something is wrong. EH630 is more of a 'check' point than an 'aiming' point. Now I would like to mention two things about the Glide slope:
1) You should always intercept it from below because it actually can emit false beams above the 'official' Glide slope. This has the risk of the autopilot latching onto one of these higher and thus steeper beams and descending dangerously fast. By coming in from below you are guaranteeing that you are intercepting the lowest and therefore 'official' Glide slope
2) A Glide slope is only certified to be accurate out to 10nm, 8 degrees either side of the localizer. To my knowledge you cannot legally track it outside of these limits. Because this is a very narrow window, would it not make sense to get established on the localiser first?
Which brings me back to Jason's screenshots. He is heading to the point of the GS intercept with a 75 degree angle and only vor/loc enabled. The subsequent turn looks like it would put him above the GS so when he goes for the APP button he could hit a false GS. Also the localizer only becomes active at 5 degrees off the centerline. At 6.2D that means you are roughly only 0.5nm away from EH630 when the autopilot starts to turn to the localizer. It looks to me like these sharp and last minute intercepts are confusing the autopilot somewhat.
Even if the autopilot did manage to cope with it, you are still making very large and sharp manoeuvres to 60 tons of flying metal carrying 180 people very close to the ground. Most airlines would require you to be stable at 1000ft above the ground or its a mandatory go around. This means on localiser (+/- half scale deflection), on glideslope (+/-half scale) and on speed (typically +5/-0kts but may vary). On 18C you should be at 1000ft at 3D. This means you are just giving yourself 3 miles to stabilize. At 180kts that's 1 minute to turn 75 degrees, capture the localizer and glide, configure flaps and gear, bleed off 30-40kts of speed, set power to maintain the GS and trim as necessary. Never mind whilst talking to ATC and performing checks as well. See where I'm going with this? A longer final makes life much easier for both you and the autopilot, so there's a better chance it will do what you want it to do.
As for the SP/ZWA I must admit I didn't notice at the time (I made the example quick and dirty...). The Aerosoft plate also shows ZWA is the ILS indentifier so I can only conclude that either the PMDG database has got it wrong or its been changed since that plate was made (the plate is dated 28/05/09). I remember someone telling be that schiphol 'cheats' and references all ILS DMEs to their central VOR through an offset. The VOR is identified as SPL, but I don't know if that's theory is true or how that would affect this particular problem.
).