HELP - 747-8 glide slope capture and release during ILS approach

ok, after many attempts, I can’t figure out why I fail to capture the glide slope.
I am in APP mode, can see the pink localizers indications and the small g/s confirmation I am under the gs All 3 Auto Pilot command buttons are illuminated. When meeting the GS, the nose starts to dip, LOC mode comes on, then GS dissapears and the nose comes back up after about half a second to fly level with only LOC illuminated.

Please… what am I missing that kicks me out of the glides slope lock??? Been at this a few days now and air traffic in Sydney and Melbourne are growing to dislike the KLM airliner when the see it lol

Flight Dirctor (both) - on
APP - ON <— this then switches off and the LOC mode activates
Auto Thrust - ON
FMC ILS approach selected

frustrated I finally turned to this, which makes it seem I am set up properly (I think) for the 747-400

Press the APP button again as soon as it disengages and you should recapture the G/S again.

thanks. I shall try it. (though pretty sure I already have).
On approach now, so will find out soon!

edit: didn’t need to hit it twice. Captured it this time. I honestly can’t see what I did differently.

now… to see if it is repeatable lol

You’re not missing anything. It’s broken/buggy.

if its anything like the 787 bugs:

level out about 2nm or more before the glideslope interception.

switch from LOC to APP (at an appropriate distance)

lower the AP reference altitude to e.g. 500ft above the runway.

press the VS twice and set it to -100ft/min (press once if its not buggy)

VS being active at the low value of -100ft/min is enough to keep altitude hold off for a prolonged period with minimal ground impact risk and thus allowing APP to capture and follow the GS at interception, rather than the APP mode turning OFF at the moment of interception.

thanks. and insightful, I was hoping I was not fighting a bug.
I finally did a successful GS capture… I am just gonna run with the idea maybe I came in a bit lower this time… maybe.

off to make it repeatable now, then can go and refine other bits and pieces… eg full auto thrust on landing and FLARE

and it did it again… sigh.
will try to introduce the VS being active

I have this issue on the 787 and the 747

Exactly as the op states


another 4 attemptts. one worked perfectly. One failed, one I used the VS mode as another user suggested and it worked, another I had to press the APP again several times for it to work.
At this point, I am not sure if it is buggy, glide slopes are just tricky to catch or if my configuration is wrong. Ahh well… practice practice practice and reading until I get a reliable procedure

For me, set VS -100 active and wait for GS clicked-in is the most secured way to win this current bug.

1 Like

It is buggy, exactly the same way as the 787 is buggy. Since these are both modern Boeing designs, the underlying code emulating the systems for each plane is probably exactly the same, subject to some aircraft-specific configuration file. So the fact that both aircraft misbehave in the same ways should not be surprising.

the 787 is a buggy aircraft in real life… so the flight sim is actually simulating it with a great deal of accuracy.

1 Like

Look, your username makes it clear you’re a ‘Bus-driving fanboy, but you should leave that nonsense on the forums instead of here where people are struggling with a buggy sim.

Geez dude… that was just a joke… I use humour to lighten up the mood from the same struggle that everyone is experiencing too.

I already knew the 747-8 was a buggy mess since launch. So I haven’t flown it ever since… I stick to the 787 and the A320 where I can still manage to work around the bugs. That doesn’t mean I’m not struggling with it. I’m struggling with the bugs too. But unlike some people, I actually work on investigating the bugs, and apply different methods and alternatives to make it at least, flyable. And I’m happy to share my findings and help anyone who’s still struggling with it. All you need is an open mind to different alternatives and flying styles, and it wouldn’t be as much of a struggle anymore.

1 Like

thanks for all the responses. It is hard trying to figure out complex systems when they are fighting against you! At least I know it’s buggy and I migh have a workaround.

So apologies to everyone at YSSY today… KLM Heavy coming in on a likely non-percision approach!

These constant preps, takoffs and approach attempts are helping a great deal, though missed glideslopes is creating for some nasty landing… but they are improving! - sadly the whole thing needs to be restarted to test that 5 seconds…

this ism is really missing the replay and save exact location function. Dealing with this stuff would be so much easier

edit: since I have used the VS mode, once at -100ft the other instance VS was activated, but with 0ft, and it captured the GS. So now I can actually practice landings, as I apparently urgently need to do! lol

I’m so happy to hear this lol.

As for the Glide Slope issues.

Best to use VS with APP…it captures the GS and you can fine tune your approach.

I am suspecting that, in actuality, the 747 isn’t meant to be APP approached all the way.

The VS stops working when approaching designated altitude (HOLD) and so if you are following a protocol such as altitude 500 above runway elevation then you are going to KNOCK OUT your VS.

I now always set the Altitude to BELOW the runway elevation by about 500+ feet in order to keep the VS functioning. I know I’ll lose it at the last few hundred feet but by then you’re not going to correct VS but throttle-up and miss approach if anything is not stable.

This way, on approach, I can continue to finetune the descent with VS.

I suspect that the APP stops working for the same reason as the HOLD is priming to take over behind the scenes. The APP probably piggy-backs off the VS system so for whatever reason that VS is lost approaching HOLD elevation, that is probably ruining your APP glide slope.

This can be tested by lowering your altitude-to-hold to much lower and see if you hold GS longer.

The 747 was designed during the Apollo missions.

I’m 99.9% certain it takes the computer about 1 minute to unload the VS program and load the HOLD program. (Functions…really)

Because of this, if emulated correctly, it would make sense that the HOLD function unloads the VS from memory and there is a gap until the HOLD takes over.

Even though he may have been joking, he’s not wrong.

The Sim is apparently functioning normally, I have about 30 hours on type now and only noticed two actual bugs.

EVERYTHING else, including this, seems to operate with logic…there’s a functionality driving it…see my post just above this one.

You’re justifying workarounds for bugs in the sim - the workarounds may make the simulation work, but since they almost certainly do not replicated the way the real aircraft systems work, they are still bugs.

1 Like

I actually give a very strong reasoning for my justifications, however.

Think about it.

You have two programs running your Glide Slope.

Vertical Speed (VS) AND HOLD.

The Memory (now called RAM) on the 747 is probably smaller than your cache on your CPU. So to switch between VS() and HOLD() it has to unload one and load the other. It can’t have both loaded at the same time for quicker transitions.

The processing power on the 747 Computer is weaker than your digital stopwatch, and the Storage media is probably tape, or flash memory (early generation microchips which would be faster but limited by BUS). So the load times are probably noticeable, a few seconds or more.

It is REASONALBLE therefore to assume an accurately emulated 747 Autopilot would have a delay and loss of function when switching from VS to HOLD.

Now your question is about the APP function … APP()

It probably works based on the VS() and HOLD() functionality as well.

Therefor it makes sense (and I’ve observed) that when on APP, you lose APP and lose the Glide Slope as you get within HOLD elevation.

This explains the lag, loss of function and the remediations.

The remediation is to lower your elevation as much as reasonable to lower it below the HOLD function and expect to cancel APP and manual the last few hundred feet.

1 Like

I would go as far as to say that any simulator that captures and continues to hold an APP function, and transitions between VS and HOLD seamlessly is NOT accurately emulating the Autopilot.

Rather it’s babying the sim-pilots for convenience.

But…I’d truly like to know from real 747 pilots how the AP responds, and what the VS does when approaching HOLD.

All I’m saying is it is that this EMULATOR seems reasonable.