More VNAV Issues

[dectenor2]

You can see here that the VNAV path calculation still needs work.



The aircraft is predicting that it will not meet the AT 4000 constraint at CAMEL.

However, the calculated descent path is just too shallow. The aircraft is only descending at 500 fpm and having to add thrust to maintain speed. 

Why not calculate a steeper path that can be done as an IDLE descent and therefore also meet the constraint at CAMEL?

This is just one example of what is not an isolated incident.

Of course there are plenty of other errors with VNAV and the aircraft's altitude and speed predictions which I have reported elsewhere and am looking forward to seeing some progress on.

Edited February 23 by dectenor2

[iniSteven]

Hi Dectenor, 

We know of this behavior, it is an exception that can happen when there is a mandatory level off segment during the approach. But it's not as simple as simply increasing the descent rate, because we are simulating the CDA function of the A350.

As you know, the A350 descent profile is calculated backward from 50ft AGL up to TOD, and as part of this calculation the aircraft seeks to remove all level-off segments and achieve VAPP and configured for Landing at 1000ft AGL.

It's important to remember that our goal isn't to make a VNAV system that hits every restriction, we're attempting to make a VNAV system that behaves like the real aircraft. Airbus pilots will tell you that the real aircraft VNAV system isn't perfect, it doesn't always hit every restriction, and can sometimes have odd behavior as well.

As mentioned before we are working on improvements to the LNAV and VNAV systems to better reflect the real aircraft behavior, but in the meantime you sometimes have to give the aircraft a little extra guidance, as sometimes happens in reality as well 🙂

 

[dectenor2]

33 minutes ago, iniSteven said:

Hi Dectenor, 

We know of this behavior, it is an exception that can happen when there is a mandatory level off segment during the approach. But it's not as simple as simply increasing the descent rate, because we are simulating the CDA function of the A350.

As you know, the A350 descent profile is calculated backward from 50ft AGL up to TOD, and as part of this calculation the aircraft seeks to remove all level-off segments and achieve VAPP and configured for Landing at 1000ft AGL.

It's important to remember that our goal isn't to make a VNAV system that hits every restriction, we're attempting to make a VNAV system that behaves like the real aircraft. Airbus pilots will tell you that the real aircraft VNAV system isn't perfect, it doesn't always hit every restriction, and can sometimes have odd behavior as well.

As mentioned before we are working on improvements to the LNAV and VNAV systems to better reflect the real aircraft behavior, but in the meantime you sometimes have to give the aircraft a little extra guidance, as sometimes happens in reality as well 🙂

 

Great to hear from you.

Many thanks for a considered and reassuring response!

Best as always.

[iniSteven]

Of course, anytime! Hope you've been enjoying the plane 🙂

[dectenor2]

23 minutes ago, iniSteven said:

Of course, anytime! Hope you've been enjoying the plane 🙂

Absolutely enjoying it. There are just some frustrating things!! - Sounds a bit like life in general doesn’t it!!

You mention above about the VNAV work, take that area of the aircraft as an example.

Fundamentally it’s solid and does the job. There are just lots of niggles that for those of us that like to keep an eye of various indications are the cause of the frustration.

Things like mismatches between the colour of the asterisks on the F-PLN page next to the altitude predictions, and the altitude predictions themselves. The ALT prediction can show you meeting a constraint but the asterisk is amber rather than magenta - and then it can be the opposite colour circle around the constraint on the ND, again a mismatch.

Or the FCU selected altitude arrow again showing a mismatch with the altitude predictions on the F-PLN page.

Or the FCU and FMA showing MACH when the aircraft is actually targeting a speed and vice versa.

Or the speed restriction being shown on the DES PERF page being the wrong one.

I mention just these things as examples, there are more.

It’s not that the aircraft is doing anything wrong as such, it just these indication things that once you see them just niggle away.

But yes, absolutely still enjoyable. Just can’t help but feel if these little things underwent a bit of scrutiny that frustration would disappear and it would be so much easier to focus on the many positive aspects of the aircraft and the aircraft would have a much wider appeal. Those like myself who like to take things a little deeper and keep one step ahead of the aircraft would suddenly find everything working really well just as those do now who are more content in just letting the aircraft do its thing and not paying to much attention to the various displays and indications. I think this would boost the product really significantly and help it fulfil the massive potential it has.

These little issues are present on all the iniBuilds Airbus aircraft and presumably wouldn’t take much to be transferred across as it’s all the same logic really.

That was a very long winded way of saying, yes I’m certainly enjoying it, but it still needs a fair bit of refining.

Sorry for the long message, but as you can tell I’m passionate about it, you guys have done so much for the flight sim community and I’m truly incredibly thankful for that, but it would be so so awesome just step up a little gear, and sort these little details out!

[Jc75]

Yes - I’m also getting into this aircraft, love the modern differences over the typical Boeing / airbus but do feel the vnav, handling feels bit unrealistic.

question for real pilots, would it be realistic that if descent was commenced let’s say 2 miles past TOD to get back on path it would command a descent of over 7000 fpm? Seems a bit excessive(and ear popping)