AI A/C at St. Barths question

carib_flyer · Post by **carib_flyer** » Sat Apr 21, 2007 3:32 pm

There is a blurb in the docs for St. Maarten regarding getting AI A/C to land at St. Barth's without hitting the hills etc.

"AI aircraft landing at St.Barts crash through the hill or overrun the runway.
This depends on the individual AI aircraft, specifically on the Flight Dynamics within the
respective Aircraft.cfg file. There are a number of values that can be tweaked by hand to
clear the hill and stop properly on St.Barts’ short runway."

Can someone tell me what to tweak?

Thanks,
Mike

jvile · Post by **jvile** » Sat Apr 21, 2007 6:38 pm

You can visit Jan Martins web site.

Jan is a 3rd party designer of .air/.cfg's and explains in various documents what each entry does to the behavior of a AI Planes flying characteristics.

http://de.geocities.com/janswebsites/fl ... index.html

carib_flyer · Post by **carib_flyer** » Sat Apr 21, 2007 7:09 pm

Thanks very much Jim, I'll check it out.

Mike

carib_flyer · Post by **carib_flyer** » Sat Apr 21, 2007 7:26 pm

Sorry Jim, nothing there that I could see that would steepen the descent angle to land at St. Barth's...I'll keep looking.

Mike

jvile · Post by **jvile** » Sat Apr 21, 2007 7:38 pm

Go down to the bottom of the page and open the word document titled

FDE Changes v5acof.

The first part gives you information on what changes effect aircraft behavior.

carib_flyer · Post by **carib_flyer** » Thu Apr 26, 2007 9:04 am

Jim,

Could I impose on you to help me with this? I know that it should have something to with descent angles, but I'm not sure where it is in the information.

Thanks,
Mike

RFields · Post by **RFields** » Thu Apr 26, 2007 9:10 am

Some folks have stated that the Henry T BN-2A lands correctly at St Barts.

They have replaced their Caravan and Twin Otter AI with the Islander FDE.

Remember that the contact points and lights sections have to match the model - not the FD setup.

paavo · Post by **paavo** » Thu Apr 26, 2007 10:15 am

RFields wrote:Some folks have stated that the Henry T BN-2A lands correctly at St Barts.

They have replaced their Caravan and Twin Otter AI with the Islander FDE.

Remember that the contact points and lights sections have to match the model - not the FD setup.

It's the other way around for me, Henry's twin otter clears the hill perfectly, and his BN 2 goes through the hill.

carib_flyer · Post by **carib_flyer** » Thu Apr 26, 2007 10:59 am

Same result as PAAVO. My HT BN2 goes though the hill and off the end of the runway (just into the sand before turning...not into water).

Mike

jvile · Post by **jvile** » Thu Apr 26, 2007 11:00 am

Mike

I am not a FD designer because that is a subject in itself. There is a compromise between Plane FD's and approach code. Approach speed of a plane vs the FPM descent rate vs where the descent should start (approach code) all play a part in how the AI plane is going to descend with elevator pitch control.

If every AI Plane flying into Barts had the FD's of the DHC6 then everything would probably be ok because that is the middle of the road AI Plane used as the standard to set the approach code to. The DO228 approach speed is faster so it holds altitude longer where as the Islander is a bit slower so it starts the descent sooner.

I know that many fly with the MS default real world weather but that weather engine does not maintain surface winds. 3rd party weather engines work much better if real world weather is going to be used.

If the winds are set to 110 degree heading and a minimum of 8 kts all the AI planes I worked with clear the hill. The Islander needs a little boost of head winds to help its lift FD characteristics.

Mage · Post by **Mage** » Thu Apr 26, 2007 11:05 am

You might also need to check the flight rules for the aircraft flightplans. Jim wrote some approaches for St Barths (the curve over the bay and probably a steep fake ILS into r/w 10 too. These will only be followed if your aircraft are on IFR flightplans. If they approach VFR they'll fly nicely direct to the island as in real life (IFR takes them on wide routings for vectors to the fake ILS) but on VFR the ai pilot flies a visual approach, ignoring Jim's carefully-written approaches and probably flying thru hills.

jvile · Post by **jvile** » Thu Apr 26, 2007 11:16 am

Thanks Mage

That is exactly what I did

carib_flyer · Post by **carib_flyer** » Thu Apr 26, 2007 3:58 pm

Thanks Jim and Mage....I should be able to figure something out with your info.

A couple of observations on my end:

My VFR flights to St. Barth's approach the island then do a right pattern about 3.5 miles to the west, then come in for landing. My HT BN usually is at about 4-500 ft about a mile out and just clips the hill but lands ok.

My IFR flights with a Twotter makes the fake ILS approach (broadcasting to traffic he is at 3nm making ILS 10 approach). He never makes it, crashing into the sea or the hill (lost sight of him in front of the hill) and never comes out. I also noticed him doing some major porpoising on final.

Jim, I see that there is a barths_approach.bgl file in the scenery dir. Did you do that and if so, how do you force it to make that approach as Mage notes?

Thanks,
Mike

Mage · Post by **Mage** » Fri Apr 27, 2007 4:17 am

carib_flyer wrote:My VFR flights to St. Barth's approach the island then do a right pattern about 3.5 miles to the west, then come in for landing. My HT BN usually is at about 4-500 ft about a mile out and just clips the hill but lands ok.

That's because it is flying a pretty typical VFR pattern and paying no heed to terrain. Making the flights IFR will cause them to follow a fake IFR procedure that keeps them away from terrain. After that, provided the aircraft FDEs are suitable, they'll fly far more interesting approaches.

carib_flyer wrote:My IFR flights with a Twotter makes the fake ILS approach (broadcasting to traffic he is at 3nm making ILS 10 approach). He never makes it, crashing into the sea or the hill (lost sight of him in front of the hill) and never comes out. I also noticed him doing some major porpoising on final.

That sounds like you're using a model other than Henry's, maybe the Premier Aircraft Design User-flyable version? Either the PAI one (based on an earlier version of the PAD Twotter) or Henry's should have the sort of FDE that can cope. I use the stock FDEs that come with Henry's Islander and Twotter and both function just fine. You might need to source repaints for the Ai models and use those, or take a close look at the FDE you're using if the planes are actually HTAI or PAI versions.

carib_flyer wrote:Jim, I see that there is a barths_approach.bgl file in the scenery dir. Did you do that and if so, how do you force it to make that approach as Mage notes?

I can field that one for Jim - you need to decompile your flightplans or find more recent ones (if you're switching models you'll be doing this anyway). In the Flightplans_... text file find each flight that has VFR in the line and change them to IFR. Recompile the plans and your planes will operate under IFR and be directed to the fake ILS each time. The flight to an island might not be as realistic (long wide routings to intercept at 30 degrees to the "localizer") but the part closest to the airport will look much better.