The website and forum will be intermittently unavailable while we're making some security updates.
File uploads to the download hangar are also disabled until further notice.

Issue with JY's F35B STOLV Ai's

All things Military AI that don't fit anywhere else.
Post Reply
617sticks
Cadet
Cadet
Posts: 3
Joined: 24 Sep 2018, 06:39
Version: FSX

Issue with JY's F35B STOLV Ai's

Post by 617sticks »

John, glad to see your still very active.

After seeing some images from friends in Lincoln in relation the Black Tie dinner at Marham for the reformed 617 Squadron I decided to download your Marham scenery and the F35B Ai's so as to go 617 at RAF Marham flying.

Great detail but the F35 turned out to be a flying shocker.

After taking off, and doing a part circuit it suddenly more or less stopped and went nose first into the ground.

What did I do wrong?

Cheers, RMcL, Adelaide, South Australia
Attachments
F35_Marham_07.jpg
User avatar
Firebird
MAIW Admin
MAIW Admin
Posts: 12111
Joined: 11 Aug 2006, 21:04
Version: FS9
Location: EGLL

Re: Issue with JY's F35B STOLV Ai's

Post by Firebird »

OK which model and fde are you using, and how were you following the aircraft?
I have a reason for asking as John has got quite a section in his manual about flight mode limitations.
Steve
_______________________________________________________
Image
Quid Si Coelum Ruat
_______________________________________________________
617sticks
Cadet
Cadet
Posts: 3
Joined: 24 Sep 2018, 06:39
Version: FSX

Re: Issue with JY's F35B STOLV Ai's

Post by 617sticks »

Firebird, thanks for your reply.

In the past I have used a lot of Johns Ai's and found them top notch.

In this case, yes I read the manual, but noted there really was nothing supplied that related to RAF Marham other than 7 UK paints in the FSX/JYAL F35B STOVL folder. No Ai plans that I could see.

I therefore did my usual thing over 8 years of flying Ai's, selected just one Ai, UK Paint XM135 (clean) to test how they flew before launching into multiple Ai's. I wrote a quick plan in AIFP2 to do a circuit and land again.

I changed the parking code from UK to 617S to allow the Ai to start and return to Johns RAF Markham (Tornado base)

I have also used his retro Marham and flown a refuel plan with his Victor Tanker at 20000 ft and a F3 Tornado waiting at East Kirkby to catch up. View attached.

I watch all my Ai's often from start to finish using the standard FSX traffic list drop down viewing procedure available in the cockpit. I often fly with Ai's as well and take images using the same process, 1000's of them over 8 years. The attachment was done the same way back in my XP days.

Only thing I forgot to do was copy the effects.

I am currently using Steam FSX and w7 pro 64 bit with nice results that are better than the boxed version. No oooooooms after 6 hours continuous flying for the 75th anniversary of the Dambuster raid last May.

The fault occurred after the F35 got permission to land (I listen to ATC) and when it put its U/C etc down. It seemed to me the issue arose with the nozzle full down long before it would be needed. Any real increase in thrust would turn the aircraft over which it appeared to do. The Ai had not even turned on the heading of the runway to start an approach and was still at about 2000 feet.

So that's my story as I see it. Today I will download the package with A models in it and see what happens.

Thanks
Attachments
Jet_ReFuel_06.jpg
617sticks
Cadet
Cadet
Posts: 3
Joined: 24 Sep 2018, 06:39
Version: FSX

Re: Issue with JY's F35B STOLV Ai's

Post by 617sticks »

Today I added a test F35A to my RAF Markham.

Programmed it with only a 2 minute difference between the 2 Ai's

I was of course able to follow both to see what happens during flight.

THE F35B TODAY DID A COMPLETE CIRCUIT AND LANDED at 61 kts in an almost vertical landing.

It turned out using the cross runway and caused a go-around of the F35A

The F35A landed at about 150kts and took the whole runway before turning out.

SO I DO NOT KNOW WHY THE ISSUE OF YESTERDAY.

SORRY TO HAVE RAISED AN ISSUE THAT HAS SELF CORRECTED.

Cheers, With thanks to anyone reading this post.

The attachment shows the UK 617 Sqd F35B on the turn to runway 24 at RAF Marham
Attachments
F35_Marham_08.jpg
User avatar
John Young
MAIW Developer
MAIW Developer
Posts: 4206
Joined: 12 Jul 2008, 15:15

Re: Issue with JY's F35B STOLV Ai's

Post by John Young »

Thanks for that information.

Is your test flight plan VFR or IFR?

Did you find any problems with the SRVL FDE - it's the mode the 4 F-35B's used when arriving at Marham from Beaufort?

There's no relationship between the nozzle position and the thrust generated by the FDE and the nozzle can't tip the aircraft over - it's just an animation triggered by speed.

John
User avatar
Firebird
MAIW Admin
MAIW Admin
Posts: 12111
Joined: 11 Aug 2006, 21:04
Version: FS9
Location: EGLL

Re: Issue with JY's F35B STOLV Ai's

Post by Firebird »

I am glad that it is working today.
This is the most protracted fde I have worked on, mainly due to the differences in the family that necessitating the differences in each fde. A long, long time was spent flight testing to try and resolve some issues which explains the note.

We also found that the FS9 AI engine operated differently according to the runway length. When flying from John's test airfield near Leuchars the aircraft landed at a normal AoA and stopped quickly but at Leuchars it landed very high AoA and stopped very quickly. I think I spent about 3 weeks on trying to figure out what was going on before I conclusively proved that the F-35B fde landed faster on the short take off ramp due to FS9.

Due to my decision to try and make all versions similar in the air it did mean that some stuff was done for the B that meant it was knife edge stuff. It is very tight for it to remain in control for some parts of the flight, the transition to finals mainly, but it did have the effect that TNGs did not work reliably, but as normally you wouldn't do TNGs on a ship it was an issue that we could live with.
Steve
_______________________________________________________
Image
Quid Si Coelum Ruat
_______________________________________________________
User avatar
hawk_sh
Lieutenant Colonel
Lieutenant Colonel
Posts: 880
Joined: 16 Nov 2008, 22:26
Version: FS9
Location: near EDDF

Re: Issue with JY's F35B STOLV Ai's

Post by hawk_sh »

During my work on the FS9 NAS Patuxent River 2018 scenery while setting it up for John's F-35 models, I observered these nose-dive crashes a few times.
It only happened to the F-35B STOVL model when in the base leg of the landing pattern when flying local patterns
(takeoff and landing at the same AFCAD or at different AFCAD files at the same location).
I only observed it when using time acceleration (4x).
However the issue was not repeatable. I estimate there were one or two of these crashes out of ten landings.

There is a comment in my Pax River package manual under "known issues with the F-35B"
"I have observed some occasional crashes of the F-35B STOVL model when in the base leg of
the landing pattern when using time acceleration. The F-35B STOVL model then quickly
decelerates in the air and drops down to earth vertically like a stone nose first."


As Steve mentioned in his earlier post the F-35B FDE is quite critical (knife edge stuff as he called it).
It is advised not to use time acceleration during some stages of flight because strange things might happen.
As far as I remember there is also a comment in the F-35B manual regarding that.

During my countless hours of testing the F-35Bs at Pax River I also came to the conclusion that there might be some further unknown (at least to me) and maybe even random factors influencing the flying behaviour of AI models.
This became especially apparent for me with the critical STOVL and VTOL models/FDEs of the F-35B.
I tried to set up the runway length for the skijump takeoff of the STOVL model and also make the VTOL model touch down at excatly the same spot every time.
It was impossible. What had worked perfectly for a few days showed a slightly different behaviour the next time without any modification to the model/FDE, AFCAD file or flight plan.
Hartwig
_______________________________________________________________

Image
Post Reply