TSR2 Flight Dynamics

[padburgess]

Hi
I have just spent a frustrating week playing with the flight model of the Alphasim TSR2. This started out as a simple plan of mine to validate the fuel storage and usage of the model as I have been using the aircraft a bit lately and planning a few trips with it.
Wanting to do semi-realistic trips over Eastern Europe and Canada with the aircraft I wanted to make sure she had the right fuel on board (in the right places) and the usage was close to what it would probably have been.

The flight model that I have for her gave approximately the right range for the weight, operating height and speeds used, but I felt like being pedantic.
First up I trolled the net and found all the cut-aways I could, including quite a good one that showed the positioning and size of the internal fuel tanks. Bonus.
So using AirWrench, I checked the numbers. First surprise was the graphic of the aircraft balance. Then some of the other numbers didn't quite stack up. I decided to change things to what they should be. Thought it wouldn't be too hard as I only intended to put things right. So I created a copy of the aircraft folder and renamed it TSR2_Experimental, parred the options down to one and got into it.

How hard could it be? By the way, that is a truly stupid stupid stupid plainly cretinous statement.

Moving things was easy.Get a ruler, measure the line drawings and transfer the numbers to AirWrench. Read the spec sheets and put those numbers into the right variables. Simple.

Next step, go for a test flight, and crash on take off. Tweak the file and again, tweak and again, tweak and bloody again, bloody tweak and die ( I must stop this repetitive behaviour but I have been damaged by the gods for messing with air files). Minor tweaks and more flights/crashes. The beast just would not fly. I spent every night for the last week playing with the numbers in AirWrench to no avail. She just would not fly. I just managed to get her into the air twice at the end of a 12000ft runway on full burner but the handling was diabolical.

Then last night in desperation ( I was getting to the point where I might have to cry uncle or just cry) I did a bad thing. I took the air.cfg file from my RA-5C Vigilante and copied and renamed into my TSR2_Experimental folder. No other changes to either the aircraft.cfg or the air.cfg. Checked the numbers in Airwrench again, saved it and went for a flight.

Yup, a flight. She took off in about the right distance with a full load of fuel, climbed well and handled beautifully. Now i know that the aircraft.cfg over-rides the air.cfg but there must have been a table in the original TSR2 air.cfg file that did not like the changes I made, but the Vigilante air.cfg file is quite happy with them.

Below are the line drawings I used and screen shots of the original balance diagrams and the ones I am now using. I still need to play with things a bit and check the numbers properly but so far I am now happy.
Can anyone please confirm what I believe or give me some pointers as to what may have gone wrong?

Cheers in advance.
Paul

http://www.sim-outhouse.com/sohforums/attachment.php?attachmentid=28940&stc=1http://www.sim-outhouse.com/sohforums/attachment.php?attachmentid=28945&stc=1http://www.sim-outhouse.com/sohforums/attachment.php?attachmentid=28947&stc=1http://www.sim-outhouse.com/sohforums/attachment.php?attachmentid=28948&stc=1

[Blackbird686]

I re-vamped the FDE for the FS9 Alphasim TSR2 manually... after gathering as much as I could from available resources. I found that given the limitations of FS, it was not possible to get the TSR2 to fly in the sim as it did in real life so I had to make adjustments accordingly. What I came up with gave me the performance I was looking for with consideration to things like wing load to weight ratios and fuel consumption (Adding payload, clean configurations... etc.). The FS9 TSR2 had a feature where you could add (visually) loadouts to the wings as well as internal weapons with a key stroke. It did nothing as far as changing weights and balance characteristics to the flight model. I usually went in and changed up the fuel and payload numbers if I wanted a more realistic experience with the actual handling of the plane in flight but on occasion... realistic weight given payload made the TSR2 nearly un-flyable and like you say... once in the air the big jet lost nearly all of its matchless coordination and flew like an anvil with short stubby wings. Frustrating...

With a clean configuration however, I could easily fly on one engine, and pretty much catch and outrun anything. The handling was stable and positive and fuel consumption was very close to the real deal.

BB686:US-flag:

[Paul Golding]

Hi
Can anyone please confirm what I believe or give me some pointers as to what may have gone wrong?


What you found is correct Paul, the content of the cfg file will override the air file.

I haven't messed about with flight models for a while, but I seem to remember that on loading an aircraft, any missing information in the cfg file is effectively populated by the air file. So, you could use a blank cfg file, with just the aircraft description section included so FSX can 'see' it, and then load the aircraft. Take a look at the cfg file to see if it worked. If it does, you then have the enjoyable task of looking for the stuff that needs to then be overwritten in the previous cfg file. Or, I could of course be talking sh*t......worth a try though.

[Dev One]

My version in both FS9 & FSX has a problem with lack of thrust, seems to take off OK but by the time I've got to 16k' & cut the afterburner & throttled back, I can no longer get above about 400 KIAS unless putting after burners on. It will then very slowly increase to about 700 KIAS.
Need to have a good look at the engine parameters as well as roll with rudder methinks when I have time & patience......& thats just for starters. Possible reason for its withdrawal?
Keith

[mal998]

Some of the airfile parameters do in fact work with the aircraft.cfg. Sections such as aileron control, sections on yaw and quite a few other airfile parameters will affect the way the aircraft behaves.

Taking a quick look at the aircraft.cfg, it looks like the engine thrust number is incorrect. Static thrust is shown to be 19600 when some reference material shows staic thrust at 22,000 lbs each engine. Also looking at the section for "afterburner available" in the aircraft.cfg, it is showing a "1", when in fact the way the sim reads afterburner is not "one" or "zero" but instead by the number of afterburner stages available. So right off the bat there are two things that can be improved.

Also it's worth remembering FSX has major issues with airspeed dropping off as altitude increases, so top speeds at 40,000ft are never as fast as they should be.

The fuel tank locations, in the model are in the wrong locations, where the reference material shows tanks located fore and aft in the fuselage.

I will have a look at the FDE and if I think I can improve performance I'll give it a try.

[Sundog]

Also, just a word of advice on adjusting air files; Get your propulsion file as accurate as possible first. Once you have that dialed in, then begin messing with the aerodynamics. Otherwise, you'll end up just going in circles. Also, listen to what Mal said, as there are errors in how FS does certain things and the only way I know to get around them accurately is to write your own code to supercede theirs.

[Blackbird686]

Taking a quick look at the aircraft.cfg, it looks like the engine thrust number is incorrect. Static thrust is shown to be 19600 when some reference material shows staic thrust at 22,000 lbs each engine. Also looking at the section for "afterburner available" in the aircraft.cfg, it is showing a "1", when in fact the way the sim reads afterburner is not "one" or "zero" but instead by the number of afterburner stages available. So right off the bat there are two things that can be improved.

Looking at my static thrust setting I had increased it to 20,000, vs. the 19,600 but the afterburner setting correction is a good thing to know. Maybe I'll roll the ol' gal out of the hangar and have another look.

BB686:US-flag:

[mal998]

I have been looking around but I think the biggest obstacle for this bird might be the lack of reference material. When doing FDEs I always look for hard-to-find data that will tell me things like degree of travel for ailerons, elevators rudders etc. It helps to fill in the basic numbers before starting work on the aircraft's flight respones in the sim. So anything I do on the TSR-2 will be purely guestimates. If I can get it flying to my own satisfaction I'll put the files up for critique and then try to follow reasonable request for changes. Remember, everbody has their own take on how an aircraft "should behave" being that most of us, if not all, have never flown the majority of these planes in real life. Purely subjective. If anyone has data on this bird give me a shout.

[srgalahad]

Also it's worth remembering FSX has major issues with airspeed dropping off as altitude increases, so top speeds at 40,000ft are never as fast as they should be.


One problem comes from what may be used as the source data, and many "sources" don't differentiate between, for example, True Airspeed and Indicated Airspeed. IAS does become lower with increasing altitude (due to less pressure acting on the pitot system) and TAS increases due to lower air resistance/drag (to a point) but without knowing what is quoted or reported, it leads to trying to make wrong values work.

Thinking of control deflections, etc. some aircraft operating at high(er) speeds also had things like airspeed-limited control systems to reduce deflection as IAS increased to reduce control forces or drag issues. This might include limiting aileron movement and/or replacing it with spoiler roll-control at speed. Gets pretty complicated within the limits of airfile/cfg development in FS.

[Dev One]

[QUOTE=Blackbird686;980067.... but the afterburner setting correction is a good thing to know.

BB686:US-flag:[/QUOTE]
Interesting...both FS9 & FSX SDK's say that the afterburner setting is binary (that as far as I'm concerned is either a '0' or a '1'), but looking at a couple of other peoples models ( F15 & Mig 25) both have values of 5 or 6!
Confused in France...
Keith

[Sundog]

TSR.2 Specs;

Weights
Normal Loaded: 103500 lbs
Max. Overload: 120970 lbs
Max. Landing: 57200 lbs

Performance:
Vmax: M1.75 (At 40000 ft.) (Based on actual flight test and analysis of what would be in production) BAC wanted to limit the Max. Speed to 1.5
Max design speed: the lower of 730 kts or M1.75 (probably the q-limit) (Based on actual flight test and analysis of what would be in production)
Sea Level Cruise: M.9 (Based on actual flight test and analysis of what would be in production)
Rate of Climb-sl: 50000 feet per minute (This does not mean it can get to 50K ft. in one minute. It's rate of climb at sea-level only, not time to climb)
Service Ceiling: 40000 ft. (Based on actual flight test and analysis of what would be in production)
Radius of action: Hi-Lo-Lo-Hi :750 nm (860 statute miles) on internal fuel at Seal level (Based on actual flight test and analysis of what would be in production)
Lo-Lo: 600nm (690 sm)
Hi-Hi at M1.7: 450nm (515 sm)
Ferry Range: 2500nm (2875 sm)

Take-Off Roll (ISA): 3300 ft at 95900 lbs (Based on actual flight test and analysis of what would be in production)
Landing run, wet runway, dispersed conditions: 2250 ft (Based on actual flight test and analysis of what would be in production)
Terrain Following Alt: 200 ft under manual control (Based on actual flight test and analysis of what would be in production)

Max G: RAF wanted a 6G limit, BAC wanted them to lower it to 4.8G since the wing failed the static test at 8.5G.

Powerplant:
(2) Bristol Siddeley Olympus 22R
Dry Thrust: 22000 lbs
Reheat Thrust: 30610 lbs

Wing:
Span: 37.14 ft.
Area: 702.9 sq. ft.
Aspect Ratio: 1.96
LE Sweep: 58.32 degrees
TE Sweep: 2.5 degrees
Tip Chord: 4.28 ft.
Root Chord (at Fuselage): 27.26
Thickness/Chord Ratio: 3.6%/5% (I believe that's clean and dirty config.)
Tip Anhedral: 30 degrees

Flaps:
Chord aft of hinge: 15% of local chord
Inboard limit: 23.78 % of semi-span
Outboard limit: 80.61% of semi span
Range of movement: Down to 50 degrees

Tailplane:
Area: 144 sq. ft.
Aspect Ratio: 2.1
LE Sweep: 60 degrees
Span: 17.4 ft.
Tip Chord: 1.53 ft.
Root Chord: 15.0 ft.
Range of movement: +10 degrees to -20 degrees
Thickness/Chord ratio: 4%
Anhedral: 4 degrees

Tailplane Flaps:
Chord aft of hinge: 15% inboard of local chord, 31% outboard
Outboard limit: 71% of semispan
Range of movement: +30 degrees

Vertical Tail:
Area: 105 sq. ft.
Aspect ratio: 3.0
LE sweep: 35.6 degrees
Tip Chord: 3.0 ft.
Root Chord: 14.99 ft.
Thickness/Chord ratio: 5%
Range of movement: + 15 degrees

Intake Cones:
Range of movement: 17 to 25 degrees

Note: Based on the flight test cards, it looks like some of the procedures were as follows;

Take off: Select retract gear before 240 knts to be up by 300 knts. make sure lights out at 315 knts
Flaps up by 300 knts

Decent and Approach: Normal descent (airbrakes out) 0.9M/350 knots
Slow descent (airbrakes in) 0.9M/ 320 knots

Landing Circuit
Refs:
Flaps 20 degrees 250-220 kts level flight, gear up, a/brake retracted, rpm 85%
Flaps 35 degrees 220-180 kts level flight, gear down, a/brake retracted, rpm 90.5%
Flaps 35 degrees 180 minimum glide path, gear down, a/brake extended, rpm 95%
Procedure:
Turn flap blowing ON then select 20 degree flaps at 250 knts (max.)
Gear down at 240 knots
Flaps 35 degrees below 225 knts
Approach 180 knts
Touchdown 170 knts
Drag chute: open at 150 knts

Two engine go around: Max Dry Power, Air brakes in, maintain config, accelerate to 220 knts and re-enter circuit.

It should also be noted that it apparently was easy to handle in the circuit and didn't have any handling problems at low speeds.

[mal998]

Excellent! Well done Sundog! Where, may I ask did you get this excellent data? These numbers are just what I love to see. I will start plugging them in and as soon as I have a prelim set of files I will post them up here for consideration. Too bad we don't have an updated VC and 2D. The panels could sure use some work. I'll ask Phil if he still has the original source files. Maybe we'll get lucky.

[Dev One]

Request for reduction of speeds could be due to structure temperature considerations. First 10 (Type 571) built using X2020 lithium alloy, which was quite brittle, but subsequent pre prod a/c (Type 579) material changed to L73, but at a better age hardening condition. ? T561?.
Keith

[delta558]

The afterburner in plain FSX is binary - on or off, 1 or 0. Acceleration allows you to go a bit further with up to 6 stages of reheat, hence higher numbers seen in some aircraft.

[Sundog]

Excellent! Well done Sundog! Where, may I ask did you get this excellent data? These numbers are just what I love to see. I will start plugging them in and as soon as I have a prelim set of files I will post them up here for consideration. Too bad we don't have an updated VC and 2D. The panels could sure use some work. I'll ask Phil if he still has the original source files. Maybe we'll get lucky.


Hi Mal,

They're from this excellent book, TSR2: Britain's Lost Bomber (http://www.crowood.com/details.asp?isbn=9781847972118). It covers the initial designs, such as the P.17, some of which are configurations I'd never seen before. It covers all of the details of OR.339, the development, testing, etc. It has many of the variants being studied, such as the big wing TSR2, before it was cancelled. It's easily the best book on the subject I've ever read.

[mal998]

Ah yes. I almost had a download of that book earlier today but couldn't get it. But I did just got a hold of TSR.2 Crew Notes so I now have most of what I need to update the FDE. Will keep those interested posted on progress.

[Sundog]

Request for reduction of speeds could be due to structure temperature considerations. First 10 (Type 571) built using X2020 lithium alloy, which was quite brittle, but subsequent pre prod a/c (Type 579) material changed to L73, but at a better age hardening condition. ? T561?.
Keith

The main reason was due to the Autostiffener. They weren't sure if it would work and they would need it above M1.5. Basically, the stability required above M1.5 would have required a larger vertical tail, but that would add too much weight and drag. So the idea was to use a system that would sense lateral g and use a smaller vertical tail to react to the lateral forces and stabilize the aircraft. Basically, what that means, is above M1.5 the TSR2 was unstable in yaw, and they were planning on using a flight control system, named the autostiffener, to keep it stable above M1.5.

The other main reason for lowering the top speed down to M1.7, from M2, was due to the massive increase in weight of the aircraft and the lower thrust from the powerplants than originally planned.

[rcbarend]

So I created a copy of the aircraft folder and renamed it TSR2_Experimental, parred the options down to one and got into it.
Paul,
Maybe trivial and maybe it doesn't apply here:
But:
- Did you make a renamed copy of the aircraft folder structure in the /Airplanes folder itself ?
- And if so, did you change the Title= definition in the aircraft.cfg ?

Because if not, FSX might not use your changed .air file at all.

I'll explain.
For loading a certain aircraft (-variation), either via a saved flight file, or aircraft menu, FSX uses the definition of Title= in the aircraft.cfg to identify an aircraft(-variation).
Meaning that, every Title= definition in ALL aircraft.cfg files in ALL first subfolders of the /Airplanes folder MUST be unique.
If not, FSX uses the first Title= definition it finds, and all files from the aircraft folder in which that aircraft.cfg is located.
I.o.w. it might not load your modified .air file, but the original (unmodified) .air file because it finds the Original aircraft.cfg first.

Rob

[Dev One]

Hi Mal,

They're from this excellent book, TSR2: Britain's Lost Bomber (http://www.crowood.com/details.asp?isbn=9781847972118). It covers the initial designs, such as the P.17, some of which are configurations I'd never seen before. It covers all of the details of OR.339, the development, testing, etc. It has many of the variants being studied, such as the big wing TSR2, before it was cancelled. It's easily the best book on the subject I've ever read.

Sounds as if that book is a lot better than the Tim McLelland version I have, will have to put yours on my Christmas list!
Keith

[padburgess]

Rob

Yes I gave the "new"variant/entry a unique title and made sure the original folder did not have that variant listed. I haven't had a chance to play with this for the last couple of days but am about to get into some test flying to see how my changed bird flies.
Thanks to all who have posted ideas or advice. It will be interesting to see how this thing pans out. I have the engines set for 19600lbs of static at present so I may increase that to 22,000dry and 30,000wet in a bit.

Paul

[COBS]

Rob

Yes I gave the "new"variant/entry a unique title and made sure the original folder did not have that variant listed. I haven't had a chance to play with this for the last couple of days but am about to get into some test flying to see how my changed bird flies.
Thanks to all who have posted ideas or advice. It will be interesting to see how this thing pans out. I have the engines set for 19600lbs of static at present so I may increase that to 22,000dry and 30,000wet in a bit.

Paul

Paul

Be careful increasing the thrust , check each increase by examining the length of Take off run.
Keep your take off run at a realistic length .

With the ALPHASIM F-111 , I increased the thrust to 20,850 and that was OK ,
then I increased it to 25,100 as the F-111F had that thrust in RL , the aircraft took off like a rocket ship , it resulted
in an unrealistic short take off , too short to live with .
I reverted back to the 20,850 figure.

Cheers
Karol
PS ;
TSR2: Britains lost bomber is a fantastic book , highly recommended for those that like detail and nuts and bolts stuff ,
best book ever on that bird.

[Sundog]

Paul

Be careful increasing the thrust , check each increase by examining the length of Take off run.
Keep your take off run at a realistic length .

With the ALPHASIM F-111 , I increased the thrust to 20,850 and that was OK ,
then I increased it to 25,100 as the F-111F had that thrust in RL , the aircraft took off like a rocket ship , it resulted
in an unrealistic short take off , too short to live with .
I reverted back to the 20,850 figure.

Cheers
Karol
PS ;
TSR2: Britains lost bomber is a fantastic book , highly recommended for those that like detail and nuts and bolts stuff ,
best book ever on that bird.

Remember, thrust and drag are related. One of the most difficult pieces of data to secure is an accurate engine deck. That's the performance table for the engine used when designing aircraft and they are very proprietary. I only have one or two of them. But they give you the performance of the engine at different rpm levels and altitudes for accurate calculations of performance. That's why I say it's always good to try and get engine data first, make sure it's modeled accurately, then move on to the aerodynamics. But FS doesn't calculate thrust accurately enough, especially with afterburners, so I've been looking at making a thrust gauge that would over ride FS's.

Anyway, my point is, if you get the thrust right, and your airplane behaves like a rocketship, that means your drag is too low. Unless you're actually modeling a rocketship. ;). I was working on an FDE for the F-101B a few years ago (I lost the most up to date version, so I never completed it) but one of the best pieces of data I had were actual acceleration times at given RPM/Thrust settings for the F-101B on take off, which really allowed me to dial it in. That's the other type of information that's really great to look for when making FDE's. A combination of accurate engineering data and detailed flight reports will be your best references when developing FDE's.

[mal998]

Here is an alternate version of flight dynamics for the TRS.2. Very tricky to get this right as the original weights and balances were somewhat off. Hope you find the new version to your liking.
mal

[Dev One]

Mal, Just tried it in FS9 & it throws up I'm afraid. A quick look cannot find the problem. Get the splash screen but then FS9 closes. Only the Prototype textures show on splashscreen, others just grey.
Not tried yet in FSX because previous version didnt work in FSX. Not had time to explore further.
Keith

Edit,
Correction; After a very short flight it does work in FSX a lot better than before, thanks a lot, but question - any reason known why HUD is opaque above about 75% throttle?
K

[mal998]

Whoops, I should have mentioned new FDE for FSX only, sorry 'bout that. I have no idea why that is happening to the HUD. My guess is it has something to do with the afterburner fx files or something to due with the HUD gauge or bitmap in the TRS.2 gauge folder. Perhaps someone who knows more about panels can shed some light. Remember this puppy was originally built back in '05, a mere 10 years ago.

At any rate, hopefully the new FDE will breathe some new life into a really cool plane and make it a little more fun to fly.

[Sundog]

Thanks Mal, I'm looking forward to checking it out.

[DaveWG]

Correction; After a very short flight it does work in FSX a lot better than before, thanks a lot, but question - any reason known why HUD is opaque above about 75% throttle?
K

I've seen this sort of effect before with FS9 ports. It's the material settings used in the HUD that turn opaque when the lights are on. I assume there's a hidden gauge to switch the AB effects on at 75%.

[COBS]

Mal

Thank you for the new FDE for the TSR2 in FSX, your efforts are very much appreciated .
It is a stunning aircraft , what a shame that it never got to show it's real potential in service .

Cheers
Karol