Su-25T Advanced Flight Model

[169th_Cobra]

I'd like to start a discussion on the new Advanced Flight Model (AFM) and some of the features that people feel may be incorrect. In particular the Su-25T as i haven't bothered with the vanilla Su25 since 1.02.

Please note: i dont want this thread to degrad into a bashing ED thread but more give constructive feedback on the AFM for ED to improve on it.

 

My personal opinion on the AFM is that it feels to be "overmodeled". (i never thought i'd have to use that word.) but it comes about after last night where myself and another 169th member were doing a bit if Su-25T practice to learn the systems in cavok conditions with nil wind and neither of us could land the jet without blowing the front tyre. Also note that both of us are professional pilots!

 

Some of my gripes with the AFM:

 

tyres blow far too easily. - i'd be happy to drop or reduce this feature, even if it is realistic, as for me it just detracts a whole lot more than it adds to the immersion to the game. You are constantly worrying about blowing a tyre just taxiing out

 

no pitch or roll control when on the runway during take-off. - with full back stick at 200km/h you are unable to lift the nose up at all, and likewise in a crosswind one wing lifts but you are unable to keep it level on the runway with ailerons.

 

heavy controls - i'll be the first to conciede that i have never flown a Su-25T but from my experience in flying quite a few different aircraft there aren't any that i have flown that were quite as piggish as this one! happy to hear other peoples inputs with flying experience on this particular topic.

 

As i said i'd like this to be a discussion about the new features of the AFM, but please try to refrain from shooting my personal opinions down with "well you've never flown a Su-25T" as i know this and am the first to admit it, but with over 4000hrs of flying commercial jets i feel i have a little credability about flight models in a general sense.

However, if you can give me facts that the models in Lock-on are correct, ie the su-25's blow an enormous amount of tyres and to combat this, they use this technique ..... then please, i'd love to hear it.

 

Please keep it civil as i know this will be a sensitive subject.

[S77th-GOYA]

Good topic, Cobra.

 

The only thing I would add is whether or not the tires should have a bit more grip on the runway. After landing, I've yawed around so that I'm 90 degrees from where I should be, and even at very, very slow speeds the T just seems to keep skidding. I realize it's a heavy beast, but it just has a feeling of wrongness about it. And yes, this is on the runway with fully inflated tires.

[169th_Cobra]

Thanks for the input Goya ... i think you have another valid point. Perhaps this may also be the casue to the popping tyres? If the nose wheel didnt skid so much perhaps it wouldn't pop?

[Jester_159th]

Well as far as the tyres go, I can't agree with you. I've burst them quite a few times on landing but can put it down to mistakes on my part 100% of the time (rough touch-down or holding the wheel brakes on too long at too high a speed etc). I've never blown a tyre taxiing. Not trying to be funny here..but what speed are you trying to taxi at? IIRC in the readme (or manual..can't remember) it says you should remain below 20km/h when taxiing...otherwise I can fully understand how the tyres would blow...especially cornering a fully loaded Frog.

 

No pitch or roll on the runway: You might have something here. It does seem to "unstick" pretty drastically after take-off speed is reached rather than gradually as speed increases.

 

Heavy controls: Not really. Have a look in the mission editor at the weight that thing is lifting when fully loaded. Combined with the extra fuel and avionics systems it carries (compared to the Su-25) really it's underpowered, so I would think that heaviness in the controls is to be somewhat expected. The difference is performance between a fully loaded 25T and an unarmed one is fairly dramatic, so IMHO I think this one might be down to the payload weight etc. However I'm at the same level of real life experience as yourself, so this is purely a subjective opinion. It would be great to see some figures on this as you suggest, but how accurate are the available figures going to be for a military aircraft currently in service?

 

Excellent discussion topic BTW, Cobra.

[169th_Cobra]

thanks Jester.

 

Yep blowing tyres whie taxiing i seemed to have sorted out and your right in that intially i was going to fast. However, i still standby my point that i feel it is overmodelled somewhat and it is noticable that you get a lot of slide if you put too much rudder in for the speed your going. I'm talking about taxiing speeds (up to 30-40km/h)

 

As an example our current Flight Manual for the 737 says to take 90 degree corners at 10kts (ie taxiway intersections), sweeping corners at 15kts and maximum recommended speed in a straight line is 30kts. Notice that these are in knots, not km/h and not only that the speeds are very conservative and i have seen quite faster with no consequences.

 

What i would expect to happen at the speeds i mentioned above if, you over controlled, is for the nose to dig in and turn as you commanded it and more than likely tip over and dig a wing in. However in real life you get many more clues and feedback that you are going too fast for a corner and you dont need to concentrate on it as intently as you do in Lock-on for fear of blowing a tyre.

 

I agree with you that given time we will be able to work out ways around the tyre blowing issues but my point is that i have never had to deal with this before in my aviation career. What i would like to see is the tyre popping tuned down a bit so that you are only punished if you make a gross error and not if you just land and have a mild amount of rudder in.

 

cheers.

[Tracker]

Having "Wheelbarrowed" a couple times in real planes I knew right away what the steering trouble was in 25T -- its nose heavy. It seems to be modeled with a 'fly by wire' as I see definate delays in the response curve in the rudder controls which will inevitably result in oversteering. I have heard rumors that rubber quality was substandard on the real 25T's resulting in failures. But what intriques me is that on two occasions I actually broke the steering linkage trying to turn to hard. The tire didn't blow, rudders still worked, but no steering (Don't try turning in excess of 50kph).

 

The lift problem is not imagined. Full flaps and 300kph fully loaded this bird should at least float into ground effect (don't ever plan on a softfield takeoff proceedure here) but it flat doesn't. Sim experience shows that lift occurs at approx 320Kph Full Flaps and even then it snaps up rather than floating (which scares the heck out of me with those bombs hanging low back there). The lack of lift in the model no doubt is related to not having control surface correction in a crosswind. Apparently the programmers have left us with needing much more airflow for the ailerons to begin functioning.

 

I like that fact that the AFM allows for the weight differentials of ordinance. But there is a bug somewhere as my A/C has rolled heavy to one side with balanced load (yes I checked the fuel tanks too), Autopilot disengaged (yes checked all lights off) and plenty of airspeed (450kph) for level flight. Even after correcting to level (with great difficulty) the A/C nearly flipped when relaxing the controls. This has happened a couple of times so far.

 

The A/C seems to handle heavy when loaded, but closely approximates the Su-25 when unloaded so I'm not sure about the model yet. I am however going to have to create banding for my trim so I can use my HOTAS instead of Keyboard (uhggg :rolleyes: ).

 

I've never learned anything worthwhile without some level of frustration, and I ALWAYS want to know the reason (even if its my fault) thus I am sort of enjoying the challenges of the SU-25T even if they aren't quite right :icon_wink Wish I could get the thing up to angels 15 (~5000m) without having to do 5ive 360's. :tongue:

 

All things considered, its a beast to fly and I am getting better. Haven't popped a tire in a while, but have broke the linkage and sunk the thing after take off a couple times.

:cool:

[Ironhand]

I don't know guys. The AFM feels pretty damned good to me. I've only blown tires when landing too hard which I think you might expect. And the takeoffs are smooth--the a/c usually floats up. Here's an example of what I mean: Su-25T Takeoff-Heavy. The only time the a/c seems to "jump" off the runway is when I'm carrying no weapons and a light fuel load.

 

Tracker, the times the a/c has rolled unexpectedly for me is when either a stall has taken me by surprise (which you seem to have ruled out in your case) or when trim settings are incorrect. Is the middle trim light by any chance dark when the aircraft rolls unexpectedly on you?

 

Rich

[christopher m]

su-25t

 

I find this fm to be very demanding but very rewarding also.

I,ve learned to takeoff with half flaps, pull back at 280kmph and get smooth

takeoff everytime.

Landings are difficult yes but not impossible.

This new fm has taken me longer to learn than any other before it.

If I overcontrol with full bomb load i also go into spin but it is my fault.

I have set up my jstick to flat responce NO CURVES, this was recommended

by developers as best for this fm.

It might use some tweeking somewhere but I'am not sure where.

If it had options for less difficulty this would prob be good enough for some

people. but I would still use full difficulty.

Thing is this FM requires great patience or the ability to learn patience.

This I learned from lots of accidents.

 

Regards,

[EvilBivol-1]

Just to be sure, did you guys remove the curvature from your input axis in the options screen?

[169th_Cobra]

awesome feedback .. thanks guys.

 

Yep, i run a different profile for the Su-25T vs every other aircraft. Bit of a pain as i have to come out of the game to load up CH Manager but that is a whole other story.

 

i dont know if i mentioned it before, but i am definately loving the challenge of learning the Su-25T and trying to master all its new weapons and systems.

 

thanks again.

[Ironhand]

thanks Jester.

 

Yep blowing tyres whie taxiing i seemed to have sorted out and your right in that intially i was going to fast. However, i still standby my point that i feel it is overmodelled somewhat and it is noticable that you get a lot of slide if you put too much rudder in for the speed your going. I'm talking about taxiing speeds (up to 30-40km/h)

 

As an example our current Flight Manual for the 737 says to take 90 degree corners at 10kts (ie taxiway intersections), sweeping corners at 15kts and maximum recommended speed in a straight line is 30kts. Notice that these are in knots, not km/h and not only that the speeds are very conservative and i have seen quite faster with no consequences.

 

(Remainder snipped)

 

Those are intersting numbers. I did the conversion (1 knot=1.852 km/hr) and 10 knots=18.52 km/hr, 15 knots=27.78 km/hr, and 30 knots=55.56 km/hr. While these are faster than I normally taxi, I just took a taxiing tour around Krymsk airbase because it has a nice selection of sweeping and 90 degree turns in its taxiways.

 

With full fuel and weapons, using the numbers you provide for the 737, I had no problems. Just wanted to pass that on. You mention, too, "that you get a lot of slide if you put too much rudder in for the speed your going." Is this the result of a joystick setting? I was getting no noticeable slide (skid?) using your numbers. Joystick settings affect skid, etc of the tires as well as in-flight behavior of the a/c. FWIW, my joystick is an X-45.

 

Rich

[Guest ruggbutt]

Not to stray too far OT but your training tutorial in 1.1 rocks Ironhand. Thanks for your work.

[EvilBivol-1]

Yep, i run a different profile for the Su-25T vs every other aircraft. Bit of a pain as i have to come out of the game to load up CH Manager but that is a whole other story.

 

Not to be pesky, but I'm not sure whether we are on the same page. I'm not familiar with channel manager, so this may be irrelavent. What I meant was the in-game Input Options, where you can edit axis curvature for X,Y, and Rudder input. Before, for SFM planes most people would set significant curvatures for these. But for AFM planes, it is HIGHLY recommended (I would even say required) to maintain linear input and a very minimal dead-zone. This is because the FM itself is designed to model the correct control inputs (including the axis curvatures) as in the real jet. Having additional curvature sort of overlaps the built-in AFM ones and screws up the model. I imagine the same thing would apply if you are setting you input option through an outside program, such as CH Manager. I would guess that both CH Manager AND internal Lock On axis settings should be made linear.

 

On a side note, an interesting point about the AFM, which isn't fully covered in the manual.

 

The AFM models gyro-error, which can result in erroneous navigation, as well as flight data instrumentation during the flight. As you probably understand, the longer the flight, the larger the percent error. If starting on the tarmac, you should be sure to give the gyros at least three minutes to spin up before taxing out for the flight to ensure maximum accuracy.

 

Lastly, have all you guys already tried flying inverted, or under negative G for any extended amount of time? :)

[ZoomBoy27]

one thing to note about turning on the runways and rampa-areas is the very narrow wheel base for the Su-25T. Combine that with a very heavy payload and you're going to have trouble.

I think it was one ting commented on by real-life SU-25T flyers.

 

As for "takeoff stickiness" I just wait until I reach 340 kmh and it gracefully glides up by itself.

[169th_TweetPuke]

I'm the other guy Cobra discussed the flight model with in that session in the -25T where he had so much trouble with tires. I never had the problem of blowing tires but I do have some concerns about the flight model or, possibly, the handling characteristics of this beast.

 

Much of my initial dissatisfaction was due to stick setup. I have a FSSB mod Cougar that used to have a curve built into the stick and another slight curve in game to refine the feel. Many problems were resolved by removing both curves. It makes every other aircraft in LO a bit twitchy but that's not a bad thing once I got used to it. Still, there's something not exactly right.

 

It's very hard to say without having flown the aircraft whether it's FM or handling qualities. The areas I'm not sure about are:

1. Takeoff - Several people have mentioned the high stick forces necessary to rotate. I agree that they are a bit high but I don't think this is necessarily wrong. It's not unusual to require high stick forces to get the nose off of the ground. This can even be a design feature of an underpowered, high wing loaded aircraft like the -25T. My technique follows real life and manual/tutorial suggestions: rotate to pitot tubes on the horizon at about 250 kph and allowing it to fly off. That usually occurs at 280-300 kph for me. I think the problem many people have is in maintaining that attitude, especially if holding crosswind controls at the same time. This is partially a function of the limitations of our control syatems but may also point to experience with a higher fidelity FM.

2. Roll - There is a tremendous amount of inertia in the roll axis and not very much roll authority to the flight controls. It takes a lot of stick to get a roll going and stopped. It rolls slowly even with high stick pressures. I like the implementation of inertia. While I think it may be a bit too high, it is well done. The main complaint I have is in control authority, it's poor. A heavy 747 has a similar roll rate. That's fine in an airlines but is unsatisfactory in an attack aircraft. The fairly high stick forces further exaggerate the effect. Now, this may be how the real jet flies. If so, I can't believe it ever made it out of flight test.

3. Pitch - I have similar comments to those I made about the roll problem. At first, I thought it handled much like a mechanical control system but I understand the elevator is hydraulically boosted. That seemed very strange again. It didn't make sense to me until I thought about this being an underpowered, high wing loaded aircraft. This could be a design characteristic to temper the effect of a hamfist pilot. I can accept it this way but, again, I find it hard to believe an air force would accept such poor pitch authority in an attack aircraft.

Here's an example of why I think something is not right. I like making weapons passes fast. That's a no brainer. When I come off the target and plan a reattack I have to reposition. I use a whifferdill to do this. The idea is to run out of the target area a short distance (stay within visual range of the target, threats permitting) ease the nose up to bleed speed and decrease turn radius and increase turn rate in the reposition maneuver. As the speed gets below ~450 kph I roll into 90-135 degrees of bank and add backpressure to get the nose below the horizon so I don't slow to less that 370-400 kph while pulling it around through 180 degrees of turn. I come out of the turn nose low with building airspeed for another pass. It's a great technique that retains energy and keeps your speed up in the pass.

It also is difficult to do in the -25T. There is a fairly limited airspeed range between low speed stall and high speed buffet which acts similarly to a stall (I don't think this is right at all). This limits the pitch attitude you can get before you run out of airspeed and then the turn rate is abysmal as you're always right on the edge of low speed or an accelerated stall. I've tried several different airspeeds, pitch, and bank variations with pretty similar results.

4. Airspeed envelope - I'm not too concerned that a heavily loaded, underpowered -25T can't go faster than 450-500 kph (242-270 knots which is slower than a loaded A-10!)although I find it curious. I do find it very strange that it maxes out at around 700 kph (378 knots) and that when it does it buffets and acts like it is in a stall. That doesn't make aerodynamic sense. I could accept that it buffets or that it hits a sudden drag increase that prevents it from going any faster. It's the loss of pitch and/or roll control authority that I doubt.

The causes of these behaviors can be many. Thrust, drag, inertia, control authority, or stall modeling could each have a part in it. Or it could be this aircraft truly just barks and bays at the moon. Past Soviet designs have been notorious for horrendous handling characteristics and control quality. (The MiG-19 comes to mind as the best example.) So, it's definitely possible in my opinion.

Now, all of the criticism aside. I think this is by far the best flight model I've seen in a desktop sim. I think the stall modeling is still weak, too computerish ie only on or off, but everything else is top drawer. Like I said earlier I also think much of the criticism of how this thing flies can be laid at the feet of our stick and rudder systems. I'm very interested in any explanations you testers might have.

[coldcrew]

I think most people are ignoring the weight of the plane when they take it for a spin. You'll have a really hard time taking off under 200 without flaps on a full load. The best way to take of is full take off flaps, when you reach 200 or more put pressure on the stick and hold and it will lift of easily. On attack runs with a heavy load I have it set on half flaps, it just turns better and it is possible to do 90deg banks, just keep the nose below the horizon (but you lose a lot of altitute).

 

The funny thing about taxing is back in 1.02 and 1.1 I saw too many people online just blast their way on the taxiways over grass and just manage to take off. Many people are used to their 50-70km taxi speed I guess.

[EvilBivol-1]

I think the stall modeling is still weak, too computerish ie only on or off

 

Well, I'm not really the one to argue, considering my flying time is limited to a few airline hops and a single private flight (although one with a ~5G turn and quite the aileron roll... :)), nor am I all that well informed on the intricacies of the AFM. Having said that, the great thing about it is just that... no "on or off" scripting. Everything is being calculated in real time. There might be possible errors or inaccuracies in the equation formuals, or the values used in them, but I've yet to see the designers agree to any serious fault, despite numerous concerns like the ones above. And it isn't because of shame, either. ;)

 

What I don't think everybody understands is that the AFM is not strictly a "flight" model. Its an aircraft model, meaning that it is constantly (and dynamically) modeling the plane as a machine, within which various component systems are also dynamically modeled. The hydraulic systems (port+starboard+back up for each, I believe), the pneumatic, the flight control surfaces, the hydro and pneumatic actuators and linkages, the engine, even the outside environment are all accounted for. I should be careful and mention that this is still a simulation, so any number of factors are excluded, for example wheel-brake temperature, but despite this, we have, IMO, one of the coolest PC flight models (ehm, "aircraft" models :)) to be had.

 

Its interesting that you find a lack of roll authority. From the day that I downloaded the demo of 1.1 (I was not a tester yet...) I thought of the roll behavior as the best visible display of the model's accuracy. This is because one of the things that caught me off-guard in my single private flight was the sharp, almost violent ability for the plane to begin and stop a roll. I saw the same thing in the demo, which, as I mention above, was a pure delight. Although, now that I'm writing this, I realize that I have't looked for this in the actual release and in the -T version. Will have to give that a try and come back here to report.

[Shepski]

Hi guys,

 

There were many occasions during testing of the AFM that I myself said to the AFM developers, "I can't belive that the RUAF would accept such a pig of an attack jet". :) They confirmed via real Su-25 pilot's words that the Su-25T, especially due to having the same engines as the older Su-25, was in fact a "pig".

 

The reasons that were given to us after commenting on various characteristics...

 

Taxi and lack of nosewheel grip... the Su-25 has very little weight on the nosewheel compared to the mains which is why it seems to skid so much. Su-25 pilots told them that they must take turns at slow speeds. The skidding and poor steering authority was much worse in earlier builds and we were very persistant in ensuring that it was improved in the final build and they did find an error in a CofG calculation which was corrected. On landing, especially a Xwind, make sure the nosewheel is centered before it touches down and always try to land as soflty as possible in order to not blow tires.

 

On take off... add some nose up trim prior to the take off run and the nose will gently lift off and will be easy to hold at the take off attitude.

 

Maneuverability... Su-25 pilots extend a long way out prior to turning back in in order to be set up on a stable and fast approach... you can't fly this thing like the A-10 when it comes to multiple close range attack passes. Gain a lot of speed before course reversals... like 650-700 km/h if pulling up into a loop in a loaded jet. The Su-25T has an aft CofG compared to most attack jets which is why it seems to reach the buffet so easily... the bloody thing needs more thrust but it unfortunately it uses the same engines as the previous Frogs.

 

I'm sure there will be further tweaking on the AFM as time goes on but don't expect too much. Remember to think of it as a pig or like you're flying a waterbed around the sky. :)

[Jester_159th]

Thanks for the input Shepski.

 

One question though:

 

Considering that the 25T has an aft CofG and is light on the nose, howcome the nose is reluctant to lift until it seems (to me) to suddenly let go at rotation speed? I would have thought that with those characteristics the pilot would be more likely to have to fight to keep the nose wheel on the ground during acceleration to takeoff speed. Or is it just that I've been out of school for so long now that I've got my physics all wrong (And I have to admit that this would be a major possibility!)

 

Thanks for the tip on trim for take off. I'll give that a shot ASAP.

 

I have to say though that the 25T certainly has the most "real" feeling of any flight sim aircraft I've come across. It certainly is a challenge to master (and I think I've a fair way to go yet before I could claim to have done that). It's fast becoming my favorite plane in LOMAC without a doubt....And I used to hate ground-pounding!

[169th_TweetPuke]

Thanks for the info Shepski. A CG that is close to the location of the main gear would in fact contribute to a low weight on the nose gear and reduce the traction the nosewheel can generate. This would account for the "twitchiness" while taxiing. I still don't see this as a particularly big deal. It's as simple as watching your speed while taxiing and not getting into high speed turns.

 

However, I would expect an aft CG to have an entirely different effect on pitch stability and control. Let me explain. The CG in most aircraft without computers to maintain pitch stability is forward of the center of lift. This configuration requires the horizontal tail to generate a "down" force to zero the sum of forces and maintain level flight. A pitch increase is created by raising the aft edge of the elevator or horizontal slab which decreases the angle of attack of the horizontal tail generating additional down force on the tail which raises the nose. It's all very much like a seesaw. The distance between the CG and the center of lift is critical to pitch stability. A greater the distance requires more down force to be generated by the horizontal tail to effect a pitch increase. The result is greater pitch stability as the horizontal stab will have to generate higher loads to make even small pitch movements. The closer the center of lift is to the CG the less the pitch stability. Small movements of the horizontal stab can now generate large pitch moments. This relaxed pitch stability is what makes possible the tremendous maneuverability of recent generations of fighter aircraft. Advanced computer flight control systems are absolutely necessary since mechanical systems and a pilot cannot keep up with the greater sensitivity.

 

So there's my problem with the aft CG answer. An aft CG would increase pitch sensitivity and stick forces would be lighter if everything else is equal. It also shouldn't affect buffet as buffet is a function of the lift characteristics of the wing and/or, possibly, horizontal tail positioning relative to the wing or external stores. The onset of buffet is quicker for heavier loading and higher drag indexes but those are really still a function of the lift characteristics of the wing. Also, the buffet seems to be programmed to occur only with the onset of stall. I'm sure it's a limitation of the FM and isn't an error. I'd just like to see these jets buffet the way they do when you load up the G in real life, without the loss of lift due to stall. But then we wouldn't ahve anything else to hope for in the future.

 

Shepski, the best information you had is that the -25T has to be flown in real life like it does in the sim. That's the best indicator of how good the simulation and FM really is. I can accept that. I'm just thankful I don't have to take this hog into real combat with it's bad habits.

[169th_Cobra]

Evilbivol-1 - yep we are on the same page. CH Manager is the program used to load a profile for CH sticks. Rather than setting the curves in-game you set them in this program.

 

Ironhand - thanks for taking the time to test out a few of those speeds. Your absolutely right in that if you nurse it around you can certainly get far more speed out of it than i ever thought possible. Your also correct about the setup of my joystick play a part in it. I use the mini-stick on a CH Throttle as my rudder control which is certainly not the most user friendly way of doing it.

 

Shepski - as usual, exactly the sort of information i was hoping for!

 

as a captain once said "stick with it!" which is what i'm doing and i'm slowly getting used to it.

 

Appreciate everyones comments.

[Prophet_169th]

I am just starting this thread. So I havent read all 3 pages yet. So if it has been mentioned, sorry.

 

For the Su25T, use a linear response curve. You will see a dramatic improvement in take off, and general handling.

 

At least that is my experience.

 

*edit* LOL, looks like this was nailed on the first page.

[Prophet_169th]

I must be doing something wrong too, I cant get much more than 450km/h fully loaded.

 

Maybe I just need to get Tweek to give me some leasons =)

 

The weekend is near....

[DayGlow]

It was hard to fly the planes at first, but removing the curves from the axis really makes a difference to how the plane handles. The only thing is that I left a curve in my rudders. I use Momo wheel peddals for my rudder and I found with no curve that they were way too twitchy. I was only using a fraction of movement in the peddal. A curve really smooths it out.