Roll reversal

Hi,

New to the Spitfire and warbirds in general.

I am experiencing some heavy roll reversal while manoeuvering.

I try to be gentle on the stick, easy with the elevators and don't take too much AoA but still struggling to fly decently (needless to say the AI FW is flying like an eagle).

Any suggestions?

Suggestion #1: Don't look at the AI. They use a different flight model, use different physics and fly perfect.

Suggestion #2: Give us a few more details like: How much did you pull at which altitude with which speed? This way we can give you proper suggestions.

I was flying between 10-13000 feet, speed 200-240 mph no less than 170 mph, I pulled 3-5G.

I find it quite difficult to maintain a low AoA while turning, bad habit with the jets maybe (I had the same with the F-14).

I can post a tacview file, don't know if it's gonna be of some help.

Use axes curves..

Sent from my Redmi 5 using Tapatalk

I do.

Use axes curves..

 

Sent from my Redmi 5 using Tapatalk

I can fly fine without any curves. But I'd say try it with and see if you like it.

I do. Curves at 15, a tiny bit of deadzone.

Sounds like you're wing stalling. It's really easy for new pilots in the Mustang to do, but not impossible for the Spitfire.

My advice, don't pull so hard, or so fast and so hard. Also, coordinate your turns. It's amazing how much tighter you can turn a plane with just a little bit of rudder to keep the ball (or needle for the Spitfire) centered.

Hey xavnl,

What joystick hardware do you have?

Hey xavnl,

 

What joystick hardware do you have?

Warthog with extension.

Okay. I was gonna suggest my fairly extreme curvature settings but as you have an extension that will change things.

I would recommend using a custom curvature; the real Spitfire has a lot of elevator travel that you simply do not need to access in flight. In fact you'll routinely find that you start reaching critical AoA at 30-40% of the full stick travel, depending on your airspeed. In the real aircraft this equates to several inches of displacement. On your stick however this will be more in the couple of cm and the margin between in control and out in fractions of a mm.

Start at 0 on your custom curvature then work upwards in smaller geometric steps (think in terms of 2 or 3 rather than 10) until you get to the last three fields then manipulate these to get an exponential curve that tops at 100.

This will give you a more realistic stick displacement up to critical AoA but still allow you access to full elevator displacement at the very limit of stick travel (useful for making taxying marginally less unstable).

I'm using the T-16000M here, only 2 months into it. I haven't needed to touch any curves in the DCS settings for any aircraft.

I prefer it this way, too. Otherwise, I get a more twisted feeling for the aircraft. I don't want vastly different planes to feel the same. I can better feel the difference between controlling the Mustang and the Spitfire.

Also, the other thing that bothers me about curves is when you're trying to aim WHILE turning. I had the hardest time with a huge curve (and saturation at 66 for pitch) with my old T-Flight. It was nearly impossible. When turning, the curve will make things far worse. No curves, and the controls are more balanced, especially in the extremes.

Having less curves will also really help prevent snap rolls/roll reversals when turning. If you turn hard, and maintain that turn, you're fine. But if you need to adjust while pulling a tight turn, that's where a curve can be trouble. A tiny adjustment to your joystick will be a drastic movement ingame, and the quick and sudden change can send you into a snap roll. With no curve, it's more 1 to 1, and a slight change is also a slight change ingame.

Just my thoughts. I use a short T-16000M, so I'd assume a Warthog with an extension should be much easier, and it's just a matter of getting used to the plane.

The Spitfire feels way different from the other aircraft, just because of how sensitive it is. I'd recommend just getting more hours and flight time in it, and learn to adapt to it.

Whilst I agree with Zach partially on the issues with setting curves - you essentially rob Peter to pay Paul trading more delicate control around the datum for increased sensitivity towards the limits, and doing so too aggressively will cause as many problems as it solves - in the Spitfire's case the range of pitch control that you use in flight is very small compared to the actual travel of the elevator/stick pitch control.

The real stick pivot to top dimension was in the region of 900mm - in most off the shelf joysticks it's easily 1/4 to 1/3 of that.

Hence for the same angle of deflection you'll have 1/4 to 1/3 of the displacement.

This in combination with the physical characteristics of the Warthog Yo-yo alluded to will be giving you a harder time than even the real pilots had to deal with.

The purpose should be to give you physical stick displacements that are at least analogous to what the real pilots deal with; now, that's not easy given the disparity in stick lengths, travel and feel but that kinda comes with the territory when simming, and this ergo results in some compromises.

In that regard I would strongly recommend a custom curve with a constant shallow gradient for 70% of the stick travel with a sharp curve at over the remaining part of the range (which you won't access in flight anyway).

Thanks for your comments gents.

I went for a custom made curve with 5 increments.

More stable in the turn and no wing stall.

Cheers.

@DD Agree to disagree. Whatever floats your boat, I guess.

Hey Zach,

The default linear 0% curve would be perfect if you had an exact scale replica, with the correct throws, of the stick for that aircraft.

However most of us don't. Our lateral displacements - particularly fore and aft - are significantly compressed.

Now if you can manage at 0% curve, then good on you - I fully agree that any curve, particularly on the pitch axis will cause some issues for the very reasons you outline, and for that reason I tell everyone if you have to have curve - because some do - make it as small as you can possibly manage and do everything else you can otherwise (stick extensions, practise making smaller & more refined displacements) to reduce your need for the curve.

But the Spitfire is a special case: it's unlike any other Warbird in DCS in that the power of the elevator authority is such that, according to the NACA report, you need only move it 3/4" (19mm) to reach critical AoA.

Attempting to fly it on a 0% curve and a traditional PC gaming joystick of only 20-30cm length is artificially difficult as the displacement covered by that range to critical AoA is going to be in the range of 5-6mm. Ergo making small corrections in pitch for gunnery and landing are harder than in reality and are more likely to result in the virtual pilot over controlling and in turn overcorrecting

I have used the 3/4" (19mm) of aft stick travel to critical AoA and attempted to replicate this with my MSFFB2 custom stick curves. I feel I've got it pretty close - lack of micrometer notwithstanding!

Compare that to my P-51D pitch curve of a 20% curve throughout the pitch range and they still feel vastly different aircraft, the responses still being very unique to each aircraft.

If your T16000 at 0% curve works for you then all well and good - and I salute your steadiness of hand and your ability to make subtle pitch adjustments!

But the Spitfire a special case: it's unlike any other Warbird in DCS in that the power of the elevator authority is such that, according to the NACA report, you need only move it 3/4" (19mm) to reach critical AoA.

 

Attempting to fly it on a 0% curve and a traditional PC gaming joystick of only 20-30cm length is artificially difficult as the displacement covered by that range to critical AoA is going to be in the range of 5-6mm. Ergo making small corrections in pitch for gunnery and landing are harder than in reality and are more likely to result in the virtual pilot over controlling and in turn overcorrecting

I don't think this can be ignored.

The fact that a 5 or 6mm stick movement with a tyical gaming PC-stick can resault in critical AoA is very important.

I personally found the spitfire impossible to handle at 0 curve, with an out-of-the-box stick. This is the reason why. I had people on various forums referring to the "curves" as a cheat!

It's bonkers how mis-informed some flight simmers are about the differences between the ergonics of a real aircraft versus a PC setup.

Similar can be said for visuals/ spotting/ aircraft identification too. . . .

I don't think this can be ignored.

The fact that a 5 or 6mm stick movement with a tyical gaming PC-stick can resault in critical AoA is very important.

I personally found the spitfire impossible to handle at 0 curve, with an out-of-the-box stick. This is the reason why. I had people on various forums referring to the "curves" as a cheat!

 

It's bonkers how mis-informed some flight simmers are about the differences between the ergonics of a real aircraft versus a PC setup.

 

 

Similar can be said for visuals/ spotting/ aircraft identification too. . . .

It's even worse than this and mixes with what Yoyo said. The problem is that the centering force of our joysticks causes us to apply so much force that once we overcome the centering force, we easily overshoot the save deflection range in the spit. That's why a warthog extension and/or curves are almost a must for flying the spit without risking wing rips and blackouts all the time.

Great data, so for this amazing bird what is a good realistic ish curve in regards to our joystick travel for a standard hotas warthog?

Being able to pull Gs and black-out is actually a huge advantage over its competitor which has the silly 'stick forces' implemented. You have to test curves on your own, and most important, you have to let muscle memory kick in if you want to control that crate. You'll eventually find that thin threshold which suits you best.

i want to pointone thing flying in spit around 200mph and pulling 5Gs it will bleeds speed in matter of seconds and goes in to stall very quickly

if you want to do some aerobatcis you want initial speed above 300mph and you need to watch speed if it drops you need to ease on stick

i want to pointone thing flying in spit around 200mph and pulling 5Gs it will bleeds speed in matter of seconds and goes in to stall very quickly

if you want to do some aerobatcis you want initial speed above 300mph and you need to watch speed if it drops you need to ease on stick

Actually this characteristic bleed of speed with Gs / positive AoA is marked, and I believe realistic in DCS vs another ww2 sim... where I can easily pull and wait until I am in trouble for quite a while longer...

´

 

Actually this characteristic bleed of speed with Gs / positive AoA is marked, and I believe realistic in DCS vs another ww2 sim... where I can easily pull and wait until I am in trouble for quite a while longer...

you can pull even 8g for a quite time but not at 200mph

5g at 200 mph too much imo at levbe flight

Sounds like you're wing stalling. It's really easy for new pilots in the Mustang to do, but not impossible for the Spitfire.

My advice, don't pull so hard, or so fast and so hard. Also, coordinate your turns. It's amazing how much tighter you can turn a plane with just a little bit of rudder to keep the ball (or needle for the Spitfire) centered.

This - Try flying straight and level, then kick the rudder. You will notice the airplane start to roll and turn. This is how a lot of smaller propeller aircraft are flown; You turn the aircraft (both yaw and roll) with the rudders. The stick basically just controls your speed, but you will have to adjust your bank as you start to roll and turn still to keep the ball/needle centered