G-limiter override characteristics

[BuzzU]

That's what I did. I maxed out in a turn and then pulled the lever. Isn't that how you'd use it normally? In a hard turn in a DF and you want a little more to bring the nose around.

[M1Combat]

Read it again.

 

 

He's talkin about when you pull the lever while the stick is centered, then pull the stick.

 

 

He's saying that when you get to where the limit "would" be, with the lever already pulled, then the G's jump. I don't recall this being the case before so I wanted to test but just can't put time into it right now.

 

 

 

 

What he's NOT saying is that he's pulling the stick to full travel, then pulling the override lever. That's not what he's talking about. It should jump in that scenario.

[M1Combat]

Alright I just tested it.

 

 

It seems to me that what's happening WHEN YOU PULL THE G LIMIT OVERRIDE FIRST, THEN PULL THE STICK is that it does indeed get to the limited G load, and then hang there as you pull more and more on the stick, then it "snaps" to a higher G load.

 

 

I think what's going on is that even when holding the G override it's not explicitly commanding the override. It's forcing you to ask for a specific amount beyond the limit before the override actually kicks in and releases the limit (or moves it... more correctly). If you relax the G back down to just over the limit, then pull... it's smooth. But, as soon as you "reset", you will again see the "snap".

 

 

 

 

Whether the aircraft acts like this in real life would best be answered by someone like Lex... but I'm not certain he quite understood the issue initially.

 

 

So the correct question would be...

 

 

 

 

While holding the override button/lever... Does the real limiter force you to pull the stick well beyond where the regular limit is before it actually releases the limit, thereby making the release quite abrupt? Or... Does it move smoothly passed where the G limit would be if you are holding the override button when you get to where the limiter would be?

Edited May 13, 2019 by M1Combat

[BuzzU]

I wish Lex would chime in and explain how it should be used.

[Lex Talionis]

G limiter overide should simply act as if there is no limiter.

 

If you have the stick at a position where you are not requesting max G from the FCS , and you paddle off the G limiter, there would be no percieved jump in controls as the FCS is not limiting anything yet. If you then pull into a region that exceeds the maximum G limit as calculated by the FCS, the aircraft will respond as if the limiter is non existant.

 

If you have the stick in a position where you are requesting max G from the FCS, and you paddle off the G limiter, the jet would then "jump" as the restriction would then be released and the FCS would give you everything you are demanding.

 

All that said, to try and isolate how this should "feel" is incredibly problematic as you are comparing anicdotal information. It is all very subjective from the way a gamer has their controls set up, the physical differences in controlers alone, to the varying difrences of how individuals perceive.

 

ED really needs to model damage correctly (on all aircraft). A proper over G damage model will remove the focus on the nuances of what should be a rarely used function of the aircraft. A much larger issue in my opinion, they need to fix the difrences in G limiting at hi speeds. 7.5 G at 700 knots is 7.5G at 250 knots. G is G no matter the speed.

 

Hope this helps:)

Edited May 13, 2019 by Lex Talionis

[BuzzU]

Thanks, Lex.

[M1Combat]

It does help Lex... thank you :).

 

 

 

And it sounds like DCS doesn't quite have it right IF the Blk20 had the same programming that whatever you flew did :).

 

 

 

It sounds like it works correctly in the case where the stick position would put the load over the G-limit and you THEN press the override. You "jump" to the higher load as commanded by the stick position.

 

 

It sounds like it does NOT work correctly in the case where you pull the override, then pull the stick. Currently what happens is that you get to 6.5 or wherever the limit is at based on current weight (I was mostly empty so 6.5 for my test), it sticks there for a bit (keep in mind we already have the override activated), and then "jumps" to a couple G higher. Not always to max...

 

 

It seems as though the limiter logic is forcing you to exceed the limited G-load (via stick pull) by some amount before it releases the limiter. Because you've pulled the stick a little ways into the limiter when it does finally release... It "snaps" into a couple/few more G.

 

 

As far as isolating how it should feel... We're just looking at the reported G on the hud :). When the limiter is off (override pressed from before you hit the limiter) it hangs at 6.5 as you pull past 6.5 requested, then snaps to a higher number once you've pulled the stick a ways passed where 6.5G was. Watching the turn rate indicates that the hud indicator is correct and the turn rate does indeed stick at 6.5G, then snap to a higher turn rate once you've pulled enough.

 

 

 

 

We need to make a trak and post in the bug section or at least get someone to mark this as a non-bug if they know it's working right for the software package/version they are simulating...

Edited May 14, 2019 by M1Combat

[MonnieRock]

I Concur

 

 

ED really needs to model damage correctly (on all aircraft). A proper over G damage model will remove the focus on the nuances of what should be a rarely used function of the aircraft. A much larger issue in my opinion, they need to fix the difrences in G limiting at hi speeds. 7.5 G at 700 knots is 7.5G at 250 knots. G is G no matter the speed.

 

 

I concur 100%.

 

 

Happy Simming,

Monnie

[M1Combat]

Aright... 3 tracks. Turn on the control view to see what I was doing with the stick. You'll see in the top trak that the aircraft hangs at the limit (about 6.5) for a tad before it breaks through. The other two work as you'd expect and as Lex says they ought to. The top one with the override pulled before the G pull should just smoothly (or as smooth as I pulled the stick...) run on up to max G.

 

 

Override pulled BEFORE getting to the limiter.

https://forums.eagle.ru/attachment.php?attachmentid=210151&stc=1&d=1557799671

 

 

 

Override not pulled at all.

https://forums.eagle.ru/attachment.php?attachmentid=210152&stc=1&d=1557799671

 

 

 

Override pulled after reaching the limiter.

https://forums.eagle.ru/attachment.php?attachmentid=210153&stc=1&d=1557799671

18G OverridePulled.trk

18G Limited.trk

18G PulledonLimit.trk

[Hummingbird]

G limiter overide should simply act as if there is no limiter.

 

If you have the stick at a position where you are not requesting max G from the FCS , and you paddle off the G limiter, there would be no percieved jump in controls as the FCS is not limiting anything yet. If you then pull into a region that exceeds the maximum G limit as calculated by the FCS, the aircraft will respond as if the limiter is non existant.

 

If you have the stick in a position where you are requesting max G from the FCS, and you paddle off the G limiter, the jet would then "jump" as the restriction would then be released and the FCS would give you everything you are demanding.

 

All that said, to try and isolate how this should "feel" is incredibly problematic as you are comparing anicdotal information. It is all very subjective from the way a gamer has their controls set up, the physical differences in controlers alone, to the varying difrences of how individuals perceive.

 

ED really needs to model damage correctly (on all aircraft). A proper over G damage model will remove the focus on the nuances of what should be a rarely used function of the aircraft. A much larger issue in my opinion, they need to fix the difrences in G limiting at hi speeds. 7.5 G at 700 knots is 7.5G at 250 knots. G is G no matter the speed.

 

Hope this helps:)

 

Thank you, this confirms that there is an issue then.

 

e.i. if you start by holding down the paddle before commanding any pitch (stick is centered), and then proceed to gradually pull aft stick to increase G's then there should NOT be any sudden jump in G onset once you reach the normal FCS limit, instead the onset of G's should continue smoothly past that point in accordance with how slow or fast you're applying aft stick.

 

A jump in G's should ONLY occur if you pull the paddle only after already commanding sufficient aft stick to immediately exceed the normal FCS G limit, like for example if you're pulling full aft stick in a turn and then depress the paddle.

 

Hopefully we will get a fix for this as a sudden jump of 2-3 G's when attempting a controlled high performance turn is very disturbing, esp. if it happens during an attempt at a good gun solution.

Edited May 16, 2019 by Hummingbird

[Hummingbird]

Please fix this ED, it's incredibly annoying

[Lex Talionis]

... they need to model damage a bit more accurately and all these things that are done in the game and not real life will become a non issue

 

 

When pilots use flaps in the f14 and over G them they should become frozen, maybe asometricly, and the aircraft inadvertently roll or the wing incure damage if it tries to sweep ....

 

The f18 should incure various structural problems when over Ged ....

 

etc ..

 

..this nuance is what makes the sim more a game at times .

Edited May 20, 2019 by Lex Talionis

[M1Combat]

For sure... I'm not trying to imply that this should take #1 priority :). For me it's a minor annoyance at most. I didn't even really notice until Hummingbird pointed it out.

 

That said...

 

 

ED relies on all of us in some part to find issues so they can put them on their to-do list :). Hopefully this one gets there at some point, but yeah... I'd rather see proper damage modeling on the list too and they're making big progress on that right now. We're in good hands I think :).

[Lex Talionis]

.... certainly

 

To more emphasize the concern ...

Asometric Gs should also become a big player in BFM . You may see Gx on the gauges, but the second you start rolling the jet under load, the up-going wing is experiencing considerably more Gs than where the G sensors is in the aircraft.

[Exorcet]

... they need to model damage a bit more accurately and all these things that are done in the game and not real life will become a non issue

 

 

When pilots use flaps in the f14 and over G them they should become frozen, maybe asometricly, and the aircraft inadvertently roll or the wing incure damage if it tries to sweep ....

 

The f18 should incure various structural problems when over Ged ....

 

etc ..

 

..this nuance is what makes the sim more a game at times .

The thing with this is that some of these issues only cause visible symptoms over time. You can over-G an airframe repeatedly in a dogfight with no issue, but if this keeps happening sortie after sortie, it will fair prematurely. The latter isn't of concern to the pilot in the here and now and I'm against unrealistic inflating the consequences to make the pilot care (but server options or something like that is fine).

 

 

My take is that the paddle on the F-18 should work like it does on the real aircraft first and foremost, and the consequences for using it should also be realistic. If players want to incentivize certain behavior, then options should be provided for that purpose.

[Lex Talionis]

Understanding failure modes is an issue, metal fatigue of large pieces is only one aspect. All my concerns are only structural.

 

Flap slats tracks warping such that they will not articulate can and has happened often in a single over G (non hornet), LEF torque tube limiter failure (this plagued legacy's for a long time)

Weapons jamming, etc...

 

If the paddle were to work correctly in the hornet, it needs to be in context because "correctly " also means you would just about never use it for fear of the possible results . Literally never unless you were about to hit the ground and needed to arrest your rate of descent so you have better chance to survive the ejection.

 

Paddle switch makes things very arcadey. (Personal opinion)

 

But its nuance. This can be quit the rabbit hole .. (shrug)

Edited May 20, 2019 by Lex Talionis

[M1Combat]

That's a super good point about the roll load. I never thought about that.

 

I think if I were an FCS programmer I would certainly at least TEST maintaining load on the "up" wing and reducing load on the "down" wing to create roll moment instead of increasing load on the up wing. It would certainly move the roll center towards the "up" wing though, so... Maybe that would be bad :). I'm neither an aero-space engineer or a pilot so... Not my ball of wax :).

 

Sounds worth testing to me though... Every advantage is an advantage...

[Lex Talionis]

That's a super good point about the roll load. I never thought about that.

 

I think if I were an FCS programmer I would certainly at least TEST maintaining load on the "up" wing and reducing load on the "down" wing to create roll moment instead of increasing load on the up wing. It would certainly move the roll center towards the "up" wing though, so... Maybe that would be bad :). I'm neither an aero-space engineer or a pilot so... Not my ball of wax :).

 

Sounds worth testing to me though... Every advantage is an advantage...

 

Asometric Gs also plays into spacificaly how we fly the jet. The pilot manages what the FCS does not. We avoid loaded rolls when ever possible. It is always "roll to place lift vector, then pull, roll then pull ..." so on

[Exorcet]

Understanding failure modes is an issue, metal fatigue of large pieces is only one aspect. All my concerns are only structural.

 

Flap slats tracks warping such that they will not articulate can and has happened often in a single over G (non hornet), LEF torque tube limiter failure (this plagued legacy's for a long time)

Weapons jamming, etc...

 

If the paddle were to work correctly in the hornet, it needs to be in context because "correctly " also means you would just about never use it for fear of the possible results . Literally never unless you were about to hit the ground and needed to arrest your rate of descent so you have better chance to survive the ejection.

 

Paddle switch makes things very arcadey. (Personal opinion)

 

But its nuance. This can be quit the rabbit hole .. (shrug)

I'm fine with instant failures so long as they're realistic. I just don't want things overtuned to try to force behavior, but it sounds like we're on the same page.

[Lex Talionis]

Exactly.

 

To try and be more precise in my explanation, what i am trying to illustrate is the spectrum of failures from minor to catastrophic.... the 14 for example, prior to total wing departure there should be. flaps freezing, maybe freezing asometric so it induces a roll the pilot now needs to fight, then the wings no longer sweep, ....then it becomes so fatigued it tears off. The "little things" should fail well before a wing departs(which i believe is a bit unrealistic. May warp the heck out of it causing some veryunique control issues)

 

...and this isn't just opinion, this is typically how failure modes works.

 

...i digress.

[Svend_Dellepude]

F-86 has those symptoms when over-G'd.

[Hummingbird]

A clean F/A-18 should be able to take 9 G's without warping, same as F-14's with a typical operational load went to 12 G's without warping or a single crack to be found. Point being these fighters are capable of taking incredible loads if necessary, well above operational limit. The problem is if it becomes a daily occurrence, then eventually structural fatigue will set in.

 

However in DCS we start in basically a factory fresh aircraft everytime, thus we don't need to worry about the aircrafts age or structural stress history (there's a reason these things are logged in RL).

 

But we're going terribly OT here, the main issue atm being that the FCS actually isn't working as it should. That said it is definitely also of structural importance that the pilot is able to apply a controlled gradual increase in G's past the normal limit when he depresses the override paddle, and that he doesn't suddenly experiences a 2-3 G jump which could be very dangerous.

[Knives]

Usually metals are designed to its yield strength and not to its ultimate strength.

 

Because after it reach its yield strength the behavior would change from elastic to plastic. And the failure becomes unpredictable.

 

So when you over G or go beyond the design limits, you will never know which part has yielded. Because there are no physical symptoms. Unless you dismantle every part and measure it and compare it with its original manufacturer drawings.

 

But when you see cracks that means that part has reached its ultimate strength well beyond its yield strength which was designed for.

 

Fatigue is a different story. There are several mathematical models to discribe it and being used for design and maintenance (this is why logs are important to compare the usage with the design mathematical model) . But at the end it is a ghost, you know it is there but you cannot see it or measure it.

 

So in general, the design limit for mechanical parts is 70% to 85% percent of its yield strength (not ultimate) to ensure longevity of the part.

 

However, some parts are designed to yield (for example bolts) but the reasons are different.

[Lex Talionis]

Swipe at the horse one last time ...

Asymmetrical Gs greatly exceeding indicated Gs, safety margins exceeded at the extremities with rolling pulls at max G (if you roll while pilling 7.5Gs, you can easly exceed double didgets at the wings in the hornet) , components with varying G limits being exceeded to where they will no longer track properly (flaps on the 14 drops G loads to 4 symmetrical) etc, are all probable.

 

There is no such thing as a number that as long as the G meter does not exceed you are safe and can fly the aircraft however you wish. Remove safety "features" only

exacerbate this. No matter how ya slice it this is a recreation so there will always be inaccuracies that can be exploited. Modeling asymmetrical Gs, although a big deal, is probably quit hard to recreate. I am probably being to critical. Apologies.

 

Thanks for proof reading GB. My phone typing skills are not up to par

Edited May 21, 2019 by Lex Talionis

[M1Combat]

Swipe at the horse one last time ...

Asymmetrical Gs greatly exceeding indicated Gs, safety margins exceeded at the extremities with rolling pulls at max G (if you roll while pilling 7.5Gs, you can easly exceed double didgets at the wings in the hornet) , components with varying G limits being exceeded to where they will no longer track properly (flaps on the 14 drops G loads to 4 symmetrical) etc, are all probable.

 

There is no such thing as a number that as long as the G meter does not exceed you are safe and can fly the aircraft however you wish. Remove safety "features" only

exacerbate this. No matter how ya slice it this is a recreation so there will always be inaccuracies that can be exploited. Modeling asymmetrical Gs, although a big deal, is probably quit hard to recreate. I am probably being to critical. Apologies.

 

Thanks for proof reading GB. My phone typing skills are not up to par

 

I think as a community we've found that being a bit "overcritical" as you say is the best way to assist ED in getting the level of fidelity they themselves are after :). Obviously they're the ones with bills to pay so they're the ones that get to decide which features we get or don't get... but one thing I've found over the years is that they most certainly seem as passionate as we do (asymmetrical roll loads or proper G override switch behavior) about getting as much correct as they can.

 

As I said :)… I think we're in good hands, and... just to clarify...

 

 

I don't feel like the G-limit override functionality is more important to model than asymmetric loads during roll. Only that it should be on the list somewhere along with asymmetric loading during roll :). I think the primary issue with modeling those loads would be more about modeling them in a consistent manner across all of the airframes. I don't mean so that they do exactly the same thing at each G load, that would be easy... but so they all act comparably to reality when in context of each airframe... But I doubt any of the manufacturers in question (Boeing, Sukhoi etc...) are going to be very interested in sharing real numbers :). I'm also betting that anyone else that knows would be a bit tight lipped as well. And rightfully so :).

 

On ED's approach to it... From what I've seen them do in the past I'd say they're most likely unwilling to model badly based on lack of data... so at least if all planes are missing the same features of a damage model... then they're comparable in their own way.