FW190 vs. BF109: Turn Radius (2011)

[drgondog]

Wow. I can almost see the skin ripple on the wing when you talk torsion. In torsion, the section looks like a thin-walled tube with all the load carried by stressed skin, yes?

That is the beauty of the shear web whether the connector for beam caps to convert bending loads to axial forces on the caps and transfer the shear stress to the opposite cap... or to stabilze a torque box (wing or fuse) and enable the shear from one bulhead/longeron panel to another and channel load paths to thingy's like longerons - such as when you stomp a rudder and roll a 51 in a gigh speed dive - and pray your tail doesn't 'depart'

When I said 'show me the drawings, the rivet diameter and spacing will tell me all I need to know about the design limit loads (and load path) when combined with cross sections of the beam/stringer/doubler/longeron, etc

Springs ONLY if you want to develop as sense for the deflections and rotations at different nodes in the model with drastically reduced elements along your simplicity notions.

 The taper I proposed as a means of getting all the properties inot some convenient f(X) form (the dream of a closed-form solution) - but I think I see what you mean about the torsional characteristics... I mean - the beam is not going to rep that thin-walled tube as well as your beam plus torque box. I'm thinking that's because the inertial props can be completely divorced in your rep.

As for Transit structure, it's probably not as bad as a wing but it still involved a fabricated thin-walled structure. Otherwise, your humble input strikes me as really kewl.

The closest analogy I suppose is an air liner/cargo hull design designed to fly in a straight line at low (relative) speeds
Is there any chance you're retired?If so, you might have the time for a development like that...

I still dabble as a management consultant that looks into ugly sales and operations deficiencies and produce glorious process (and sometimes leadership) changes.. and candidly I am a better practitioner than a developer. There are theoretical aero/math wizards that have forgotten more hovering over the john than I will ever know about actual relaxation/iterative solution methods.

The rest, re: loads - I'd also be surprised if its all that sophisticated too. Many of the conditions you describe as difficult still frequntly occur in game, though. I look at that crazy tail slide. Isn't that a fine example where you've probably got detached and/or reversed flow and bad pressure dist on one side of the fuselage? Is that all discounted in force development? I seem to be able to get good speed reductions out of yaw oscillation, but suspect that's just based on a linear model of Cd w/r sideslip... In fact, I''d guess it's all just linear range stuff but have no idea.

I don't play the game and haven't experienced the responses to stick or attitude so I really have no clue.

Finally - life advice goood. The grasshopper, though, is more like an ant. I just sort of like work.

I did well when making a career change to IT Services - but I miss flying and the airframe business (military and civilian) - living well and hard has a tendency to cause deficiencies in critical internal systems when not managed properly - long story/short regarding why I don't fly anymore.

[Stoney]

You two have now effectively departed from Stoney's envelope with all the engineering gee-whiz stuff... 

HTC does model the lift distribution of each wing, complete with pitching moment and all the drag moments at various stations along the span.  Somewhere around here is a thread with a very telling image of an F4U in flight with the flight model shown graphically.  Maybe Moot knows the linky for it--I can't remember and didn't bookmark it.

[drgondog]

You two have now effectively departed from Stoney's envelope with all the engineering gee-whiz stuff... 

HTC does model the lift distribution of each wing, complete with pitching moment and all the drag moments at various stations along the span.  Somewhere around here is a thread with a very telling image of an F4U in flight with the flight model shown graphically.  Maybe Moot knows the linky for it--I can't remember and didn't bookmark it.

Stoney - I will retire gracefully - at least temporarily.

What technique did HTC use to develop the lift distribution - pick an airfoil for the basic pressure dictribution and integrate based on a span-wise algorith to simulate the effect that twist and taper applied to more closely approach elliptical?  - then resolve to a single lift vector that they scaled based on N? Did they fool with the rudder and elevator since that data is also available (although relative AoA in a dwonwash for a high G turn would be tricky)

[Stoney]

Stoney - I will retire gracefully - at least temporarily.

What technique did HTC use to develop the lift distribution - pick an airfoil for the basic pressure dictribution and integrate based on a span-wise algorith to simulate the effect that twist and taper applied to more closely approach elliptical?  - then resolve to a single lift vector that they scaled based on N? Did they fool with the rudder and elevator since that data is also available (although relative AoA in a dwonwash for a high G turn would be tricky)

I'm not sure exactly how they model it, but it appears, from the picture, that each station on the wing, there is a lift component, drag component, and I think a pitching moment.  I know that the prop wash over the wing is modeled, and I assume that similar things are going on on the empenage (it wasn't in the picture).  Its a vector based model, so it has the associated drawbacks, but is still very impressive, especially with respect to the fidelity it achieves with the real life examples.  Dale even modeled an RV-8 to fly in-game to compare to his personal RV-8 to reality check it.  I'd guess, except for some of the playability issues (like the engine management) its probably as close to the real thing as you can get without some sort of CFD type engine.

[drgondog]

Pretty impressive - I wonder if it captures the Aw Sh--! moment on the F4U wing when it loses the inboard left gull lift distribution just before stall?  Doesn't matter it still sounds impressive.

[PJ_Godzilla]

Impressive and tractable... that last is really important. If he's already discretized the wing and has loads, well, that sounds a lot like the jumping-off point we used to always get to with a rigid body model. We'd be able to achieve a certain level of agreement to test, then we'd start wondering about compliance effects. Sometimes, we'd run into crazy ones - like tire memory (tires aren't simple at all), but mostly, it was the usual suspects; frame flex causing compliance-based over/under steer effects, that kind of crap. There's an analog there, given that the tire is like the wing, the slip angle like alpha. Beyond that, it'd be great to hear from the coaders on these matters. In some ways, I could see them really not caring. Otoh, what's correlation worth... to a "game" and a small subscriber base? What's low-cost correlation worth to anyone out there with no access to deep pocket places like Bell or MSC or Ford? There's got to be a cross-sell opportunity there.

[drgondog]

I'm Still Impressed

[Ardy123]

I'm not sure exactly how they model it, but it appears, from the picture, that each station on the wing, there is a lift component, drag component, and I think a pitching moment.  I know that the prop wash over the wing is modeled, and I assume that similar things are going on on the empenage (it wasn't in the picture).  Its a vector based model, so it has the associated drawbacks, but is still very impressive, especially with respect to the fidelity it achieves with the real life examples.  Dale even modeled an RV-8 to fly in-game to compare to his personal RV-8 to reality check it.  I'd guess, except for some of the playability issues (like the engine management) its probably as close to the real thing as you can get without some sort of CFD type engine.

I remember a picture Dale posted, the wing was broken up into 16 lift vectors. I wonder if sub-dividing it further would make a noticeable impact.

[PJ_Godzilla]

I remember a picture Dale posted, the wing was broken up into 16 lift vectors. I wonder if sub-dividing it further would make a noticeable impact.

As Stoney mentioned, and I think it was in our last battle with Gaston, Thor, and the minions of darkness, there was a nice example of a turning 190 with vector graphics on glorious display. One of the memorable things to me was the accounting for differences in velocities relative to freestream of the prop blades. This, in itself is no huge deal - the freestream component of the relative air velocity w/r the prop foil is clearly larger to the outside of the turn - but I was impressed.

As for your question, I suspect they set a target for correlation and hit it - and used the crudest discretization that delivered in order to minimize computation intensity.

That's where I'm coming from with the elastic stuff - how crudely could it be done meaningfully? Would the graphics support it?

[moot]

I remember the dev/debugging screenshot that HT (or Pyro?) posted a while back.  But I don't have it anymore, deleted it with everything AH a year+ back.  Someone's bound to have it though, if it was important enough to eyeball it.

This might have been it:
http://bbs.hitechcreations.com/smf/index.php/topic,197316.msg2309082.html#msg2309082

[Stoney]

I remember the dev/debugging screenshot that HT (or Pyro?) posted a while back.  But I don't have it anymore, deleted it with everything AH a year+ back.  Someone's bound to have it though, if it was important enough to eyeball it.

This might have been it:
http://bbs.hitechcreations.com/smf/index.php/topic,197316.msg2309082.html#msg2309082

That was it Moot, but it looks like Pyro deleted the pictures.  Ehh...well.  And Dragondog, they do model the "gull wing stall" on the F4U.  These pictures would have shown you that.

[Old Sport]

Maybe something like this?

Looks to me like every foot there's a vector.

[drgondog]

As Stoney mentioned, and I think it was in our last battle with Gaston, Thor, and the minions of darkness, there was a nice example of a turning 190 with vector graphics on glorious display. One of the memorable things to me was the accounting for differences in velocities relative to freestream of the prop blades. This, in itself is no huge deal - the freestream component of the relative air velocity w/r the prop foil is clearly larger to the outside of the turn - but I was impressed.

As for your question, I suspect they set a target for correlation and hit it - and used the crudest discretization that delivered in order to minimize computation intensity.

That's where I'm coming from with the elastic stuff - how crudely could it be done meaningfully? Would the graphics support it?

For an example - when I modelled the AH-1 Cobra I used combinations of Beam/plate or rod/panel - the rod panel nicely agreed with 'traditional hand crank analysis' for stress, and lousy dynamic agreement- while the beam/plate was lousy stress wise it was very good for dynamic response.  One was too elastic while agreeing stress loads (and imposing excess deflections) while the other was 'under stressed' but in good agreement with deflections and natural frequency.

NASTRAN eventually 'fixed' the rod/shear panel' model response on harmonics but while I was fooling with it, plates were always 'overly optimistic' for stress analysis.

Summary - if the really complicated, and far better representation of complex geometry, model has intriguing failures on dynamic and elastic response - I am not overly optimistic that a simple beam approach is useful...

[PJ_Godzilla]

Summary - if the really complicated, and far better representation of complex geometry, model has intriguing failures on dynamic and elastic response - I am not overly optimistic that a simple beam approach is useful...

There are really two issues here. Bear in mind, the elast stuff we did in ADAMs was pretty intensive at the time. We were tying up what were state of the art SGI stations to run 'em (I had two of my own in a corner of PArklane restricted from prying eyes and dubbed "The Cage"), and a leaf spring (another example of a simpler elasto we'd do) would often take on the order of hours. I guess the two questions are how simple and how fast... The first pertains to how much the structure could be simplified while still retaining correlation, the second pertains to how much "moore's law" we've gotten since the late 90's.

Otherwise, I was stunned to find you don't play. Why not? It's one of the best diversions I've found. I don't have time to become a "real" fake pilot but still enjoy it. If I get a couple of hours in per week, I'm pretty happy.  You could fly with JV44 if you like.

[drgondog]

There are really two issues here. Bear in mind, the elast stuff we did in ADAMs was pretty intensive at the time. We were tying up what were state of the art SGI stations to run 'em (I had two of my own in a corner of PArklane restricted from prying eyes and dubbed "The Cage"), and a leaf spring (another example of a simpler elasto we'd do) would often take on the order of hours. I guess the two questions are how simple and how fast... The first pertains to how much the structure could be simplified while still retaining correlation, the second pertains to how much "moore's law" we've gotten since the late 90's.

Otherwise, I was stunned to find you don't play. Why not? It's one of the best diversions I've found. I don't have time to become a "real" fake pilot but still enjoy it. If I get a couple of hours in per week, I'm pretty happy.  You could fly with JV44 if you like.

Too much of an addiction.. I would fly with whatever flies the P-51B I suspect.  I have too many diversions as it is between building gunstocks, bird hunting and compiling complete 8th AF lists for air/ground scores/macr's/accident reports, etc in my spare time,

I just finished Vol I Our Might Always - a two part history of the 355th FG, TFW and current FW and trying to decide between self publish or Schiffer (and get hosed, royalty wise). It is now 400+ pages with tables and not counting photos (0-2000)