P40 Propeller gyro effects (unusual forces noticed) - now an investigation into propeller-induced wobble phenomena

[19//Moach]

this is to continue on my ongoing research about the seemingly excessive forces experienced on the P40 due to gyroscopic precession of the propeller

 

 

a few test flights were conducted, to try and isolate a source for the abnormally unstable dynamics of the kittyhawk - and shortly into that, it started becoming clear that the gyro forces delivered by the propeller are more pronounced on it than in any other plane featured in the simulator

 

 

at max continuous setting (2600rpm) a hard pitch-down maneuver (not up, to avoid stalls interfering) - done by nosing up, then applying full forward stick, with neutral aileron and rudder resulted in the plane yawing 180° before the dive could be brought fully vertical

 

this was conducted several times with similar results, and inevitably led to loss of control at some point or another into the maneuver

 

 

 

then, a series of short pitching maneuvers with neutral aileron/rudder were done, and revealed that at 3000rpm the P40 exhibits a near 1:1 pitch-to-yaw coupling behavior.  this lessens with lower revs (linearly, the math suggests) - and the plane demonstrates more familiar forces around 2400rpm, judging against other planes in the sim

 

 

when performing a full elevator-only pitch down, it was found to be impossible to prevent a full spin developing, with settings of 2600rpm and above - at 2400 and below, the plane is naturally stable enough to resist the resulting force without departing

 

 

I have been grinding the numbers with the help of this calculator here: http://www.gyroscopes.org/math.asp

 

 

unfortunately, it has proven remarkably difficult to find a source listing the actual mass of a Curtiss propeller blade, so calculations had to be ballpark-approximated

 

 

with the 11 feet diameter of the P40 propeller, at 3000rpm, a torque of just over 2000kg/m could be computed for a pitching angular velocity of 30 degrees per second. using a pessimistically estimated gyroscopic mass of ~200kg (I do not believe this would be the case, those being hollow aluminum blades, tapered towards the tips)

 

the average (more loaded than not) P40 weight is just around 3000kg -- though it is not trivial to translate this into the relevant inertia tensor figure, due to the uneven mass distribution around the airframe, we can still use a rough approximation to exercise the point here

 

 

3000kg with a 9.68m length can give us a crude inertia tensor (Mass * (radius²) / 2) for the p40 along the longitudinal axis, which we can use for a rough shot at how it'd resist our estimated yawing torque

 

 

my rough calculations, using these guesstimated values suggest a yaw response ~5% as strong as our input pitch should be expected as a result of the propeller gyro effect (which seems about right, considering how the other planes behave)

 

not exactly the kind of forces which should be able to turn the plane around on itself

 

 

evidently, this is not the behavior observed in the sim, where the trial series of short pitching maneuvers indicated a tendency of all pitch movements being accompanied by a similarly strong yaw - resulting in a diagonal path being traced by the nose against the background, from the view of the pilot

 

 

this leads to the conclusion that the propeller of the P40 in the game, at 3000 rpm, produces a gyro force significantly greater than our approximation suggests - such as what would be needed for it to be capable of matching the fuselage inertia tensor at a near 1:1 ratio - thus producing the observed yaw reaction where the nose moves diagonally from pitching up and down without rudder correction

 

 

keep in mind - this does not account for the effects of airflow and slipstream in any way - but we don't need to go any further to see that the behavior we're getting from this plane is very simply "too weird to be correct"

 

 

 

P-factor (aka: asymmetric disc loading) is probably also at play here -- we could go into calculations over it too, but that'd require even more specialized data on that propeller, which I currently haven't been able to find - if you have it, please share -- anyhow, it wouldn't produce a pitch/yaw coupling effect, due to precession, the greater load on downward side of the propeller would generate a pitch-up moment in response to AoA... this has so far not been observed in any unusual way in the sim

 

 

feel free to help out with the math as I could have made a mistake somewhere, if you can get us a more precise calculation, by all means, go ahead 

 

 

but these results seem in accordance to what has been observed in the simulator, and go a long way in determining a possible cause for the puzzling mysteries of the P40's unconvincingly poor flight performance -- so much discussion has gone over the aerodynamic characteristics of it, maybe the real issue lies with the propeller instead

Edited March 16, 2017 by 19//Moach

[19//Moach]

a further thought on the P40's unusually high gyro forces -- let's compare it to the Yak (whichever version) - it has a very similar propeller configuration as the P40, and has an rpm regime that is pretty much identical ( 2200~3000 ) as well

 

 

now - the yak-1 has a 3.60m diameter prop, whereas the p40 has 3.35m -- and the yak weighs from about 2400 to 2800kg, give or take - the P40 weighs between 3000 and 4000kg depending on loadout....

 

 

curiously however, the in game Yaks don't seem to experience anything close to the degree of yaw forces the p40 gets in reaction to pitch - and pitch in reaction to yaw...  even though it has a larger prop on a smaller, lighter airframe spinning at the very same speeds

 

 

now - remember that propeller gyro forces are not affected by the torque going into the prop - the angular velocity of the blades is what generates these effects - so the fact that the Yak has less horsepower is quite irrelevant in this regard 

 

 

obviously something is not quite right -- the Yak propeller comparison makes it very clear that the gyro forces on the P40 are largely exaggerated, even without grinding down the precise numbers, it is a simple logic:   the two planes cannot reasonably behave so differently in response to the same maneuvers with such a similar prop -- furthermore, the Yak should feel considerably larger gyro effects than the P40, given it's proportions - yet the opposite is true

 

 

further examination of the P40 revealed a remarkably decent turn performance can be achieved, if one uses generous rudder corrections against any pitch movements - it has even proven capable of out-turning a 109 (which I could reliably repeat in multiple cases), just as widely acclaimed -- yet the workload required for this is beyond anything experienced in any other aircraft we have - it basically requires helicopter-like axis coordination, and it does not forgive the slightest error - one false move and it will spin, regardless of airspeed

 

also it has to be done at 2600rpm maximum, not because of engine limits, but because above that the gyro torque becomes too extreme to effectively counter.... and also, it explodes. but that's not the biggest problem you'll have

 

 

 

speaking of blown engines:  a control series of tests has been performed too by deliberately destroying the engine for a zero-rpm gliding run -- the plane performed most predictably and did not display any yaw tendency against pitch in any way - it also proved considerably unwilling to enter a spin, matching all the characteristics described by the operating manual very precisely

 

the glide test series ended abruptly when a wayward tree suddenly jumped into my flight path - the other trees were unharmed

Edited March 14, 2017 by 19//Moach

[BlitzPig_EL]

This is fascinating stuff.

 

I have been searching my books, but can find no data on the weight of the Curtiss prop as installed on the P 40.

 

I'll keep on the hunt for that info.

[unreasonable]

Some info on prop weight - cannot find for P-40 as usual, but for Spitfire, so you can extrapolate. From https://en.wikipedia.org/wiki/Supermarine_Spitfire_(early_Merlin-powered_variants)

 

 

"The early Mk Is were powered by the 1,030 hp (768 kW) Merlin Mk II engine driving an Aero-Products "Watts" 10 ft 8 in (3.3 m) diameter two-blade wooden fixed-pitch propeller, weighing 83 lb (38 kg). From the 78th production airframe, the Aero Products propeller was replaced by a 350 lb (183 kg) de Havilland 9 ft 8 in (2.97 m) diameter, three-bladed, two-position, metal propeller, which greatly improved take-off performance, maximum speed and the service ceiling.

 

In June 1940 de Havilland began manufacturing a kit to convert their two pitch propeller unit to a constant speed propeller. Although this propeller was a great deal heavier than the earlier types (500 lb (227 kg) compared with 350 lb (183 kg)) it provided another substantial improvement in take-off distance and climb rate."

 

Presumably the extra weight was in the constant speed mechanism, ie close to the axis of rotation. Anyway, looks like 200kg is actually ball park right for an upper limit. 

 

As for the rest - a bit short of time to learn the P-40 and test right now, I would love to see a track/video of your P-40 turning 180 degrees on itself after a 90-100 degrees nose down movement!

[19//Moach]

well - my not quite top-notch GTX580 card, albeit capable of 1080p/ultra at 60fps, is alas, one generation too old for quality video capture (no shadowplay for moach  )

 

 

so here's the next best thing: ingame track recordings

 

 

on those short flights, I have demonstrated the P40s reaction to pure pitch maneuvers, without correcting with the rudder (which you'll see me do otherwise, to recover from the "consequences" that follow)

 

 

the results are somewhat bizarre - consider just how much the Yak doesn't do this kinda stunt, and then try and puzzle just how much more prop-to-plane the smaller, larger-prop'd Yak has, compared to the "Buick of the skies" that is the Kittyhawk

 

it does not make sense that the Yak (and LaGG too) experiences much lower propeller gyro forces than the P40, the physics of a spinning propeller unmistakably mandates that the Yak should have a more pronounced reaction, for it is lighter, and with the same RPM has a wider diameter prop

 

there is no aerodynamic phenomena which would account for this most incommensurable response on the P40, so estranged of the other planes in how they react to the same maneuvers

 

 

 

I can confidently say, just by the obvious result of comparing the rest of the IL2 fleet against the P40:   this is not how it should fly

 

 

aerodynamically, I've found that with massive rudder correction, the expected performance (out-turning a 109) can be reliably achieved -- but as I remarked before, this requires input coordination of the sort you would require in a helicopter....

 

and from my experience with rotor-wing sims, I also reckon the P40 in BoX is generally more intense for a pilot to handle and requires more pedal pushing than an MD500 does (the LittleBird is also less prone to trying to kill its pilot -- which it still does, of course, since it is a helicopter) -- while this doesn't prove anything, it's nevertheless weird to regard a fixed wing airplane to be more worksome to control than a thing that supposedly "beats its way up the air"

 

 

 

the conclusions I'm drawing from this are surprisingly simple:

 

- the overpowered gyro torque from the propeller is the true cause of the resulting instability and poor performance

- the error is evidenced by observing that the other planes in the sim get far less of this gyro effect, including types which should get even more of it than the P40 does

 

there is no reason to assume the aerodynamics of the plane are at fault and/or contribute to the performance issues in any manner besides that which is trivial

 

if the gyro excess is effectively countered (not trivial), the plane delivers very satisfactory performance, then matching the historical claims about its capabilities - otherwise it makes for a sideslip attitude which turns the fuselage into a very formidable air-brake

 

also, the engine can be manhandled in "combat" range for way longer than the specified limits indicate - what will constantly cause catastrophic failure is to run at revs above 3000rpm for even a few seconds -- it can tolerate 50" boost at 2600rpm for well over a minute, if no overheat (which rarely happens, since the cooling is "magic")

Edited March 14, 2017 by 19//Moach

[unreasonable]

Thanks for tracks will have a look.  Be interesting if anyone else with good P-40 hours can validate this comparison.

 

Just a note of caution - it may not be a problem with the calculation of the precession force but the overall stability of the plane in flight, it is not just about weight. After all a flying plane is not a gyroscope hanging from a piece of string, and P-40s were known to have iffy tail/rudder issues, so I think it is a little early to be sure that there is a single gross error.  Order of magnitude error can happen even to the most professional people - just a decimal place - but I would still be surprised if that were the case. I suspect it is more of a combination of more subtle issues.

 

At least, however, you are homing in on something easy to check (for the developers). They can just look at their code and say "Nope, precessional forces modeled just as for all others"  or "Whoops!". 

[Brano]

a further thought on the P40's unusually high gyro forces -- let's compare it to the Yak (whichever version) - it has a very similar propeller configuration as the P40, and has an rpm regime that is pretty much identical ( 2200~3000 ) as well

 

 

now - the yak-1 has a 3.60m diameter prop, whereas the p40 has 3.35m -- and the yak weighs from about 2400 to 2800kg, give or take - the P40 weighs between 3000 and 4000kg depending on loadout....

 

Yak 1 engine has max 2700rpm and its prop doesn't have 3.6m diameter. VISh 105SV has 3m in diameter. What you are reffering to is most probably propellor VISh-105K used on IL-2s.That one is 3.6m

[19//Moach]

Yak 1 engine has max 2700rpm and its prop doesn't have 3.6m diameter. VISh 105SV has 3m in diameter. What you are reffering to is most probably propellor VISh-105K used on IL-2s.That one is 3.6m

 

well, I suppose my sources (google) were wrong... or maybe it was a non-standard yak that had this prop listed on the website I got it from - either way - 3m diameter is just 35cm less than the P40s prop - and the yak is still a good tonne less massive than the hawk, so the comparison stands true

 

il2 also has nothing close to these yawing effects we get on the P40 -- nor does the LaGG, or any other plane, regardless of the propeller...

 

also, at 2700rpm the yak still has a lot less of this phenomena to it than the P40 at the same (or even lower) rpm

 

from outside the sim, the supremely detailed and splendiferously accurate A2A mustang for FSX also confirms that even with the large 4-blade job of the '51 - the plane still manifests a much softer yaw tendency on pitch than our P40 here has - so that's one more source to take for comparison

 

 

there's not much else to it - the gyro forces from the p40 propeller are just something "unique" 

 

 

fixing the plane, is a matter of turning this one parameter down to match the standard set by the other planes we got - this would very possibly solve most, if not all, disputes about its handling characteristics (there's still the over-cooling and the startup items to look into, but those are separate issues)

Edited March 14, 2017 by 19//Moach

[unreasonable]

 

 

fixing the plane, is a matter of turning this one parameter down to match the standard set by the other planes we got - this would very possibly solve most, if not all, disputes about its handling characteristics (there's still the over-cooling and the startup items to look into, but those are separate issues)

 

Now here i would disagree quite emphatically. This parameter should only be "tuned down" if it is actually wrong. The precessional force is is relatively trivial calculation - either the team have made a data or calculation error or they have not. The only people who are in a position to say whether this is true are the developers, by looking at their formulas.

[BlitzPig_EL]

A couple of the BlitzPigs work, or have worked, in aviation, they are on the hunt for the prop info.

[unreasonable]

Furthermore, having looked at your first P-40 track, it does not show what you claim at all.

 

Take the first track (2600 rpm) and go to outside view and align the camera behind the plane so that you can see the direction relative to a field. (Better than using the cockpit compass since this likes to be level and takes time to settle).

You will see there is some yaw - to the right - on the initial pitch up to about ~40 degrees above the horizon, but nowhere 1/1 with pitch. I estimate about 5 degrees, someone would have to put a protractor on the screen, I do not have one handy. 

 

Is that an appropriate amount? I do not know, but it is most certainly not a 1-1 relationship.

 

As the nose drops back to the horizontal the yaw reverses so that the heading at zero pitch is almost unchanged from the original heading, as expected. 

 

From then on your ailerons are not neutral: quite clear in the external view.  Bank + elevator = turning, ie changing heading - which means nothing else from here in this first maneuver is useful to measure yaw. You are at up to 45 degrees of bank here. Perhaps you could do this test with a huge dead-zone in your aileron control to make sure you are not putting in small movements. 

 

In the second pitch up you stall the aircraft - the air speed indicator is quite clear - of course you go all over the place. After that you are thrashing around never establishing a constant heading or speed so I stopped watching.

 

Second track - you stall the aircraft in the very first pitch up.

 

This is intended as constructive criticism BTW - you are trying to test and calculate as opposed to just emoting, which I am sure is the only way the developers will ever take any notice. 

 

But if you are expecting a consensus here that the pitch/yaw coupling in the P-40 is grossly out on the basis of these tests you are not going to get it.

[19//Moach]

those tests were indeed not the best demonstration I could come up with - it's not a very easy thing to isolate, as there are many factors that get in the way as you actually fly the plane

 

 

never mind it then - I don't think those flights ever did capture what I was seeking to demonstrate, now that I go over them again... it's tricky, since I get a feel for the plane which does not translate in the tracks, and maneuvers that at first seem like they captured what I wanted don't necessarily do so when I watch the playback again

 

 

 

but I still stand by my theory that the gyro forces are unlike we have on any of the other planes -- I've put it into practice in combat, and found that by very carefully matching every pitch input with a generous amount of rudder, the plane delivers some very appreciable turns - even at low speeds, where it tends to want to go for a spin when flown with the "normal" techniques that we use on the other aircraft without much further thought

 

 

so just disregard the track recordings - it was curious to see how the plane resists pitching down by twisting top-down and entering an inverted spin, but I really never meant for those recordings to be used as empirical data in any way (somebody asked to see what it did, so I posted it mostly for curiosity) -- I would have been much more careful flying through them if I had actually intended them for such use

 

it is probably best that everyone tries it by themselves - that way you can actually get a sense of the forces at play, which you don't by just watching the twists and turns that result from the process... it's simply not the same - these are forces that are felt by a pilot much more than they are seen by anyone else

 

 

the key is - don't stick to what you know the plane should do -- instead, bite the bullet and let it do exactly what you're afraid it might - try to suppress your reflex of correcting it (not very easy in practice) - this was how I noticed the effect in the first place

 

 

a dead-zone test might do well to really isolate this behavior, but it will still be muddied by the interactions of asymmetric stall and P-Factor and a million other little things - anyways, it seems that my x55 stick isn't precise enough for me to rely on it being neutral when it feels like it should be - someone with a less shitty device might have better results there

 

 

 

and worth remembering too -- while hard data would be a blessing, we unfortunately don't have the tools to actually do a proper breakdown of forces when flying in game... if we had some means of "flight debug" display, then precise measurements would be just a click away -- without that, I don't see what else we got besides the "general feel of it" - and whatever numbers we can scrounge up from documents and whatnot... but as far as ingame behavior goes - we just don't have the tools to treat it as an exact science as we'd like - so we have to go by judgement calls about what "feels" more or less right 

 

as far as this goes - the "feel" is that the P40 does exhibit an unusually large precession response compared to the other planes, and reacts most exceptionally to maneuvers and conditions that otherwise yield genereally consistent results across the rest of the fleet.

 

this to such a degree that it requires a very distinct technique to handle safely (let alone effectively) -- it is, well... to put shortly:  rather eccentric

Edited March 14, 2017 by 19//Moach

[Farky]

but I still stand by my theory that the gyro forces are unlike we have on any of the other planes

 

 

I disagree. At least Bf 109F and G, Fw 190 and MC.202 are very similar to P-40E as regards the gyroscopic moments.

 

I was testing this a little and it looks like this left wing drop is more about torque effect than anything else.

 

In hard quick push airplane has tendency to roll, not to yaw rotation. It also depends on the airspeed at which the maneuver is performed. At low speed, even La-5 or MiG had significant tendency to roll. LaGG is different because has "ineffective" elevator - hard quick push results in a very slow maneuver.

 

From what I've seen during my testing, I think that "wrong" is rather Yak or Bf 109E than P-40E. They have very little tendency to rotation, maybe with the exception at low speed, which does not make much sense to me. It should be same or similar to other airplanes. Or maybe not, I don't know.

 

----------------------------

 

P-40E propeller weight - 383 lbs. (173,726 kg)

 

[19//Moach]

From what I've seen during my testing, I think that "wrong" is rather Yak or Bf 109E than P-40E. They have very little tendency to rotation, maybe with the exception at low speed, which does not make much sense to me. It should be same or similar to other airplanes. Or maybe not, I don't know.

 

just what I was thinking - they can't all be correct - at least one side of this relationship must be wrong - be it for more or for less....

[Holtzauge]

Interesting theory Moach and it’s always nice when someone has ideas and presents them in a constructive manner and even backs them up with calculations.

 

However, that being said, I tried a few flight myself now in the P-40 and to be honest, I cannot see that the in-game P-40 suffers from more gyroscopic forces than the other planes? In fact to me it seems more like too fast pitch input stalls one wing faster than the other leading to disturbances in roll which we know in IL-2 has a known connection to yaw which leads to the IMHO infamous BoX yaw-roll coupling that AFAIK the developers are thankfully planning to address. But concerning the P-40 then sorry no, I don’t see that it sticks out here really? In fact it reminds me a bit of the G2 in terms of wobblyness so I would not be surprised if someone said the same IL-2 developers were behind both FM’s!

 

Anyway, I agree with Farky: It does not seem like the P-40 really sticks out in this department and although I was lazy and only tested the Me-109G2 for comparison, I think the G2 also wobbles around in yaw when you push the stick back and forth much like the P-40 so based on my admittedly very quick and dirty testing in my subjective opinion I can’t really see that the P-40 really stands out here.

 

OTOH since I was up flying the G2 and P-40, I also tested level flight just adding rudder and the effect of rudder is truly remarkable: Both crates hardly yaw at all but start rolling like crazy and clearly this is totally off! No way an IRL aircraft will respond like that IMHO: Sure, the yaw will initiate some rolling but the rolling we see now? I remain very sceptical to put it mildly….

 

Concerning the prop weight I think this will be no bigger than around 30 Kg per blade: Assume prop length 3.36/2 m, biggest width circa 0.3 m, blade thickness assumed 6% and blade in solid aluminium alloy at 2700 Kg/m^3. Now assuming this as rectangular slab I get a blade weight of 29 Kg per blade. Sure, it gets thicker at the root but then again the blade is not rectangular in cross section so a rough ballpark guesstimate of the total blade weight for 3 blades would still be in the order of 90 Kg IMHO. In addition, even if the P-40 used a hollow steel blade I would be surprised if that was substantially heavier.

[AndyJWest]

It may possibly be relevant that the P-40 has a fair bit more dihedral than the other BoS/BoM fighters, which will probably result in any yaw induced by gyroscopic precession leading to more roll than seen otherwise. And quite possibly make one wing stalling as a result of yaw more likely too. 

[19//Moach]

yes, a theory it is - and perhaps it still needs refinement - there are simply so many things that could lead to this behavior - which can best be defined as you had called it; "wobble" 

 

 

this is contrary to real world pilot claims about this exact aircraft - here's one for instance: https://youtu.be/gl4BCNqecJ8?t=282

 

 

the p40 in the sim as we have it is not what I'd call "easy to fly" - in fact, it is perhaps the most challenging thing I've ever flown that didn't have the wings spinning on top of it (helicopter) and/or happened to be a haphazard sub-airworthy contraption I stuck together in Kerbal Space Program....

 

anyways - our p40 - whyever it is this way - is not a "beginner plane" - it does not forgive, it does not forget - any slight of concentration is punished with one of the Hawk's infamous stall/spins (which do seem to be correct, according to accounts) - but in normal flight, it tosses wildly to the sides with yaw and roll spasms as you pitch

 

if not the gyro effect, I'm nevertheless positive that the core of the issue still lies with the propeller - and the rest is incidental in result thereof - this I can affirm due to my previous experiment of deliberately causing the engine to seize - freezing the propeller and removing all possible spinning-thing related interference -- the plane glides in accordance to accounts - it is indeed quite easy to fly in those conditions

 

 

and alternative explanation for the "wobble" is perhaps what interpretation the game has given to P-Factor -- a thing about it which easy to confuse, is that the higher-loaded downward blade is subject to precession - thus the P-Factor yields a force that is effected at the lowest point of the propeller disc - and not at the side at which the blade load is greatest

 

which means - P-Factor in reality induces a pitch-up moment in respect to AoA - not a yaw (this is the same concept as a helicopter retreating-blade stall - instead of dropping to a side, a chopper will try to pitch up as the asymmetry of lift becomes more pronounced)

 

 

if this has not been considered, and our simulator takes P-Factor forces without considering precession, it would very possibly lead to similar behavior as we are observing

 

so that is another theory which is probably worth looking into

 

 

and yes - it is relevant to point out this could be affecting all aircraft, albeit in widely (and rather inexplicably) varying degrees

 

 

 

as for the last post - I have also figured the roll component of our wobble is simply being induced by yaw, as the plane rotates, one wing moves faster... you know how it goes

Edited March 14, 2017 by 19//Moach

[Holtzauge]

@Moach: I don't think anyone is disagreeing with you that there is a disconnect between how the P-40 flies in IL-2 and how it flies IRL. However, as I see it the effect that you are trying to nail down here affects other Il-2 aircraft as well and if you look at IRL cockpit videos from the Me-109 for example (have been posted may times in this forum or see Me-109 G4 red 7 on YouTube for example ) you can see a similar discrepancy: There is no wobble here as well IRL so really, I don't think the problem you have identified in the Il-2 P-40 is specific to that particular plane but currently more of a general shortcoming in the BoX modelling of certain planes that's all.

 

Anyway that was the point I was trying to make but don't let that stop you from trying to fix the P-40 and by all means do carry on investigating because if you manage to fix this I for sure would appreciate it as I'm sure the rest of the forum will as well.

Edited March 14, 2017 by Holtzauge

[19//Moach]

feel free to disagree as you please - in fact, my original theory of gyro forces being the culprit, though seemingly sound enough to start with, is probably incorrect, just as suggested here

 

currently, I am much more inclined to believe that the game has a general misrepresentation of P-Factor, possibly omitting the precession effect on the resulting forces of such - this theory so far provides a better fit for the observed phenomena, whereas the gyro concept failed to account for the apparent link between engine power output (boost) and the experienced torques

 

a P-Factor model which neglects precession will be 90° short along the prop rotation from the real thing - and you'd have results such as:

 

with AoA, a yaw tendency towards the upward side of the prop appears - the correct would be for this to be a pitch-up force instead

with side slip, the plane will try to pitch up/down depending on which side you slip into - whereas IRL the propeller would generate a pro-slip force

 

as you move the nose around as one would in a dogfight (violently) - you can expect a nasty wobble to result from the inadvertent pitch/yaw coupling this imposes - the greater the blade load, the more pronounced the effect -- not necessarily in direct relation to total power-2-weight ratio either -- the forces will be greatest on propellers which take large "bites" and rotate relatively slow

 

those would be the MiG, the P40, The G2.... you know - all the "muscle planes" -- curiously, this is the same "group" of planes that exhibit the most noticeable wobbling

 

 

so the key thing to check is:   does the simulator apply the P-Factor forces correctly with relation to gyroscopic precession? -- if not, then this is probably the main source of the wobble issue

Edited March 14, 2017 by 19//Moach

[Farky]

... In fact to me it seems more like too fast pitch input stalls one wing faster than the other leading to disturbances in roll ...

 

 

At the beginning I had two "suspects", either engine torque effect or too fast pitch stall. So i did few "flight tests" without torque - i kill the throttle (0%), no power = no torque. Left wing drop in hard quick push maneuver disappeared. I therefore think that reason for left wing drop is torque effect.

 

 

so the key thing to check is: does the simulator apply the P-Factor forces correctly with relation to gyroscopic precession? -- if not, then this is probably the main source of the wobble issue

 

What do you mean by relation between P-factor forces and gyroscopic precession?

 

Anyway, I think that P-Factor and gyroscopic precession are in game negligible or completely missing. However, I can be wrong of course, this would require more testing on various airspeeds etc. When I did "no torque tests", i kill throttle from normal (continuous) mode (37inHg/2600rpm). Propeller was able maintain 2600 rpm for some time (at 200+ mph IAS if I remember well) and without torque I have not encountered any really noticeable rotation, either in climb or in dive. If propeller (engine) is really source of "wobbling", main reason is the torque (torque effect). Or propwash.

 

 

... the gyro concept failed to account for the apparent link between engine power output (boost) and the experienced torques ...

 

 

What do you mean by gyro concept?

[unreasonable]

 

Anyway, I think that P-Factor and gyroscopic precession are in game negligible or completely missing. However, I can be wrong of course, this would require more testing on various airspeeds etc. When I did "no torque tests", i kill throttle from normal (continuous) mode (37inHg/2600rpm). Propeller was able maintain 2600 rpm for some time (at 200+ mph IAS if I remember well) and without torque I have not encountered any really noticeable rotation, either in climb or in dive. If propeller (engine) is really source of "wobbling", main reason is the torque (torque effect). Or propwash.

 

 

Not completely missing - I thought Moach's first maneuver in first track showed precessional forces rather well, but only during the initial pitch up to 45 degrees then down to level again. During this time the wings are still level, but the plane yaws several degrees to the right during the pitch up then back during the reverse. This is much easier to see in outside view than in cockpit - the compass only moves a couple of degrees in this phase of his test but in outside view you can see the heading move more.  

 

In BoS there seems to be a clear yawing force as you lift the tail quickly during a take-off run - I do not think I am imagining it.....   During general flight at normal WW2 speeds I would not expect it to be particularly noticeable. 

 

I recall during RoF FM discussions that precessional forces were often thought to be undermodelled - but they are clearly there and it may just have been that the planes are too stable, or that taking accounts of flying the Camel as the norm for how precession should affect the typical feel of flight is not a good idea.

 

The pitch/roll coupling seems to be another RoF hangover - the thing is that when you read modern accounts of flying WW1 planes, you often see phrases like "to turn you first put the aircraft on it's side using the rudder..."  aileron design being generally fairly ineffective, despite or because of being enormous. So the overall effect seemed to match up with many accounts, despite perhaps being based on an incomplete model - as all models must be. 

[Farky]

You are right, gyroscopic precession is there. It is clearly noticeable during aggressive pull up, nose goes right (or left in some airplanes of course). P-factor is probably there also, for example in climb with Bf 109E-7 nose goes constantly to the left. It's hard to say if there are all four left turning tendencies, it's very difficult/impossible to isolate one phenomenon from the other using "Eyeball Mk.I" and "Chainsmoker's shaky hands" devices.

[Holtzauge]

At the beginning I had two "suspects", either engine torque effect or too fast pitch stall. So i did few "flight tests" without torque - i kill the throttle (0%), no power = no torque. Left wing drop in hard quick push maneuver disappeared. I therefore think that reason for left wing drop is torque effect.

 

Yes, that one wings stall before the other is probably due to torque and maybe also to some extent due to the rotating from the propeller slip stream. I remember that there was initially problems with the BoS Me-109 stalling when people tried to initiate dives: The in-game Me-109 would go into a negative stall and roll and yaw when you pushed the stick forward tried to initiate a dive. Of course if you push too hard IRL that would also happen but this occurred at rather moderate pitch inputs in-game IIRC and was after a while fixed by the developers. To me the moral of this story is that the BoX FM is a malleable beast and things can be and are fixed. Since the P-40 also behaves strangely in this aspect it makes sense that what was once done for the Me-109 could probably also be done for the P-40 as well. The problem here is of course in proving that what we have today is off but I for one find it very hard to believe that the P-40 IRL behaved like it does in the sim today so hopefully the developers can be convinced to do a similar fix on the P-40 as was done on the Me-109.

 

Not completely missing - I thought Moach's first maneuver in first track showed precessional forces rather well, but only during the initial pitch up to 45 degrees then down to level again. During this time the wings are still level, but the plane yaws several degrees to the right during the pitch up then back during the reverse. This is much easier to see in outside view than in cockpit - the compass only moves a couple of degrees in this phase of his test but in outside view you can see the heading move more.  

 

In BoS there seems to be a clear yawing force as you lift the tail quickly during a take-off run - I do not think I am imagining it.....   During general flight at normal WW2 speeds I would not expect it to be particularly noticeable. 

 

I recall during RoF FM discussions that precessional forces were often thought to be undermodelled - but they are clearly there and it may just have been that the planes are too stable, or that taking accounts of flying the Camel as the norm for how precession should affect the typical feel of flight is not a good idea.

 

The pitch/roll coupling seems to be another RoF hangover - the thing is that when you read modern accounts of flying WW1 planes, you often see phrases like "to turn you first put the aircraft on it's side using the rudder..."  aileron design being generally fairly ineffective, despite or because of being enormous. So the overall effect seemed to match up with many accounts, despite perhaps being based on an incomplete model - as all models must be.

You are right, gyroscopic precession is there. It is clearly noticeable during aggressive pull up, nose goes right (or left in some airplanes of course). P-factor is probably there also, for example in climb with Bf 109E-7 nose goes constantly to the left. It's hard to say if there are all four left turning tendencies, it's very difficult/impossible to isolate one phenomenon from the other using "Eyeball Mk.I" and "Chainsmoker's shaky hands" devices.

I’m also pretty sure both the P-factor and gyroscopic precession is modelled: They are for sure in DCS and one of the developers there has had long discussions with the community about the effects and in order to get the quality we have in the BoX FM I don’t see how the developers could have achieved it without modelling the propeller blades as “small wings” with mass, i.e. using some sort of bound lifting line or vortex panel model for all lift generating surfaces and if you do it like that it has the bonus that all effects like precession, P-factor etc. should come for “free” as far as I can see. With “free” I of course mean that you don’t have to script things and slipstream rotation, torque, unsymmetrical stall effects, P-factor and precession will all come naturally since the model is a good enough representation of reality that they will be captured. However, the problem here I think lies in tuning all the variables for these components in this model so that you fulfil all the boundary conditions like speed, climb, turn rate etc. while at the same time getting the flying characteristics right. Don’t know if this is how they model it but if it is, then I can understand that the tuning can be time consuming……..

[19//Moach]

What do you mean by relation between P-factor forces and gyroscopic precession?

 

 

what I mean is this: https://en.wikipedia.org/wiki/P-factor#Effects

 

the text there says: 

 

Confusion about the effect of P factor exists due to gyroscopic precession. The asymmetric thrust produced by the descending blade produces a force to the propeller disc at an average location of 3 o'clock. The resultant force due to precession therefore acts on the propeller disc at 6 o'clock, causing a pitch up tendency as opposed to a left yaw tendency, as traditionally thought.

 

 

there is much controversy and a wide assortment of misleading information about the real effects of P-Factor on airplanes - presumably, the controversies survive due to the difficulty in isolating this from other (often stronger) phenomena which contribute to behavior that is easily misattributed to this cause

 

so, let us make things easier to understand by scaling up our thought experiment - while a propeller is relatively small, and the effects of gyroscopic precession on it when asymmetrically loaded are easily overshadowed by larger factors, (like spiral slipstream)  a helicopter rotor makes this much clearer to grasp

 

so, here's a helicopter rotor for us to learn from:  https://youtu.be/eTjGTxSevHE?t=67  -- I strongly recommend you watch that entire video series on helicopter controls

 

one can easily see the connection between this phenomena in helicopters and propeller airplanes - they are in fact, closely related, if not the very same thing, see our helicopter again: https://youtu.be/_pbdwueqGp4?t=174

 

 

the much larger rotor (and lack of a tail surface "under" it to interact with the slipstream) now makes the real nature of these phenomena very much easier to identify -- the mechanics of gyro precession are very well understood in helicopter design, so much so that cyclic controls are designed to account for it, and apply pilot inputs 90° ahead of the required position to achieve the expected results, see: https://en.wikipedia.org/wiki/Phase_lag_(rotorcraft)

 

 

similarly, as pointed out before - a retreating blade with less lift, will not induce a left bank, but actually a pitch-up force will develop (effectively reducing the helicopters forward speed, and ultimately limiting its maximum achievable velocity) - see: https://en.wikipedia.org/wiki/Retreating_blade_stall#Flight_performance_during_a_retreating_blade_stall

 

 

 

I thus reckon that (as with many things in aviation) a vast array of incorrect literature exists about the REAL effects of propeller load asymmetry (aka: P-Factor) -- more than too likely, the confusion is fueled by the much larger effects of engine torque and the spiral slipstream interacting with the airframe 

 

but when you consider how a helicopter reacts to changes in lift around different sides of the rotor, it becomes abundantly clear that most "common knowledge" about this phenomena is largely mistaken, when it comes to propellers

 

 

from this, we can thus confidently draw a conclusion and affirm that:  

asymmetric thrust on the downward side of a propeller will NOT cause a yaw to the opposite direction, but an upwards pitch instead

 

 

 

with that in mind, I have done experiments in game which suggest the model used to represent P-Factor (at least on the P40) will incorrectly induce this same mythically promoted right yaw motion in relation to AoA

 

 

this can easily be verified by doing a steady slow flight holding the plane at altitude by raising the nose higher until the yaw tendency becomes obvious - this happens close to the point of stalling, so be careful when doing it

 

the plane will inexplicably start yawing to the right (which it does not do with the propeller stopped) - and left rudder is required to keep straight 

 

it is possible that this yaw is being contributed to by engine torque, and by the rudder catching the slipstream as well - however, this in is greater accordance to the observed tendency to drop the right wing and ultimately spin out that way on hard pull-ups, another easily performed experiment you can do in game

 

as you sharply introduce a larger AoA, by diving in (to around 300mph) and firmly pulling back on the stick, you may notice a strong rightwards force, which will require a good deal of rudder to overcome (and if properly countered, yields a turn radius fitting of the plane's historical reputation, i.e. it can turn inside the 109) 

 

the "no-precession P-Factor model" theory offers a solid explanation to why this happens, as well as suggesting a factor which if changed, could very likely solve the problems we're experiencing in this regard

 

conveniently (or not), a correctly modelled P-Factor would also provide a small but much appreciated "free pitch-up boost" when pulling up, which should help you turn and/or stall faster, depending on how you handle it

 

 

 

therefore, it can be reasoned that the most likely candidate for a fundamental cause of the general "wobble issue", as well as the P40's unhistorical "handling deficiencies", is the possibility that the game fails to factor in gyroscopic precession in its modelling of P-Factor effects, instead applying the forces directly at the point where they affect the propeller 

 

and quite likely, the game could also have this process optimized by "skipping" the calculation of these localized forces, and applying the results thereof directly as torque, acting on the airframe as a whole - this would make perfect sense to have in a desktop simulation, and should not noticeably detract from the flight dynamics if the modelling were to correctly observe all aspects of the effect as it calculates these torques

 

 

regardless of this optimization being the case or not, a precession-free P-Factor model will still have its effects wrongly rotated 90° along the longitudinal axis in a counter-propeller-wise direction, resulting in the same behaviors observed in our experiments and also providing a sound explanation to the stability issues and wobble tendencies reported here

 

and while most evident on the P-40, this is very likely the same major cause of the infamous "wobble" behavior experienced in various other types in our sim

Edited March 16, 2017 by 19//Moach

[unreasonable]

Another grand unified theory. I have to say that I trust the developers to have the basics of their physics right even if we all know that some of the input data they have to work with is incomplete or suspect.

 

The P-40 yaws to the right as you pull up the nose (and left as you push it down) because of what looks like entirely correct precessional forces: as you demonstrated earlier.

 

I have just spent 30 minutes or so flying the P-40 and as far as I can see if you keep it in stable nose up flight at low speed  - hands off, trimmed tail down - any roll or yaw  is very tiny. So even if it is "wrong" - which I very much doubt - it would have a trivial effect on handling.

 

You responded to my previous challenge for tracks and calculations with tracks and calculations, so brownie points for that. Also points for having seen the feedback and accepting that it does not show what you thought is showed.

 

So same again with this most recent hypothesis please. (Although I am beginning to think that you are just trolling us  )

Edited March 16, 2017 by unreasonable

[19//Moach]

Another grand unified theory. I have to say that I trust the developers to have the basics of their physics right even if we all know that some of the input data they have to work with is incomplete or suspect.

 

The P-40 yaws to the right as you pull up the nose (and left as you push it down) because of what looks like entirely correct precessional forces: as you demonstrated earlier.

 

I have just spent 30 minutes or so flying the P-40 and as far as I can see if you keep it in stable nose up flight at low speed  - hands off, trimmed tail down - any roll or yaw  is very tiny. So even if it is "wrong" - which I very much doubt - it would have a trivial effect on handling.

 

You responded to my previous challenge for tracks and calculations with tracks and calculations, so brownie points for that. Also points for having seen the feedback and accepting that it does not show what you thought is showed.

 

So same again with this most recent hypothesis please. (Although I am beginning to think that you are just trolling us  )

 

 

I assure you I do not mean to "troll" anyone 

 

if you have found disagreeing results by repeating the experiment on your side, please explain how it is different from what I have noticed  - I can only account for what I perceive on my end, with my (perhaps erratic) joystick setup

 

and what I have come to notice is that, particularly with the P40, there is a tendency for yawing in response to pitch movements and/or AoA which is unfamiliar to the other planes - now, as to what extent it is difficult to tell, since we do not have any tools to measure ingame forces in a way that would negate the need to work by "feel"

 

none of this would be in argument for one end or another if we had a "flight debug mode" option, in which the forces acting on the plane could be inspected as a track recording is replayed - alas, without that, it is all we can do to try experiments and trying to see if we agree on the results - which can vary wildly across each one of us

 

 

have you not noticed any tendency to yaw increasing with AoA, or in response to pitch movements? - this would mean you're getting results different from mine, which can be due to a huge number of things... 

 

 

as far as I can tell from my end, the yaw is there - and that suggests either there is something to this hypothesis or there is something interfering with my flight controls (on some planes more than in others, with no changes made between them)  - others have remarked they observed yawing as well in this thread, so there must be something to it which may be worthwhile for devs to go about investigating

 

 

the slow flight experiment has revealed indeed a small amount of yaw - I never said it'd be huge - but it is a yaw nonetheless and the experiment was done just to verify that it was there, large or small - I cannot be sure if it derives from P-Factor (which would be wrong) based on that experiment alone - it could easily be due to engine torque - the rudder and ailerons were trimmed to neutral for this, and I did not move either control as the plane slowed down, only corrected with pitch to hold level flight

 

on the hard pull-up test, however, a somewhat stronger yaw force is felt - which on the P40 is unusually severe when compared to the other planes (though some also exhibit similar behavior, somewhat inconsistently perhaps)

 

 

I cannot tell by that last post if you've found contradictory results on your own experiments with a possible "precession-free P-Factor model" or whatever else you may be trying to suggest... actually, I really don't know what your point was there.... 

 

 

I have spent the last few days scrutinizing the behavior of the P40, and it does show a larger tendency for cross-axis coupling than we generally see on the other planes - yet exact figures cannot be had, as the game simply doesn't provide this information for us to verify the things we may "feel" about the plane

 

 

the only thing to take out of this is a suggestion for devs to "have a look and see" - there is a vast amount of incorrect information published about this particular aspect of how P-Factor affects an airplane - it does not require a huge stretch of imagination to reckon that just maybe, an incorrect source was used in the making of the sim (many other titles have done this the same way)

 

so yes, this is but an educated guess on a probable cause for a number of things we have read about various times in this forum - I cannot say with any certainty whether or not this is indeed the case, but it would be unwisely dismissive not to ask this one simple question:

 

 

 does P-Factor produce a pitch or a yaw in the game? 

Edited March 16, 2017 by 19//Moach

[JtD]

The issue with the P-40 is the low yaw damping, and that's pretty much it as far as handling is concerned. All aircraft show a significant amount of cross coupling, but the P-40 is extremely poor at damping things. For instance, if you push-pull-push a Yak, you'll have the nose make pretty much a line from top left to bottom right, if you do this in a 109, it will do a clockwise ellipse, whereas if you do the same thing in a P-40, it will do pretty much a clockwise circle. Why? Because the Yak is extremely well damped, the 109 still OK, and the P-40 just sucks. If you like, any deviation from a straight line is lack of damping, any deviation from vertical are gyroscopic effects. The latter is similar, the former completely different.

 

In addition to the yaw damping, the P-40 also suffers from an extremely sensitive rudder, which it had in real life, but only at very low speeds. According to the docs I posted in the other topic, you'd need about 300lb on the pedal to achieve something like 5° of rudder deflection - at speeds as low as 200 mph IAS. In game, you can still thrown the plane about just using the rudder at 300mph IAS. Which is why in real life, P-40 pilots had their already mentioned muscular leg, while in game without any physical effort, we can easily push the P-40 beyond the limit.

[AndyJWest]

19//Moach, if you want to claim that P-factor is affected by gyroscopic precession, I suggest that you find a better source than a Wikipedia article edited by an entirely anonymous contributor, backing up the claim with no evidence whatsoever.

[unreasonable]

I assure you I do not mean to "troll" anyone 

 

if you have found disagreeing results by repeating the experiment on your side, please explain how it is different from what I have noticed  - I can only account for what I perceive on my end, with my (perhaps erratic) joystick setup

 

 

I do not mean to be rude - I have spent quite a bit of time looking at your tracks and trying to replicate various effects in game myself, since you have taken the time to provide them. As I said, I can fly slowly nose up with hands and feet off and there is a very slight right roll and yaw, easily corrected with trim changes - rudder and aileron trims start the Quick Mission in neutral.

 

What I mean is that claiming that some specific effect is a) the overall reason for the P-40's horrid behaviour in game and/or b) that the global FM contains gross errors or omissions in it's physics formulas is a big claim that can only be backed up by solid analysis and demonstration. Neither of your claims in this thread are anywhere near that threshold, IMHO.  As you say, we simply do not have the tools in game to measure effects separately: and what I observe in my own crude tests does not support your theories. 

 

On the other hand, if you were to present your hypotheses a little more cautiously, or ask why the P-40 sometimes seems to do X,Y,Z sometimes, someone like JtD will come up with a useful insight or even an answer. That might be a more fruitful approach than thinking you can identify basic errors in the FM by flying around. So I am not trying to put you off investigating analytically - far from it!

[Farky]

what I mean is this: https://en.wikipedia.org/wiki/P-factor#Effects

 

the text there says: 

 

Confusion about the effect of P factor exists due to gyroscopic precession. The asymmetric thrust produced by the descending blade produces a force to the propeller disc at an average location of 3 o'clock. The resultant force due to precession therefore acts on the propeller disc at 6 o'clock, causing a pitch up tendency as opposed to a left yaw tendency, as traditionally thought.

 

This text is wrong, P-factor and gyroscopic precession are not related at all.

 

Check this very nice video about left turning tendencies, good stuff -

 

 

 

does P-Factor produce a pitch or a yaw in the game?

 

 

Yaw of course, which is correct.

[19//Moach]

what continues to puzzle me about this 'left yaw" business is:

 

 

why/how is this different from a helicopter rotor? 

 

how come when you have a heli rotor under asymmetric forces the results are affected by precession, but when you apply the exact same concept to a propeller, then they are not?

 

 

what is it that's different between the two?  one side has a higher load, and the rotor/propeller is spinning, I don't see any reason why they're not the same way - by this logic, the wikipedia text should actually be correct

 

unless there's something else to it - we're dealing with two cases where a rotating object has a larger force on one side than another - in one of them we have this force acting 90 degrees ahead, due to precession, in the other we do not... for no other reason than it being omitted from the explanation

 

or something else is missing here

 

can anyone explain then, why does the rotor have precession affect the resulting forces while the propeller does not?

Edited March 16, 2017 by 19//Moach

[unreasonable]

Pursuing your thought experiment - from one of your links:

 

"In a retreating-blade stall, however, only the retreating half of the helicopter's rotor disc experiences a stall. The advancing blade continues to generate lift, but the retreating blade enters a stall condition, usually resulting in an uncommanded increase in pitch of the nose and a roll in the direction of the retreating side of the rotor disc"

 

There are two effects described here - the pitch up and the roll. The roll is down on the retreating side of the rotor disc - ie to the side that is exerting less lift.  The division between retreating and advancing sides of the rotor goes forwards - for a propeller disc it is a vertical line. So for a propeller an analogous force to the roll of the helicopter would be a yaw. So even this example certainly does not show that asymmetric prop lift would not cause a yawing action. It suggests that it would - but it is hard to get one's head around this without a nice scale model in front of you.  

 

Would it also cause a pitch up similar to that experienced by the helicopter?  Will have to read up a bit more.  

 

edit Whole threads of current helicopter pilots disagreeing on how their contraptions work...  

Edited March 16, 2017 by unreasonable

[19//Moach]

a simpler example even is the cyclic control of a rotor - the input linkages are designed to affect the blade 90° ahead of the point where the forces are expected 

 

so in order to pitch up, a helicopter increases lift on the forward side of the rotor, see here: https://youtu.be/eTjGTxSevHE?t=67

 

 

besides it being simply overlooked from any explanations about P-Factor, I don't see why this would not be the case for a propeller in an analogous way -- if the helicopter reacts to more lift on one side by pitching up instead of rolling, then simple logic suggests a propeller with more thrust on one side should act in a similar way, producing an upwards pitch force

 

 

I have yet to encounter any sources that provide an actual reason for this to be different (most of them simply don't mention any of this) - unless there is another factor that cancels out precession on the propeller and not the rotor, then it should be fair to assume that what the wikipedia text suggests is actually true, despite lacking a citation to back it up - logic is on its side

Edited March 16, 2017 by 19//Moach

[unreasonable]

Moach, all the sources I can find on retreating blade stalls say that the helicopter will tend to pitch up and roll down on the low lift side. Including your link. So it is not a simple either/or in this case.

Edited March 16, 2017 by unreasonable

[Holtzauge]

But isn’t it clear that if you pull back on the stick and pitch up the nose, i.e. introduce a pitch acceleration and resulting velocity, you will get both a yawing and pitching effect from the propeller?

 

Beginning with the prop blades when they are in the horizontal plane: On one side the aoa of the blade will due to the pitching velocity the acceleration causes experience an increase, while on the other side a decrease. So here we get a yawing force and hence a movement in yaw.

 

Looking at the blades when they are in the vertical, the same pitching velocity will again result in an increase on one blade and a decrease on the other so we get a component in pitch as well.

 

Then on top of this you get the reverse effect when you decelerate the pitching speed and stop at a new aoa and added to that you will get the classical P-effect of a propeller operating at a higher aoa.

 

Add to this the gyroscopic precession, roll, pitch and yaw effects of the rotating propeller slipstream, the aerodynamic derivates for the angular accelerations and speeds in both yaw and pitch for the wing and tail assembly and to top it off  a correct model of the planes moments of inertia on all axes and then you are good to go. A piece of cake right?

 

So I don’t think it’s as simple as it’s “only propeller effects” or a “yaw dampening” problem and nothing else. As I said before: I believe they have a generic FM model used for all planes that gives all that for “free” but the problem lies in tuning all the parameters so that you get both the performance and the handling aspects right.

 

As far as I’m concerned I hope that this is a tuning problem so that they can tune this out of the P-40 and Me-109G2 because as it is today I stay away from both because I don’t find the wobbly FM of either believable or enjoyable to fly. The alternative, that the FM model they use has some inherent weakness that is un-tuneable and will leave us with those planes in more or less their current status, I don’t even want to contemplate…..

 

However, I still have a nagging suspicion that different BoX planes are done by different developer teams and I would not be surprised if it turns out that the same team that did the Me-109G2 did the P-40 but that would be pure speculation of course……

 

BTW, maybe we should employ the diplomatic prowess of President Merkin Muffley to present our case?

[AndyJWest]

a simpler example even is the cyclic control of a rotor - the input linkages are designed to affect the blade 90° ahead of the point where the forces are expected 

 

so in order to pitch up, a helicopter increases lift on the forward side of the rotor, see here: https://youtu.be/eTjGTxSevHE?t=67

 

 

besides it being simply overlooked from any explanations about P-Factor, I don't see why this would not be the case for a propeller in an analogous way -- if the helicopter reacts to more lift on one side by pitching up instead of rolling, then simple logic suggests a propeller with more thrust on one side should act in a similar way, producing an upwards pitch force

 

 

I have yet to encounter any sources that provide an actual reason for this to be different (most of them simply don't mention any of this) - unless there is another factor that cancels out precession on the propeller and not the rotor, then it should be fair to assume that what the wikipedia text suggests is actually true, despite lacking a citation to back it up - logic is on its side

 

Helicopter rotor blades are articulated. Propeller blades aren't.

Edited March 16, 2017 by AndyJWest

[Farky]

why/how is this different from a helicopter rotor? how come when you have a heli rotor under asymmetric forces the results are affected by precession, but when you apply the exact same concept to a propeller, then they are not?

 

Because P-factor isn't result of a force applied on "spinning disc", therefore there is no gyroscopic preccession. It is a result of the tilted relative wind effecting propeller, blades aren't forced to anything. Higher lift (thrust) on down going blade is the "unforced" consequence. Rotor ("spinning disc") on helicopter is under asymmetric forces, because you are forcing blade on one side produce more lift than blade on the other side. Does that make sense to you?

 

This is for me unfortunately pretty difficult to explain in details even in my native language, let alone with my really poor English.

[unreasonable]

@Farky - I agree this is difficult, but your written English is a lot better than that of most native speakers....

 

"Moach's Conundrum":  I am not convinced that your answer actually solves it. So here is me thinking aloud - certainly not claiming any authoritative knowledge!

 

Thinking through the two case:

 

1) Helicopter. The sides of the disc have the same orientation to the relative wind but are moving at different speeds relative to the wind as soon as the helicopter moves forwards. If the AoA on each side were the same, a helicopter could hover, but when moving 

forwards the speed difference would cause a lift difference leading to asymmetric lift. So the cyclic mechanically alters the AoA of each blade through it's rotation so that the AoA on the retreating side is increased - and decreased on the advancing side - to 

equalize the lift.

 

When you go too fast, the AoA of the retreating blades has to be increased so far to maintain lift that they start to stall. Result - involuntary asymmetrical lift.

 

This results in both a roll towards the retreating (low lift) side and a pitch up, explained in the sources by precessional movement of the net asymmetrical force.

 

2) Plane - AoA of all the blades is fixed relative to the plane at a given time, but not relative to the wind if the aircraft is pitched. By pitched here I mean held at a constant pitch position. The downwards side has a higher effective AoA and so produces more lift, 

the speed relative to the wind being (almost) identical.

 

On the analogy of the helicopter case, this would produce both the yaw as described - equivalent to the helicopter roll - and a pitch through precessional force.

 

This is clearly a different force from the pitching up yaw demonstrated in Moach's track. That happens where you exert a force on the prop disc by using the elevator. Here the prop disc is exerting a force through asymmetric lift. 

 

The objection - "Because P-factor isn't result of a force applied on "spinning disc", therefore there is no gyroscopic preccession." - seems wrong to me. I would think that the prop disc is exerting a force on the airmass, so it must have an equal an opposite force exerted on it.

 

Hence Moach's conundrum.

 

Another way of looking at it is that the force on the prop disc is not the only thing to worry about - the asymmetric lift means - by definition - that there is asymmetric acceleration and pressure in the airmass behind the prop disc, hence asymmetric force on 

the fuselage and tail. So is the yawing force acting on the prop itself actually yawing the plane? Or is it more the asymmetric force of the air mass acting on the fuselage and tail - obviously with a much larger moment. I mean if you wanted to yaw a plane would you push backwards one side of the prop disc or push sideways on the tail?

 

Perhaps the force pushing on the prop is precessionally turned 90 degrees - but due to it's very small moment the effect is hardly noticed in a plane, and the yaw is induced by a force on the tail. In contrast in a helicopter the moment is large and probably the asymmetry in lift is too: the total lift force generated by a helicopter rotor has to carry the weight of the craft, not so for a aeroplane.

 

So perhaps it is not the force on the prop as such which yaws the plane, which is the observed behaviour, but the secondary effect of the air mass?

Edited March 17, 2017 by unreasonable

[19//Moach]

Perhaps the force pushing on the prop is precessionally turned 90 degrees - but due to it's very small moment the effect is hardly noticed in a plane, and the yaw is induced by a force on the tail. In contrast in a helicopter the moment is large and probably the asymmetry in lift is too: the total lift force generated by a helicopter rotor has to carry the weight of the craft, not so for a aeroplane.

 

So perhaps it is not the force on the prop as such which yaws the plane, which is the observed behaviour, but the secondary effect of the air mass?

 

 

 

as far as I have read, and managed to wrap my brains around, this is what's probably happening

 

the few and far between sources which will even acknowledge this aspect of the effect, seem more or less in accordance that the largest part of yaw tendencies observed on propeller planes is really caused by the spiral slipstream, plus torque and basically everything else that goes with it - and P-Factor is (according to these sources) then the smallest of contributors to the overall forces encountered, to such extent that it is often negligible in practice

 

which I suppose just might mean that the pitch up is there, indeed - however, it is fairly minimal, and the other forces at play are generally so much larger that it goes unnoticed...

 

 

I did read a few bits here and there that had mentioned it perhaps becoming relevant on aerobatics (whereas most literature about P-Factor seems to be limited to what happens during takeoff) - if anyone can find a good study on propeller forces acting on high-performance stunt aircraft, this person should win the Medal Of Awesome* 

 

*may or not be an imaginary award

 

 

this is indeed a tricky subject... how much it would affect the game is a whole other can of worms - warbirds are kinda like stunt planes, in a way... with more guns, usually - this could turn out to be anything from "completely trivial and inconsequent" to "kind of a big deal" in how it affects handling in the sim

 

I really can't tell with the information we have at hand so far

[Farky]

At the end of the day, it really doesn't matter how it works. P-factor is left turn tendency (in airplane with clockwise turning propeller of course), nobody in aviation will tell you something else, including manuals. Theory that P-factor is affected by gyroscopic precession is either revolutionary and in the past 100 years no one noticed that P-factor actually produce a pitch and not the yaw, or this theory is wrong.