Flaws with P-47 Flight Model

[RugerDog]

I've noticed the following two issues with the P-47/Accusim package:

1) When reducing propeller RPM only with the Curtis electric propeller model, manifold pressure is somehow reduced at the same time as if the prop lever and throttle were interconnected (incorrect behavior)? However, it is possible to increase or decrease manifold pressure with the throttle whilst the RPM stays constant (correct behavior). The correct procedure for instance in setting climb power after takeoff would be to reduce manifold pressure (via the throttle) from takeoff power (52") to climb power (42"), and then reduce propeller RPM to 2550. After having set the manifold pressure to 42" via the throttle, reducing propeller RPM to climb setting will also instantly reduce manifold pressure at the same time (incorrect behavior). The only way to set climb power like this is to reduce manifold pressure, reduce propeller RPM (which wrongly also reduces manifold pressure!), then reset manifold pressure again. Guys, this can't be right?

2) The P-47 gear/flaps drag modeling is way, way, WAY off. Sorry but you guys kinda goofed on this big time. Extending flaps has the effect of lowering the nose on the real P-47, but on the Accusim version the flaps are acting like drag chutes and create far too much drag. The nose angle with deployed flaps on the Accusim P-47 actually seems higher than in a clean configuration. The gear is also causing grossly exaggerated drag as well, and the combination of these two flaws requires enormous power settings to be maintained during final, coupled with unrealistic descent rates and glide angles. I'm amazed this has gotten by everyone as it is a really big mistake on the part of A2A, wow.

I hope you guys can address some of this in a future patch.

[CodyValkyrie]

1) Reducing RPM should not be reducing MP. Is it possible that during a climb, you are reducing RPM and noticing the manifold pressure lower because of the higher altitude gained? This was never mentioned in the Beta and I never myself experienced this behavior. This should not be acting this way regardless, and I am at a loss of why you are seeing this result.

2) The P-47 had intensive drag from both the landing gear and the flaps. As noted in the manual, it is not suggested to lower full flaps until on final, and to not make any signifigant turns. If you make a shallow turning approach to maximize visibility, with good timing she is a kitten to land. Manifold pressures should be about 30 inches or so, with 2550 RPM. Using these settings, if you watch what you are doing and place full flaps before you hit the fence, she should land pretty nicely. Watch a lot of P-47 videos and you will often see guys bouncing her. A good pilot can nail it pretty good though.

Remember, in a P-47, you have to fly her into the ground. This was the quickest diver out there, and that large radial in the nose didn't help. Having good power when coming in for landing is a necessity. You can't chop the throttle and bring her in. She glides as well as a brick.

I feel fairly confident we got these right, however I'm always willing to be wrong and am looking forward to hearing from some of the other A2A guys who specialized in these sections of the flight model.

I'll be the first to admit I am on a less technical level than some of my airplane brethren here.

[CAPFlyer]

Mr. "RugerDog",

Hi... glad you joined the forum. Do you not find it quite presumtuous of you to make such a post with very strong statements without first introducing yourself, much less giving us some idea of your background and how you can so authoritatively make those statements?

I do not wish to be harsh, however there are far too many people out there who claim to "know" how an aircraft flies without even actually having seen one in real life. Before one can be taken totally seriously, it would be helpful if they explain why their information is important, where it came from, and why it is contrary to multiple published technical documents and manuals on the subject, including documents prepared by pilots, for pilots, like the handling notes.

Also, I will make a statement on your first point. When not equipped with an automatic turbocharger, this is something that happens to an engine with a constant speed propeller, especially when operating below the critical altitude for the turbocharger. I remember during climb with the R2800 on a CV-240 how the first and second reductions of power required a reset of power once RPM was set. By doing this, you are slowing the engine sufficiently to change the required venturi opening to maintain pressure in the manifold, resulting in a drop in MP with a drop of RPM. When an engine is equipped with an automatic turbocharger (i.e. one with a waste gate) such as on the T182RG, the waste gate can compensate for that reduction in pressure by closing slightly and increase the pressure of the air charge going to the engine. On a supercharged engine with a manual turbocharger like on the P-47 or on a supercharger-only engine like on the CV-240, there is no way to account for the load change except to increase the throttle, something that must be done manually.

The primary purpose of setting MP/throttle first and then RPM when decreasing power is to prevent the propeller from being moved against it's upper stop so that the prop is always within its governed range during operation. The same is true when increasing power. If you increase MP/throttle without first increasing RPM, you place the prop against a stop and it becomes ungoverned, something you don't want.

[lonewulf47]

I've noticed the following two issues with the P-47/Accusim package:

1) When reducing propeller RPM only with the Curtis electric propeller model, manifold pressure is somehow reduced at the same time as if the prop lever and throttle were interconnected (incorrect behavior)? However, it is possible to increase or decrease manifold pressure with the throttle whilst the RPM stays constant (correct behavior). The correct procedure for instance in setting climb power after takeoff would be to reduce manifold pressure (via the throttle) from takeoff power (52") to climb power (42"), and then reduce propeller RPM to 2550. After having set the manifold pressure to 42" via the throttle, reducing propeller RPM to climb setting will also instantly reduce manifold pressure at the same time (incorrect behavior). The only way to set climb power like this is to reduce manifold pressure, reduce propeller RPM (which wrongly also reduces manifold pressure!), then reset manifold pressure again. Guys, this can't be right?

I'm going to answer this one as for the rest I'm way too unexperienced with my 15'000 (real) flight hours in 70 different A/C types . I'm sure you read about the setup of the Razorback's engine. If not I will give you a brief summary. the R-2800 has - like most of its colleagues - a fixed geared supercharger and an exhaust driven turbocharger. During take-off you will advance the throttles to reach 52 inches of manifold pressure at the nominal 2700 RPM - all with the use of the supercharger only. You neither don't nor should use the turbocharger upon take-off. There are more important things to do at this stage of the flight. While reducing to climb power you reduce the MP to 42 inches and still have 2700 RPM. Now guess what happens to the fixed geared supercharger . You reduce the RPM to 2550 and hence also reduce the superchargers speed. Now give me ONE reason why MP should NOT drop ....? Do not compare this to modern (eg. GTSIO 520 or similar) waste gate controlled turbos. Those work in a completely different way. But I'm sure you know that....

Oskar

[RugerDog]

I'm going to stand by both of my original observations. It is truly amazing how people on here suddenly become mechanical and aerodynamic engineers in the forum when in reality they themselves don't really know what they're talking about. Insurance salesman and IT guys will boast of having "1000's of hours" of real-world time in various aircraft with the same powerplant, etc, lol, when the reality is that they have probably never even sat in a real aircraft other than a airliner as a passenger. Boy, doesn't stop them from being all-knowledgeable though, wow.

The way manifold pressure is being reducing by prop rpm changes on the Accusim P-47 is basically synchronized with throttle behavior, i. e, movement of throttle or propeller lever by the exact same degree will yield the exact same reduction or increase in manifold pressure change. This is not right.

The drag modeling of flaps and gear is completely exaggerated; if the real P-47 had that much drag on during landings, it would be a flying deathtrap. Lowering flaps SHOULD HAVE the effect of generating lift as well as drag, and at the same indicated airspeed would have the effect of lowering the nose of the aircraft on landing approach. The Accusim P-47 essentially deploys drag chutes when flaps are deployed, resulting in nose high attitude and very high power requirements on the glidepath. In reality the P-47 can nicely fly a normal approach with moderate power and very decent over-the-nose visibillity with full flaps.

The Accusim P-47 requires a dive-bombing approach and excessive thrust to land safely with full flaps.
This is extremely unrealistic.

What a shame, I had high hopes for this package. Perhaps if A2A were to collaborate with an expert flight dynamics programmer they could make substantial improvements. There is always hope!

The systems modeling and sound package are indeed truly amazing though.

[ICDP]

RugerDog,

When I read your first post all I saw was someone giving his opinion (right or wrong) without showing any evidence to back up your claims. When you received responses from experienced pilots giving valid reasons for your alleged problems you responded with insults. Essentially you belittle the input of other (highly experienced) people and come across as condescending. This is really not the way to get people to take you seriously.

You obviously created your account to present your opinion on the P-47D FM. The least you could have done was state your qualifications and experience along with your sources so we could properly evaluate what you had to say. After all you could be entirely correct in your claims but the way you have presented them and your second post makes you appear to be a very obnoxious person. I suggest you present your sources for your FM issues and try a less confrontational approach to presenting them and debating them.

I could be entirely wrong and you could be a really nice guy but your first two posts did not help to endear you (or your opinion) to the people who read your post.

[Hog Driver]

Very well said ICDP.

RugerDog If you have any information, research or literature that can support your claims, by all means please feel free to share them here or with A2A, the only thing that could happen is that the aircraft will become better.

My thought is, if it can get better then let's make it better.

[RugerDog]

The only way to produce a realistic flight dynamics model in a simulation is to have a real-world qualified pilot on the aircraft type fly the simulation itself and give the flight dynamics programmers feedback. A2A obviously did not do this and instead decided to make a game arcade type flight model based on their own research. Computer programmers frequently try to build code like this that performs to published specifications instead of evaluating the actual flight results in the simulation.

In other words, A2A built a mathmatical 'formula' P-47 flight model without having qualified real pilots evaluate the results. The fact that the 'experienced and knowledgeable', self-proclaimed 'pilots' who have responded to this thread have not themselves picked up on the blatant flaws with the drag modeling tells me right away about their actual aircraft experience. Sorry, but I called 'em like I see 'em.

Finally, let's visit the supplied A2A manual on P-47 Landing technique and I quote,

"10.After your flaps are down and you roll out of the turn onto the landing (approach) leg, your speed should be about 125-135 mph IAS.
Don't keep so much power on that you'll be making a power approach. However, keep enough power on to keep your engine clean."

That last line, "don't keep so much power on that you'll be making a power approach....." speaks volumes. Gentleman, I challenge you to try this yourselves with the Accusim P-47, make a non 'power' approach as described in the A2A manual and see what happens. The fact is you'll make a nice hole in the ground without considerable power setting on approach and landing with the Accusim P-47. Yet, why does the A2A manual suggest not making this so-called kind of power approach? HINT: Because A2A transcribed the actual landing technique in the supplied manual from a real-world P-47 flight manual. As we can now see, using real-world technique on the Accusim P-47 won't quite work out right.

Why? Because the Accusim P-47 drag modeling just about requires rocket boosters on approach to stay on glidepath with gear and flaps down. The REAL P-47 has a heck of a lot of drag in landing configuration, certainly, but the ACCUSIM version has exaggerated this by leaps and bounds.

Again though, don't take my word for it as I obviously don't know what I'm talking about. Go try to land the Accusim P-47 just like it describes in the A2A manual. Just be sure the fire equipment is in place and your life insurance is paid up beforehand. I rest my case.

[CodyValkyrie]

I would like to put a word in on what I feel are some pretty interesting comments.

First and foremost, the flight models are, were and always have been built by PILOTS. Many on the A2A team have flown, owned planes and/or still retain their current certifications. I'll admit that some of us have only been "around" said planes. In my case, I have flown planes. Scott himself owned a plane.

We have a very lengthy process for the creation of our aircraft. Let me explain some of those particular parts.

Gathering of materials
-As many flight manuals as we can get our hands on. For this project, we had no less than 4 different manuals we referenced
-Flight reports and maintenance records where available
-Technical specifications, ratings etc for engines, subsystems, drag ratios, range charts, flight profile charts, etc.

Hands on experience
Whenever possible, and especially in the case of the P-47, we try to get hands on with the planes in question. This includes often attaching cameras to real planes, taking ride along flights, getting into the cockpits and snapping HUNDREDS of pictures, getting the feel of the individual levers, instruments, etc. I can validate this with the 3.8GB of pictures I took just a couple of weeks ago for an A2A project where I spent over 6 hours with a set of planes, taking well over 500+ pictures.

If possible, we always try to take planes up or get "ride alongs." Naturally, this isn't such an easy proposition.

In this process we do all our recordings from scratch, including startups, positional sounds, etc.

Pilot interviews
This is one of the most criticall phases to developing an A2A plane. The charts and specs are always important, but hearing it directly from the REAL men who flew these machines is what brings it all together. We spend hours at a time on the phone, in person and otherwise asking questions, confirming reports, and generally getting to know all the little ins and outs of a particular aircraft.

If we have pilots who currently fly said machines, have flown them or have a lot of experience around them, we attempt to get them involved in the beta testing. We often do independent testing with these folks in order to relieve the pressures of beta testing and to get direct results.

Testing, testing and... more testing
Many of the beta testers in our group are extremely competent pilots, mechanics or other professionals who also happen to be of the mentality of rivet counters. In many cases, our testers have particular experience with specific airframes, engines or otherwise. All testers are generally screened, and often interviewed. Because of this we retain a VERY specialized group of individuals who are not only experts, but people who are willing to criticize us where we need to be criticized. The process is rather excruciating, but we feel that unless people are honest, we cannot produce what I feel are excellent products.

Final tests
Before anything leaves the door, everything is cross checked by real world pilots, often those with particular airframe time. In the case of the 377 for example, we had a C-97 captain who certified and "looked over" the whole process to ensure we provided an experience akin to that he himself experienced flying the big plane.

Through all this process, you will find that not only do our models exhibit quality flight models, but they perform on levels that often times other developers are unable to themselves achieve. You will also find that our models where applicable slip, spin and snap roll. Yank on the Spitfire and hit full rudder sometime and you will see what I mean

Why am I defending my position? I want to be clear so everyone understands the amount of dedication that goes into our aircraft creation.

In the case of the P-47, such as landing, our documented and verbal sources indicated that the P-47 had significant drag with flaps and gear fully locked and full dirty. We had documented sources with drag coefficients that we referenced as well during the process. For example, as in the real P-47, you will find that if you close the intercooler flaps fully, it will actually create MORE DRAG than if they are flown in their neutral position. This was found from service reports and drag charts provided by Republic on those particular surfaces.

On a side note, regarding the landings. As I said, keep roughly 25-30 inches on approach. Don't lower full flaps until you have completely settled on final. Keep airspeed 125-135. Don't chop the throttle! If you want a wheel landing, come in with a little more power.

Hundreds of users of the A2A P-47 have been landing the plane successfully. If you need tips, we are willing to help.

With reference to keeping the engine clean, it is a reference to shock cooling, spark plug fouling and keeping the engine warm enough so the carburetor doesn't get too cold. Could you imagine an icing condition in a P-47 on a landing? That will create a nasty crater! The idea is not to come in with no power, but just enough to keep the approach established and controlled. As I said, fly the plane into the ground. Radial engine aircraft create an immense amount of drag, which is inherent by their absolutely horrid drag producing engines. It is almost like placing a flat wall on the front of your aircraft, attaching a spinner to it and trying to fly. When power is reduced quickly, or physical factors are introduced such as the landing gear and flaps, speed will taper off quickly. The goal is not to come in full dirty with the engine wide open on a power approach, but only to provide enough power to keep her airborne and controlled.

25-30 inches MP is a far cry from a full 52 inches MP, or as the 2800 can produce, up past 70 inches MP. Even in a real plane, unless you are under emergency or training conditions, coming in with power chopped is a VERY BAD and unsafe practice. At least with most aircraft. You always want to have the ability to bail yourself our in a bad situation, and you always want to be in control. Think WELL ahead of the plane. Any good pilot will tell you that. Don't chase the instruments. Again, any good pilot would say that as well. These factors are only multiplied in a plane as big as a P-47.

That said when there is a failure on takeoff, the recommendation directly from Republic was to not attempt a go around due to the high drag. They actually recommended flying straight and ditching on runway heading if necessary, rather than attempting a dangerous return to base because of the extremely high power that needs to be applied just to keep her in the air.

I would like to leave you with a question... Is it possible that you are expecting the plane to behave a certain way that you feel it should? If so, are you willing to consider that often times people are left with misconceptions that linger far beyond the realities of the planes themselves? The best example I can give is the P-40. Growing up I expected, regardless of the looks, that the plane was a dog. Sure, at high altitude it truly was. The reality however is that the P-40 was an extremely agile fighter aircraft that even the best enemy pilots gave a certain amount of consideration when going toe to toe with one.

EDIT;
We are proud of how far we go and so we naturally like to show it, here are some of our youtube videos that show just how close we get,
http://www.youtube.com/watch?v=JmlnGtuylKA

http://www.youtube.com/watch?v=mx6l7GMc ... annel_page

http://www.youtube.com/watch?v=6xeuaBsq ... re=channel

[RugerDog]

Cody, thanks for this reply. I appreciate the amount of research you guys have done, but the end result just isn't right.

Watch how one of your own developers lands the P-47 here:

http://www.youtube.com/watch?v=Rag1q9OG ... PL&index=4

Scott delays lowering the landing gear until the very last minute because he admits the drag is enormous, then after lowering gear on short final guns the manifold pressure up (he has to), and dives hard for the runway. Had Scott lowered the gear and partially lowered flaps further out on approach, he would have had to either maintained about a 6 degree glideslope or shoved the throttle up to about 45-50 inches to maintain a 3 degree glideslope with gear and flaps partially extended. Were he then to select full flaps on short final, he'd be around 50 inches of manifold pressure to stay on glideslope.

Basically one must either dive at the runway or use an extreme amount of power to fly a normal glidepath angle in the P-47. Frankly, Scott's technique of delaying landing gear extension until almost over the fence works out pretty good with the Accusim model. Realistic? Nah, but it's just a game in the end and it's not right for me to expect too much from the sim.

As for the manifold pressure/RPM issue I initially posted about, you mentioned that neither you nor any of the beta testers had noticed this happening before. The 'experts' who joined in on this thread said it was supposed to do that, however. Ummmm, what? Do you see my point?

I will say this though: the systems modeling, Accusim, virtual cockpit, and sound package is astounding. If only you could have a real P-47 pilot work with the simulator and gain his or her feedback. Interviewing and researching are wonderful things, but translating data into computer code without some kind of hands on validation of fidelity will never work properly. The mechanics with the Accusim P-47 are nothing short of brilliant, but the flight model itself casts a shadow over that accomplishment unfortunately. It can be fixed though and I hope A2A will eventually come to the same conclusion.

Thanks again for your reply.

[ICDP]

When I am landing the P-47 I usually have around 20"-25" of manifold pressure to maintain speed and glideslope. The glideslope does not seem excessive though the runway is not usually visible. I don't know if this is excessive power required to maintain speed and glideslope but it is not unique to the P-47. For example I find that the CH Fw190A requires even more power to maintain speed and glideslope.

For reference I have read the following from a flight report on the P-47N (taken from the WWII Aircraft performance site) http://www.wwiiaircraftperformance.org/ ... 88406.html

" J. Approach, Landing and Wave-off.

Approach is made with 50% rated power, full flaps, 0° rudder and aileron trim and approximately 5° nose up elevator trim. Recommended IAS for approach is 140 mph with a minimum of 120 mph. There is no ground looping tendencies due to the locked tail wheel. Going around is not dangerous but flaps should not be raised below 400 ft. and then gradually."

It is clear from this that a power off approach was not recommended in the P-47N. The 50% recommendation seems to match up quite well with the A2A P-47D.

Taken from the same web site, this time referring to the P-47C. http://www.wwiiaircraftperformance.org/ ... rials.html

" i. In a glide with engine idling, engine tends to load up and difficulty is encountered starting up engine again. "

It is clear from this point that idling the engine during a glide was not recommended.

I am not a pilot, but I don't need to be to be able to interpret these two points. It is clear that 50% rated power was required during approach and idling the engine during a glide was not recommended. These points were taken directly from flight test reports relating to the P-47.

[lonewulf47]

I'm going to stand by both of my original observations. It is truly amazing how people on here suddenly become mechanical and aerodynamic engineers in the forum when in reality they themselves don't really know what they're talking about. Insurance salesman and IT guys will boast of having "1000's of hours" of real-world time in various aircraft with the same powerplant, etc, lol, when the reality is that they have probably never even sat in a real aircraft other than a airliner as a passenger. Boy, doesn't stop them from being all-knowledgeable though, wow.

Well, usually I'm not going to discuss anything with anyone on that extremely low level neither do I need to have my flight experience judged by anyone else but me. Anybody else might recognize the difference between your judgement and mine. As I don't have personal flying experience on a P47 (neither do you I presume) I do not judge other peoples work with attributes like 'far too high' and 'completely wrong' or similar just because I admit that I'm not able to judge it thoroughly. But you as the only expert of course are allowed to do so. I heartily agree. You have of course also thousands of flight hours on old fashioned supercharged engines. Sorry to say that I have only a few hundred. So again even on the tech field you are of course the expert. Well then: I'm soo happy to finally see the ultimate expert showing up. So I can easily withdraw from the expert's board and leave my place to you, Mr. KNOW-IT-ALL. Funny enough I've seen experts come and go and only a few finally remaining on an acceptable level. I see from your post cited above that you consider me to be big-mouthed. I personally consider you the greatest expert on the world. But maybe we're both wrong....

Oskar

[Gypsy Baron]

Oskar, you are too kind...

It turns out that I know a pilot that has plenty of time in P-47's and P-39's as well,
flying combat missions over North Africa and Italy during WWII. Unfortunately he has
been MIA in the forum where he and many other flightsim types hang out. Should he
surface in the near future I will definately point him in this direction.

Paul

[CAPFlyer]

"Ruger Dog"

As a way to kinda "bookend" my participation in this discussion, I will give you a name. This man has my upmost respect and has more than enough time in most warbirds to be considered "competent" in how they fly to be able to evaluate various aircraft. He has been helping various developers, including A2A for some time, and his words to me on all of A2A's products (including the P-47) was -

"I will say that [their] research into the products [they] design for the sim community are very well researched and I have found very little in [their] work that I would correct." (BTW, for clarity, I edited the quote with brackets because the original quote was in a discussion about Scott, but he later said the statement was true of the whole team.)

The name of that person - Dudley Henriques. Look him up online. He's out there. His accomplishments are well established. If he says this plane is good (and he does) then it's good enough for me.

[mJolner]

Have'nt felt the need to post in a long while but this discussion caught my interest. As I am sure all Flight Simmers/Pilots are, I too am concerned with the fidelity of a flight model. Even with the limitations of FSX, I have high hopes for as much "reality" as possible or in the least the ability to the fly the ship as documented with given parameters.

A2A/SHockwave have always done their very best to provide as realistic as possible a presentation given the limitations of flight simulator and for that hats off from me. The recent P47 was beyond my expectations and will be a staple for me for much of the forseeable future.

That being said, I must admit that I agree with Ruger Dog.

I will say that I am a pilot with limited hours and was only a guest in the trainee seat of a T6 Texan, the closest I have ever come to flying a Warbird. My only knowledge of the P47 is from the numerous documentation and accounts available. To those with more experience you have my respect and apologies.

The late Jeff Ethel, may he rest in peace, with his wonderful Roaring Glory series provided me with some insight into the world of flying warbirds including a P47N (I think) Tarheel Hal. In that video Jeff clearly states that full flaps at 150 brought the nose down enough to provide a clear view of the runnway even given the Thunderbirds long nose. With the A2As excellent presentation I cannot do this even if I raise the eyepoint reference. Jeff does not mention about his MP setting when landing. In the video Jeff lowered gear well before final but at that time he had RPM set to 2550, he did wait until final to lower full flaps.

Now I am trying desperately not to come across as arrogant or authoritative but in my opinion with all the fault that that implies is that the current A2A P47 FM calculates too much drag in a dirt config, along with too little nose down pitch with full flaps. Since I know nothing really of aerodynamics other than that wings make planes go up, I will atleast say that the latter of my two complaints should atleast be more thoroughly expounded. If maybe Scott or one of the developers can provide more insight as to why the plane behaves this way I am an eager student.

So I would ask, (sorry to hijack Ruger's thread) that some provide a short video demonstrating how a documented normal decent and landing can be achieved with the A2A P47. Can anyone honestly say it is possible. If so, then there is something wrong with my procedures or my installation.