Warhawk

[Ivan]

The first screenshot shows the main gear struts bleeding through the flaps. Note that the same thing is happening from the far side strut as well. A workable solution in the original model was to add flap parts to the landing gear but as insignia displaying up and aft. When the flaps were reshaped, the insignia flipped to facing forward and down which made them useless. When a couple extra parts were added in other areas and the insignia direction was flipped, the texturing here completely fell apart. I ended up removing the pieces because with the texturing incorrect, it was worse than the bleed it was trying to cover up.

The second screenshot shows a couple of the reasons for the extra parts count. The spinner on a P-40 doesn't really come to a sharp point. The end is spherical with a 1.5 inch radius. Adding one more section to the spinner structure added 12 polygons. Building the wing guns as parts rather than as texture added one more part for each gun and one for glue to attach each one to the wing.

The third screenshot shows a notch or cutout at the trailing edge of the flaps. The flap on each side is made up of 4 separate parts. To have a cutout like this on a textured piece generally requires at least 3 parts. The earlier Warhawk had the inboard edge of the cutout aligned with the join between wing and fillet.
The wing / fillet join needed to be moved slightly 0.15 foot inboard to cure another bleed and required that an additional part be added to the flap on each side along with the glue part to locate it.

I don't know if I found all the issues yet....
- Ivan.

[Ivan]

The front of the landing gear fairings got a bit of a rework to cure a small sparkly where the panels did not overlap.
I finished the SCASMing this morning and adjusted the collision bubble. I did some minor texture tweaks this afternoon also. It seems like even some of the references I am using do not agree with photographs. The photographs are guaranteed accurate, but don't show enough detail to get all the major panel lines.

Sometimes for animation tests, it helps to have a slow computer so that the transitions don't just flash by.

This afternoon, my 10 year old nephew took it for a fighter intercept mission. I believe he was chasing B-17s. The sim remains paused on one of the game machines in the living room.

- Ivan.

[Ivan]

The military rating for the V-1710-39 was 1150 HP at Sea Level with a Manifold Pressure of 45.5 inches Mercury.
In the last releases of this aeroplane, I had not included any War Emergency Power but since then, I have found a manual that does give a WEP rating. In addition, there is the document from Allison which permits up to 60 inches Mercury on the -39 engine.

The aircraft manual gives the WEP rating as 1470 HP at Sea Level with MP of 56 inches Mercury. In preparation for tweaking, I first added the WEP change to the AIR file and bumped the critical altitude (I don't this field does anything) down to 12,500 ft. Since it is pretty much impossible to actually test at Sea Level, I run at a minimum altitude of 500 ft for consistency. Generally this means that the engine power will be around 5 HP or so above the SL rating.
I was a bit surprised when the first test showed 1487 HP which is close enough for my purposes. All that remains is checking the power at each altitude and conducting a bunch of performance tests.

Sometimes we get lucky.
- Ivan.

[Ivan]

Although the latest version of the P-40E is more dimensionally correct, there was something about the look that was not as good as the original.
I believe that tonight I fixed one of the issues.

It isn't easy to see and depends a bit on the angle of the photograph, but there is a very slight rounding of the underside of the Radiator Fairing before forward of the cooling flaps. The open flaps often disguise this because with the flaps even slightly open, the underside looks quite flat.

Fixing this added another 4 Parts to the project
Unfortunately, the textures also needed a slight modification and the Animation needed to be redone.
The worst part is that this aeroplane needs another trip through the rather tedious SCASM process.

- Ivan.

[Ivan]

There was something about the shape of the Carb Scoop that bothered me.
I believe this version is a bit closer to the actual shape. It may not show it, but I spent a bit over two hours building and re building to see if I could get the shape the way I thought it should be.
I don't know how good of a match it is now, but it looks closer to me.

I was wondering why I was having such trouble staying on the runway at take-off.
After a few tries, it finally occurred to me that the issue was mostly my lack of understanding:
We all know that steerable tail wheels (which the P-40 has) do not lock in CFS.
As such, they are VERY sensitive at the start of the Take-Off run.
As soon as the Tail Wheel lifts, the torque swing must be counter-acted by the Rudder.
I was not catching the change quickly enough. When I did catch this, I found that very little Rudder input is required to hold the aeroplane straight while in reality, there should have been barely sufficient control.
Next task is to reduce the control effect at low airspeeds.
Elevator control seems too high as well.

- Ivan.

[aleatorylamp]

Hi Ivan,
It does look rounder!
Rounder is always nicer in AF99 if one has the parts to do it!
Cheers,
Aleatorylamp

[aleatorylamp]

Hi Ivan,
Your Warhawk not only looks really cool with interesting details, but also flies with lots of character - quite a challenge!
Nice work!
Cheers,
Aleatorylamp

[Ivan]

Thanks Aleatorylamp,

The Warhawk is one of my favourites. I am still not quite sure WHY it flies as well as it does, but it does fly as I intended.
The beast does have its peculiarities, so reading the Pilot's Manual (in this case, the ReadMe File and CheckList) is worthwhile.

The appearance isn't greatly different than the original from 2005, but this one has much more accurate dimensions.
Was it worthwhile to spend weeks to rebuild EVERYTHING? Maybe and maybe not, but I like this aeroplane so much I had to do it.

The latest Panel Line addition is a strange one and Smilo is the fellow to blame here!
He commented earlier than he liked "Plain Jane" P-40s without all the gaudy paint.
I didn't quite do that with my U.S. Army paint scheme, but I did think a bit about what would happen if one were to paint the Spinner the same colour as the Fuselage / Cowl. I had never done this but realised that there was no separation line between the Spinner and Cowl. Most P-40s probably left the factory painted that way.

Adding a line was easy, but I also tried a small (0.01 ft) gap which I did not include in the final model in this release because I am not sure I like it. The gap is probably smaller than the real one but is a bit distracting.

The screenshots show what I mean. Imagine if the Spinner and Cowl didn't have a line for separation.

- Ivan.

[aleatorylamp]

Hi Ivan,
About the spinner colour I just looked on the net and found spinners in red, yellow, beige, dark green, light green, brown, red and khakhi together... There were even a couple of colour shots with the plain factory colour that you mention, where the separating line is hardly noticeable at all. They actually all look good. Was the spinner colour then a matter of taste for each individual pilot? As they say in Spanish "for tastes, they made colours" - Para gustos, se hicieron colores!

The shark´s mouth seems to have been one of the most popular nose-arts for this plane - beat the enemy by scaring the **** out of them when they only see you coming! It was even on some sent to China. The tiger one was rather tame, I thought.

By the way, I am mystified as to how you avoided bleedthrough with the 3-pointed star-shaped division in the air intake? I am also mystified as to how you got the plane to fly so temperamentally in the .air file... I wonder if you could possibly indicate where in the .air file one can define the strength of the torque effect that pulls the aircraft to the left.
For example, would it be technically possible to eliminate it for the case of either concentric counter-rotating propellers or a tandem mounted push-pull setup?

Cheers,
Aleatorylamp

[Ivan]

Hello Aleatorylamp,

There were so many different ways the Spinners were painted that if you think of a combination, it was probably done.
The only things I don't believe I have ever seen on a P-40 are Spiral designs and non symmetrical designs that would require an animated Spinner to display properly.
Colonel Robert Scott of the AVG was reported to have had his Spinner repainted between missions to give the Japanese the impression that they were being attacked by more aircraft than there really were.

Regarding the Tiger Shark mouth design, it started with RAF 112 Squadron in the African Campaign. It was covered in Life Magazine and the AVG apparently got the idea from those photographs. The AVG became famous. Fewer people have heard of 112 Squadron.
Some of the Shark mouth designs look "Fierce" as my son calls it. Some just look silly like a smiling drunk. My designs are not copied off of any particular aircraft. I just drew them to represent my impression of what one SHOULD look like.
The same applies to the Hawk 81 P-40 I released a while back.
A Frontal view of the Shark Mouth is particularly interesting to me because I have seen a photograph of a modern P-40 from that angle and my version looks VERY close in my opinion. To get that view right requires not just the paint to match but the nose contours have to be pretty close as well.

Regarding the Intake Dividers, I like the way they turned out. This version actually has textured Radiator faces unlike prior versions. The way this was done was VERY expensive from the AF99 resource standpoint.
IIRC, the Radiator Face, Intake Interior Walls, and the Dividers AND the roof of the Shark Mouth area are all in the Body, Main group with the display priorities set so that from Foreground to Background we have

Dividers
Roof
Interior Walls
Radiator Face

They are all Glued at the Radiator Face
The exterior is in both the Body, Main group AND in the Nose Group
If you try out the Aeroplane without texture files, it will become much more apparent what I actually did.

It is actually a touch more complicated, but the basic idea is that if you are behind the CoG, the display order of the interior parts doesn't matter because all will be hidden by the exterior of the cowl.
If you are ahead of the CoG, the interior parts are displayed in proper order and then covered by the duplicated exterior.

This is one of the goofy things about AF99 in that there is no way to specify something as ALWAYS in the background.
AF99 always expects a viewing plane spec to make the determination and will add one if there isn't one.

Regarding "Temperamental" flight characteristics, This beast is tuned to the best of my abilities to match pilots reports. It wasn't always the nicest handling aircraft especially with all the trim changes, but the combination wasn't bad. If you watch airshow displays, P-40 pilots fly a lot more maneuvers (especially vertical maneuvers) than a typical high speed pass that you see with other warbirds.
The very "Pointable" characteristic is just a side effect and I wish I knew how exactly to quantify and reproduce it. Strange thing is that my FW 190A also has this characteristic but even more so (as it should).

- Ivan.

[aleatorylamp]

Hi Ivan,
Very interesting, thanks, and psychologically effective, no doubt!
Thanks also for the description of your air intake construction.
I´ll investigate in the .air files and look for indications as regards the "Pointable" characteristic. I remember you once asked a rhetoric question on what to do with concentric counter-rotating props.
Cheers,
Aleatorylamp

[Ivan]

Hello Aleatorylamp,

In one of your earlier replies you commented that rounder is better.... Usually it is, but some things are supposed to be angular instead of curved. I found a P-40 model for another simulator that had probably 10 times the polygons of mine but had enough general shape problems that I believe it is a much inferior model. The author should have looked at more photographs and done a bit more than just making a "Tube" and squishing it into a shape resembling parts of a P-40.

I think it is about time to take this P-40E out for some more test flying and experiments.

- Ivan.

[Ivan]

Hubbabubba made a very good suggestion a while back that the next P-40 perhaps should be a Merlin powered version.
Since I have what I believe to be a pretty well dimensioned P-40E, the next one might be a Short Tail Merlin P-40F.

The first step with any of these modifications is to do a little research to note what the changes might be.
The most obvious change is the removal of the Carb Scoop, but there are many more changes than that.

This again is where we run into the issue of contradictory data from different sources. If they contradict, which do you trust?

The P-40E that was just completed was built to an overall length of
31.73 Feet or 31 feet 8.75 inches. This does not agree with William Wylam's drawings, but was listed in a couple technical manuals and also in "America's Hundred Thousand" by Francis Dean.

The P-40F Pilots manual lists the
Early P-40F at 31 feet 7 23/32 inches
Late P-40F and all P-40L at 33 feet 3 23/32 inches.
The P-40F Service Manual lists the length of
Early P-40F at 31 feet 7 3/4 inches which is close enough to be considered identical.

So far so good. Multiple sources state that the Extension to the Tail was 20 inches and this is an exact match.

Now comes the fun part:
The P-40N Erection and Maintenance Manual states that the overall length of the P-40N was 33 feet 3.7 inches
Other Long Tail Allison P-40s are also listed as 33 feet 3 23/32 inches or 33 feet 4 inches.

The problem is this:
If the Tail Extension on a Merlin P-40 iw 20 inches, then does it make sense than the Tail Extensions on Allison P-40s is only 19 inches?

Or is the nose section on later P-40s shorter which is not supported by photographs. If anything, the later Allision P-40s had LONGER noses..... Now if the Nose is longer, something else must be shorter, but where?

Hmmm.... Makes you wonder.
- Ivan.

[Ivan]

.....And just when you thought it was safe, I just found that one of my reference drawings probably had a slight error.
The actual drawing wasn't bad. The annotation on it was incorrect.

The drawing listed the P-40E Thrust line to be 3.75 inches above the FRL
The newer drawing found in looking for data for the P-40F states that it should be 3.078 inch.
The really ironic thing is that this is pretty much the difference between the current model and the prior model so I am putting things back the way they were before.... Sheesh!

The change would not really be visible (about 0.06 foot difference) but I have to do it anyway for the P-40F.....
Corrections in the quest for perfection. It never ends....

- Ivan.

[aleatorylamp]

Hi Ivan,
Yes, it is annoying when one´s annotations on a blueprint have a glitch, but small ones often go unnoticed on the visual model.
What causes more of a problem though, and that´s when you have to "eyeball" it, is when factory blueprints are not available. I have often found the drawings and sketches available for any given model to have some differences, not only depending on the author if there are several sketches available, but differences with photos of the planes.
These differences, can mostly only be corrected by comparing the photos with screenshots of the model - if you get angles and distances right. I have sometimes had to correct rudder height, nose or chin curves, and even elevator widths and lengths this way. Of course by how far the result is accurate with reality is impossible to tell and it is only the general impression I can go by.
But I can understand your strive to get it perfect! I get similar sensations when I am building something!
The best of luck!
Cheers,
Aleatorylamp

[Ivan]

In looking over the Drawings and the mismatch, what I see now is that the actual change to make a correction to the Thrust Line from 3.75 inch above the FRL to 3.078 inch above the FRL is actually only 0.04 feet.
It is a little less because of the rounding errors necessary to represent items listed in Thousandths of an Inch to the AF99 resolution which is only 0.01 Foot or 0.12 Inch.

The next step is to actually make the correction and fix everything else this will break.
It doesn't look like much of a change in the screenshot, does it?
I am almost tempted not to do it except my eyeball has been telling me since I built this version that the nose was a touch too high.

The quest for "Perfect" dimensions never ends!
- Ivan.

[aleatorylamp]

Hi Ivan,
Usually the pixels on the drawings that I have to work on for old models are thicker than the differences in your blueprint! Quite honestly, if my humble opinion serves, such small differences for me would qualify as negligible, and I really woudn´t worry.
Cheers,
Aleatorylamp

[Ivan]

Hello Aleatorylamp,

I know the feeling of working with very imprecise drawings.
That is how my first models started and even more recently, some of the models were created from much less than perfect drawings.
The issue here is that I KNOW the correct dimension here and this discrepancy is the largest one that I know about thus far.
The Carb Scoop that I most recently worked on was a result of this. I could not get the shape right with trying to keep a smooth curve and keeping within the dimensions and now I know WHY I had that issue.

Not knowing is a good excuse, but here I DO know. It also appears that ALL of the P-40s had the same Thrust Line so correcting it once here corrects it for all the descendants as well. It may take a while to get there though....

- Ivan.

[Ivan]

This screenshot shows a slightly revised Template for the new Parts locations and moving some of the actual pieces of the aeroplane.
Shifting the Thrust Line down 0.05 feet appears to be the best solution.

There are four basic areas to address:
1. The Carburetor Scoop (Probably the best reason to shift so many Parts)
2. The Contour Line of the entire Fuselage back past the Cockpit to ensure there are no strange bends
3. The actual Cowl Panels.
4. The Radiator Intake and pieces on the lower side of the Cowl.

This looks quite tedious, but hopefully this will not take too long.

- Ivan.

[aleatorylamp]

Hi Ivan,
An aircraft with such an appealing design as this one is definitely worth the effort to get as exact as possible if within the capability of the modeller and the limits of his patience!
Cheers,
Aleatorylamp

[Ivan]

Hello Aleatorylamp,

I don't really know if it is worthwhile, but it is what I am doing. I need a break from the Gauge Project because I am pretty much stuck. I have changed the code to what I understand it needs to be and yet I am not getting the results I want. I will continue at some point, but several days of zero results tells me I should change subjects.

Here is a comparison of the revised nose with a set of factory drawings.
The match isn't too bad at this point but there are a couple changes that are pretty obvious.

1. The Cowling Flaps are way too short in my model
2. The contour of the Carb Scoop is quite a bit off. Mine has too much of an arch.
The opening on mine agrees better with photographs than the factory drawing does though.
3. The Front of the Radiator needs to be significantly deeper.
4. The Rear of the Radiator has a gradual sweep upward that my model does not.

There are many more small mismatches, but the major components line up pretty well with the rest of the drawing.
A very interesting issue here is that my model lines up much better with these drawings han they do with photographs of actual aeroplanes! It also looks much better in the simulator than here in these screenshots.

- Ivan.

[aleatorylamp]

Hello Ivan,

There is something that I have recently been asking myself when it comes to factory drawings and actual photos of planes: To what extent could changes be introduced into a given production batch, differing from the factory drawings of the initial design? These changes would perhaps only have drawings, if any, reflected on supplementary subsections to the main ones.

With the different photos available of any given model I sometimes have the uncanny sensation that I am being tricked by optical illusions, ...or are the small differences from one unit to another in effect real?

Moreover, to what degree would a factory engineer be capable of simply adjusting certain panels on a specific unit, to fit over some alternative part being fitted that was different just because of a logistics problem, especially in the wartime production rhythm?

So then, what are we doing as virtual modellers? Producing models as per factory specification drawings, or producing the planes that actually flew? ... or are we just splitting hairs?

I would muse that it is something in-between: We can apply our own "magic", to use your expression, so that suddenly, Presto!: Our models fit both concepts!

Cheers,
Aleatorylamp

[Ivan]

My goal as usual is to produce "My Own Impression" of the aeroplane in question.
When possible, use all resources available but NOTHING is infallible other than a photograph but those are subject to a lot of factors and to interpretation.

My latest issue is that I am hitting a place where the drawing and a couple specified dimensions do not match. Which should I believe? I am also getting small inconsistencies depending on where I get the wire frame image from. The one from DPED and the one from AF99 seem to differ slightly in proportions.

Again, it will be a best judgment thing. Flight models have always been a best judgment thing for me.

- van.

[aleatorylamp]

Hi Ivan,
I´ve just checked the difference you mentioned in the Dped and AF99 blueprints, and I was quite shocked when I saw the magnitude of this difference!
Just to show, I made a comparative joint screenshot, and compared it to Paul Matt´s drawings. It appears that the AF99 blueprint is quite a bit too long, and that the Dped one matches the drawings. I hate to think about what connotations this could have in all these years...
Good luck!
Aleatorylamp

[aleatorylamp]

Hi Ivan,
I just thought I´d investigate the Aircraft Animator blueprint too, and it turns out that dimensions coincide 100% with Dped blueprints and Paul Matt´s drawings in the case of the AT-9.
Cheers,
Aleatorylamp