P-39C Airacobra
Contribute Now
Goal amount for this year: 10000 USD, Received: 9,535.00 USD (96%)
Welcome Back, Unregistered     Today's Date:       Current Time: 

Please Support the SOH "2020 Charter Membership Drive"

Accepted Payment Types

Results 1 to 10 of 10

Thread: P-39C Airacobra

  1. #1

    P-39C Airacobra

    This is another Airacobra that is loaded on my Development Machine.
    After releasing the update (Version 0.52) to my rework of Eric Johnson's Airacobra a few days ago, I was looking around to see what other versions of the P-39 I had on the machine and came across this rather interesting package by Indianola.

    It is obviously also based on Eric Johnson's 3D model but is intended to represent the P-39C in an imaginary paint scheme similar to what the prototypes may have worn.

    A little History:
    The P-39C was the first 'production' model of the P-39 Airacobra.
    It carried less armament than later P-39's and carried no armour or self sealing fuel tanks.
    Because of the lack of equipment, the P-39C was determine not to be combat ready and production switched to the more capable but heavier P-39D after only 20 P-39C were built. P-39C were re-designated RP-39C (R for Restricted Use) and sent to training units.

    The P-39C armament differed from the P-39D in the following ways:
    The 37 mm cannon carried only 15 rounds instead of 30.
    Two .30 cal MG were mounted between the .50 cal MG in the cowl.
    Each .30 cal cowl MG was supplied with 300 rounds.
    Normally there was no wing armament but one .30 cal MG could be mounting in each wing .

    Attached Thumbnails Attached Thumbnails P-39C_Airacobra_Guns.jpg   P-39C_Improvements.jpg  

  2. #2
    The P-39C is a quite interesting package.
    It appears to be a conversion of a FS98 flight model with a very rudimentary Damage Profile as part of the CFS conversion.
    As can be seen in the screenshot of the firing of all the armament, there is a motor cannon and four wing machineguns.
    There are no guns on the cowl as there should be on a P-39C.

    The panel is quite attractive but also shows some of the DP problems.
    There are 500 rounds of cannon ammunition and 1000 rounds of gun ammunition.
    The cannon rounds weigh about 2 pounds each, so if the weight per round were correct, that would be 1000 pounds of ammunition for the cannon which actually only carried 15 rounds. (The cannon ammunition in the DP only weighs about 1 ounce per round.)

    The lack of a heavy cowl armament completely hides the nastiest part of the Airacobra and the stability problem that gave the Airacobra its reputation for poor handling.

    The fuel selector appears to be from the stock P51D which means that it has a dead center tank.
    There is no center tank on the Airacobra.

    This appears to be a nice package with some problems in the DP and flight model that do not appear to be very difficult to correct.
    The question is whether it is worth the effort to make these corrections.

    - Ivan.
    Attached Thumbnails Attached Thumbnails P-39C_Indianola.jpg   P-39C_Armament.jpg   P-39C_Panel.jpg  

  3. #3

    About Two Hours Later - P-39CEJ

    One of the very nice things about not significantly changing the shape of Eric Johnson's model is that just about any of the other paint jobs will fit onto my modified model without a lot of work.
    The canopy frame section maps onto a different texture file and that needed some slight editing but it was not difficult.
    The missing textures (that I added to prevent multiple pieces mapping to the same files) came from the D model.

    The DP file was also lifted from the P-39DEJ with the maximum ammunition load for the .50 caliber cowl guns reduced from 270 rounds down to 200 rounds. I do not know if this is accurate, but I have not seen any references list anything more for the P-39C.
    Also, there needed to be some room in the nose for the 600 rounds of .30 caliber ammunition.
    One Wing Gun Station was deleted and the other was modified to become a Nose Gun Station with firing rate of the guns reduced from 1200 RPM to 950 RPM to account for synchronization losses.

    Unfortunately on my first pass through the DP, I neglected to delete the bomb load.
    I do not believe the P-39C was equipped to carry a bomb.

    The Damage allowed for various systems also has not been reduced from that of the P-39DEJ.
    The P-39C had unprotected fuel tanks of 85 Gallons capacity which were replaced by self sealing tanks on the P-39D and capacity was reduced to 60 Gallons each.
    This meant that the fuel tanks on a P-39C were probably very similar in vulnerability to those of the typical unprotected Japanese Zero Fighter.

    Data for updating the AIR file came from Detail & Scale Volume 63:
    Empty Weight 5070 pounds
    Gross Weight 7180 pounds
    Maximum Take Off Weight 7300 pounds

    From these weights, it is clear that the P-39C could not carry any significant external stores.
    Bell also had a tendency to list Gross Weight with only a partial fuel or ammunition load.
    My belief is that the 7300 pound Maximum weight is with full stores and 7180 pounds is not.

    From this maximum weight and deducting all the removable stores and a portion of the engine oil (90.75 pounds),
    the AIR file weight should be around 6065 pounds.

    - Ivan.
    Attached Thumbnails Attached Thumbnails P-39CEJ.jpg  

  4. #4

    Something Removed - Something Added

    The overall appearance of the paint scheme for the P-39 BMC was what attracted me to begin with but there are a few details that are not quite right.

    The First Screenshot shows Blue lines on the carburetor scoop behind the cockpit.
    I believe these lines are on the edges of the textures of the aft fuselage in order to create blue highlights.
    The bluish colour can be seen at very shallow angles. It is a technique I have used in models in the past.
    This also shows why having different objects map very close in the same texture file may not be a good idea.
    If the colour is different between the objects, there may be an unintentional highlight.
    Generally a margin of about 3-4 pixels seems to avoid this problem.

    The problem here is that the Carb Scoop maps to the same texture as the upper aft fuselage and the highlight technique cannot be used there without causing some side effects.

    The other item worth noting in the First Screenshot is the NACA style scoop and vents that are located just behind the propeller spinner. These were features that were added to late model Airacobras and did not exist on the P-39C.

    The Second Screenshot shows some texture changes to the underside of the P-39C.
    The "U.S. Army" was lifted directly from the P-39D and of course the mapping is identical.
    The Exhausts for the central Coolant Radiator and side Oil Coolers has also been added near the trailing edge of the wing.

    This image also shows a problem with the dimensions of the 3D model.
    The Wing Root intakes feed air to the coolers in the belly of the Airacobra.
    This was one of the problems of the aeroplane: There wasn't room for any larger radiators or oil coolers as engine power increased in later versions, but it WAS a very clean and low drag installation with nothing projecting out into the airstream.

    The inboard intakes feed the Coolant Radiator and the outboard intakes feed the Oil Coolers which are directly in line behind them.... About where the Wheel Wells are located in this model.
    The Wheel Wells are located too far inboard in this model. In reality, most of the Wheel Well was located in the detachable outboard Wing sections with just a small section including the inner Gear Doors in the wing center section.

    The Third Screenshot shows the P-39CEJ after a fairly quick SCASM treatment to remove Wing Guns, and Antenna Mast.
    Actually the Antenna Mast was repurposed to cure bleed with the Carb Intake Opening. This was a bleed that I had not noticed when the intake and opening were both dark as on the P-39D.
    The "proper" way to fix this is in AF99. The more interesting way to address the issue is with SCASM edits.

    - Ivan.
    Attached Thumbnails Attached Thumbnails P-39C_TexturesRemoved.jpg   P-39C_TexturesAdded.jpg   P-39CEJ_SCASMEdits.jpg  

  5. #5

    Editing the P-39D as well

    The P-39DEJ also had the same disappearing Carb Intake problem but the solution of repurposing the Antenna to cure the bleed obviously would not work. We need the antenna mast there.
    The quick solution was the SCASM equivalent of adding a Glue Part and duplicating the Intake Opening.

    The First Screenshot shows where the Intake Opening would normally disappear (fixed via SCASM).
    The fuselage colour is changed so the opening is more visible.

    The Second Screenshot shows where the AF99 Glue Part would go to address this problem.
    The Part is larger here so its location is obvious. I should put this into the Project so that the change becomes permanent and is picked up in any further re-compiles.

    - Ivan.
    Attached Thumbnails Attached Thumbnails P-39DEJ_IntakeFix.jpg   NewGlue.jpg  

  6. #6

    DP Updates

    In working on the P-39DEJ, I had been so concerned with flight performance and handling that I had not actually done any sanity checking of the DP files.
    The Weights in the DP were correct as were the Armament Locations so that the Aeroplane balanced properly, but the hitting power of the guns and the location of the Parts in each box were not accurate.
    In looking at changes needed for the P-39C, it became obvious that the correct starting point was to go back and rework the P-39D.
    The 37 mm M4 cannon was a very big hammer. For a start, I made it about the equivalent of the German MK 108 cannon.
    The 37 mm was a much larger shell but the German shell was very thin walled and carried a lot of explosive.
    Either way, a single hit on a fighter sized aircraft is probably enough for a kill.

    Without any armour, the P-39C will be a pretty fragile beast but it is a lot more agile than the P-39D.
    Loops are much easier and stalls are gentler even though stability is just as poor when the stores are expended.
    This should not be a surprise; The P-39C weighs about 600 pounds less than the P-39D.

    - Ivan.

  7. #7

    Hi Ivan

    Interesting read... i like the comment "SCASM equivalent of adding a Glue Part"....well i'd say more like mig wielding... but glue is good too...

    Good to see somebody using or/knows scasm.

    Cheers mav

  8. #8
    Hello Mav,

    Been doing little tweaks on my models with SCASM for years.
    I can't claim to be an expert, but the typical model does get a lot of edits before it is finished.
    The internal POV is changed to match the values in the AIR file,
    Extra pieces are added to kill some of the bleeds from an interior viewpoint,
    Canopy Frame is flipped inside-out and added to the interior view,
    Some animation is adjusted so that it better matches what is supposed to happen such as angles of Flaps and sequencing of Gear Doors.
    (Aircraft Animator lacks flexibility and has a few bugs.)

    I used the term "Glue" for viewing planes because that is how the development package Aircraft Factory 99 refers to them.
    I am actually pretty close to finished on a spreadsheet for calculating the parameters for a "VectorJump" from the vertices of a Glue Part in AF99. I calculated some of that for the "MIG welding" but it didn't adjust the distance because it was not necessary.

    The maths are not that difficult but I am way rusty because I haven't been there in about 35-40 years. I just got back there recently because that is what my Son is now studying in school.

    - Ivan.

  9. #9
    The Spreadsheet for calculating SCASM VectorJump parameters is mostly done but with a couple minor issues.
    One of the first was to recognize that the cross product of used to produce the equation of the viewing plane MUST produce a vector that faces AWAY from the origin.

    As can be seen from the attached Screenshot, the calculated parameters are pretty close, but the distance parameter is just slightly off.
    The distance unit appears to be 1/512 meter, so a fraction of that is actually a very small number but the fact that the numbers do not agree bothers me.
    This screenshot was done with a project that does not involve moving the Center of Rotation.
    For some reason, the parameters are a little further off when the Center of Rotation is altered as it was in this series of Airacobra derived from Eric Johnson's AFX.
    So.... While the spreadsheet works for my own projects, it is only an approximation for these Airacobra which means there is still some tweaking that needs to be done.

    - Ivan.
    Attached Thumbnails Attached Thumbnails GlueCalculation.jpg  

  10. #10
    Or maybe there is NO Tweaking that needs to be done.
    I went back to look at the calculations for adjustments for Center of Rotation.....

    The attached Screenshot shows the differences between the calculated parameters for a VectorJump and the AF99 generated SCASM code. (It was moved 3.75 feet forward and 0.75 feet down for this test.)

    Please note that the first three parameters for the VectorJump are the SAME as for the prior post when the Center of Rotation was not modified. Only the last parameter for distance of the viewing plane from the origin changes.

    In looking at the calculations for the equation of the Viewing Plane from three points, it becomes pretty obvious that the relative location of the points to each other determine the angle of the plane in space and that moving the Center of Rotation only changes the distance of that plane from the origin. Because the angle of the plane doesn't change, the a perpendicular vector (normal) to the plane also doesn't change, so the first three parameters SHOULD NOT CHANGE.

    So now that we know that the calculations are correct and that Aircraft Factory 99 is doing something different, what should we do????

    This is actually not the first time we have seen the Viewing Planes generated by Aircraft Factory 99 do something strange.
    During the "Design Study" for the P-38 Lightning, once the Parts count had reached a certain point, the Glue between the Inner Wing and Nacelle seemed to reverse itself and stop functioning correctly. I did not track down the exact location of where the VectorJump was failing because at the time, I had no means of addressing the problem.
    Perhaps now I do have a means of fixing these kinds of problems.

    - Ivan.
    Attached Thumbnails Attached Thumbnails CenterOfRotation.jpg  

Members who have read this thread: 85


Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts