One of the reasons X-Plane (just as DCS World and IL-2 when I used those sims) is my go to platform for flight dynamics modeling is in the simulation of powerful prop aircraft with a tailwheel, like the DC-3 / C-47 which is one of my preferred fixed wing aircraft in simulation and in the real World...

Now, regarding options for the DC-3, I have used 3:

- LES DC-3 which I purchased for the first time for X-Plane 10 if I'm not mistaken, and then upgraded for XP12 a few months ago when the new version became available;

- VSkylabs DC-3 Airliner

- Aeroworx C-47 (free and simply GREAT!!!)

I can find good reasons to keep the three installed and I've been trying to use them all, but I have stumped, in all of the 3, against at least one puzzling feature:

- At takeoff and initial climb power, but not only because it can also happen at cruise power, all of the 3 present a very evident tendency to bank right (starboard) ??? This is weird since I would accept bank (n and yaw) due to torque effects trying to lower the por side wing, but for any unknown reason they all bank right, actually requiring the use of aileron trim (to the left !!!).

Being a clockwise rotating props aircraft torque would cause left bank and yaw at high power regimes, specially at  lower speeds and even more if configured. 

After opening the aircraft ACF files I can find, in both the LES and the VSkylabs versions that the developers opted for setting some ground fixed aileron trim tab "to the right" = positive in X-Plane values. The AWX DC-3 has 0.00 for fixed trim tab but still banks right too (???)

Any clues / ideas ?

There are a few other details OI would like to debate but let's start with this particular one.

Any contributions really welcome !!!!

BTW: contarily to the "competition", X-Plane had that excellent representation of ground physics, and taxiing these DC-3 models is a true joy, being able to apply rw techniques an be subject to the rw traps if I let it go off control 🙂  

Edited July 24, 20241 yr by jcomm

I have both LES and VSL ones and tried the freeware one too... I find it to be true, but don't really know why. I also do know that Airfoil Labs is supposedly working on theirs. If/when that comes out, I am pretty sure it will be the ultimate rendition of that aircraft as their other stuff is spectacular. But that one's been in the works for a long time now, I believe. 

Just now, BostonJeremy77 said:

I have both LES and VSL ones and tried the freeware one too... I find it to be true, but don't really know why. I also do know that Airfoil Labs is supposedly working on theirs. If/when that comes out, I am pretty sure it will be the ultimate rendition of that aircraft as their other stuff is spectacular. But that one's been in the works for a long time now, I believe. 

Thx @BostonJeremy77 !  I wasn't aware of the Airfoil Labs project. Will try to keep tracking it.

Thank you also for confirming my observation.

In the case of the AWX it's even more difficult to understand since, as I pointed out above, it's ground fixed aileron trim tab is set at 0.00 in the ACF file.

2 minutes ago, jcomm said:

Thx @BostonJeremy77 !  I wasn't aware of the Airfoil Labs project. Will try to keep tracking it.

Thank you also for confirming my observation.

In the case of the AWX it's even more difficult to understand since, as I pointed out above, it's ground fixed aileron trim tab is set at 0.00 in the ACF file.

Check out their website. They have some screenshots! 

Interesting issue - to be explored more. DC-3 airways is basically the host for AWX's C-47 and you have to be a member to get access to latest updates etc. Worth thinking about for some!

Anyway I have them all at the moment so here are some screenshots just for the visuals"

LES DC3

VSkylabs

AWX

The AWX has the most old fashioned cockpit. the LES is basically a modern avionics set and the VSkylabs is different again, just like in real life I guess. 

Performance - still trialling them all when I can.

None of them get the sound right to my old ears - the C-47 DC-3 was incredibly noisy inside, loud or deafening almost I find the sounds for all of them muted or lacking that bust your ears effect that the real DC-3 had. 

VSKYLABS and AWX are slightly missing the mark on the nose. The upper curvature from the windows downward looks much more accurate on the LES version.

 

9 hours ago, jcomm said:

After opening the aircraft ACF files I can find, in both the LES and the VSkylabs versions that the developers opted for setting some ground fixed aileron trim tab "to the right" = positive in X-Plane values. The AWX DC-3 has 0.00 for fixed trim tab but still banks right too (???)

Any clues / ideas ?

Too lazy to check, but looking at photos of C-47 props, the left engine's vortex hits the underside of the wing ouboard in an upward motion while the right engine's vortex hits outboard in a downward motion. In my book, this overall produces a right rolling motion.

 

10 minutes ago, Bjoern said:

VSKYLABS and AWX are slightly missing the mark on the nose. The upper curvature from the windows downward looks much more accurate on the LES version.

 

Too lazy to check, but looking at photos of C-47 props, the left engine's vortex hits the underside of the wing ouboard in an upward motion while the right engine's vortex hits outboard in a downward motion. In my book, this overall produces a right rolling motion.

 

I see your point, and indeed in twin props this "vortex" in some way can cause an apparent tendency for torque cancelation, but actually torque plays a major role... At most under a considerable range of power settings and prop pitch, torque may be less effective, but surely not overcome, specially at higher power settings like at takeoff and initial climb, by the way the slipstream hits the wings.

This made me get out my old teaching materials on twin engined aeroplanes and assymetric thrust effect and my C-47 manuals. The higher the angle of attack the worse the roll becomes and rudder is critical to keeping balanced flight, maybe there are some issues with the way the rudder is set up as well, I cannot fix any of the model issues so have to accept some blemishes! 

Taildragger configurations are a difficult beast in real life and the DC-3 is that early era where that type of aircraft was common. You had to work hard on the ground with the DC-3 as you only had assymetric power and differential braking, if you have ever had the opportunity to observe it from the cockpit - there is a a lot of wheezing and puffing (from the pilot) to push and pull the DC-3 around on the ground, the old fellas made this look simple, which it was not. XPlane makes you do the same so that is good. 

Anyway going to join JComm and do some indepth flight in them all see what transpires.

The Aeroworx C-47 has a slight asymmetry in one of the "misc wings", but I don't know if that is the cause of the banking.

6 hours ago, Murmur said:

The Aeroworx C-47 has a slight asymmetry in one of the "misc wings", but I don't know if that is the cause of the banking.

I thought about something like that being present.

Was common in most prop aircraft and I had seen other models using it before, but I was lost in "misc wings" when I opened the ACF.  

I  did check the fixed aileron trim tab though and the VSL has it set for a right rolling moment.

 

jcomm - you can hardly notice the roll and setting a little aileron trim before take off will sort it (retrim for other flight attitudes). Propeller slipstream effect is modest in terms of lift, the airflow over that portion of the wing going faster is cancelled out by drag (profile) and angle of attack, the engine slipstream is producing thrust, the key effect of the thrust is that the left hand side hits the rudder surface, the right hand engine does not, theoretically this will yaw the aircraft to the left. hence the need for a little right rudder. Because of these aerodynamic forces the DC-3 would be in a mild sideslip condition or imbalanced. The DC-3's major aerodynamic issue was adverse yaw - so are Frise ailerons modelled in XP?- I do not know. The AWX ailerons appear to me to be modelled correctly. In the real aeroplanes Frize ailerons were used to reduce adverse yaw but were not that effective but the key take away was the lack of rudder authority to control the yaw and the push forces required by the pilot. Basically the DC-3 was imbalanced as a result - the verdict you will be imbalance at a variety of speeds and rolling the aircraft but not unpleasantly so. The other thing is the ailerons themselves they have a long span compared to other aircraft as well. The NACA test report attributes this issue to 'rudder float and rudder cable stretch) maybe different hinges and control lines (for eg hydraulic) may have fixed this but the shape of the rudder and its method of attachment resulted in float and hence lack of positive control authority in an imbalance condition caused by adverse yaw. 

I find I am not overcontrolling to compensate and the AWX model is well behaved and the effects are mild indeed!

There are a couple of tools in XP where you can see the forces at play overlaid on the aircraft. Interestingly (I am not sure of the interpretation) but the right wing is not symmetrical in its readings compared to the left on the ground (Local wind direction effect I suspect given the streamlines) but take off and climb it does not to my eye indicate any signficant difference between the wings either. Cockpit controls confirm this with minimal rudder or trim required to fly straight. 

I do know with my joystick I can inadvertently hold on rudder force after or during take off and relaxing that pressure, ergonomic bias, fixes the issue straight away. 

 

Edited July 26, 20241 yr by coastaldriver

Thx @coastaldriver.

From the NACA reports we can read:

"Directional and lateral control power: A series of tests were made with one engine inoperative to determine the minimum speed at which the airplane could still be flown with zero sideslip and bank. With the airplane trimmed in the take-off configuration at an indicated airspeed of

90 miles per hour, the pilot cut the right engine at an airspeed of

110 miles per hour. He was then able to decrease the airspeed to 90 miles per hour while maintaining zero sideslip. Therefore, it is believed that the requirement that the rudder be capable of holding the airplane with zero yawing velocity and not more than 5° of bank at all speeds above

85 miles per hour in that configuration was satisfied."

There are other interesting features regarding pitch and directional stability that you can read on that important reference doc for comparison with the three models.

Search for NASA Determination of the Flying Qualities of the Douglas DC-3 Airplane... (NACA-TN-3088)

Edited July 26, 20241 yr by jcomm

8 hours ago, jcomm said:

Thx @coastaldriver.

From the NACA reports we can read:

"Directional and lateral control power: A series of tests were made with one engine inoperative to determine the minimum speed at which the airplane could still be flown with zero sideslip and bank. With the airplane trimmed in the take-off configuration at an indicated airspeed of

90 miles per hour, the pilot cut the right engine at an airspeed of

110 miles per hour. He was then able to decrease the airspeed to 90 miles per hour while maintaining zero sideslip. Therefore, it is believed that the requirement that the rudder be capable of holding the airplane with zero yawing velocity and not more than 5° of bank at all speeds above

85 miles per hour in that configuration was satisfied."

There are other interesting features regarding pitch and directional stability that you can read on that important reference doc for comparison with the three models.

Search for NASA Determination of the Flying Qualities of the Douglas DC-3 Airplane... (NACA-TN-3088)

Jcomm - interesting - an accident report into a DC-3 ditching into Botany Bay some years back revealed that single engine in the take off configuration required a min speed of 95kts, the DC-3 in that case also could not develop max rated power properly anyway at max weight it could not maintain 95 knots and they ditched basically 300 ft after take off. 

Noted also that there is only one aileron trim tab and it is in on the right wing for the C-47 DC-3. At max weight the Stall Speed for the DC-3 is 57 knots in full flaps gear down landing configuration. Flaps up and gear up the Vs was 67 kts, The speed they are referring to in this report is the Vmca it is 78 kts, The limit of Vmca is a factor of rudder size and fin area DC-3 has big rudder!

Take the AWX out and stall it see whether it rolls left or right on you! See what speed readings you get! Watch the balance as the speed changes see what it does!

I've found the cause of the right rolling moment... On at least the AWX and Vskylabs models there's a prominent lift vector when the engines are running, or even simply windmilling, originating from the port side of the nacelle of the port engine.

That asimmetry of lift is very considerable, and always present if the props are turning. 

@Bjoern had mentioned the way the slipstream hits the wings in the port and starboard sides being different and that's probably it.

But none of the official documents I've read about the flight characteristics of the dc-3 refer to such an effect actually causing right rolling moment when at most a left one was expected due to the two powerful props that rotate CW...

In the past I recall having seen this effect in other XP twin props.

Have also found a difference in misc wing dimensions between starboard and port wings in the Vskylabs dc-3 that I will try to further investigate.

At least the reason for that huge rolling moment is found 😁 in the VSL and AWX models. Now I will further investigate the LES dc-3, but as far as I recall that one also has the fixed aileron trim tab set for a right rolling moment, which might further add to the tendency...

P.S.: interesting answer to a similar thread I have started at the .org, by @danhopgood1 author of some really great free aircraft available from there :

https://forums.x-plane.org/index.php?/forums/topic/311499-right-rolling-moment-in-all-dc-3c47s-ive-used-in-xp12/&do=findComment&comment=2745684

Edited July 27, 20241 yr by jcomm

Continuing my flight tests, I picked the AWX for a few test flights and played around with stalls and spins and!!! it has a deadly spin !

The AWX DC-3 can easily be made to enter a completely irrecoverable spin. Load the default aircraft, with the default weight and CoG setup, climb to a safe altitude of no less than 10k feet and enter a left (or right) spin by fully deflecting the rudder when entering a wings level stall... Try to recover and report back, if you survive 🙂

I am trying to find evidence of this, and already found at least a reference to an accident where we can see the pilot of a Skydiving DC-3 nose diving for quite a while to exit the stall / spin that resulted from a high AoA situartion during the skydiving launch:

DC3 Stall (youtube.com)

There's also this interesting wind tunnel report from a scale model of the DC-3: Free-Spinning-Tunnel Tests of a 1/23.75-Scale Model of the Douglas DC-3 Airplane - NASA Technical Reports Server (NTRS)