Board index FlightGear Development Aircraft

Improving Douglas DC-3/C-47 Project need help!  Topic is solved

Questions and discussion about creating aircraft. Flight dynamics, 3d models, cockpits, systems, animation, textures.

Improving Douglas DC-3/C-47 Project need help!

Postby tonghuix » Tue Jan 30, 2018 6:10 pm

I am a lover of DC-3/C-47. Recently, I have flied Flying the Hump and Breitling DC-3 World Tour, and found the Douglas DC-3/C-47 in FlightGear is so good. But still something to be improved.

Why improve it?

Firstly, I found it lacking of DME receiver, so I added a DME receiver for it. And by searching the model and its author, I found VSKYLAB's DC-3, it based on this model and modified a lot, improve plenty much, but NO CONTRIBUTE BACK!There was two way, one is asking them contribute back, another is work on myself. By reading their note:

Emmanuel's model was licensed to the VSKYLABS Douglas DC-3/C-47 project for further development, and The VSKYLABS copyright and disclaimer notes are fully applied to this package.


I am very angry! I contacted Emmanuel and ask him contribute back but no reply, I also tried to contact RAF team, but their forum all French, I cannot understand any infomation. So I planed to improve by myself.Also this Douglas DC-3/C-47 project has at least three version, FGAddon version, RAF team version and 3rd party hanger version. So I also plan to merge those all version to one. And there is no update from May of 2016 (other version stopped at November of 2016).

What have I done yet?

Before, I only use Blender for 3D printing, no any experience for Game modeling, and UV, what I have done is just learn-and-try.

1. First of all, I found it is no DME, so I added a DME from DHC-6 project. Good news is DHC-6 project is GPLv2, so I could using any code from it. And add more radio equipment (COM2, NAVs, ADF and a transponder). As the result, it could now adopt modern navigation!

Image
Image

2. Then I improve the fuseledge reflection for ALS, it works fine and pretty.[*]After finishing my plastic model, I found the connection between wing root and fuseledge are too abrupt, and wing tail is wrong compare with real DC-3. So I improve it by adding a 'patch' to make the connection more smooth and re-create the wing tail and wing mount. Now it is more smooth, and less drug will produced. I could say the exterior is FIVE stars!

Image
Image

3. Re-modeling wipers, and merge a wiper tube from other version (wipers need hydraulics to work so the tube is vital). but I do know nothing about UV, NEED HELP to improve it!

Image

4. Add a flaps indicator from DHC-6 project, just modified the texture.

5. Add interior shadow for ALS, using code from DHC-6 project, but re-generate shadow files.

6. Other minor updated, antennas, lights...etc.

Image

TO DO and help is needed!

Following are what I plan to do recently, all needs help!

1. Fuseledge is too 'fat', real DC-3 is thinner. But the thing is not so easy, a lot of works to do not just 'shape' it.
2. Liveries need updated. Due to add a wing-root 'patch', UV map updated, so some livery may have bug, I do believe most livery works perfect but only a few need to be modified.
3. Fly-by view and Model view work incorrect. Need to fix.
4. Movable side window.[*]Emergency Exit at afterward of overpanel.
5. Seats for military interior.
6. Autopilot (Sperry III) not work, 3D models are ready, need coding and make it alive.
7. Aileron and rudder trim, the 3D models are ready, just make them alive.
8. Windshield effects, such us rain/snow, interior reflection.
9. FDM to Jsbsim. Another version implied it, but needs more things todo.
10. Much more...

My development repo: https://github.com/tonghuix/Douglas-Dc3, a blender file, with three-view of DC-3, included to help improve it.
FG Manual Chinese Translation https://github.com/tonghuix/getstart-zh
tonghuix
 
Posts: 119
Joined: Sun Nov 29, 2015 6:11 pm
Callsign: CPA112
IRC name: tonghuix
Version: next
OS: Debian Testing

Re: Improving Douglas DC-3/C-47 Project need help!

Postby GabrielYV » Tue Jan 30, 2018 7:53 pm

Hi, I can help you but the base file is missing from that plane, I unload the plane and when I look for it on my aircraft list it is not available, Please check it
New Boeing 737-800YV (Improved version) https://github.com/YV3399/737-800YV

Gabriel Hernandez.
GabrielYV
 
Posts: 215
Joined: Sun Dec 13, 2015 7:06 am
Location: Venezuela/Mexico
Callsign: YV3399/YV123B
Version: 2020.3.13
OS: Mac OS

Re: Improving Douglas DC-3/C-47 Project need help!

Postby tonghuix » Wed Jan 31, 2018 8:36 pm

@GabrielYV, Thank you for your help!

I just update and fix it, and I tested it on 2017.3 and 2017.4, both work fine.
FG Manual Chinese Translation https://github.com/tonghuix/getstart-zh
tonghuix
 
Posts: 119
Joined: Sun Nov 29, 2015 6:11 pm
Callsign: CPA112
IRC name: tonghuix
Version: next
OS: Debian Testing

Re: Improving Douglas DC-3/C-47 Project need help!

Postby GabrielYV » Wed Jan 31, 2018 10:09 pm

Ok, Dont worry. I love to help.
thanks for fixing it, then I will try it and I will see what I can do for this plane, Does the dual control work? I think that the dc3 has the best dual control in flightgear but only that the co-pilot does not stay inside the cockpit when the plane moves :(
Regards, Gabriel Hernandez (YV3399)
New Boeing 737-800YV (Improved version) https://github.com/YV3399/737-800YV

Gabriel Hernandez.
GabrielYV
 
Posts: 215
Joined: Sun Dec 13, 2015 7:06 am
Location: Venezuela/Mexico
Callsign: YV3399/YV123B
Version: 2020.3.13
OS: Mac OS

Re: Improving Douglas DC-3/C-47 Project need help!

Postby tonghuix » Thu Feb 01, 2018 8:55 am

Thank you very much!

I don't test the dual control yet, maybe it is still a bug,
FG Manual Chinese Translation https://github.com/tonghuix/getstart-zh
tonghuix
 
Posts: 119
Joined: Sun Nov 29, 2015 6:11 pm
Callsign: CPA112
IRC name: tonghuix
Version: next
OS: Debian Testing

Re: Improving Douglas DC-3/C-47 Project need help!

Postby tonghuix » Thu Feb 01, 2018 4:42 pm

It is very strange, some model work perfect at other aircraft, but when I move it to DC-3, it not work such as the knob of DME...
FG Manual Chinese Translation https://github.com/tonghuix/getstart-zh
tonghuix
 
Posts: 119
Joined: Sun Nov 29, 2015 6:11 pm
Callsign: CPA112
IRC name: tonghuix
Version: next
OS: Debian Testing

Re: Improving Douglas DC-3/C-47 Project need help!  

Postby Fritz » Wed Sep 16, 2020 3:41 pm

This is my first comment since 2017... real life really got into my way! I haven't "played" FG for more than three years, and only about a week ago I installed the new version (2018.3.6 to be more precise) and had to spend some time re-learning to operate the simulator and "my" DC-3.

I say "my" DC-3 because it's the plane I'm usually flying, and because over time I made some modifications to the PAF version, starting with fixing the tail wheel issue discussed in this forum in a lengthy thread in 2016 (viewtopic.php?t=29667) . I also made it "safer" to fly by adding a trim indicator ("stolen" from the Super Constellation), and control lights for "parking brake", "tail wheel locked", and "autopilot on". This is not authentic of course, but neither is the modern radio or the RMI, and the cockpit of every DC-3 still flying today is different anyhow.

In the past days I changed more things I had already planned to do in 2016/17. Most of it was fixing instrument lighting issues (for example the gear indicator lights should only work if the master battery switch is on, and the magnetic compass lighting was completely wrong). Also, for fun, I added a fasten-seatbelts switch and control light in the cockpit. it doesn't really do anything because there isn't a fasten-seatbelt sign in the cabin (there probably should be one, at least in the civil passenger versions). Maybe I'll add a screenshot of my current cockpit.

All the changes I made so far are comparatively simple (I'm not a FG expert) and affect xml files only. Apart from adding or duplicating existing switches, lights, or instruments, I haven't done any 3D modelling.

I also started to think about the autopilot. As you have noted it doesn't really work, and what is working isn't working like the real thing (no Sperry A-3 modeled in any flight simulator seems to get it right as far I can tell from the internet). I found several pages of the "real" operating manuals on the internet, not all of them but enough to understand how it works and what the knobs really should do. But correcting the "mistakes" and implementing all of the functions would be a huge project incompatible with my real life at the moment.

So I decided to start with a "simplified but more realistic" version using the existing 3D model. The first steps would be to change the functions assigned to the knobs and buttons, and to swap the two directional scales: The upper one should be the one to set the desired course and the lower one is the gyro compass. The knobs in this simplified version would operate the generic FG autopilot. This isn't perfect for this, for example it controls direction via the ailerons, while the Sperry A-3 uses the rudder. But for flying the simulator, this doesn't really matter. Sticking to aileron control, the wing leveler would be useless, so It can be ignored in this "simplified version".

Bonus functions would be restricting the direction control to small changes, blocking the controls while the AP is engaged, making the gyro compass adjustable (like the real things are) and indicating the vacuum level (at least change between 0 and 4.7) . Adding the missing elevator alignment indicators would need changes to the 3D model which is beyond my current capabilities.

So far, I didn't do any work on the Sperry except collecting information. I also missed a DME and 3D models for most of the radio equipment, but I haven't done any work to change this either.


Btw, I don't think the model looks too fat. Maybe it's an illusion created by the tail fin extension towards the front? I can't imagine the original creators of this superb model getting the basic measurements wrong.
Fritz
 
Posts: 283
Joined: Tue Apr 26, 2016 11:04 pm
Location: Bavaria, Germany, near ETSL
Version: 2018.3.6
OS: Windows 7 Prof.

Re: Improving Douglas DC-3/C-47 Project need help!

Postby tonghuix » Wed Sep 16, 2020 10:51 pm

I think the most important thing is the exterial 3D model.
FG Manual Chinese Translation https://github.com/tonghuix/getstart-zh
tonghuix
 
Posts: 119
Joined: Sun Nov 29, 2015 6:11 pm
Callsign: CPA112
IRC name: tonghuix
Version: next
OS: Debian Testing

Re: Improving Douglas DC-3/C-47 Project need help!

Postby Ovazor » Thu Sep 17, 2020 6:45 am

The first steps would be to change the functions assigned to the knobs and buttons, and to swap the two directional scales: The upper one should be the one to set the desired course and the lower one is the gyro compass. The knobs in this simplified version would operate the generic FG autopilot. This isn't perfect for this, for example it controls direction via the ailerons, while the Sperry A-3 uses the rudder. But for flying the simulator, this doesn't really matter. Sticking to aileron control, the wing leveler would be useless, so It can be ignored in this "simplified version".


That´s what I have done with my DC-3 years ago. For now, in my DC-3 the two left knobs both control the heading with the ailerons and the one in the top right the pitch, both with the inbuild standard FG autopilot.

Other fix I have made is to make the A/C load sliders to actually change the total mass and its distribution in the yasim-model.

And, IIRC, the parachute jumpers weighted 125lbs each.
Flying the Dakota
Ovazor
 
Posts: 76
Joined: Wed Feb 12, 2014 7:53 am
Location: Finland
Callsign: OH-OVA
Version: 2020.4.0
OS: OpenSUSE LEap 15.2

Re: Improving Douglas DC-3/C-47 Project need help!

Postby Ovazor » Thu Sep 17, 2020 6:48 am

Here is the thread about payload bug

https://forum.flightgear.org/viewtopic.php?f=4&t=37395
Flying the Dakota
Ovazor
 
Posts: 76
Joined: Wed Feb 12, 2014 7:53 am
Location: Finland
Callsign: OH-OVA
Version: 2020.4.0
OS: OpenSUSE LEap 15.2

Re: Improving Douglas DC-3/C-47 Project need help!

Postby Fritz » Wed Oct 07, 2020 11:41 pm

Meanwhile I've completely redone the autopilot, much more than I initially had planned. It's not based on the generic autopilot anymore, the 3D model has indicators adjusted by the pitch and roll knobs, and the AI (gyro horizon) has a roll indicator like it should have. The model isn't visually perfect yet because I didn't change the existing ac3d files and textures significantly, and the real thing has a "window" im the the right side of the AI where the pitch indicators are located. Maybe later...

Important for now is that it can be operated correctly and without using the generic autopilot dialog. The autopilot now does, what the real thing does, and I tweaked the PID controllers to give it a softer reaction to the knob changes than the generic autopilot does. It's really more fun that way, although I obviously don't know how the real thing feels.

Here's a picture of the current version, sporting a revised roll indicator scale, follow-up indicators for roll and pitch, a working suction gauge, a gyro compass that has to be manually set to the magnetic direction, and an almost completely redone, geometrically perfect compass texture. Maybe I'll redo the numbers too, because they are much "fatter" than on the real thing.

Image

I even found the prototypical on/off switch (actually it's a hydraulic valve and not a switch, there is no electricity in the Sperry A3 except for the lighting). Interestingly the valve is part of the original PAF DC-3 model where it is described as a "long range tank" in the XML file (without any function implemented). I only had to change the texture of the label and the property changed by the switch.

Image

As you can see, my DC-3 meanwhile is a mix of old and new instruments (using the recommended T-arrangement), but if you look at real cockpits of airworthy DC-3, they always look like this (or worse!). Currently I'm in the process of enlarging the overhead "radio box" to accommodate two NAVs and an ADF, using the Collins models matching the already installed radio. My goal is not having to use any dialog windows except for simulator setting, fueling, and loading. That's why I replaced the altimeter, because the one originally fitted doesn't have a window for adjusting QNH.


Edit: Here is the cockpit of the DC-3 "Congo Queen". Until recently it still had the Sperry A3 installed, but now it sadly has been removed and replaced by electronics. The mentioned valve is still present though: https://www.jetphotos.com/photo/9156626.

Edit 2: Picture of the cockpit with the A3 still fitted: https://www.jetphotos.com/photo/7018858
Fritz
 
Posts: 283
Joined: Tue Apr 26, 2016 11:04 pm
Location: Bavaria, Germany, near ETSL
Version: 2018.3.6
OS: Windows 7 Prof.

Re: Improving Douglas DC-3/C-47 Project need help!

Postby AndersG » Thu Oct 08, 2020 8:00 am

Did you check the A2 autopilot of the Short Empire flying boat? It is a JSBSim autopilot. I don't expect it to be good but it was the best I could do at the time and from the documentation I had. It is not complete with the surrounding systems, though.
Callsign: SE-AG
Aircraft (uhm...): Submarine Scout, Zeppelin NT, ZF Navy free balloon, Nordstern, Hindenburg, Short Empire flying-boat, ZNP-K, North Sea class, MTB T21 class, U.S.S. Monitor, MFI-9B, Type UB I submarine, Gokstad ship, Renault FT.
AndersG
 
Posts: 2524
Joined: Wed Nov 29, 2006 10:20 am
Location: Göteborg, Sweden
Callsign: SE-AG
OS: Debian GNU Linux

Re: Improving Douglas DC-3/C-47 Project need help!

Postby Fritz » Thu Oct 08, 2020 12:24 pm

No, I didn't, but I certainly will! Functionally, the A2 and A3 should be about the same. To be honest, I've only flown about five different planes in FG (the Short Empire is not one of them) and had a look a some additional planes to find instruments or to see and learn how things are done. I did have a look at the B-17 which had the A3 in early versions, but the cockpit is far from complete and doesn't have an autopilot at all.

What I basically did was to take the simple "pitch hold" PID controller of the generic autopilot (Aircraft/Generic/generic-autopilot.xml), adapted it for roll and direction, added filters for slowing it down, and tweaked a few values. I never did anything like this before in FG, so there was some trying and guessing, but it seems to work fine.
Fritz
 
Posts: 283
Joined: Tue Apr 26, 2016 11:04 pm
Location: Bavaria, Germany, near ETSL
Version: 2018.3.6
OS: Windows 7 Prof.

Re: Improving Douglas DC-3/C-47 Project need help!

Postby WoodSTokk » Thu Oct 08, 2020 10:10 pm

I hope you have time to wait for the Collins devices. I have allready remodeled and animated the Collins devices on the Citation II but i want a second update (include axis objects and upgrade the radio for 8.33kHz). I hope i find spare time before Christmas.
WoodSTokk
 
Posts: 1077
Joined: Tue Oct 17, 2017 3:30 pm
Location: Milky Way/Sol/Earth/Europe
Callsign: SX-W57
IRC name: WoodSTokk
Version: 2020.4.0
OS: Debian Bullseye

Re: Improving Douglas DC-3/C-47 Project need help!

Postby Aboim2 » Sun May 30, 2021 10:51 pm

It makes me very happy to know that this project is still rolling! I want you guys know that DC3 is one of my favorite planes!
I know how to model 3D in Blender and have some knowledge in Python, not much in YASim bit willing to know.
I am planning to build my own aircrafts.
If you need help with something, just let me know!
Aboim2
 
Posts: 58
Joined: Fri May 28, 2021 11:23 am
Location: Lisbon, Portugal
Version: 2020.3.13
OS: Windows 11

Next

Return to Aircraft

Who is online

Users browsing this forum: wlbragg and 16 guests