• Which the release of FS2020 we see an explosition of activity on the forun and of course we are very happy to see this. But having all questions about FS2020 in one forum becomes a bit messy. So therefore we would like to ask you all to use the following guidelines when posting your questions:

    • Tag FS2020 specific questions with the MSFS2020 tag.
    • Questions about making 3D assets can be posted in the 3D asset design forum. Either post them in the subforum of the modelling tool you use or in the general forum if they are general.
    • Questions about aircraft design can be posted in the Aircraft design forum
    • Questions about airport design can be posted in the FS2020 airport design forum. Once airport development tools have been updated for FS2020 you can post tool speciifc questions in the subforums of those tools as well of course.
    • Questions about terrain design can be posted in the FS2020 terrain design forum.
    • Questions about SimConnect can be posted in the SimConnect forum.

    Any other question that is not specific to an aspect of development or tool can be posted in the General chat forum.

    By following these guidelines we make sure that the forums remain easy to read for everybody and also that the right people can find your post to answer it.

FS2004 Boeing 747 Classic

FSMuseum

FSMuseum

Messages
233
Country
us-kentucky
Yes, I know... FS2004?!

For those of you who haven't seen me in the Gauges\FDE forums, this might come as a shock. I started this project around 6 years ago and have only just recently gotten major traction with it. I absolutely adored the old RFP 747 and finally endeavored to make my own, in the spirit of it, but with much more complete systems and more modern graphics. Yes, I do use more modern flight sims, but I still have a deep love for FS9 and may be stuck in it forever at this rate... :D

This panel that you see is currently plugged into my custom flight dynamics, which I am confident are as near as possible to the real 747 in every conceivable way, even managing to get around several bugs in the default Flight Dynamics system with some clever manipulation. The model utilized for this project is the CLS 747, as I have no modelling experience with aircraft to make my own.

Everything you see here is XML, and please understand it is far far from finished as you can probably see in a few areas, but I am very proud of how far I have come.

Some fun features of this overall project:

- Fully featured FDE with realistic damping, control, stability, and mach effects. Everything from dihedral effect, ground effect, mach pitch and roll effects, lift, drag exact engine thrust, engine spool and start times, etc. has been extremely finely tuned according to Boeing and NASA documents for all phases of flight from mach 0 to mach 1 at any altitude.
- Extremely fine tuned load station data directly from Boeing W&B docs, with exact locations for MAC, LEMAC, Cargo, PAX, fuel tanks, and landing gear including suspension (as best as I could anyway, the CLS model has a bug which causes the main gear to have no suspension)
- Current systems simulated are electrical and fuel. Both work exactly according to Boeing spec with correct loads and transfer rates. Fuel flow is simulated using DD's fuel dump gauge and completely bypasses the default FS system. The fuel flow scalar in my cfg is 0!
- EPR, fuel flow, and EGT are all calculating outside of FS default systems using exact specifications from JT9D-7A maintenance manuals. They are accurate to within 0.5% for any thrust setting!
- In an effort to make the aircraft feel 'alive' I have randomized the values of several things on the panel every time it is loaded. Airspeed bugs, gross weight indicator, EPR bug settings, EGT temps, EPRL, and other future components will load at random but realistic values every time the aircraft is loaded to make it feel more like a real aircraft in use.
- Custom sounds included switches, knobs, equipment fans, packs, electrics, vibrometers, etc.

Eventually I want this to be as close to a study level 747 classic as possible... This aircraft and the Concorde are definitely my favorite aircraft of all time.

I have spent a lot of time trying to make sure I capture the 'feel' of the 747 from every crew member's perspective, and I would like to extend a thanks to everyone on the forum here who so far has helped make this a reality, as well as the late Mr. Ralph Tofflemire and all who once worked for Ready for Pushback who inspired this project with the original RFP 747 panel and aircraft.

Enjoy this preview, and ignore the debug information everywhere :duck:

(Original images were 4k, downscaled to 1080p, so ignore any artifacts)

Progress1.png


Progress2.png
 
Last edited:
While not a huge fan of the 747, I do love a good FS9 2D panel, in fact I much prefer them to the VC's. Looks like you've done a great job here. Well done.
 
Really nice work, Chris!

I had the same motivation 3 years ago but for FSX and with very limited xml knowledge:

Simviation: Microsoft Flight Simulator Addons

Simviation: Microsoft Flight Simulator Addons
simviation.com
I have tried my best but all in all it has by far not the same system depth like the RFP! But i had much fun and have learned a lot!
 
Some more work in progress shots.

Most of the main panel is now complete, and includes night lighting at two levels (backlighting only, and full flood lights). The color of the floodlights is still a work in progress. But it is there at least!

The autopilot now has all functional components complete except for a few switches, and I am beginning work on the INS. The INS will simply be the CIVA model but with custom bitmaps to match the rest of the panel.

I have completely redone most of the code and aerodynamics involving the engines, correcting a few errors I made previously and including some new data I found. Total thrust, N1, N2, EPR, EGT, and Fuel Flow now all exactly match both my NASA papers and my PW JT9D maintenance schedules in all phases of flight to within a 1% degree of error.

Further, I have done much flight testing with the airplane to confirm that the stabilizer provides exactly the right amount of trim in all phases. The stabilizer settings, for all phases of flight I could test against in the NASA paper, are within 0.2 degrees for all phases, with most being less than 0.1 degree difference. This includes takeoff settings, landing approaches, flap configurations, etc.

Every little thing I can do accurately brings me much joy :)

VL2pXzw.jpg


k7VSvv1.jpg
 
I say it looks amazing. And the fact that you want to make it even better than the RFP deserves a big applause!

I guess it will be also compatible with the Posky birds, right? I'm moving to P3D4, but I have FS9 still installed just for birds like the DF 727, RFP and now this one.
 
lmhariano said:
I say it looks amazing. And the fact that you want to make it even better than the RFP deserves a big applause!

I guess it will be also compatible with the Posky birds, right? I'm moving to P3D4, but I have FS9 still installed just for birds like the DF 727, RFP and now this one.
Click to expand...

Thank you very much!

As of right now I have no plans to move it to the POSKY birds, mainly because I have decided I am dissatisfied with the CLS model and have begun learning how to make my own. Stay tuned for that development...

The model is also the big thing preventing me from distributing this aircraft publicly in a form that is easy for a non-savvy user to install. With my own model, this won't be a problem.

I may be late to the FS2004 game but I say it still has plenty of life left in it... Works well enough for me, anyway :)
 
alex_94 said:
Is this still being worked on? :)
Click to expand...
Jon_aus said:
Yes I also would love to know how Chris is getting on with this.
Click to expand...

Wow, I can't believe it's been almost 3 years since I updated this thread... so much to do, not enough time!

Yes, this is still being worked on, and a LOT of progress has been made since the last update here.

The fully-custom PID-based XML autopilot is (almost) complete, with all control formulas and gain values being directly from the Sperry SPZ-1 documentation. This autopilot is possibly one of the most realistic classic autopilots ever made in FS2004, when paired with the extraordinarily-accurate flight dynamics model. The flight director works just as the real one does as well. Fully-functional autoland is also simulated. There are still a few bugs and minor features to add to the system, but the majority of it is complete and working correctly. Everything you see on the autopilot panel functions just as the real one does.

Aside of that, the pressurization and pneumatics systems are complete as well, and are coded based off of actual heat-exchange formulas, CFM rates, pressure differentials, etc. Essentially there is an entire system simulating the air in the airplane coded from scratch, and all of the appropriate sources are simulated accurately as well. Re-circulation fans will help cool/heat the cabin faster, all of the packs can be controlled in full-auto or full-manual, and the rate of pressurization affects the temperature of the air entering/exiting the cabin as well. The pack auto controllers can do all of the heavy lifting for you, or you can set them to manual and have fun trying to manage it yourself.

The electrical systems have been expanded and include the most accurate data for the KW usage of all the aircraft components I could find. It is hard to do, but you can overload a generator and cause bearing failures!

Lately I have been spending time working on some other projects that I might show here soon, but here and there I have been nicking away at things. You might also notice the much more complete state of some of the flight instruments as well. The ADI works off of the CIVA INS accurately, spinning up and erecting at the appropriate times during the alignment phase (thanks Marco for a lot of help with the XML bridge!). Other minor systems are being taken care of slowly as well.

The eagle-eyed among you might also notice slightly different EPR settings on the engines and fuel flows that don't match - the former is a feature of the panel, accounting for the slight inaccuracies of the gauges/off positions of the throttles (it does not actually affect the thrust at all!). The latter is more or less a 'bug', caused by me updating the fuel flow equations to better match my data from PW, I have simply been to lazy to fix it on all engines at this stage.

Speaking of the engines, there is a highly sophisticated EGT model in place that (as far as I am aware!) accurately accounts for turbine seals changing shape with temperature, increased cooling rates with motoring the starters, wind direction and speed during start, and consequences for extreme or prolonged over-temps.

Originally I was planning to model the 747-100 with JT9D-7A engines, but I have switched gears and I am now representing the 747-200 with JT9D-7R engines instead for eventual release, basically offering the most capable version of the 747 classic first, and other versions to come later. I do plan on eventually creating accurate CF6 and RR variants as well, but those will undoubtedly come later and probably not be as exactly accurate as this PW version is due to lack of exact thrust data for those other engines. However, I do have enough data that I can get a very close approximation, probably close enough for anyone here (though the jury is still out on me, ever the perfectionist...)

Lots of other features are there as well but I will let them speak for themselves in due time!

Don't mind the debug menus, INS pop-up, or mis-placed gauges. Most of those come from re-working the panel bitmaps to look more realistic and I have just been lazy with them :) I have also borrowed the center window post from the Felis 747 for right now since I have not been able to track down a high enough quality photo of it from the right angle.

dpVOoPG.png


JQaZc84.png

nIUWbpr.png
 
Last edited:
Reading everything on the amount of realism and real world physics you have considered and knowing about the complications in xml coding I am just left with awe . . .

Michael
 
It just finally got to me a few months late that Active Sky Evolution no longer works. A big blow to anyone who still loves FS9 (like me) but thankfully, FSGRW still seems to work okay. One day I will finally migrate to newer sims. But until then...

Been working a lot on these flight instruments lately. I programmed the Central Instrument Warning System (complete with the annoying buzzer sound it makes when it detects a comparison problem between instruments!), the lights flash at exactly the correct timing according to my maintenance manuals, 1.5 Hz or 2/3 second per flash. A minor detail, but one that I am proud to say I included.

The altimeter works off of correct standby and CADC modes: in standby, the appropriate correction will have to be added to the altimeter, whereas the CADC can compensate for you (best keep the power going!).

The HSI has seen a lot of attention: it too now requires that the INS be at least out of alignment mode 8 to provide accurate heading information from the directional gyro. I have also modeled an accurate directional gyro precession: depending on your speed and relative direction, the gyro will very slowly precess over time and give less accurate heading information. What's more, INS 1, 2 and 3 all precess at ever so slightly different rates (slightly different latitude nut settings). However, this is only applicable if you turn the corresponding compass switches from SLAVED to DG, thereby removing the compass flux valve compensation from the INS directional gyros. Of course, an accurate alignment must be made on the ground before the DG's can be useful for navigating.

The ADI turn coordinator ball is also completely custom. I was watching a lot of videos of classic 747 landings from the cockpit and noted how sensitive it seemed that the ball could be. After some research, I also discovered that the way that FS calculates it's own turn coordinator ball position seems to be flawed under certain conditions. Therefore, I whipped out my physics knowledge and wrote my own script for the ball that based on lateral acceleration and yaw rate (accounting for the fact that the ball is some 40 meters ahead of the center of rotation). Essentially treating it as an inclinometer instrument on a long stick held out from the center of rotation. Lo-and-behold, it turns out to be much more sensitive, but responds the way I would expect such an instrument to.

The altimeter setting (as well as the autopilot heading, speed, course, and altitude settings), randomizes each time the panel is loaded to something that is reasonable for each one. Below the value is actually very near the maximum that it can randomize to, 30.50 inHg. Helps the panel feel a lot more 'alive'.

And lastly, the FMA: the series of green and amber indicators at the top. This is the model that was included on DUAL autopilot aircraft with FFRATS installed. I spent a good deal of time trying to get the colors and lighting just right on them (according to my manuals they are always backlit regardless of instrument panel lighting). I also exactly matched the fonts in use, as I have done for all other instruments now. I intend to model a triple autopilot aircraft at a later point probably.

Oh and also: the airpseed indicator. It is well known that the equation used to get indicated airspeed in FS2004 (FSX and P3D too) is incorrect: there seems to be a compressibility term included in the calculations that should not be present (or at least, could be compensated for by a CADC like the 747 has). It's important to note however that actual airspeed is not affected by this 'bug', only the reading on the gauges. Therefore, I have implemented the same IAS/CAS and Mach number equations that NASA used in their own 747 simulator. The results are surprising: as much as a 10 knot difference in some instances in normal flight. The autothrottle I have programmed will account for this as well.

Speaking of the autothrottle, I have still not located any good information for the specifics of how the FFRATS computers would calculate throttle positions, so I have created my own code for the EPR mode and use the default IAS and Mach hold modes to do the rest. It's not perfect, but it works surprisingly well, good enough for me (which is rare for me to say...).

Anyway, enjoy the screenshots. More to come.

ANNdcf0.png

cZPOblo.png
 
Last edited:
The ball was really only used for trimming the aircraft when asymmetric - and for simulator instructors to use as evidence that you weren't doing it properly!

FFRATS was often not used (in my airline) as the flight engineers reckoned they could do a better job - and they were usually right. I think it was mandatory for an autoland approach (CAT 3 in 3-AP aircraft, Cat 2 in the others). It was rare to see a throttle quadrant with all the throttles in line in the cruise. 'Bunch of bananas' was the usual FE phrase!
 
Tim Collins said:
The ball was really only used for trimming the aircraft when asymmetric - and for simulator instructors to use as evidence that you weren't doing it properly!

FFRATS was often not used (in my airline) as the flight engineers reckoned they could do a better job - and they were usually right. I think it was mandatory for an autoland approach (CAT 3 in 3-AP aircraft, Cat 2 in the others). It was rare to see a throttle quadrant with all the throttles in line in the cruise. 'Bunch of bananas' was the usual FE phrase!
Click to expand...

Interesting, Tim!

The ball is such a simple 'stupid' instrument, something in me could not let the issue rest until I had something that was physically accurate. Such is the way I am about developing these days. Time consuming, but rewarding!

The FFRATS, from most descriptions I've read, seemed a little squirrel-y at best. I guess things got a lot better in terms of autothrottle design with the advent of EEC and FADEC in the 737 Classics and 767.

Speaking of my perfectionism, I revisited the FMA and re-did the indicators and lights. My manuals say the indicators are always backlit but all of my photo-references taken during the day don't seem to reflect this (or the lighting is just really dim in the day), so I revised them to look more like what I see in the clearest and most color-accurate photographs I have located for the unit. I also redid the AT and AP warning lamps, to mimic the appearance of multiple colored bulbs inside each light. I have only located clear photos/video of the TRIPLE style FMA with the lights illuminated, but I figured they would work pretty similarly on the DUAL indicators.

PpVXZUw.png


k50tCnG.png
 
Some more screen shots for you! You might notice that all of the debug panels are now finally out of the way and that the INS is no longer blocking the view. It finally has a home, but that panel is not yet ready for viewing :)

Enjoy these two shots of one of my favorite features of the original RFP 747 that inspired this project, re-imagined: the approach panel view point. The panel is obviously not done yet, the overhead elements that are in view are borrowed from the Felis 747 until I can get my own photos/graphics made. Handy clickspots to be added to make switching views as intuitive as possible.

gfaiZBx.png

0GZZes5.png
 
You must log in or register to reply here.
Back
Top