Conspicuous by Their Absence

[aleatorylamp]

Hi Smilo, Hi Ivan,

I´d mistakenly mentioned AD2k2 at first because I hadn´t realized the latter was not for CFS1, so I was correcting myself - I know you know, but I didn´t know... Ha ha!

In the North of England there´s a Geordie story that goes:

When the ship "Anna" sailed into Newcastle on a misty morning on the Tyne,
the port authorities shouted "What´s ye name?",
and the answer was shouted back: "Anna!",
and the port official shouted again "What?"
The answer came shouted back: "Anna!",
so the port official shouted "A na ye na, but I divn´t na!".

Anyway, not to worry, dear fellows, I was not asking for or about anything, only musing about my possibly delving into AD2000. This is however, unfortunately rather debatable, and not only for lack of time - which all of us seem to have! I think I need a somewhat more graphic tool to build aircraft, but no mesh extrusion either... I suppose my limitation here is that I tend to get lost in abstraction.

Smilo´s not the only one with memory problems: I´d completely forgotten about how difficult AD2k2 was after the RFO tutorial. I quite enjoyed it, but when I started on the Gotha, the enjoyment slowly waned, and I was not even using seals (equivalent of AF99 glue templates), because at that moment I was building for FS2002. For CFS1, AD2000 looks even more difficult because it involves the viewing plane handling issue that Ivan has just mentioned.

Anyway, Cheers!
Aleatorylamp

[Ivan]

My Analyze over the Ocean,
My Analyze over the Sea,
My Analyze over the Ocean,
Oh Bring Back My Anatomy.

I figure I should always try to address the biggest current limitation to development first.
At the moment the Propeller "thing" is being addressed if I can get off my lazy behind to actually work on it.
This would make it possible to finish up a couple projects.
There is also an aircraft longitudinal trim issue that I ran into with the FW 190A which prevents its re-release.
After that, there is the need for some multi-engine gauges and some gauges for testing.

I tend to procrastinate on those things for which the technique is known but just require the execution.
From the standpoint of accomplishing things, this is bad because I only work on a project long enough to resolve all the difficulties which may not be enough time to actually finish the project when the remaining tasks seem too much lie drudgery....

- Ivan.

[aleatorylamp]

Hi Ivan,
I had to laugh out loud! Great!
With aircraft experiments I used to do before virtual aircraft existed I found that as soon as I could see how it would work, and that indeed it would, I´d leave it and never finish it up!
Anyway, musing about other "tools" does no harm, especially when they are more difficult that what one is used to... the enjoyment factor being crucial, without which it would be "toil" and not "hobby".

So, I´m sticking to AF99, which is just as well because I like it. I think that the only other tool that would come into question would be FSDS1, that came out when I bought AF99 in the year 2000, I think, but I know of nobody who has used it and have found nothing on it on the web. Incidentally, I made a mistake about the price though, it cost me the equivalent of 100 bucks back then, and FSDS1 was about 175.

Anyway, I have made tremendous progress with the Gotha G.IV. I thought more in depth about glue-sequencing on the different fuselage sections containing gunners and guns, and your comments to that respect, and managed to solve 90% of the problems. Now I´m ironing out little hairline splits here and there on the wings.

Textures on the fuselage are now also better - I´ve repeated the elevator textures for the top and bottom horizontal surfaces of the fore and aft fuselage, and have discovered that they don´t wash out so much because they spread out beyond the fuselage side-limits - i.e. the pattern isn´t crushed to the narrowness. Nice to discover how to get something better!

OK, then. Have to cook lunch!
Cheers,
Aleatorylamp

[No Dice]

Shame on you Ivan,

You know where to go for anything for CFS1, AD2000 and 2002 can be found here:

http://www.thefreeflightsite.com/Design.htm

Dave

[Ivan]

Hello No Dice,

Thanks for the reminder, but I have already downloaded it from your site twice.
I think Smilo also sent me a copy before that.
The issue is one of organization and getting a few things finished before jumping into an entirely new thing.
As I commented earlier, a new 3D design tool is not my greatest block right now.
The biggest is probably the inability to program my own gauges because all the AIR file stuff can be worked put with a bit of time and dedication.

Hello Aleatorylamp,
FSDS is what I originally bought as a design tool and it does not work for CFS1.
The Abacus folks were kind enough to send me a copy of AF99 as a replacement.

- Ivan.

[No Dice]

Ivan,

If you ever get a free moment, maybe you could test the FSDS to AF99 and tell me if it works ??

http://www.thefreeflightsite.com/Design.htm

Any efforts are much appreciated,

Dave

[aleatorylamp]

Hi Ivan,
Well... I never knew FSDS1 was only for "more advanced" FS´s - then it´s just as well I didn´t buy it!
It was nice though, of the folks at Abacus to send you AF99! I wish they could get hold of the source files so somebody wise could fix the glitches.
Cheers,
Aleatorylamp

[Ivan]

Hello No Dice,

If this requires a working copy of FSDS, then we are out of luck. If not, then I will take a look at it but as I commented earlier, Anna Honey is doing another World Tour at the moment (Shanghai as of last night), so my free time is VERY limited. 5 or 10 minutes to type a response here isn't hard, but downloading, setup and test needs a bit more than that.
I also don't need to be really awake to type a response here, but real testing requires a bit more care.

I am sure there is probably a copy of FSDS floating around my house, but I also know I have not done anything with it since I first got it which I am guessing was around 2002. I would not even know where to begin looking.


Hello Aleatorylamp,

I have never tried to contact Abacus with this proposal. Perhaps they would agree, perhaps not.
If the application source is C code it would be almost ideal.
Keep in mind though that there is still a very VERY major logistical problem:
At the moment, I am not programming gauges that I need not because I do not understand what to do. I even have reworked a very basic gauge.
I am certain I don't really understand the workings yet, but I also have no means of really experimenting further.
That is because I do NOT have a working MS Windows C Compiler that will generate an executable that will run on a Windows 98 machine.
With the way AF99 works, it is absolutely certain to use Windows library functions and my current C compiler for W98 can not handle that situation.
The most it will handle is non graphical ANSI C.

The likelihood of all this lining up in a workable fashion is about the same as me becoming a billionaire by next week.
When it happens, I will do the planning on how to best spend all that money.
At the moment, I believe time s better spent on more practical things.

- Ivan.

[aleatorylamp]

Hello Ivan,

I contacted Abacus in 2010 to ask a few technical questions about AF99 workings, and I got a very amiable response expressing a) surprise that someone was still using this programme, and b) that unfortunately nobody on the staff knew anything whatsoever about AF99 as the person/people responsible for it had long ago left the company and the product was no longer carried, so I did not pursue the issue - i.e. ask them if they knew the programmer´s whereabouts.

In answer to their surprise, I sent them a few screenshots of a couple of my models, to illustrate what could be done with AF99 and AA despite their age. I had just completed a nice-looking, accurately shaped Boeing Dreamliner, so I included a screenshot of that too, to push my point.

So then, it looks like as things are, I (or rather we) will have make do with these tools and their accessories (and their "little" shortcomings), which is not too bad either... and the Hiero-Gotha G.IV is progressing nicely too!

There are a few contradictions in the available Hiero-6 engine specs, which I´ll have to "adjust" (manipulate!) to get the desired engine performance, so I will probably be posting one or two technical questions to see where would be best place to make the necessary alterations.

Incidentally, you mentioned (already twice) that you were programming, or had to programme, some multi-engined gauges. Very interesting, as these are very much lacking! I managed to get some really nice vintage authentic looking Gotha and Staaken gauges done for FS2002 by one Claudio Mussner in Brazil who used Easygauge, but they won´t work for CFS1, unfortunately. There was a very interesting inclinometer and a dual-engine RPM gauge - as well as a dangling pencil...!!

OK, then! Hopefully you will get a bit of free time these days to do some enjoyable and practical things!
Cheers,
Aleatorylamp

[Ivan]

Hello Aleatorylamp,

No Dice seems to have been convincing enough so that Abacus is making AF99 available on special order.
Perhaps I should send them some screenshots of my recent projects also?
I believe good work can be done within the limitations of AF99.
I have already said it many times, but trying to prove that a nice looking aeroplane could be built entirely within AF99 is why I started building to begin with.
It is similar to a rather attractive girl my son and I keep running into. She has a beautiful fact but with one exception: She has huge hairy mole on her cheek. I forgot who asked her why she doesn't just get it removed.
Her comment was that she had gotten very used to it because she has had it all her life.
My comment to you all is whether she would be the same girl without the mole?
Kind of philosophical and silly, huh?

In the case of AF99. I suppose I would fix it if I could but would probably change things so that it would build a CFS MDL instead of a FS98 MDL.
The question here is how deep would a redesign go? I can see this getting into a very very long project.

- Ivan.

[Ivan]

This thread pretty much describes the issues:

http://www.sim-outhouse.com/sohforum...S98-CFS-Gauges

My W98 C Compiler is Delorie Gnu C. It works well enough for text based ANSI C which accounts for most of my programming.
It has no hope whatsoever of building a Windows application.
There are many compilers that will build a Windows application.

Microsoft Gauges come with a HUGE SDK with libraries, header files, etc.
The problem is that all of the compilers with the exception of probably MS Visual C will make their own definitions for objects and types that they need but the definitions in THEIR header files will most likely not match those of the Gauge SDK.

Thus when you try to compile a gauge, the compiler barfs (technical term) when it encounters SDK typedefs that conflict with its own.
In addition, the names and contents of the header files will not agree. There are probably a few thousand references across a hundred or so files.
If you dig down far enough as I did, you come across basic type mismatches in fundamental header files themselves which means that one compiler may simply not work with the other's "extensions" to ANSI C.
It could be something so simple as one header defining a new type as an Unsigned Character while the other defines it as a Character.
Looks simple enough, but if every library function of the compiler expects one type and gets the other, then the whole world blows up.

The gauges I need are Tachometers and Manifold Pressure. Additional Trim gauges would be useful as well.

What I did accomplish was to use MinGW on my old dead Vista machine to compile a dumbed down tachometer for the second engine of my B-25 Mitchell. I dummied out some things, redefined others and hard coded things that referenced libraries and eventually got it to compile.
The problem was that while it ran on a Windows 2000 game computer, it crashed to desktop when I tried to run it on Windows 98.

From what I can tell, MSVC version 5 is probably what I need to find. MSVC version 6 MIGHT work as a cross-compiler but won't install on W98.
Neither version is available at this time so I believe I am stuck.

Ideas?

- Ivan.

[aleatorylamp]

Hi Ivan,
-Pity about the gauges problem. It seems to be more difficult for CFS1 and FS98 than for newer simulators.
-Very interesting, what NoDice is trying to accomplish with Abacus and AF99!
-As for the hairy mole on the beautiful girl´s face, I would say definitely remove it if you can do so without ruining her face or altering it... as non-invasive as possible!
-You commented:
Quote: "...but would probably change things so that it would build a CFS MDL instead of a FS98 MDL.". Do you mean the CFS virtual cockpit and the trembling view here?
-I agree entirely with your comment:
Quote: "I believe good work can be done within the limitations of AF99.
I have already said it many times, but trying to prove that a nice looking aeroplane could be built entirely within AF99 is why I started building to begin with."
They do look good!

Well... Here are the 2 technicalities I was wondering if I could bounce off you, about the Austro Gotha G.4 Grossflugzeug.
1) Engines: These were Austrian Hiero-6, normal, 6-cyl, non-high-compression.
Displacement: 943 cu. in. (157.2 cu.in.per cyl).
Compression ratio 4.93:1
Power: 200-230 Hp @ 1400 RPM @ sea-level.

There´s no additional information to justify the power-range. Austro-FIAT also built this model, of which I found a British report - they assumed it to produce 220 hp. Curiously, displacement was 1014 cu.in (169 cu.in. per cyl), so the more general information before is somewhat incomplete - they did increase displacement a bit to get the higher power.

The problem in the .air file is that RPM and Hp readings are impossible to get right simultaneously. No manipulation of propeller efficiency/thrust or engine torque/friction or Zero Lift Drag parameters produce tallying results. Either RPM are too high, or Hp too low.

Experimenting with displacement of 200 cu.in. per cylinder seems to balance Hp and RPM better.

Looking at specifications for other similar engines, those producing 230 hp normally have greater displacement, and alternatively, with similar displacement, power is lower. For example:
Austro-Daimler-6: 917 cu.in, comp. 5.00:1, giving 200 hp @ 1350 rpm
Benz IV 6-cyl : 1149 cu.in, comp. 4.91:1, giving 230 hp @ 1400 rpm

2) Angle of incidence: 6 degrees! I fear this is too much for the .air file to handle, as RoC is 3 times too high. Aircraft performance is more correct if I enter 3 degrees for the AoI.

What would be your opinion about changing the specs to achieve a more fitting performance?
Take your time with the answer, I know you don´t have much time lately, and there´s no hurry!
Cheers, and thanks in advance.
Aleatorylamp

[Ivan]

Hello Aleatorylamp,

My opinion is that all of the parameters in both your engine AND your wing angle of incidence can be met.
The aeroplane will have a very odd CL Graph (Table 404) but I believe it can be built.

I need a little time to prove the engine part. The airframe and angle of incidence requires a LOT more information than you supplied.
What would the idle speed be? I am going to put it fairly low but will not try to tune it since I don't have a number to aim for.

- Ivan.

[No Dice]

Aleatorylamp or anyone,

Is it possible to change the gray to white on Aleatorylamp posts, very hard to read.

Dave

[Ivan]

Hello No Dice,

You are using CombatFs/CFC default colour scheme (Lower Left at bottom of the page).
SOH scheme works better for me.

- Ivan.

[aleatorylamp]

Hi Ivan,
Thanks for your post, I didn´t expect it so soon! I hadn´t thought of CL Graph Table 404!!
OK, so the parameters are not "out of scope" for the sim, which was the main reason for my query.

Here is some more information:
1) An additional deduction:
If the displacement was 943 cu.in. for the 200 hp version, and 1014 for the 220 hp version, then for the 230 hp one it would be 1121 cu.in, which would make it easier to fine-tune.
2) Prop diameter was 10.2 ft. and idle speed 450 rpm (yes, fairly low!).
3) Restly specifications for the plane are
-Max. speed: 85 mph (74 kt) at sea-level,
82 mph (71 kt) at 3000 ft
75 mph (65 kt) at 9000 ft
-Ceiling: 15000 ft loaded with 770 lb, 21000 ft unloaded
-RoC to 3000 ft: 428 fpm loaded, 500 fpm unloaded
-average RoC to 9000 ft: 150 fpm loaded, 333 fpm unloaded

-Weights:
Dry weight: 6833 lb
Fuel: 250 USG (1500 lb)
Medium Bomb load of 330 lb
Ammo: 2000 rounds, 156 lb

Oops! Edited update:
-Wing:
Span: 933 ft
Area: 966 sq ft
Chord: 188 (biplane)
AoI: 6 degrees


Thanks a lot for your help! I´ll see what I can do with Graph 404
Cheers,
Aleatorylamp

[Ivan]

A Wingspan of 933 ft ???

I think you're going to run into problems there.....

- Ivan.

[aleatorylamp]

Hi Ivan,
Oh, dear! Sorry: Inches... that´s what it was: 77 ft 8 in.

The tweaking of Graph table 404 is already going quite well, and I´m getting better RoC´s, and I´ve been adjusting the FD to get a better approximation to the performance I´m aiming for.

Present data for 500 ft:
Hp: 230 (perfect)
RPM: 1480 (too high, should be 1400)
Horizontal speed: 74 ft (perfect)
RoC: about 560 fpm (Good)

Performance for 3000 ft and 9000 ft seem to be quite coherent as well.
The only way I can think of to get the RPM down at the moment, is to increase the cylinder displacement, but from what you said, there must be a better way!

I was wondering if it is a good moment to send you the plane for you to have a look into the FD, or perhaps it will be better to leave it for a while until you have more time on your hands.

Cheers,
Aleatorylamp

[Ivan]

I guess I don't need to do any engine testing after all if you already have the right numbers.
As for dropping RPM, how does increasing the displacement do this????
You would want to increase the propeller power coefficient instead, but probably not by very much at all.

The last two evenings I turned on the development computer but got caught up in chores and just turned it off again a few hours later without having done a thing.
The Ki 61 Hien just waits patiently as it has since 2006.

One of the things I believe I should tune is maximum propeller efficiency. The stock P-51D hits 90% or 91% at its peak.
I believe the proper value should be no more than about 85% maximum, but adjusting this causes a whole mess of changes in a lot of places.

- Ivan.

[aleatorylamp]

Hi Ivan,
OK, then, it sounds like it won´t be too difficult then, and I won´t have to bother you!


What I´ve experimented, is that by increasing displacement, Horsepower goes up more than RPM, and later, to reduce power to the correct Hp again, I increase prop thrust, and maybe also friction, and then RPM comes down further, reaching a more correct value.


I am also using the propeller power (thrust) coefficient, to adjust things, but that also affects Hp - that´s why it´s a bir difficult, but I´ll get there in the end!


Edited update:
I´ve just tried a small prop-thrust-coefficient increase again, combining it with a slight friction reduction, and this time I´ve managed to reduce the RPM by about 50 RPM. I´m almost there! Interesting that you insisted on small adjustments - Good! - maybe I won´t have to cheat with the Cylinder Displacement after all!
By the way, I also noticed that if the thrust coefficient is reduced too much (below .03700) it starts having the opposite effect, and actually reduces power/rpm instead of increasing it further - which I suppose is coherent too.


Cheers and thanks a lot!
...and good luck with the Ki 61 Hien!
Aleatorylamp

[aleatorylamp]

Hi Ivan,
It worked! Just like you said! (you´ll think: "Of course...")
Wow! As per specs: 74 kt, 230 hp, 1400 rpm, at sea-level.

Thanks again for your never-ending patience and help!
Cheers,
Aleatorylamp

[aleatorylamp]

Hi Ivan,

I´m adjusting the AoA vs. lift graph, and have of course found that raising or lowering different points on the graph affects rate of climb. However, I don´t understand exactly what I´m doing, as the AoA´s indicated in the Beckwith control gauge and the AoA in the graph table don´t coincide.

When the plane is in horizontal flight, at 500 ft, the AoA reading is 57 degrees, and at 10000 ft, requiring pronounced trim-up, the AoA reading for horizontal flight is 77.7 degrees.

Presumably, adjusting CL for the higher AoA´s can help differentiate RoC´s for higher flight levels as opposed to the lower ones, as a more pronounced AoA is needed for level flight higher up.

Then, the reading is zero in a pronounced 45 º dive...

I wonder if you could perhaps clarify what exactly the AoA on the Beckwith guage measures.
In the beckwith Gauge, stall happens more or less after 100 degrees, and in the Graph table, it is just after 16 dregrees. I´m afraid I´m a bit lost with the numbers here!

Thanks very much in advance for your help!
Cheers,
Aleatorylamp

[Ivan]

Hello Aleatorylamp.

From what I can tell, the AoA gauge does not appear to work and I have never used it.

- ivan.

[aleatorylamp]

Hi Ivan,
OK thanks, no problem!
Cheers,
Aleatorylamp

[aleatorylamp]

Hi all!
Here´s a link to the new upload:
http://www.sim-outhouse.com/sohforum...id=19&id=19654
I hope you like it!
I´ve just uploaded a 1918 Austro-Gotha G.IV . It is a revised version with guns and gunners, CFS1 .air file, Dp files and SCASM-corrected virtual cockpit view.
These machines served with the Austrian Airforce in Aviano and had a pair of Austrian 230 Hp Hieronymus engines, with which they managed to cross the Alps even though they were not of the high-compression, high-altitude engine type as the Mercedes or Maybach ones.
Enjoy!
Cheers,
Aleatorylamp