Possible new Caudron G.4 Bomber/observer

[aleatorylamp]

Hello Folks,
I was trying out the 1915 Caudron G.4 bomber-observer model I had made for FS98 twelve years ago, testing it for
bleeds, and despite all the struts, spars and wires, it seems they are within manageable limits, possibly even
being improvable.

I´m planning to rig up the Caudron G.4 for CFS1 by putting in the two swiveling Lewis machineguns it had, one firing
forwards and one backwards, and equip it with the bombload of 3 internal 83 lb bombs. The guns make this bomber
an excellent candidate to be used with TG2, so it should be fun.

As per one source:
" ... Bombs were dropped by the observer through metal trapdoor shaped like door of a rolltop desk. Observer, using
the "oculaire" or eyeball technique, could ease the door back to take pictures or drop 3 slender bombs that were hung
on leather straps in the cockpit. ...".

Then, from a technical point of view, this slow-flying aircraft (82 mph top speed) had very interesting engines, which
work nicely in the CFS1 .air file. They are Le Rhône rotaries, whose animation also comes through very well.

Driving 10-foot-diameter, fixed pitch wooden propellers, the engines were rated at 80 Hp at 1200 RPM, although factory
specs state that maximum performance was 92 Hp at 1300 RPM. The necessary engine and propeller adjustments in the
.air file yield a very satisfactory result at S.L. of 82.3 mph with 92 Hp at 1301 RPM and 28.9 Hg.

Its climbing and altitude performance (13000 ft ceiling) was also very appreciable for the time, and this also comes
through very well in CFS1.

Here´s a screenshot.
I expect there could be some interest for an upload...
Cheers,
Aleatorylamp

 

[hertzie]

Looks & sounds good! Great there are still CFS1 designers! Hope you'll manage to give it a working vc.

Thanks for all your good work, MUCH appreciated!!

hertzie.

 

[aleatorylamp]

Hello Herzie,
Thanks for your interest.
The Virtual Cockpits I can make don´t have working instruments.
The panel-texture is only the bitmap made from a panel screenshot.
I´m still working on a few elements to clean up the VC view, and the
observer´s head will be made to turn with the rudder, maybe!
Then, I´m also planning to put in the propeller blurs like the ones
that Ivan makes.
Here´s a provisional pic.
Cheers,
Aleatorylamp

 

[hertzie]

Hi Sir,

You mention a "thing" that I've always been curious about: how come that original Microsoft CFS1 aircraft, as the Spit and the Hurri, have working VC-gauges, and so many "non-Microsoft designs" (hope I'm obliged to call them so) don't have them ?

Any ideas?

All the best,

hertzie.

 

[aleatorylamp]

Hello Herzie,
I didn´t answer before as I´ve been busy re-arranging and re-building the Caudron G.4, to get
enough parts free for the machine guns. It isn´t easy because of the rotary-engine cowlings, that
are all individually displayed insignia parts, (otherwise the engine-animation display won´t work),
and this lowers AF99 compilation limit.

Also, cowlings still require a concave component visible from the inside, and there weren´t any parts
left over for this. However, rationalizing construction by putting all the different wires into the strut
components, (at the cost of losing the different colouring), as well as sacrificing animated control-surfaces,
freed enough parts and components, so I´m making a lot of headway.

Anyway, regarding your comment on Virtual Cockpit Working Instruments:
Strange... Upto now I hadn´t seen details or comments on the subject, other than the general comment
you made some time ago.

I remember NoDice making some beautiful gauge bitmaps, and Ivan programming the N2 gauge for the
He-162 "Sparrow" (Eeeek! A jet!!). Then, Ivan also programmed a very complete set of rather excellent
multi-engine gauges, and Smilo has done good work on panel instrument layout, but upto now, I don´t
recall any talk about functional VCockpit gauges, so I don´t know if anyone has investigated if this is
possible with CFS Add-On aircraft.

It sounds like an intriguing line of reasearch. One only clue to start off with seems the following:
There is a paragraph present in all CFS1 stock aircraft panel.cfg, which is absent in standard FS98 ones.

It is labelled [Vcockpit01], sets up some viewing-window parameters, and locates the gauges, which
seem to be the same gauges as for the standard panel.

Of course, there is no panel bitmap, as [Vcockpit01] uses the virtual cockpit view. Also, obviously,
gauge-location coordinates are different due to the larger field of view.

Initially, one could expect that just copying the panel folder of a stock aircraft into a non-stock
aircraft would work, but it doesn´t. Preliminary tests I´ve done just show the virtual cockpit view
corresponding to the model in use, without any instruments.

I wonder if anyone knows anything on this subject.
Have you tried any strategies yourself?

Cheers,
Aleatorylamp

 

[aleatorylamp]

Biplane Struts / Wings bleed problem

Hello all,
There´s one specific type of bleed-problem I´ve been trying to deal with for years now, again and again,
to no avail, namely the interaction on biplanes between wingstruts and landinggear struts, and the wings.

The best solution is always to place the top wing in Canopy-High Wing, and have the struts in Wing Mid left/right
and/or Inner Wing Mid Left/right, and the lower wing in Wing Low.

However, there always seems to be one or two angles at which what is shown in the screenshot happens, and as yet,
it seems totally unavoidable. Additionally, the more the wing curves downwards, (obviously), the worse it becomes.

Other groupings with glue-sequencing without using Canopy High-Wing, can prevent this problem, at the price of worse
things happening with other elements such as engines, cowlings, and also struts, but in different ways.

So, it seems that we are landed with the lesser evil here, and some viewing angles simply MUST be avoided. A pity,
because otherwise, these old vintage contraptions come out looking quite good in the simulator. Too bad, ...unless of
course, there is something I´m missing, that perhaps someone has succesfully tried out that I haven´t!
...but then, not many build biplanes nowadays...

Cheers,
Aleatorylamp

 

[hertzie]

Hi Aleatorlamp,

Yes, I've tried a lot of things without any results. Like crazy things as putting the complete Spit panel map in a WW1 plane, which gives a perfect 2D Spitpanel, but the planes original static VC. Or messing around with the HEX-editor, also no succes. So there must be something that has to be done in the original mdl.-design, I suppose, but as I am not a designer I have no idea and as we don't have the original Microsofts sourcefiles I suppose we'll never know. But I'll put a post on the designers forum, you'll never know what will happen.

And I have a plea, hope you're not getting mad at me, but would you please be so kind to change the backgroundcolor of your posts? It is so hard to read because it's light grey text on a little bit darker grey background, so almost no contrast.

All the best,

hertzie.

 

[aleatorylamp]

Hello Hertzie,
My posts are all dark-grey text on white background, so that´s OK here.
However, I believe it could be something to do with your user settings.
Perhaps Smilo can explain how to put that right.

As regards V-cockpit working gauges, I´m afraid I haven´t got a clue about them.
Sorry!

Anyway, I found a different livery for the Caudron G.4, with more appealing colours.
No. 7 appears to have been quite a famous unit from Escadrille 47, together with No. 9.

Here´s a screenshot. Slowly the bleeds are getting better too, but I haven´t got enough
free parts for the drum on the Lewis guns. With Parts Count at 148.8 now, it will be difficult,
but at least with the more rationalized construction style now, I got beyond the 145% AF99
compilation limit, which is quite satisfying.

Cheers,
Aleatorylamp

 

[smilo]

my guess is you're in CombatFS/CFC view mode.
try this, scroll to the bottom of the page,
in the lower left corner click on the Quick Style Chooser window
and select --SOH Default(fluid)

 

[hertzie]

Thanks Smilo! Had never noticed the key nor that I was in a different mode.

hertzie

 

[smilo]

my pleasure, sir...enjoy

 

[aleatorylamp]

Getting there...

Good morning, folks!
I´m glad that worked.

Regarding the two machineguns on this machine, I found some rather interesting historical data:

"To protect against attacks from behind, some G.4s were fitted with an additional gun mounted on
the top of the upper wing and pointed rearward, but this proved to be ineffective, and it was frequently
removed from operational aircraft. A number of G.4s had a second gun mounted immediately in front of the
pilot on the deck of the nacelle, but more often the pilot and observer simply carried hand-held weapons
to respond to attacks from the rear."

Furthermore, and the unit with the red and white 7 on the cowls only had the nose machinegun.
So, I had enough parts left over to make the drum on the Lewis-gun.

For TG2 then, I´ll enable a rearward-firing automatic parabellum hand-gun to be used from the pilot´s position.
Here´s a new screenshot now.

Cheers,
Aleatorylamp

 

[hertzie]

I did put a post on the designers forum with my question about non-working VC-gauges, but so far no replies.

hertzie

 

[aleatorylamp]

Hello Herzie,
Hmmm... It´s probably something nobody knows anything about (yet).
The strange thing is that stock aircraft have it, but add-on aircraft seem to be excluded from
this feature - at least I haven´t seen any with it.

I don´t think it´s something managed by the .mdl file. Looking into it with SCASM de-compiler,
I haven´t seen anything related to panel instruments, so it is probably something managed by
the CFS .exe file itself.

For a VC, the .mdl file deals with model parts made by the modelling program, and doesn´t do
any calls to gauges. It would perhaps seem strange that stock aircraft .mdl files were to be
so different from add-on aircraft .mdl files.

On the other hand, stock aircraft .mdl files have a different format which is not recognized by
SCASM, so maybe the issue IS something to do with the .mdl file, but stock .mdl files seem to
be a closed environment.

I wonder...

P.S.
I just tried writing in a [Vcockpit01] paragraph into an add-on aircraft panel.cfg, and as
was to be expected, it didn´t bring anything. Then, I tried deleting the texture line in the
paragraph - for the Stock P47d, it´s: "texture=$P47d". This blocked out the view of all the
VC gauges with individual boxes, so one is led to believe that this line makes transparent
windows for the gauges. Then, how to define the texture lines that start off with the symbol $
for the [Vcockpit01] paragraphs, is also a mystery.
Strange things, the stock aircraft VC´s.

Cheers,
Aleatorylamp

 

[aleatorylamp]

Hello Folks,
It´s nearing completion - so here are some more pics.
Several bleeds with the tailplane, fins and wires were solved by re-instating separate elevators and rudders,
which in turn has allowed animating these control surfaces again, at no extra parts cost, as they are all
structures here.

I had to give the aileron animation a miss though - on the real airplane it was done with wing-warp! Like the
Wright Brothers. Not so easy to do with AF99, I´m afraid. I´d thought about moving a large triangular slice on
the end of the wing, but the parts aren´t free anyway.

The model is now packed with 30 structures and 30 components, and parts count is at 149.7%, by rationalizing
contstuction even further, and also checking some more vertices on all the struts and wires.

Some bleeds are unavoidable because of the way the original is built, with some parts overlapping into others,
and there just aren´t enough components or structures available to separate them all.

Anyway, I´m just working on the RoC. It should average 492 fpm upto 3200 ft, going down to 327 fpm upto
6500 ft, and at the moment it´s at 535 fpm and 486 fpm, so I have to try and adjust the corresponding graph
point here. Nevertheless, service and absolute ceilings are correct at 13100 and 14100 ft.

Anyway, hopefully everything will be ship-shape for an upload soon, or rather airplane-shape, as soon as I´ve
done the SCASMed VC-view, where I hope to include an extra rear-firing gun-view for TG2 add-on programme.
This should be fun for more than one simmer!

Here´s a shot of the VCockpit too, zoomed out a bit, with a few bleeds corrected.

Update - Engine control:
Actually flying this aircraft is quite curious. Being a rotary engine, there´s really no throttle, and once started, it
revs up to full-power. For descent, slow-down and approach to landing, they did what was called "blipping", i.e.
switching the magnetos on and off.

Now, there being only single-engine magnetos available, which is visible on the panel for the simmer´s, the engine
has to be "blipped" by pressing first the "M" and "-" keys simultaneously until the Magnetos are switched off, and
then "M" and "+" to switch them on again. This has to be done several times on the approach path to reduce speed.

I could of course make it easier and do it the conventional way as for normal engines so as not to make it a pain
for simmers, but that would be cheating, wouldn´t it?

Cheers,
Aleatorylamp

 

[aleatorylamp]

Performance issues

Hello all,
As usual, there was contradictory information on propeller diameter, depending on the drawings, ranging from 7 to 10 ft. Nothing was quoted on the pages about the aircraft, even less anything on pitch angles. I used 20 degrees and 10 ft to start off with, but then I found some vintage wooden propellers being sold, and exact details were given:

The Caudron G.4 used 2.5 metre diameter, 1.9 metre pitch propellers, which translates to 98 inches and 19.14 pitch angle at 70% blade length.

Interestingly enough, the single-engined Caudron G.3 had the same engine, but was slower. It used a 2.6 metre diameter, 1.8 metre pitch propeller, i.e. 102 inches diameter and 17.53 degrees pitch.

Further work on the propellers, to reduce the excessive RoC I was getting on the previous ones, proved futile, as was to be expected, and revealed the usual dilema: Adusting for correct level flight performance at specified altitudes and ceiling, came at the price of excess RoC. Alternatively, adjusting for correct RoC and maintaining S.L. performance, came at the price of very poor altitude performance.

The propeller´s advance ratios for the different specified speeds are so close together that speeds overlap, but as RPM are lower at altitude, it was possible to make adjustments to the Torque Graph, to separate the powers for lower RPM. This fortunately helped correct ceiling performance and RoC higher up.

However, at low and mid altitudes, level flight RPM coincide with climbing RPM, so here it is either one or the other. I chose to go for correct level flight performance, leaving RoC a bit on the high side. The flight envelope is more correct this way, I think, and maintains the airplane´s general reputation at the time for good performance, especially at altitude. Anyway, RoC is only about 40 pm in excess, which is not all that bad.

Thus, performance results have come out quite well, I would say.
Cheers,
Aleatorylamp

 

[aleatorylamp]

Model Performance

Hello Folks,
Here is some tangible data on real performance compared to model performance, also taking into account data variations between sources:

Performance Specifications:
---------------------------
Vmax:
Sea Lev: 82 mph, 92 Hp, 1300 RPM
6500 ft: 77 mph.

Service Ceiling : 13100 ft.
Absolute Ceiling: 14100 ft.

Rate of Climb:
Av. RoC from S.L. to 3280 ft: 492 fpm.
Av. RoC from S.L. to 6500 ft: 437 fpm.

Model performance:
------------------
Level Flight:
_300 ft : 82.0 mph, 92 Hp, 1300 RPM (J=0.672)
6500 ft : 76.4 mph, 63 Hp, 1223 RPM (J=0.670)
13000 ft: 66.3 mph, 39 Hp, 1084 RPM (J=0.656)

RoC anywhere from 300 ft upto 8000 ft:
- With Autopilot RoC set at 550 fpm: 535 fpm, (at 92 to 56 Hp, 73.7 to 57.7 mph, and 1230 RPM to 1038 RPM, depending on height).
- With Autopilot RoC set at 500 fpm: 488 fpm, (at 92 to 53 Hp, 74.5 to 56.6 mph, and 1238 RPM to 1018 RPM, depending on height).
Note: This is a bit high, but unavoidable if other performances are to be maintained.

RoC at 13100 ft Ceiling: 130 fpm, at 57.5 mph, 35 Hp, 983 RPM.
RoC at 14100 ft Ceiling: 33 fpm, at 61.9 mph, 31 Hp, 985 RPM.

At the moment I´m working on the Virtual Cockpit with SCASM, to put in the Gunner´s view when the spoiler key is pressed, leaving the default view as the pilot´s normal position in the rear cockpit. Update: I´ve been at it with SCASM all afternoon, but it´s not working yet - I had it working on the Ju52´s though, which even had two extra views, so this one should be simpler with only one. I´ll get it in the end, though.

Unfortunately, there will not be any working gauges.
However, at least it will be fun with the TailGunner Add-on Program "TG2", the pilot and gunner can (hopefully) be made to fire their automatic hand-guns in all directions, by the simmer pointing with the mouse at the selected enemy plane in Cockpit Chase View.

I´ll be uploading the model when I finish. As always, any suggestions, ideas and comments are welcome.
Cheers,
Aleatorylamp

 

[aleatorylamp]

Two VC views

Hello all!
I managed to get the two virtual cockpit views in for the different gun positions:
One for the pilot and one for the fore-gunner. I also found a much better way to
toggle between the two views.

The Nav-light switch "L" does not interfere with anything. It is more convenient
than selecting with the landing-gear and/or spoiler button.

It took a while to figure out the different SCASM code required, but works well!

Now I have to clean up the fore-gunner cockpit view, to correct the display order for
the different elements seen from there, i.e. struts, wires, cabin-floor, engine cowls,
pilot´s head, windshields, fuselage spars, tail, etc.

Anyway, it is coming out very well indeed!
The Pilot´s VC view appears to be perfectly fine, and requires no extra work.

Here are 4 screenshots of the pilot´s view for eye-candy that I know some like very much!!
I find biplanes always have exciting virtual cockpit views.

P.S. It looks like I still have to fix the texture bitmap for the rear engine nacelles...
Cheers,
Aleatorylamp

 

[aleatorylamp]

Spanner in the works

Hello, Folks!
The engine animation display isn´t satisfactory yet seen from the rear. Viewed from the front and from the sides,
it is rather effective-looking, but from the rear, despite the good rotating engine animation, the cowl-panels disappear:
Their inner-view is not possible as panels are insignia parts facing outwards - otherwise the engine animation won´t display...

A partial remedy with a "collection" outboard half-cowl component showing the inner cowling-half in rear view. The virtual cockpit view was
also OK, but there were short bleeds in the cowlings tops seen from the front, so the remedy was worse than the ailment.

However, parts-count was "only" 147.6 %, without AF99 compilation or SCASM problems, although the limit wasn´t far off.

Eventually, I found an elegant solution: Closing the rear-cowling with dark-grey panels was a clean way of solving the problem,
keeping a good engine animation seen from the front and lower-rear. Finally, closing the cowl-bottom with a panel displays everything correctly.
Parts-count is lower, at 145.2%, and there´s no AF99 compilation problems, but it gives an error

loading into SCASM!!
Backtracking, I also discovered that only slightly simplifying the previous well working, there´s also a SCASM error!

Thus, I´ll I have to back-track further, to a previous build, to implement my latest improvements that look better.
It SHOULD be possible to get SCASM to accept the build!

Update: Modifying an earlier, well-working model build with a some of the recent modifications, but not the latest ones,
also causes a SCASM loading error, so I´m closing in on the culprit.

It will be a while yet...
Cheers,
Aleatorylamp

 

[aleatorylamp]

"Redo from Start"

Hello all,
Strange, SCASM rejecting the Caudron .mdl files.

It happens with any of the new builds, except the one I was working on for the VC,
as long as that one isn´t modified any further. Of course this is of no use, as I have
to modify it anyway to improve the engine animation.

All earlier progressive modifications are also rejected, and it seems like the only one
that works in SCASM is the original FS98 build I started out with, which must be improved.

I also thought perhaps some components that include wires with the struts were too
complicated for SCASM, but taking all the wires out of them did not help either.

So it looks like it is a matter of "Redo from start", like the option the old Fdisk program
on the floppy had, when you were partitioning a HDD!

No rest for the wicked!

Afterthought:
"Redo from start", i.e. modifying the original build with the necessary modifications
will expectedly lead to the same result - i.e. SCASM error.
Thus, this job will require modifying the modifications,
to circumvent SCASM reluctance.

Cheers,
Aleatorylamp