pontoon angle alignment with craft and ground

[wkitty42]

is our CG a little off when wearing either of the floats or are our struts off just a little? i've been looking at some pics of the 172 with floats and those seem to sit much flatter and level than this craft... check this image at the wipaire site... http://www.wipaire.com/Wipaire-Dev/piwigo/i.php?/upload/2012/06/18/20120618121559-882f7b8a-sm.jpg it just seems like our craft is a bit more nose up when wearing the floats... i notice it all the time when i can't see the area in front of the craft in the sim... i'd like to find some photos of the craft just sitting on the water to compare with... all my flying and testing have been done with no baggage or additional weight... just the 180 pound pilot and whatever fuel is standard to start with...

[wkitty42]

here's a shot from the sim... the floats and craft angles are off compared with the above referenced picture... floats are angled down and the craft nose is raised... can this be fixed easily and ready before the freeze?

[dany93]

Currently, that is a pure effect of the gear height. Geometry. The CG cannot be off because (I think) you do not still have their weight taken into account (unless you are testing the pull request https://github.com/Juanvvc/c172p-detailed/pull/212#issue-80878085. The CG position in written in Fuel and payload. Check if the weight changes and if the CG moves with floats or not. Read above your message here https://github.com/Juanvvc/c172p-detailed/pull/212#issue-80878085, it is the subject of this pull request.

If there is an off angle, it can be fixed (by the gear height). But, differently from you, I rather feel that the aircraft nose in the sim is lower than your picture.

[wkitty42]

yes, i thought later, after i wrote this, that it isn't the CG... the nose gear needs to be a little longer to make the top of the pontoons level with and parallel to the ground... the front mounting struts should also be a little shorter to lower the nose... the pics i've looked at and the one i linked to earlier show a more parallel set of lines between the top of the pontoons and the bottom flat area of the craft from the rear of the door opening to the engine compartment... that instead of the parallel portion being of the tail behind the door opening and the pontoons... i'll try to prepare a pic if needed IF we really should take care of this... i noticed it some time back but have only now just gotten around to mentioning it :( sorry for that :( :( :(

[dany93]

Yes, now I understand what you mean for the levelled pontoons (should be more parallel to the ground) and the angle between the craft and the pontoons to compensate. But wlbragg has done (and is still doing) such a huge work on the pontoons that these details can (currently) probably be considered as secondary.

[wkitty42]

yes, he is doing a fantastic job... i wish i had half of the ability that his little finger has...

FWIW: i also mentioned this because i thought maybe taking care of it sooner than later would be better than having to redo anything else that already may need to be handled when it is all incorporated together... but i realize now that maybe i should have said something sooner when the pontoons were first brought out...

[dany93]

You may be right for the "better sooner than later". Especially because this can have an effect on the wing angle of attack at acceleration for taking off from the water, if the pontoons floating pitch is matched to the water. But by reading the current forum http://forum.flightgear.org/viewtopic.php?f=4&t=25157&sid=bcdc03969281f82aef38f451b4e8037c&p=244634#p244625 I wonder if there will not be important changes...

However, we have to keep in mind that the FDM and the 3D model are two different things. Independent. We naturally superpose the 3D model on the FDM but we could have a 3D wheelbarrow fly if we wanted.

[wkitty42]

exactly on the "sooner" part... yes, i've read the forum topic and i feel the same about important changes... i just hope that everything can be taken care of before the freeze... i'm kinda wondering about that, too... is that just a code freeze for the sim or for the base craft also? yes, that's a question better asked on the dev list... i guess i'm going to have to join that sooner than anticipated... i've been reading it via the web archive interface :laughing:

[dany93]

I guess the freeze will contain the features which work, with maybe a few bug fixes. IMHO, the aircraft as it is now is already worthy enough to be in a freeze. I think that in some time, a version "almost freezed" will have to be chosen, with only bug fixes to come on it. With a branch for the development version.

[Juanvvc]

I agree with a branch for v3.6 ("the freeze", with only bug fixes) and keep the master branch in development.

[wlbragg]

I was never happy with the front gear on amphibious. I tried to get the to full extension but it wasn't working, so that is part of it and yes it needs to be at full extension (in other words I think it has more travel to play with without any modeling edits. To compensate and drop the nose lower after we fix the gear extension, modeling wise, it would be be relatively simple to extend the top of the back struts. I would really like it if someone eventually gets to this.

How come we need to make a new branch instead of just tagging a specific commit as the 3.6 release? I don't have a problem with it either way, I'm just curious why we need another branch if the GitHub system is designed to tag a point in time (commit) as a release. Unless it is because we want to remove superfluous files like blend files from the new "release" branch. Either way is OK with me, I just hope we all agree to do it SOON!

[wkitty42]

I've renamed this issue to better fit what the real problem is... I don't know how to fix this so I have to leave it to others...

[andgi]

This alignment has a potentially large impact on take-off performance both on water and on land since it sets the pitch range available during the take-off run. If a reference on how floats are to be rigged on a c172p can be found it may provide an answer.

[dany93]

We at least have this drawing in the POH. http://www.aerodynamicaviation.com/documents/cessna-172p-1982-poh.pdf Section 9, SUPPLEMENTS FLOATPLANE, Section1, Figure 1 p.4 of 42 (PDF p.198/357). By this image, the angle does not seem so bad ? But these floats seem smaller than wlbragg's (this said for the weight and glide-ratio)

[wkitty42]

@dany93 yes, there is a difference... that pic in the manual is for floats and i was looking at the amphibious on the wipaire site linked in my first post... looking at it again, it appears that the front caster wheels are lifting the front of the craft and leveling the pontoons so their top is parallel with the ground... aside from that, i was looking also at that picture for the angle between the top of the pontoons and the belly of the engine and seating area of the craft... if it is decided that this needs some adjustment, we can hold off until the next version if it will be too much before the freeze... it was just something that i saw different and thought to bring it up for discussion...

[dany93]

I think that the important is, as you wrote, the angle between the top of the pontoons and the belly of the engine and seating area of the craft. To be good on the water, it is common to the pure floats and amphibious version. I guess the top of the pure floats or amphibious floats is not far from parallel to the water surface. If this angle is good, the top surface must also be parallel to the ground too with the caster wheels. The same amphibious give the angle of attack of the aircraft on the ground and on the water (presuming that the angle of attack is the same for ground and water acceleration before takeoff). As pointed by wlbragg, it is a matter of height for the front caster wheels on your FG picture.. Not easy for the 3D drawing of the floats themselves, a sufficiently accurate representation of the reality is necessary.

[wkitty42]

ok... so... we hold this issue open for future visitation and evaluation or?

[dany93]

The solution does not depend on me, but I would let it open.

[wkitty42]

yes, i'm leaving it open for more comments from everyone else...

[wlbragg]

I think we need to start with the front caster wheels and get their compression and extension adjusted to the correct specs. I really don't know how to do that. Once that is correct then we can look at the angle of attack of the aircraft. I think the angle of attack will be a simple fix. I'm pretty sure it will only require the extension of the two back strut attachment points to get the angle how we want it.

[wkitty42]

@wlbragg that's my thoughts, too... perhaps a closer look at some amphibious pontoons can help?

[dany93]

I think we need to start with the front caster wheels and get their compression and extension adjusted to the correct specs. I really don't know how to do that.

I do not understand the problem. These gears are boguey 21 and 22. The spring and damping coeff are in the FDM (lines 416 .....). In the FDM, their Z-position is 0, but it seems that you set it in physics.nas

var amphibious = func
{
    resetcontacts();
    resetalldamage();
       (......)
    setprop(gears~"unit[21]/z-position", -54);
    setprop(gears~"unit[22]/z-position", -54);

(Not sure this is truely set here, I don't understand well. Fix me, please....) Why not set it by

fdm/jsbsim/gear/unit[21]/z-position
fdm/jsbsim/gear/unit[22]/z-position

? It can lift the front of the amphibious, after that is a matter of 3D in Blender, then animation, to have the wheels just on the ground? With z-position = -60 the top surface of the amphibious is almost horizontal. You will probably have to decrease the length of the front struts to get the aircraft pitch angle. I know that it can be a true work in Blender....

But I may not have understood well....

[wkitty42]

i think you have it, @dany93 ... decrease front struts or lengthen back ones... either way will adjust the angle of the craft in relation to the pontoons... the question is which one to do to have proper clearance between bottom of spinning propeller and top of pontoons... yes, i know the prop is in between them but... :wink:

[dany93]

Yes and no. I rather understood that wlbragg had a problem at setting the compression and extension values for the front caster wheels in the FDM. I trust him for Blender.

[wkitty42]

yes, the casters is one... fixing that should level the pontoons... then the craft itself needs to be adjusted so it isn't so nose-high (too high angle of attack?) when the pontoons come up level...

[wlbragg]

"I do not understand the problem. These gears are boguey 21 and 22. The spring and damping coeff are in the FDM (lines 416 .....). In the FDM, their Z-position is 0, but it seems that you set it in physics.nas"

I'm not sure I understand either. I don't set any of the gear bogey points in the FDM because we don't know for sure what gear configuration is going to be used at any point in time or at startup.

As far as the front amphibious gears not being configured correctly, I don't think it can be fixed with the z-position, I think the problem was the weight VS spring/compression settings of the parts of the gear. It compresses them too much just sitting there.

It's been awhile since I experimented with it and I have forgot exactly what my issues were with trying to adjust them. Let me take a quick look at them again tonight and play with their settings so I can refresh my memory and better explain why I couldn't get them to extend the way they should.

[wlbragg]

OK, I just got through looking at this again. The issue is the spring_coeff and damping_coeff for both the front and rear amphibious gears. They are to "soft". They are gear "bogey" points 21-24. If I extend some other contact points 13-16 to take the weight off of the gear you can see the peak extension that they should be able to obtain. I know they should be under some load but not anywhere near what they are now.

Now if I remove the extra contact points and apply all the pressure on the gear you can see the compression on the gear is too great.

The other thing at play is the complication of watching for which type of gear and which type of surface those gear are on. It makes a difference on whether or not I need to use other contact points in addition to the gear to control wind drift or ground friction. That is why I control it all in physics.nas.

Then there are another issues at play that had to be taken into account. One of those was the gear would break just by the force applied to them while switching from one type to another. So I created a delay loop to allow settling out of the aircraft before I allowed the damage loop to be active.

Then when I added the amphibious gear, I guess because of the amount of spring_coeff and damping_coeff, I found an issue where it went bonkers when extending or retracting the gear, I had to use additional contact points to steady the aircraft while the gear traveled at which time it finished its travel I could remove the contact point.

This is all done in physics.nas and the more I added the more complicated it got to the point where I started hacking solutions instead of using clear, well thought out logic, in some cases. The amphibious gear was one of those cases.

But really I think the main issue with the amphibious gear is the spring_coeff and damping_coef settings. They need to be able to handle more weight, I think.

I used contact points 13 and 14 to take some of the pressure off of 21 and 22, that is a HACK and I am surprised it works as well as it does, but is not right. setprop(contact~"unit[13]/z-position", -55); setprop(contact~"unit[14]/z-position", -55); setprop(contact~"unit[15]/z-position", -22); setprop(contact~"unit[16]/z-position", -22); setprop(gears~"unit[21]/z-position", -54); setprop(gears~"unit[22]/z-position", -54); setprop(gears~"unit[23]/z-position", -48.5); setprop(gears~"unit[24]/z-position", -48.5);

[andgi]

If you consider merging the hydrodynamics ( https://github.com/andgi/c172p-detailed/tree/hydrodynamics-wip ) for the coming release it might be worth looking at solutions to the floats on water/land and alignment problem from that perspective.

Yesterday I removed the float contact points 13-20 and kept the float landing gears (formerly 21-24, now 13-16 in hydrodynamics-wip). This seems to work ok on land and water, apart from two issues: i) the floats float if the gear is retracted on land; and ii) the non-retractable float nose gears affect the water performance if they touch the water (this is a rather new feature in JSBSim). Having structure points that can be moved down to the keel when on ground might be a solution to i). Another missing part is detection of too large forces to trigger float damage (damage itself is partially implemented - set the damaged or broken property and the float will leak and eventually sink).

[wlbragg]

"If you consider merging the hydrodynamics" Maybe you've completed some more work on this, but the last time I tried it there were several issues with it. Unfortunately I can't remember what they were. They may be small things that can be easily fixed but I remember having the impression that it had a ways to go. I'll pull fresh and give it another try and report back any of my concerned. I am very hopeful that your code can eventually replace the nasal hacks we're using now.

"Having structure points that can be moved down to the keel when on ground might be a solution" That is exactly how I am handling floating and other undesirable things.

[andgi]

If you pull now floats on ground (issue i) above) is now better handled along the lines you had before, but only when there is solid ground below the contact point. The hydrodynamics system is on unless all contacts are above solid ground. [EDIT]It should now start ok on water too again.[/EDIT]

There are things that may need some more tuning, such as: the location of the forward float contact point and the friction values - there is strong a tendency to nose over when the float contacts the runway even at moderate speed. The condition of spark particle systems and the damage detection system also need attention.

Maybe we should open a separate issue for this discussion?

[onox]

@andgi I tried your branch yesterday, but the aircraft reacted really badly when touching the water during landing or during takeoff.

[andgi]

Are you sure you didn't touch down too hard? I can make acceptable landings but it is not trivial, but that is somewhat consistent with what I have read about seaplane handling. Do we have anyone with actual seaplane experience around?

If you make a more extreme landing you will certainly see some bad effects - but should probably have seen the floats tear off instead. Addition: I think it is not straightforward to judge what is proper behaviour, and, in particular, where the boundaries for normal behaviour is. At speed water is hard and the suspension provided by the floats (or, certainly, a flying boat hull) is much less than that of a landing gear. I can also add to that after quite a number of take-offs and landings back I'm getting used to how my hydrodynamics model behaves so to me the behaviour is more likely to look ok.

[dany93]

@wlbragg I am trying to understand your amphibious front gear issue. Thus, pardon and don't be surprised if I write non sense.

But really I think the main issue with the amphibious gear is the spring_coeff and damping_coef settings. They need to be able to handle more weight, I think.

In your first picture https://github.com/Juanvvc/c172p-detailed/issues/214#issuecomment-109181958 the Z-position of the wheels do not seem so bad (although not easy to evaluate when they do no touch the ground).. In the FDM the spring coefficients (bogey, 21 - 24) seem correct with 2500 and 5400 LBS/FT (damping do not matter here). 2500 is probably hard, but this is not an issue. In your second picture, the top surface of the floats is almost horizontal, the height above the ground seems close to good. Up to there, I'd say that the compression in the FDM and Z = 60 values are not far from good.

The only issue is the viewed position of the front wheels in your 2nd picture, which means a 3D issue for me (not a FDM issue). I feel that is an animation which is over-amplified. That is obvious if you observe from outside. Roll and accelerate enough, pull the yoke or press the brakes: the front wheel enters underground or go above it instead of just following the difference in compression (difference in height from the ground). Here, that is worse because the stroke is weak, 2500 LBS/FT for these front springs is a bit hard. The nose weight is divided between the two, which makes only 8 - 9% compression on it. Anyway, this spring coefficient is not the issue, the gear has a much too large 3D displacement.

The animation is in Models/c172p.xml, lines 5067 and 5137 (all for gear 21 and 22)

<!--    Retractable Gear    -->
  <animation>
    <type>rotate</type>
    <object-name>Ldamper</object-name>
    <property>gear/gear[21]/compression-norm</property>
    <factor>7</factor> <!-- initial 91.5 -->

(maybe fdm/jsbsim/gear/unit[21]/compression-feet would be better, because if you change the spring coefficient I think the animation in feet would remain good after the factor adjustment)

Just to check if I have understood your issue well, what would you think of: (for gear 21 and 22)

• FDM c172p.xml, lines 426 and 441, spring_coef =1000 (instead of 1500 or 2500)
• Models/c172p.xml, factor 7 (like my xml code above, instead of 91.5)
• physics.nas lines 277 -278, fdm/jsbsim/gear/unit[21]/z-position = -62

These animations are always very hard to adjust. Difficult to find a good methodology.

But firstly, am I beside your issue or not?

[wlbragg]

"I feel that is an animation which is over-amplified." You are probably right. The animation was not a 1 to 1 copy from the dhc2 because I resized the landing gear and the tires. So that make sense that it is more likely animation which is over-amplified VS spring. Feel free to work on it and see if you can improve on it, that whole area of development is out of my wheel house.

The pictures in #214 were really just using z-position to show what pressure (or weight) was doing to the gear. Those z-positions I used wouldn't work for final positioning, the first one had the gear off the ground by at least a foot or more and they were not the gears z-position but other contact points I was using simply to raise the gear off the ground so they would hang free with no weight on them.

[wlbragg]

@andgi, has a fix and we should be able to put this to rest shortly after he gets the pull request sent..

[andgi]

@wlbragg I'm not completely happy with how the aircraft floats when idle with the hydrodynamics system - the H(YDRO)RP is currently offset forward compared to what the buoyancy data was generated for to compensate which is clearly a hack - but OTOH the resulting reactions are okish. It is one area for further refinement, but most likely not before the 3.6 freeze.

[wlbragg]

The gear PR has been merged, Now we need to verify we are OK with the AOA and then we can close this issue.

[dany93]

Not really the AoA, but the pitch angle (stopped on the ruwway): with the default gear: 2.7° with the amphibious: 4.1° Not sure this is very important, we set the AoA as soon as taken off. The higher AoA can lengthen a bit the roll, but anyway the amphibious is not the better choice if we look for performance. On the water it remains to be seen (I know nothing....). But also, the pitch angle, hence the AoA during acceleration, is set by acting on the elevator. See http://www.seaplanes.org/mambo/UserFiles/File/AC61-21A.pdf (link by tigert).

[wlbragg]

"Not really the AoA, but the pitch angle " Sorry, poor choice of words.

So do you think the pitch need to be adjusted?

[tigert]

To me it looks like the Flightgear floats and fuselage have too big angle.

http://www.vesilentokerho.info/wp-content/gallery/kellukeasennus/oh-ctl_05-07-2013-129.jpg http://www.vesilentokerho.info/wp-content/gallery/kellukeasennus/20130625_181541.jpg

There is an angle but I think in the simulated version it is a bit too big.

[dany93]

It was not for your choice of words, pitch angle and AoA have the same value when the aircraft accelerates on the runway. Merely, the pitch angle can be obtained even when the aircraft is stopped.

Yes, I also feel that this angle is a bit too high, from tigert's second picture and this scheme in my message https://github.com/Juanvvc/c172p-detailed/issues/214#issuecomment-108519431. Difficult to evaluate by the eye, but a value could be the one with the default gear (2.7° instead of 4.1°, check in the simulator). Which gives a 1.4° difference (easy to measure). By trying to measure (more difficult) the angle between the upper surface of the amphibious and the lower part rear of the aircraft, I evaluate this angle at about 2° in FG, 4.5° on the scheme. Unless you can superimpose the scheme in Blender. On tigert's picture I'd rather say about 3° (difficult to distiguish..), which is rather close to yours. I do not know which is the best. The pitch angle is easy to get in FG, and taking the default gear's one as a reference value seems sensible.

But I know this small difference gives a true work in Blender. This was why I wrote "Not sure this is very important" (in practice) because we can act on this angle with the elevator. This is rather to look like reality. Also, don't be surprised, this change will lead to readjust the Z-values for the gears, because the reference axis from the ground is the aircraft. This should not be difficult.

[wlbragg]

"But I know this small difference gives a true work in Blender." I was just looking at this in blender and xml, No real easy way I can find to fix this unless we can change pitch of c172p.ac on the fly when pontoon is selected for gear. That would be the easiest fix, if even possible. I superimposed a RL picture of a float c172p with a screenshot of ours and the difference was not great, but it was there. Also the size (length) of out floats are longer than RL. But so was the shape of the fuselage different in RL VS ours.

[onox]

@wlbragg @dany93 @andgi Can this be fixed this week or should we move it to 3.8 release milestone?

[wlbragg]

I did this in about 2 minutes. What I don't know is if it requires any other changes like contact points or something else. All this required was select everything in Models\Effects\pontoonc172p-pontoon.ac and rotate 3 degrees, edit Models/c172p.xml (line 6) and subtract 3 degrees from pitch. We would also have to do the same in c172p-pontoon-damage.ac. Plus we would have to account for this change in all other configurations and put the 3 degrees back in on the change over. Probably needs to wait for 3.8 to be safe.

[onox]

Wouldn't it be better to adjust the pontoon part instead of the fuselage and everything else?

I'm not sure if it needs changes in the FDM.

[andgi]

It certainly needs changes in the FDM: moving all float gears and contact points and adjusting the orientation and position used by the hydrodynamics system. The tables for buoyancy were computed based on the position and orientation of the floats in the poonton model file so the change can be accounted for by the offset in the hydrodynamic pitch property - at least if the floats were rotated around the VRP/centre of the top level model coordinate system.

If you look closely on the "after" picture above you can probably see the nose gears hovering above the ground.

[dany93]

I had a look at this and it is not at all simple.

Setting the 3D attitudes is just for the 3D view (which seems very good in your picture). Setting the FDM, attitude, Z-positions are in the FDM c172p.xml and in physics.nas for the buoyancy on water, amphibious wheels on ground and c172p-hydrodynamics.xml for the buoyancy on water and for the planing attitude.

Do not confound the 3D representation, which is pure illusion, and the FDM attitude. Although they should in accordance.

I currently read (/orientation/pitch-deg, which is in accordance with fdm/jsbsim/aero/pitch-rad) (juanvvc:master 07 jul 2015)

pitch-deg: default wheels; 2.65° (I would take this as a Reference) 26": 0.53° 36": -1.70°

Amphibious on the ground (wheels): 4.1° (+1.45°) Amphibious on water, engine and aircraft stopped on water: 4.4° (+1.75°) Floats on water, engine and aircraft stopped on water: 4.7° (+2.05°). ==> Only +1.45° to +2.05° compared with default wheels. Not so far (less than the 3°) I ignore what the FDM pitch-deg values are to aim for floats on water. Small differences can be made good with the elevator, if the float reactions are not too hard.

The (most important) attitude to takeoff on water is driven by the hydrodynamics. More difficult to measure. And worst to get a RL specification. The default wheels values can help. Even on the ground, takeoff is done with the nose wheel lifted.

I respond because you are on it and I do not want to make you waiting, but at this time, I cannot say more.

[wlbragg]

dany93 right, this is just the visual representation which is relatively minor but not without some issues. The floating of the wheels in the second image was because I guessed (did this by sight) at the pitch and it wasn't done precisely.

Wouldn't it be better to adjust the pontoon part instead of the fuselage and everything else?

That is what I did, but the way they are attached in the physical world of FG code, when you pitch just the pontoons in the model and run the sim then the front of the pontoons float up the amount of applied pitch. You have to counter that degree of pitch in the c172p.xml file to compensate and bring it back down to level.

It appears to be the extent of FDM adjustments that may be the show stopper.

[dany93]

Wouldn't it be better to adjust the pontoon part instead of the fuselage and everything else?

Here is my current understanding:

For the flying behavior, the reference axis are obviously those of the FDM. But the 3D representation follow those FDM axis when they move relatively to the ground. Offsets are almost always applied to compensate for the differences between 3D and FDM axis, but as we only see the 3D representation, applying wrong offsets to obtain an illusion is easy.

In the FDM reference axis, the important points are the propeller location and the wheel contacts location, which are given in the fdm file. These wheel contacts give the FDM axis position relatively to the ground or water (and the 3D representation, if the offsets are good). Ideally, the X axis in the FDM should be parallel to its equivalent in the 3D representation, but this is not explicitly defined.

Which means that the fuselage is the reference. It is good now with default wheels, do not offset it. Its position and orientation (height from the ground or water, attitude) is given by the Z values for wheels in the FDM coordinates (c172p.xml for default wheels, physics nas for bush and amphibious wheels).

The aim was to decrease the angle of the floats representation relatively to the FDM axis. I don't know which is the best way, but you have done it. But after that, did you change the fuselage position relatively to the ground with the FDM (physics.nas, c172p-hydrodynamics.xml) or by a 3D offset angle?

All this required was select everything in Models\Effects\pontoonc172p-pontoon.ac and rotate 3 degrees, edit Models/c172p.xml (line 6) and subtract 3 degrees from pitch.

If you have done it by a 3D offset angle, all that you have done is only changing the representation of the floats and fuselage. Changing their angle relatively to the FDM axis. But you have changed nothing in the FDM, the angles (AoA, pitch) are the same. If they were wrong they are still wrong. Moreover, the 3D fuselage pitch angle with the default wheels will be wrong.

To readjust after the viewed float-angle change relatively to the fuselage (= FDM axis), you have to:

• On the ground, to change the Z values, at least decrease Z for the front gears (physics.nas if I remember well),
• On the water, at least decrease or cancel the pitch value offset (currently = -3.0) in c172p-hydrodynamics.xml, line 339.

  <fcs_function name="hydro/float/pitch-deg">
   <description>
    The float pitch with the trim change due to waves and
    the displacement wave applied.
   </description>
   <function>
    <sum>
     <value>-3.0</value> <!-- The float model is 3 deg offset in pitch. -->

From what I understand of the c172p-hydrodynamics.xml, this should make the FDM AND the 3D model to pitch down by 3°. The floats front will go down with the fuselage and the FDM front axis.

Applying an offset on the 3D model is only for the eyes, but the FDM will remain unchanged. If you change nothing in the FDM (Z for the amphibious wheels in physics.nas, offset angle in c172p-hydrodynamics.xml, the FDM will still be wrong.

Rather than applying an offset angle on the 3D fuselage representation, I think that the changes must be done in the FDM (Z in physics.nas and float pitch in c172p-hydrodynamics.xml). By changing the contacts with ground or water, the FDM axis should pitch down by 3 deg and the 3D representation (fuselage + floats) should follow. Like this, the pitch angle and the AoA in the FDM will be changed.

Very difficult to explain, I'm never satisfied. And I may have done mistakes....

@andgi help!!!

[andgi]

If the floats have been rotated up around the right centre of rotation (i.e. the VRP) then reducing the 3 deg offset with the same amount should be ok. Note also that I offset the Hydro RP in the hydrodynamics file forward by 10 inches (guessed) just to get a "decent" idle attitude, so even the old visual representation isn't exactly in tune with the FDM buoyancy.

The float contact points (for land reactions) OTOH have been placed fairly precisely from the 3d model with the transformation to FDM space considered (see bushkit.xml and c172p.xml). These will also need to be moved.

It is unfortunate that this model has such a mess of coordinate frames, thanks to whoever (years back) introduced the 3 deg rotational offset (and a bit less so the z-offset) between the 3d model and the FDM frames.

[dany93]

Thanks @andgi for having responded so quickly.

reducing the 3 deg offset with the same amount

I guess you mean in c172p-hydrodynamics.xml?