Board index FlightGear Development Aircraft

FloatGear: Skipjack Kathryn...

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

Re: FloatGear: Skipjack Kathryn...

Postby Kugelfang » Mon Aug 22, 2022 11:57 am

/orientation/track-deg suffice (or even /orientation/track-magnetic-deg)


Ah, there are the droids I've been looking for! Interestingly enough, track-deg gives a result three to six degrees less than using v-east and v-north. Track-magnetic-deg results are on the other side and shows ten to fifteen degrees greater than track-deg. (I assume that's about what the variation would be in this geographic area.) So I guess the question before me now is which is more appropriate to a late 19th- early 20th- century boat. I'm pretty confident compass manufacturers of the day were able to deal with magnetic variation. I'm not sure the watermen of the Chesapeake harvesting oysters cared or spent money on a high-end compasses!

Can we assume the sail animation is still WIP, it's rotation seem off.


Always. Right now the user can trim either sail up to 45 degrees in either direction regardless of the wind. Couple of reasons for this. 1) It's much easier to set up testing scenarios this way--I just happened to trim the sails into the wind in this instance; 2) Historically, sailors could 'boom out' a fore-and-aft sail into the wind. You wouldn't typically see this being done with the mainsail like in the picture but it was fairly common with a foresail in order to get it out of the shadow of the mainsail when running before the wind--this was known as running wing-and-wing; 3) In the back of my mind I don't want to put in constraints that would inhibit developing ships with square sails. Square riggers had the ability to turn the sail into the wind to 'back' the sail thus producing negative thrust. Not sure I'll ever get to that point, though.

i notice in your image, your rudder is hard to starboard... what happens if you set it amidship in the same scenario?


In the picture the rudder is hard to starboard because I knew that would bring the boat into the wind to a point where forward motion ends and it begins to be blown sideways. Thus, the heading and the course should be approximately 90 degrees off. Straitening the rudder at this point doesn't have much affect because without forward movement the rudder is ineffective. However, at this attitude the foresail will produce a yaw moment that slowly turns the bow away from the wind. Forward movement then gradually returns and the rudder becomes effective again. I do not yet have everything in balance. This thing is all about balancing the different forces and moments! Right now, she doesn't answer the helm well, she can't sail into the wind, and I my gut feeling is that the hydrographic moments are exaggerated.

Still a lot to do!
We were set in our ways. We believed in a good God, a bad Devil and a hot Hell, and more than anything else we believed that the same good God did not intend man should ever fly.
--Bill Tate (Crouch, The Bishop's Boys)
Kugelfang
 
Posts: 141
Joined: Sun Dec 24, 2006 12:40 pm
Location: Silver Spring, MD, USA
Version: 2020.4.0
OS: Linux Mint 20.3

Re: FloatGear: Skipjack Kathryn...

Postby Gijs » Mon Aug 22, 2022 12:14 pm

Kugelfang wrote in Mon Aug 22, 2022 11:57 am:Interestingly enough, track-deg gives a result three to six degrees less than using v-east and v-north.

I suppose /fdm/jsbsim/velocities/vrp-v-east-deg_sec is the change in longitude in degrees per second and vrp-v-north-deg_sec in latitude? Depending on where you are on the globe, travelling over 1 deg of latitude relates to a different distance (in meters, or whatever unit of length you prefer) compared to travelling along 1 deg of longitude. While the length of a latitude doesn't change that much (it's always around 111 km), 1 deg of longitude can be anything between 0 km (at the poles) and 111 km (at the equator). You can only use your calculation when there is a 1:1 relation between them (roughly at the equator), which is why S&J's suggestion to use u-fps and v-fps (feet per second) makes sense as it is independent of your location on Earth. But of course the track properties already do the calculation for you :-)
Airports: EHAM, EHLE, KSFO
Aircraft: 747-400
User avatar
Gijs
Moderator
 
Posts: 9438
Joined: Tue Jul 03, 2007 2:55 pm
Location: Delft, the Netherlands
Callsign: PH-GYS
Version: Git
OS: Windows 10

Re: FloatGear: Skipjack Kathryn...

Postby Thorsten » Mon Aug 22, 2022 5:30 pm

I'm pretty confident compass manufacturers of the day were able to deal with magnetic variation.


No, compass manufacturers can't do that - you have to have your variation chart and make the correction yourself. Only a gyrocompass doesn't suffer from that particular issue - but that drifts over time.

The bit of technology that does get this right is the GPS - it always points true north as long as it has a signal.
Thorsten
 
Posts: 12315
Joined: Mon Nov 02, 2009 8:33 am

Re: FloatGear: Skipjack Kathryn...

Postby Kugelfang » Mon Aug 22, 2022 6:53 pm

S&J's suggestion to use u-fps and v-fps (feet per second) makes sense...


I wasn't able to get sensible numbers from the -fps properties. I'm sure it was how I was handling the code (long changed and forgotten by now) but the only values I could get were 1 and 2. But then I didn't know what atan2() was either until I looked it up. The track properties are much more convenient.

No, compass manufacturers can't do that...


Huh, so many rabbit holes to go down. The animated map from NOAA at this site put it in perspective for me:

https://www.myopencountry.com/compass-magnetic-declination/

--jeff
We were set in our ways. We believed in a good God, a bad Devil and a hot Hell, and more than anything else we believed that the same good God did not intend man should ever fly.
--Bill Tate (Crouch, The Bishop's Boys)
Kugelfang
 
Posts: 141
Joined: Sun Dec 24, 2006 12:40 pm
Location: Silver Spring, MD, USA
Version: 2020.4.0
OS: Linux Mint 20.3

Re: FloatGear: Skipjack Kathryn...

Postby MariuszXC » Mon Aug 22, 2022 9:38 pm

So the declination is a term I am not familiar with, but, from my sailing training, there are:

- true north: fixed (as long as we don't change the definition),

- magnetic variation: the difference between true and magnetic north. This one is caused by Earth's iron core being in a flux and is constantly changing, but at a yearly rate and direction we approximately know from the charts,

- magnetic deviation: the difference in compass readings and magnetic north. This one is caused by metal distribution around the place the compass is installed at, and also magnetic fields from wires passing curent nearby. Therefore for every stationary mounted compass there is (should be) a deviation table, valid _only_ for one particular installation location and one particular vessel.

Some marine compasses happen to have additional movable metal incorporated in their construction, to tune/cancel out the influence of other nearby metals. Don't know if it is the case with aviation ones.

Certainly one can forget about magnetic deviation for hand held compasses, at it will depend on what metals one has in their pockets or their backpack on a given day..
MariuszXC
 
Posts: 505
Joined: Tue May 18, 2021 4:38 pm
Location: Europe
Callsign: SP-MRM
Version: 2020.4
OS: Ubuntu 16.04

Re: FloatGear: Skipjack Kathryn...

Postby Thorsten » Tue Aug 23, 2022 8:34 am

So the declination is a term I am not familiar with, but, from my sailing training, there are:


Declination is basically another term for magnetic variation - geophysicists don't like to use 'variation' for the difference between true and magnetic north because they call the change of the magnetic field in time 'variation', so they use another word - navigators usually don't talk so much about the rate of change and are fine with using variation.
Thorsten
 
Posts: 12315
Joined: Mon Nov 02, 2009 8:33 am

Re: FloatGear: Skipjack Kathryn...

Postby MariuszXC » Tue Aug 23, 2022 3:40 pm

Thorsten, thanks for explaining. Makes sense in this context.
MariuszXC
 
Posts: 505
Joined: Tue May 18, 2021 4:38 pm
Location: Europe
Callsign: SP-MRM
Version: 2020.4
OS: Ubuntu 16.04

Re: FloatGear: Skipjack Kathryn...

Postby Kugelfang » Thu Sep 22, 2022 7:45 pm

Jeesh... what a month. I 'retired' in June, took a new job at the end of August, caught Covid in my second week. Still not 100% yet but improving. But how selfish am I? I didn't share it with my wife for our 28th wedding anniversary. I don't think she'll complain but she still wears a mask around me.

Meanwhile, I've been trying to figure out why I couldn't sail Katherine upwind. Then it dawned on me... maybe the sails are stalling out. At first I dismissed this because I was thinking I was using external reactions for the sails and I hadn't put in anything for the sails to stall when the apparent wind was at 90 degrees which is where it consistently falls off. But when I got to looking at the code the sail force/moment functions are actually referred to in the <aerodynamics> section of the config file.

So... does FGFS do some default stall calculations in the aerodynamics? Would calculating the sail forces as true external reactions side-step this issue?

Meanwhile, I'm thinking of a third 'floatgear' project. But I want to do this one fully from a blank canvas so I have a better grasp of how everything works. My other boats have all started out using Anders boats as templates--and there are always parts that I don't fully understand. I'm thinking of doing a small, single sailed skiff as something of a test bed.

So... is there anything in the FGFS universe that would prevent the water surface from rendering inside the small boat. In the sailing simulator that I used to tinker with we could define a plane (basically the shape of the deck) which would prevent rendering of the water surface when viewed through that plane. I think it would be similar to define the walkable area in an aircraft. Is this getting into shader territory?

TIA

--jeff (cough, cough, cough)
We were set in our ways. We believed in a good God, a bad Devil and a hot Hell, and more than anything else we believed that the same good God did not intend man should ever fly.
--Bill Tate (Crouch, The Bishop's Boys)
Kugelfang
 
Posts: 141
Joined: Sun Dec 24, 2006 12:40 pm
Location: Silver Spring, MD, USA
Version: 2020.4.0
OS: Linux Mint 20.3

Re: FloatGear: Skipjack Kathryn...

Postby merspieler » Thu Sep 22, 2022 8:01 pm

certainly shader territory for the water.
completely different to a walkable area as that is collision...
Be the change you wish to see in the world, be an ally to all! Love is the answer.

If everything is going against you, keep in mind that airplanes take off against the wind, not with it.

Join the official matrix space
merspieler
 
Posts: 1624
Joined: Thu Oct 26, 2017 10:43 am
Location: Wish to be in YBCS
Callsign: JST935. ORI1711
IRC name: merspieler
Version: next
OS: NixOS

Re: FloatGear: Skipjack Kathryn...

Postby TheEagle » Thu Sep 22, 2022 11:16 pm

Kugelfang wrote in Thu Sep 22, 2022 7:45 pm:So... does FGFS do some default stall calculations in the aerodynamics? Would calculating the sail forces as true external reactions side-step this issue?

No. You can make a plane climb at 10000 FPM with alpha = 90 - no problem with JSBSim ! :)
Kugelfang wrote in Thu Sep 22, 2022 7:45 pm:So... is there anything in the FGFS universe that would prevent the water surface from rendering inside the small boat. In the sailing simulator that I used to tinker with we could define a plane (basically the shape of the deck) which would prevent rendering of the water surface when viewed through that plane. I think it would be similar to define the walkable area in an aircraft. Is this getting into shader territory?

Definitely shader territory - but not implemented yet. Should be straightforward though - you'd just need to add logic to the water.frag shader to set a pixel's alpha value to 0 (= transparent) if it lies within a user-defined area. No idea though how to exactly implement it ! ;)
You won't believe how many issues dissolved into thin air after reinstalling !

Cessna 210 (GitHub | Wiki)
Find a list of the other aircraft I'm developing or enhancing in my wiki profile !
Addons: Landing challenge (GitHub)
Other: FGTools (GitHub)
User avatar
TheEagle
 
Posts: 2648
Joined: Sat May 01, 2021 2:27 pm
Location: France
Callsign: F-EAGLE
IRC name: none
Version: 2020.4.0
OS: Ubuntu Studio 20.04

Re: FloatGear: Skipjack Kathryn...

Postby Thorsten » Fri Sep 23, 2022 11:17 am

is there anything in the FGFS universe that would prevent the water surface from rendering inside the small boat.


You could try to introduce a transparent surface without actually registering it as transparent - via z-ordering that might block water from being rendered.

Not sure how well that works in practice - usually we have the opposite issue (a transparent surface not registered as transparent causes weird rendering behind it...)

Should be straightforward though - you'd just need to add logic to the water.frag shader to set a pixel's alpha value to 0 (= transparent) if it lies within a user-defined area.


Well, yes - only finding that user-defined area inside the shader is the thing that is not so straightforward at all...
Thorsten
 
Posts: 12315
Joined: Mon Nov 02, 2009 8:33 am

Re: FloatGear: Skipjack Kathryn...

Postby TheEagle » Fri Sep 23, 2022 12:46 pm

Thorsten wrote in Fri Sep 23, 2022 11:17 am:Well, yes - only finding that user-defined area inside the shader is the thing that is not so straightforward at all...

Exactly.
You won't believe how many issues dissolved into thin air after reinstalling !

Cessna 210 (GitHub | Wiki)
Find a list of the other aircraft I'm developing or enhancing in my wiki profile !
Addons: Landing challenge (GitHub)
Other: FGTools (GitHub)
User avatar
TheEagle
 
Posts: 2648
Joined: Sat May 01, 2021 2:27 pm
Location: France
Callsign: F-EAGLE
IRC name: none
Version: 2020.4.0
OS: Ubuntu Studio 20.04

Re: FloatGear: Skipjack Kathryn...

Postby S&J » Sat Sep 24, 2022 8:52 am

So... does FGFS do some default stall calculations in the aerodynamics? Would calculating the sail forces as true external reactions side-step this issue?


Lift is normally calculated against alpha degs. The direction being applied 90 degs perpendicular to alpha and drag aligned with alpha. And all this occurs internally within jsbsim.

But isn't your intention to calculate 'lift' and drag due to the beta angle the wind makes with the sail ?

I don't think you can use jsbsim internals in this case . I think you would be best using external forces using WIND co-ord for the sails. BODY co-ord for the hull and WORLD co-ord for bouyancy.

That said you could use X Y Z (which uses body co-ord) for the hull forces
Last edited by S&J on Sat Sep 24, 2022 9:56 am, edited 1 time in total.
"Stay away from negative people.They have a problem for every solution." - Albert Einstein
S&J
 
Posts: 567
Joined: Wed Aug 26, 2020 6:31 pm

Re: FloatGear: Skipjack Kathryn...

Postby Thorsten » Sat Sep 24, 2022 9:23 am

I don't think you can use jsbsim internals in this case .


But of course you can, because it's a 3d Problem in JSBSim and there's also for a plane a side-force that's driven by the beta angle.

So one thing you can do is use the side force, the other would be to rotate your coordinate system with the sail.

The fact that the ship sails indicates that one of these already has been done...
Thorsten
 
Posts: 12315
Joined: Mon Nov 02, 2009 8:33 am

Re: FloatGear: Skipjack Kathryn...

Postby Kugelfang » Sat Sep 24, 2022 4:25 pm

S&J wrote in Sat Sep 24, 2022 8:52 am:...
But isn't your intention to calculate 'lift' and drag due to the beta angle the wind makes with the sail ?

I don't think you can use jsbsim internals in this case . I think you would be best using external forces using WIND co-ord for the sails. BODY co-ord for the hull and WORLD co-ord for bouyancy.

...


Definitely headed in that direction. I can see that:

Code: Select all
 <aerodynamics>
   <axis name="DRAG">
     <function name="sail-drag">
       <sum>
         <property>aero/force/D_sail-1-lbs</property>
         <property>aero/force/D_sail-0-lbs</property>
       </sum>
     </function>
   </axis> <!--END OF DRAG AXIS -->
...
</aerodynamics>


Certainly exhibits what I suspect is stall behavior. But

Code: Select all

<external_reactions>
...

<!-- SAILS THRUST -->
   <force name="sails-X" frame="BODY" unit="LBS">
   <location unit="M">
     <x>  7.94 </x>
     <y>  0.00 </y>
     <z>  1.00 </z>
   </location>
   <direction>
     <x> 1.0 </x>
     <y> 0.0 </y>
     <z> 0.0 </z>
   </direction>
  </force>
...
</external_reactions>
...

  <system name="sails">
     <channel name="sailing">
      <summer name="sailing-x">
        <input>aero/force/D_sail-0-lbs</input>
        <input>aero/force/D_sail-1-lbs</input>
        <clipto>
          <min>0.0</min>
          <max>50.0</max>
        </clipto>
        <output> external_reactions/sails-X/magnitude </output>
      </summer>
     </channel>
  </system>


Does not exhibit stall behavior. I have to 'clip' magnitude (for now) to avoid excessive acceleration which results in cascading NAN errors. So, I'll have to revisit how the sail aero forces are calculated. Right now most of my numbers are made up to achieve a behavior rather than accurately representing the physics. The numbers aren't correct but the behavior is closer to real life. As always, the physics is a challenge for me but I continue to learn.
We were set in our ways. We believed in a good God, a bad Devil and a hot Hell, and more than anything else we believed that the same good God did not intend man should ever fly.
--Bill Tate (Crouch, The Bishop's Boys)
Kugelfang
 
Posts: 141
Joined: Sun Dec 24, 2006 12:40 pm
Location: Silver Spring, MD, USA
Version: 2020.4.0
OS: Linux Mint 20.3

PreviousNext

Return to Aircraft

Who is online

Users browsing this forum: portreekid, TheEagle and 4 guests