1

L2 Halo orbits are often portrayed as a pair of “North” and “South” orbits. Could the difference in “tilt” around the Y axis be explained by the direction of rotation and the Coriolis Effect?

enter image description here https://www.semanticscholar.org/paper/Maintenance-of-earth-moon-halo-orbit-Liu-Hu/6fa14df4563cd71841a53b0017ff44264ac45cee

Below is a sketch of the Sun-Earth L2 and a closer view of two hypothetical counter-rotating halo orbits

enter image description here

Since this is a rotating frame of reference, the Coriolis Force will affect objects moving in the X-Y plane in relation to the frame of reference. Movement along the Z axis will experience no Coriolis Effect.

In sketch a) below, the green, “counterclockwise” orbit is isolated. The blue arrows indicate the Z-axis which is the rotational axis of the frame of reference. The dark green arrows represent the velocity vectors at various points on the orbit. The orange arrows represent the Coriolis Force acting at that point in the orbit. In sketch b), extraneous vectors have been removed. The dashed green line shows the expected displacement of the orbit around the Y axis due to the Coriolis Force.

Sketch c) shows similar effect on the counter-rotating red orbit, but with the tilt in the opposite direction as expected. Sketch d) shows the relative tilt of the two orbits together.

enter image description here

enter image description here

The final sketch e) shows a slightly rotated view

enter image description here

I cannot source any information on the orbital direction of “North” and “South” halo orbits. Is Coriolis Force a reasonable model for the difference in their orientation?

Woody
  • 21,532
  • 56
  • 146
  • 1
    the caption on figure 5 in https://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.20.6071&rep=rep1&type=pdf seems to emphasize the relative independence of the in-plane (ecliptic) motion and the out-of-plane (polar axis) motion, which is what your good SHM question led me to believe earlier. – Roger Wood Jan 19 '22 at 07:48
  • 2
    I have a hard time reading your own plots - there are not enough visual cues for my eyes to see how they are supposed to be oriented in 3D. – asdfex Jan 19 '22 at 09:53
  • 1
    @asdfex I wish we could somehow easily generate and embed WebGL in our SE posts. What are ways that we can show 3D orbits and other 3D things in posts? – uhoh Jan 19 '22 at 10:05
  • If I get the drawings right, the orange arrows are not Coriolis, but residual centrifugal forces. – asdfex Jan 19 '22 at 10:27
  • @RogerWood ---- but the caption says "The amplitudes can be chosen such that the frequencies become equal.." In other words, there are a range of diameters for halo orbits. Outside this range, the "orbit" breaks down. Very small halo orbits don't exist. – Woody Jan 19 '22 at 17:32
  • @asdfex ---- all the sketches (except the first, credited illustration) are from roughly the same perspective. All blue vectors are parallel to the Z axis. All orange vectors are parallel to the X axis. Green vectors are tangential to the green halo orbit. – Woody Jan 19 '22 at 17:36
  • @asdfex ---- Centrifugal force is always radial, away from the axis of rotation. Coriolis vectors are at a right angle to a plane formed by the rotation axis and the direction of motion (in a rotating frame of reference). – Woody Jan 19 '22 at 17:42
  • @Woody But the orange arrows are pointing radially. "parallel to the X axis", as you write. – asdfex Jan 19 '22 at 19:33
  • @asdfex --- Yes. But some of the orange vectors point TOWARDS the axis of rotation, which centrifugal force cannot do. A Coriolis Force vector is a cross product of the rotation rector (Z axis in this case) and the object velocity vector (green vectors in this case) – Woody Jan 19 '22 at 19:39
  • @Woody Yes, it's seems like there should be just one unique halo orbit (or one pair) where the periods of the z-axis and x-y plane orbits match. But it looks like there is another degree of freedom which is the placement of the orbit (nearer or farther) from the secondary object (the Earth in this case) and this gives a family of orbits. – Roger Wood Jan 19 '22 at 20:33
  • @RogerWood --- Right. Check out this famjly of orbits https://www.researchgate.net/figure/The-L-1-and-L-2-halo-orbit-families-and-the-NRHOs-11_fig1_319531960 – Woody Jan 19 '22 at 21:02
  • @Woody Wow! - and wouldn't it be nice if they put arrows showing the direction of rotation. – Roger Wood Jan 20 '22 at 02:17
  • @Woody I wrote "residual centrifugal force". The remaining force, after you subtracted the gravitational force. – asdfex Jan 20 '22 at 08:48
  • @Woody: I'm not sure it is appropriate to use the RLP framework (Rotating Libration Point) for assigning velocity vectors to object for the purpose of determining Coriolis forces. In reality object never has velocity component in -Y direction. "Rotation" about L2 is a convenient artificial construct for visualization of relative motion. I wish I had citation for my assertion that coriolis is a second or third order affect on halo orbit shape with body 2 gravity effect being the first order cause of tilt. I believe coriolis is a concern with telescope pointing accuracy over time. – BradV Jan 25 '22 at 19:59

1 Answers1

-1

halo orbit 'tilt' has nothing to do with coriolis but is caused by the object speeding up and slowing down as it orbits the sun.

In a Sun-Earth L2 halo orbit the object is ALWAYS orbiting the sun in the same direction as the earth at an AVERAGE orbital velocity equal to earth's. HOWEVER... the object in halo speeds up and slows down a tiny bit so that it is always in a cycle of apparently overtaking the earth or falling behind the earth. simple 'orbital velocity about Sun' calculation results in an orbit closer to sun when object is faster (overtaking earth) and orbit farther from sun when slower (falling back from earth).

Because of this behavior it is impossible to get a halo orbit with no tilt. It is possible to get a halo with 90 degree tilt (halo plane in same plane as earth orbit).

EDIT: earth gravity causes speedup and slowdown. When object is 'behind' earth the earth gravity works to pull object along, increasing its speed. In same way, when object is 'ahead' of earth the earth gravity slows down the object. Max object velocity at 12 oclock position gives minimum radius sun orbit, minimum velocity at 6 oclock gives maximum radius sun orbit for "north" halo.

Edit 2-24-2022: I need to acknowledge that I am in way over my head here. I no longer 'have the chops' to support my assertions in a scientific/mathematical way. I would like to invite ANYONE to provide another answer. Just because I can no longer 'math out' the answer does not automatically mean my assertion is wrong. Its just not supported.

THIS IS AN INTERESTING ASPECT OF ORBITAL MECHAINCS. I WISH BUZZ ALDRIN COULD HELP HERE!

BradV
  • 3,284
  • 11
  • 33
  • I think you've captured the idea that the orbit around L2 has to be retrograde. I'm not sure exactly what the definition of a halo orbit is, but I think it refers to a 3D closed orbit that includes a weak definition of stability such that small disturbances shift the orbit but don't necessarily lead to a runaway instability. I don't think this is true for the small elliptical orbits in the plane of the ecliptic or for small linear oscillations along the z-axis – Roger Wood Jan 22 '22 at 20:57
  • @BradV ---- An object at SEL2 (or a halo orbit around it) does NOT have the same average orbital velocity as the Earth. It has the same angular velocity as Earth, but because it is 1% further from the Sun, its average orbital velocity must be 1% greater that the Earth's. The object is not in a Keplerian orbit around the Sun, so you cannot use 2-body formulas for radius/speed/period. – Woody Jan 22 '22 at 22:01
  • @BradV Woody is correct. The Earth's field gradient dominates at L2, so it's better to say closer or further from the Earth. I do like the recent edit about the object lagging or leading the Earth (though I'm not sure where 12 o'clock and 6 o'clock are) – Roger Wood Jan 23 '22 at 06:48
  • @Woody.. good catch! Thanks! and Roger Wood... I wish I could recall the acronym for the non-inertial reference frame used for looking at halo orbits. Basically, X axis is through system barycenter and +x direction passes thru second body. The +Y axis is thru the L point being discussed in the direction of planet orbit. The +Z axis points in direction of system north. SO... 12 o'clock when viewing a halo orbit from this reference frame would be when object has +z, zero y location, and 6 o'clock would be at -z, zero y location. – BradV Jan 24 '22 at 04:31
  • @roger wood I'm still not sure if I buy into the idea that a halo orbit can be prograde or retrograde. The terms I've read are 'north' which is also known as class 1 and 'south' or class 2. However... I have no idea what these definitions mean in terms of direction. I'm guessing that north (class 1) means object is north side (+Z) when prograde. – BradV Jan 24 '22 at 06:49
  • @BradV I've convinced myself that these orbit can only be retrograde around the L2 point (as seen in the rotating frame). See my answer to Woody's earlier question https://space.stackexchange.com/a/57826/14419 – Roger Wood Jan 24 '22 at 07:46
  • @roger wood I now agree with you regarding retrograde about Z axis for JWST. I believe the first graphic at space.stackexchange.com/a/57826/14419 is messed up. – BradV Jan 24 '22 at 10:11
  • @woody and Roger Wood I've put together some sketches that I feel do a more 'true to physics' presentation of coriolis effects but I'm struggling with getting a clear/concise graphic of object apparent motion along Y axis for our L2 type discussion. I fully agree that coriolis effects are being manifested in JWST halo shape but only in a minor way and I still believe that the major shaping of halo and tilt is gravity effects. – BradV Feb 21 '22 at 16:40
  • @Roger Wood and Woody ... building on fundamentals... I believe we can all agree that the Z +/- as well as Y +/- motions are (primarily) earth gravity driven. Same for X +/- motions. The associated momentum vectors are ALWAYS in a gravity induced state of direction change. Can we agree on these things? – BradV Feb 21 '22 at 16:51
  • @BradV In the stationary inertial frame, GRAVITY RULES! There are no other forces. In the rotating frame the gravity is unchanged, but you have to introduce the two additional ficticious forces: the centrifugal force and the Coriolis force. Neither of these affect the z-motion, so you can argue the z-motion is always gravity driven whether or not you're in the rotating frame. The x-y motion is due to gravity, but also, if you're in the rotating frame, you have to have the fictional centrifugal & Coriolis forces too. – Roger Wood Feb 22 '22 at 04:06
  • @BradV FYI, I did look into the situation for small perturbations about L2. In the limit, if the big mass is very very very very much larger than the secondary mass, then the z-motion has a period of exactly 1/2 and the x-y motion has a period of exactly 1/sqrt(2*sqrt(7)-1). – Roger Wood Feb 22 '22 at 04:12
  • @Woody ... Last heard saying. "Whoa. Lost my eyebrows again." A while after I was goofing around with my son doing his pyrotechnic stuff I asked "Why does everything smell like burned hair?" Joe told me that 1/2 of my mustache was gone! – BradV Feb 24 '22 at 21:41