Jump to content

Killing relative velocity during rendezvous.


Recommended Posts

Quote

Furthermore, once you have achieved 0.0 relative velocity even taking into account the gravitational effects of the ship's masses and they - "not" being in the "perfectly same orbit" ships do not rapidly accelerate towards each other and apart accelerating infinitely. 

first you said that the acceleration was slow, now you say rapidly. which description is false?

keep in mind that ksp has smaller planets with faster orbits, and so the effects of drifting are bigger than in real life.

Quote

And can you explain why one time I can hold at a stable distance at 0.0 relative velocity and then in the very same save get to roughly the same point get to  0.0 relative velocity and then just accelerate past the ship infinitely with zero external force being applied and never be able to achieve station keeping?

both hotel26 and 18watt already did mention the possibility in their answers. if the two ships are sharing the same orbit, one in front of the other, they can be very stable. if they are one above the other, they will drift away much faster.

Quote

I have a video of Physics working as it should and then with the bug I am describing here from the same save, rendezvousing with the same two ships.  Perhaps you are all just not clear on what I  am describing. The same two shiips should not behave one was (the correct way) one time and then completely differently another!

perhaps. what you described the first time is perfectly compatible with normal orbital drift,  now you say that the change in speed is fast. I have experienced bugs with trajectories too, so there may be a bug at work.

Quote

 

I think with a MS  in aerospace engineering from UofM in 1985 with a focus on control systems and 40-plus years of writing guidance control code I have a pretty damn good idea of how orbital mechanics and physics work and don't appreciate being talked down to with snarky comments questioning what I know about physics.

I do not want to go off and get snarky myself.  I will make a video and post it here showing the behavior I am describing so you can understand that this is a bug and improper physics simulation.

 

we questioned what you know of phisics because you do not talk like an expert. Except for a few throwaway line that do hint at technical competence, like mentioning the physics of two spaceships orbiting each other (without whom I'd just dismiss your claims of MS as internet bravado), most of your messages come across as just ranting about things that have perfectly reasonable explanations. what you described in your first post was perfectly compatible with orbital drift, so the simplest explanation was that you were experiencing orbital drift and had no physical knowledge. even now, you boast of a MS in aerospace engineering but you never correctly describe orbital drifting.

We who are posting here have years of practice at this game. we do not have a degree, we do not know how to write guidance control code, nor we know about crafting techniques for advanced aerospace materials, realistic aerodinamic models, actual gravitational equations, or a bunch of similar stuff I suppose is studied in aerospace engineering.

but we do have a lot of practical experience running orbits, rendez-vous, dockings, transfers. I showed ksp to a friend with a phd in physics - he specialized in particle physics, but he took courses in orbital dynamics - and I was surprised at how much more knowledgeable I was than him. I would bet good money that when it comes to this practical understanding of the kind of orbital operations required in this game, we are actually more experts than people with actual degrees. in this specific field of space navigation, we are even more experts than several people working at nasa, because there's plenty of people who are not working on orbital mechanics but are instead building rovers, improving thermal shields, ruggedizing  delicate science instruments so that they will survive space, how to establish communication protocol. Or perhaps writing code.

And so we also don't appreciate being talked down to with snarky comments questioning what we know about physics either.

 

Link to comment
Share on other sites

13 minutes ago, king of nowhere said:
Quote

I think with a MS  in aerospace engineering from UofM in 1985 with a focus on control systems and 40-plus years of writing guidance control code I have a pretty damn good idea of how orbital mechanics and physics work and don't appreciate being talked down to with snarky comments questioning what I know about physics.

I do not want to go off and get snarky myself.  I will make a video and post it here showing the behavior I am describing so you can understand that this is a bug and improper physics simulation.

 

we questioned what you know of phisics because you do not talk like an expert. Except for a few throwaway line that do hint at technical competence, like mentioning the physics of two spaceships orbiting each other (without whom I'd just dismiss your claims of MS as internet bravado), most of your messages come across as just ranting about things that have perfectly reasonable explanations. what you described in your first post was perfectly compatible with orbital drift, so the simplest explanation was that you were experiencing orbital drift and had no physical knowledge. even now, you boast of a MS in aerospace engineering but you never correctly describe orbital drifting.

Their guidance code and whatever else they covered for their masters definitely won't pertain to the "space" in aerospace if they're having trouble coming to grips with the fact two orbits will behave differently to one another. I think that their qualifications are likely irrelevant to the discussion.

Link to comment
Share on other sites

18 hours ago, ctbram said:

I have a video of Physics working as it should and then with the bug I am describing here from the same save, rendezvousing with the same two ships.

...

I will make a video and post it here showing the behavior I am describing so you can understand that this is a bug and improper physics simulation.

I too would appreciate this video. It is possible you are experiencing something different than what we expect, though to be frank we expected that because of how you described it in your original post. A video pointing it out will for sure leave no room for doubt.

Make sure your video shows the entire UI for the entire run, though, so it can be as clear as possible.

Link to comment
Share on other sites

Here is the video I made that shows the bug.  It does not repeat even in the same save but it does happen again only with different ships.  It occurs in other runs not just the save file I was grabbing the videos from.  What's more, I will run the save one time and a pair of ships that failed to behave properly in one run will fail in another, and then work again in yet another run.

I have tried going to the space center and back to the ship or the tracking station and back to the ship and the bug seems to remain.

I have not tried switching ships with the target yet as dunbaratu suggested.  He has looked at the video and confirmed that the behavior looks like a bug.  We just do not know what triggers it.  However, I have seen a few posts describing the exact behavior.  So it may be rare but others seem to have seen it.

Here is a link to the video.  I tried to keep it as concise as possible in the first clip I show rendezvous and station keeping as it should work. In the second clip, I show a series of attempts to get to station keeping only to have the ship accelerate towards or away from each other after reaching a relative velocity of 0.0 m/s.

 

Link to comment
Share on other sites

First off: You stated in your video that we said the problem was caused by the gravity between the ships. To my knowledge no one said that. What we (or at least I) said is that if your two ships are next to each other and relative velocity is 0, then YOU ARE NOT IN THE SAME ORBIT.

You are CLOSE to the same orbit. VERY CLOSE. But you are NOT in the same orbit and in a surprisingly short time your ships WILL start to wander relative to each other. Yes, it's frustrating. Yes, it's counter-intuitive. But it's a real phenomenon and NOT a bug.

Now, as to what you showed in your video: In the first half, looking at KER I see your relative velocity between the two ships DID change when you weren't thrusting. You thrust them down to about 14.4mm/s and with no thrust that number instantly started climbing to about 15.2mm/s in the couple seconds you were talking. This is the wandering we're talking about.

In your second video, the numbers are far more dramatic and I agree that it's too much. I do not however agree this is a KSP bug as you have at least 4 mods installed and likely more. Have you recreated this in stock KSP with no mods? If I experienced this problem I'd do that, and then if I could recreate the problem I'd submit a bug report.

I've personally never experienced this in the stock game. I've had a similar problem with KAS or KIS or one of those but that was well over half a decade ago.

Link to comment
Share on other sites

How do you control your throttle?

Watch carefully your throttle indicator. After you made some short burns, the thrust doesn't went back to 0. If you look at the orbital parameters, they change at the same time. Stop the video at 0:34, watch the throttle and jump back to 0:14. You will see the difference. At 0:40 you can see how the thrust go back to 0, the orbital parameters and the relative speed are constant from that point. If you look carefully, you can see the exhaust gases too.

Link to comment
Share on other sites

This. The throttle. I don't know if you're trying to make fun of us, but how did you elaborate such a rant and such resistance to explanation when you... Just failed to zero your thrust ?... Seriously. I'll even guess that you made out this video knowing that you'll need to use that thrust error to mimic the issue you were describing, since at 0'38" you indeed have a "drift" that suddenly disappear by an action of yours : to cancel properly the throttle rather than having that tiny 0.5% of gas.

C'mon...

Edit : OR this is totally honest, you're using a HOTAS or something else to control your thrust, and your dead-zone is really not wide enough and you got plenty false cut-off. I might even bet on that ^^

Edited by Dakitess
Link to comment
Share on other sites

6 hours ago, DennisB said:

How do you control your throttle?

Watch carefully your throttle indicator. After you made some short burns, the thrust doesn't went back to 0. If you look at the orbital parameters, they change at the same time. Stop the video at 0:34, watch the throttle and jump back to 0:14. You will see the difference. At 0:40 you can see how the thrust go back to 0, the orbital parameters and the relative speed are constant from that point. If you look carefully, you can see the exhaust gases too.

You can see this more clearly in the KER windows up above - over the same time period, the TWR is registering as 0.02 rather than 0.00, and the total delta-V drops from 2319m/s to 2317m/s, which suggests fuel is actually being used up. 

Link to comment
Share on other sites

11 hours ago, ctbram said:

Here is the video I made that shows the bug.  It does not repeat even in the same save but it does happen again only with different ships.  It occurs in other runs not just the save file I was grabbing the videos from.  What's more, I will run the save one time and a pair of ships that failed to behave properly in one run will fail in another, and then work again in yet another run.

I have tried going to the space center and back to the ship or the tracking station and back to the ship and the bug seems to remain.

I have not tried switching ships with the target yet as dunbaratu suggested.  He has looked at the video and confirmed that the behavior looks like a bug.  We just do not know what triggers it.  However, I have seen a few posts describing the exact behavior.  So it may be rare but others seem to have seen it.

Here is a link to the video.  I tried to keep it as concise as possible in the first clip I show rendezvous and station keeping as it should work. In the second clip, I show a series of attempts to get to station keeping only to have the ship accelerate towards or away from each other after reaching a relative velocity of 0.0 m/s.

 

Even if you properly shut the engines off, you are not in the same orbit as your target and you should still slowly drift away over a period of several minutes. Being within 100m and having no relative velocity does not qualify as being in the same orbit. 

Also "stationkeeping" isn't a synonym for "staying perfectly still relative to another object". It's "staying roughly still relative to another object via active counteraction of external force". I wonder why they came up with a word which describes this. It's almost like two distant craft follow different paths.

Edited by Bej Kerman
Link to comment
Share on other sites

19 hours ago, ctbram said:

Here is the video I made that shows the bug.  It does not repeat even in the same save but it does happen again only with different ships.  It occurs in other runs not just the save file I was grabbing the videos from.  What's more, I will run the save one time and a pair of ships that failed to behave properly in one run will fail in another, and then work again in yet another run.

I have tried going to the space center and back to the ship or the tracking station and back to the ship and the bug seems to remain.

I have not tried switching ships with the target yet as dunbaratu suggested.  He has looked at the video and confirmed that the behavior looks like a bug.  We just do not know what triggers it.  However, I have seen a few posts describing the exact behavior.  So it may be rare but others seem to have seen it.

Here is a link to the video.  I tried to keep it as concise as possible in the first clip I show rendezvous and station keeping as it should work. In the second clip, I show a series of attempts to get to station keeping only to have the ship accelerate towards or away from each other after reaching a relative velocity of 0.0 m/s.

 

problem solved. when the "bug" manifests, you simply did not shut down the engine properly. look closely at the throttle.

zdbx7OS.png

dl9dyrS.png

see how the needle is not exactly on zero. see also how there's a light glow behind your engines, further showing that those engines are, indeed, working. see also how, on the data window on the left, in the first case it says acceleration 0, twr 0, and in the second it says acceleration 0.16 and twr 0.02

I don't know if you changed the default controls, but on pc standard to shut off engines is X. that will shut down the engines for real. Unless some of your mods can activate them, maybe some kind of mechjeb docking mode; I'm not familiar with those.

 

well, turns out both sides were right. you were right in that it was not orbital drift. we were right in that the game was not bugged.

Link to comment
Share on other sites

of course, it is not I am still killing relative velocity.  When the relative velocity is 0.0 I press x to kill all thrust.  But then the ship accelerates with NO THRUST 0.0 0.1 0.2... Then I have to reorient target retrogade and burn to reduce relative velocity to zero again and hit X to cut the throttle and once again the ships accelerate towards each other with the throttle cut. 

Link to comment
Share on other sites

15 minutes ago, ctbram said:

of course, it is not I am still killing relative velocity.  When the relative velocity is 0.0 I press x to kill all thrust.  But then the ship accelerates with NO THRUST 0.0 0.1 0.2... Then I have to reorient target retrogade and burn to reduce relative velocity to zero again and hit X to cut the throttle and once again the ships accelerate towards each other with the throttle cut. 

Yeah, cause two objects don't qualify as being in the same orbit just because they're within a distance of each other.

Link to comment
Share on other sites

3 hours ago, ctbram said:

of course, it is not I am still killing relative velocity.  When the relative velocity is 0.0 I press x to kill all thrust.  But then the ship accelerates with NO THRUST 0.0 0.1 0.2... Then I have to reorient target retrogade and burn to reduce relative velocity to zero again and hit X to cut the throttle and once again the ships accelerate towards each other with the throttle cut. 

The difference between speed and velocity is that speed is a scalar, the differential of distance (rate of change), but velocity is (speed & direction).  You are being shown the relative speed, which is not the same as velocity (speed and direction).

I am going to put this another way (and then I am going to give up).

I am doing 50 kmh down my local street in an easterly direction and you are doing 60 kmh in the same street in the westerly direction.  Our velocities are constant but our relative speed (rate of change of distance) is getting lower and lower until we are broadside, at which point in time (for only that point in time) our relative speed is zero and then it changes polarity and the relative speed then climbs asymptotically back toward 110 kmh.

For a moment there, the speed is zero, changing polarity even though we pass at what looks like 110 kmh, but our velocities are constant AND DIFFERENT.  The direction is DIFFERENT.  That is what causes the inevitable divergence.

VELOCITY and SPEED are not the same thing!!

                                                                    

I think the insight that will expand understanding for you now, is to realize that getting docked to another vehicle, involves two distinct processes:

  1. intercept/rendez-vous, gets you into close proximity, but not synchronized in orbit.  You will drift if you do nothing else.
  2. docking involves constant maneuvering to the dock, making progress faster than the natural drift.  Doing this is actually causing your orbits to synchronize and your distance to decrease at the same time.

That's probably a separate tutorial (and a docking aid).  Meanwhile, you've mentally stalled on what is normal drift and you simply need to proceed to learn how to continue from that situation to the actual dock.

                                                                    

I will finish by tantalizing you (for incentive): there is something called "invisible tethering".  Without docking, you can fix your proximity to another craft within a small range.

  1. get your target relative speed momentarily very close to zero
  2. then make a small burn pro/retrograde as necessary to change your orbital period (KER read-out) to be within milliseconds of your target's orbital period.

Whatever the residual difference in orbits, that means you will come back to your same position, relative to the target, with only milliseconds of drift (multiplied by your orbital speed for drift distance  per orbit, e.g. 2.2m) per orbit (e.g. 34min).

 

Edited by Hotel26
Link to comment
Share on other sites

4 hours ago, ctbram said:

of course, it is not I am still killing relative velocity.  When the relative velocity is 0.0 I press x to kill all thrust.  But then the ship accelerates with NO THRUST 0.0 0.1 0.2... Then I have to reorient target retrogade and burn to reduce relative velocity to zero again and hit X to cut the throttle and once again the ships accelerate towards each other with the throttle cut. 

As has been mentioned in the posts above yours, it seems that the issue is that the ship is thrusting for whatever reason even after you've cut engines, with your delta-v going down while your relative velocity is rising, and your TWR sitting at 0.02, which suggests it's not 'phantom forces' per se, but the engines not zeroing out when they should - you can see this in the video, the acceleration is always in the direction you'd expect the engines to be pushing you. This might be due to some weird control setup (flight sticks can act a bit strange with deadzones and all that) or some other control-related issue (does trimming thrust work? I actually don't know...), but it doesn't look like a physics bug.

4 hours ago, Bej Kerman said:

Yeah, cause two objects don't qualify as being in the same orbit just because they're within a distance of each other.

I don't think this is the issue in the second clip shown - the change is way too fast, and not really in the right direction.

Edited by GluttonyReaper
Link to comment
Share on other sites

12 minutes ago, GluttonyReaper said:

I don't think this is the issue in the second clip shown - the change is way too fast, and not really in the right direction.

The second clip is no longer relevant;

4 hours ago, ctbram said:

of course, it is not I am still killing relative velocity.  When the relative velocity is 0.0 I press x to kill all thrust.  But then the ship accelerates with NO THRUST 0.0 0.1 0.2... Then I have to reorient target retrogade and burn to reduce relative velocity to zero again and hit X to cut the throttle and once again the ships accelerate towards each other with the throttle cut. 

So they're killing the thrust this time, and for clear reasons they begin to drift.

Link to comment
Share on other sites

5 hours ago, ctbram said:

of course, it is not I am still killing relative velocity.  When the relative velocity is 0.0 I press x to kill all thrust.  But then the ship accelerates with NO THRUST 0.0 0.1 0.2... Then I have to reorient target retrogade and burn to reduce relative velocity to zero again and hit X to cut the throttle and once again the ships accelerate towards each other with the throttle cut. 

Your video, quite frankly, is bad. Because every time you claim "it's accelerating without cause" we can clearly see the TWR indicator showing 0.02. Either that or you're just thinking you cut thrust but something else is engaging it—as mentioned perhaps a HOTAS controller with imperfect dead zones.

At any rate, what you are showing us is that the mysterious acceleration only takes place when under thrust. When there is an additional force in play.

Put some RCS thrusters on your ship and deactivate the engine. Use RCS and show us again. Remember, extraordinary claims require extraordinay evidence and so far you're not convincing us that there's a bug, but rather a combination of user error, misinterpreting instrument readouts and not realizing that two ships together are not floating in space without any forces on it. (There's gravity forcing them into curved orbits. Unless they are exactly the same orbits you will experience orbital drift).

Link to comment
Share on other sites

15 hours ago, ctbram said:

of course, it is not I am still killing relative velocity.  When the relative velocity is 0.0 I press x to kill all thrust.  But then the ship accelerates with NO THRUST 0.0 0.1 0.2... Then I have to reorient target retrogade and burn to reduce relative velocity to zero again and hit X to cut the throttle and once again the ships accelerate towards each other with the throttle cut. 

in the video you posted you were NOT killing velocity. Look at 0:34, when you said at this point it should stay, the thrust is not at 0. the most likely explanation, if you were pressing x, is that some mod was interfering with that. maybe some kind of autopilot. maybe you were miclicking. maybe you changed config so that x does not zero thrust. whatever the reason, thrust is not zero.

Link to comment
Share on other sites

On 12/8/2023 at 11:28 AM, Kerbart said:

Your video, quite frankly, is bad.

I'd argue not quite that bad. We  can  extract a lot of information from what is show there .

If you pay close attention you will notice that ctbram does indeed cut the thrust when the relative velocity is 0.0m/s. At this point we may observe the relative velocity is not constant.

Exactly what he want to show us. But what catches our attention is something else:

A moment later the engine is reactivated when we are not expecting . That is because ctbram is relying on SAS to point the ship retrograde and he can't do it with speed below 1m/s.  "inefficient"?  Maybe, but it is not the "mysterious" acceleration ctbram is referring to. It distract us for a moment but as soon we notice what is happening it becomes irrelevant.

 

Let's get back to the situation where Barlin Kerman is chasing the Target and he just cut the thrust. Ctbram's question to us: why, at this moment, the relative velocity is not constant?

Considering:

K       position of Kerbin

B       position of Barlin

T       position of Target

ab      acceleration of  Barlin in the direction BK due gravity

at      acceleration of Target in the direction TK due gravity.

db    distance between Barlin and Kerbin

d    distance between Target and Kerbin

Given the angle BKT, make sense to reescribe aas:

  atcos(BKT)r  + atsen(BKT)

where is the direction BK and   the perpendicular direction to BK within the orbital plane. (For the sake of simplicity let's assume relative inclination is zero.) 

considering some numerical values  for at and BKT:

at BKT atsen(BKT)
9.8m/s2 171.033mm/s2
9.8m/s2 0,1º 17.104mm/s2
9.8m/s2 0,01º 1.710mm/s2
9.8m/s2 0,001º 0.171mm/s2

as we can see the acceleration component in the direction approaches  zero as the angle BKT approaches zero. In fact at for BKT = 0 (B, K and T are collinear) we have:

atcos(0)r  + atsen(0)s       ==    atr  

But wait. This is acceleration due gravity: 

g=GM/d2   where GM is constant and d is the distance between the central body (Kerbin) and the orbiting ship. 

we want to find the situation were ab - at = 0 

GM/db2 - GM/dt2 = 0 

1/db2 = 1/dt2

db =dt

Conclusion: to have exactly the same acceleration due gravity (amount and direction) Barlin  and Target need to be in exactly the same position relative to Kerbin. The acceleration  you observe is the acceleration of the non-inertial frame of reference due Kerbin's gravity.

 

PS: the special  case were both craft are in the same perfectly circular orbit is left as an exercise to the reader

Edited by Spricigo
typo
Link to comment
Share on other sites

19 minutes ago, Spricigo said:

Conclusion: to have exactly the same acceleration due gravity (amount and direction) Barlin  and Target need to be in exactly the same position relative to Kerbin. The acceleration  you observe is the acceleration of the non-inertial frame of reference due Kerbin's gravity.

If I understand this, without the math that requires a masters degree in aerospace engineering, in an elliptical orbit velocity differs with altitude and therefore with any point along the elliptic curve (Kepler pointed this out many centuries ago). By virtue of not being at exactly the same spot (they'd be docked), two objects in exactly the same orbit will therefore have different velocities and cannot be stationary.

Corollary: two objects in orbit that have a relative velocity of exactly  0 are therefore in different orbits and will eventually drift apart or together

Link to comment
Share on other sites

On 12/10/2023 at 2:50 PM, Kerbart said:

two objects in orbit that have a relative velocity of exactly  0 are therefore in different orbits and will eventually drift apart or together

And generally apart, for math reasons that don't quite boil down to (but are easy to visualize as) the fact that if you stand next to someone and then go a random direction, you are slightly more likely to go away from them than toward them.

Do it on paper. Place 2 dots and think about all the paths that one of them could take, and see how many go toward the other dot and how many go away.

Link to comment
Share on other sites

  • 1 month later...

I downloaded the lastest version, and I have to say, I experienced this bug too. I know orbital drifting, and I suddenly got a lot more acceleration than drifting could be accounted for.

it's something that only happens occasionally, though, so it was fixed on reload. it's just one of many little bugs that are annoying but not critical.

Link to comment
Share on other sites

10 hours ago, king of nowhere said:

and I suddenly got a lot more acceleration than drifting could be accounted for.

is an old bug, I have  experienced it myself several times. Still..just not what is happening in the videos above.

I get that you were not in doubt but for future reference the easiest way to confirm that bug is looking at the orbital period when no thrust is applied.

However, no idea how to actually reproduce the bug. (maybe it need be two bugs that love each other :confused:)

Link to comment
Share on other sites

  • 1 month later...

 

Quote

Furthermore, once you have achieved 0.0 relative velocity

Here's a thought, It may have a relative velocity in more than one direction.  Another point is sometimes you have to use your RCS thrusters instead of your main engine because it's too powerful you don't have fine enough control over it.  

 

Just a thought

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...