Jump to content
  • 0

Large Ship NERV Burn Problems - What am I doing Wrong?


Irihi
 Share

Question

I have a Laythe base docked to a nuclear-powered Intersolar Transfer Vehicle (ITV).  The whole ship is pretty massive. The ITV started out with 14,000 km/s, but only shows about 3500km/s when docked to the base. I think that might be an error due to the number of parts and various fuel tanks. There are 3 NERV's propelling the whole thing.

1. I'm planning on a departure burn to a Mun swingby for a gravity assist at about 50KM periapsis to fling the whole thing out into Kerbol-centric orbit,

2. where I'll do another burn to raise Apoapsis into Jool's SOI.

3. After that I'm going to hunt down a Joolian moon for another gravity assist to get joolian capture. 

3.5. Somewhere in there I might do a refuel with a Bop mining ship (coming along on a different transfer vessel)

4. Then some more maneuvering to get into Laythe orbit and drop the base.

 

The problem is that my 900km/s 18:22 departure burn isn't going off right. I Start the burn at t-9:11, but the deltaV immediately starts climbing, as does the burn timer. It gets up to around 1000km/s and the timer freezes at about 20:00. My orbit ends up just being an elipsis around kerbin. 

 

I've had this problem before with smaller stacks of ships, where I just had to re-do the burn and it would go off correctly. Is this a bug, and is there any way around it other than a complete rebuild of the mission to not use nukes? Can KSP just not handle pushing really big ships with really small engines?

 

I also get a really bad wobble when I try to physics warp during the burn. 4X gives me super shakes. 3X can go for a few minutes before getting some small shakes. 2X is stable. The incorrect burn seems to happen with or without physics warp.

Thanks for your help!

 

Link to comment
Share on other sites

22 answers to this question

Recommended Posts

  • 1

This sounds to me like it could be a control point issue, since your burn timer starts going negative as soon as you start. Double check and make sure your control point is facing the right way, maybe you are unknowingly controlling the ship from a backwards-facing pod. 

Other than that it might also come down to a staging error, this may be a bug but I always make sure if I am in control of a complex multi-stage ship, my engines are isolated in the very bottom stage by themselves, and I am controlling the vessel from a point that is towards the front. With how KSP calculates dV and burn time, it uses your current control point to guess which pieces you’re going to stage away when. 

Link to comment
Share on other sites

  • 0

There is a very similar question about nervs and burn time etc still being answered in the Gameplay Help section now, you might find some relevant info in there.

Link to comment
Share on other sites

  • 0

Yeah, I just didn't expect it to fall so much. I've got 3 large and 1 medium fuel fuselage pushing 2 science labs a mostly empty fuel tank (final landing burn) some thuds, a 10m heat shield, a refinery, drill... Hm.... I guess it is a lot.

Thanks for the tip on control points. That could be it! That's gotten me before. I docked using the docking port sr. As a control point  And the one on Laythe Base is facing "backwards" So it's very possible I've got a upside down control point.

 

Link to comment
Share on other sites

  • 0

 

2 hours ago, Irihi said:

Yeah, I just didn't expect it to fall so much. I've got 3 large and 1 medium fuel fuselage pushing 2 science labs a mostly empty fuel tank (final landing burn) some thuds, a 10m heat shield, a refinery, drill... Hm.... I guess it is a lot.

Thanks for the tip on control points. That could be it! That's gotten me before. I docked using the docking port sr. As a control point  And the one on Laythe Base is facing "backwards" So it's very possible I've got a upside down control point.

 

besides that thing with the control point, there are a couple more points to make here:

1) 18 minutes burn time is a lot. and starting at +9 is horribly inefficient. To make this relatively simple: look at the navsphere, the manuever marker in there. look at the prograde marker. the more distant they are, the less efficient the burn. efficiency is roughly equal to the cosine of the angle between them. at t-9, your prograde is roughly 90 degrees away from the manuever node, meaning a burn there is practically wasted.

1b) you can mitigate this by splitting the burn into several smaller burns. the first orbit you burn from +2 minutes to -2 minutes, you raise your apoapsis a bit in this way, but not enough to escape. that's good, because on the next orbit you repeat the manuever, this time you're already going faster, so you will burn some more and raise apoapsis even better. repeat for how many orbits it takes. it's much more efficient. unfortunately, this meddles with the manuever node. ideally, you should make a new manuever node at every new orbit. keeping aligned for the gravity assist is going to be nigh impossible, but fortunately you can wait 6 days for mun to be in the right place again; jool transfer windows are generally long.

you could also add more engines to your contraption; it would probably be more efficient, and easier.

2) your idea of going in solar orbit first, and then raising apoapsis from there, is horribly inefficient. you miss on all the oberth effect, and you may end up paying over twice the deltaV you'd need otherwise. the most efficient point to make the manuever is in kerbin low orbit, a single big burn for 2000 m/s will get you directly to jool.

of course, you already have problems with burn time, this will exacerbate them. i tell you, it is quite possible to make such a long burn in an efficient manner. but it takes practice. on the plus side, even if your burn is inefficient, it's still going to be better than the same burn in solar orbit

Link to comment
Share on other sites

  • 0

1. That's a great point and idea. I think I could split up the burn into 3 or so. What is the maximum number of nodes one can pre-program?

2. I think I could make my munar slingshot push my apoapsis close to jools SOI with just a few hundred dv in course correction, especially if I do it 50km over the munar surface. The problem I have is that I'm not getting course projections after I leave Kerbins SOI. Is there a way to get the map screen to project the course after more than 3 SOI entries/exits?

Id never heard of the oberth effect before. Thanks for that. 

I understand that prograde burns should happen at periapsis when possible, what about retrograde capture burns? Would those be more efficient far away (and slower) from the hyperbolic focus/periapsis?

Link to comment
Share on other sites

  • 0
8 hours ago, Irihi said:

1. That's a great point and idea. I think I could split up the burn into 3 or so. What is the maximum number of nodes one can pre-program?

There's no limit as far as I'm aware, but since they're going to be in roughly the same place it can get kind of difficult to manage.  Since you can never have a (combined) Δv for the non-escaping burns of more than 950m/s (or you will escape), the most common way would be to use three burns total, on of about 450m/s, the second to go right to the edge of the SoI, and the third for the remainder of the burn.  Of course, it will be difficult to work a Munar assist into this, which brings me to point 2

8 hours ago, Irihi said:

2. I think I could make my munar slingshot push my apoapsis close to jools SOI with just a few hundred dv in course correction, especially if I do it 50km over the munar surface. The problem I have is that I'm not getting course projections after I leave Kerbins SOI. Is there a way to get the map screen to project the course after more than 3 SOI entries/exits?

The Mun doesn't really have enough mass (in the stock game) to make it worthwhile using it for a gravity assist.  If you are going to do it (and it is definitely fun to do so), then try to make the burn as close to the surface as you can, to get full advantage of the Oberth effect (although it is much less pronounced at The Mun than at Kerbin, due to the lower mass). 

You can increase the course plot by changing the CONIC_PATCH_LIMIT setting in the config file (settings.cfg).  This can be done in game using the PreciseNode mod (or it may be possible in stock in more recent versions, but I don't know where that setting might be).

8 hours ago, Irihi said:

Id never heard of the oberth effect before. Thanks for that. 

I understand that prograde burns should happen at periapsis when possible, what about retrograde capture burns? Would those be more efficient far away (and slower) from the hyperbolic focus/periapsis?

If you are trying ro raise or lower your apo- or peri-apsis, then it is most efficient to burn at the opposite apsis (so at perapsis if tyring to raise apoapsis so as to escape from orbit).  Since in a capture burn you want to lower your apoapsis (from infinity), you should burn at periapsis.  This also gives the most benefit from the Oberth effect.

Link to comment
Share on other sites

  • 0

Hm... Isn't the kinetic energy vector of my propellant, in a retrograde burn, pointing in the wrong direction and counterproductive to throw that mass overboard at periapsis? Assuming I'm not going to do any gravity assists or areobraking, isn't it better to dump velocity at the edge of a SOI rather than deep down in the gravity well? Or am I physicsing wrong?

 

Thanks for the tip on the mun. On a Laythe landing mission, I dropped down to Kerbin through a 100km tylo swing by with just 500m/s or so, so I just assumed I could get a solid kick from the mun at half the altitude, but I guess it's a pretty small body.

Link to comment
Share on other sites

  • 0

Oberth still applies retrograde, so you lose more kinetic energy the faster you're going.

From the Wikipedia entry (the expample is a prograde burn, but it's easy enough to reverse it:

Spoiler

A rocket works by transferring momentum to its propellant.[3] At a fixed exhaust velocity, this will be a fixed amount of momentum per unit of propellant.[4] For a given mass of rocket (including remaining propellant), this implies a fixed change in velocity per unit of propellant. Because kinetic energy equals mv2/2, this change in velocity imparts a greater increase in kinetic energy at a high velocity than it would at a low velocity. For example, considering a 2 kg rocket:

  • at 1 m/s, adding 1 m/s increases the kinetic energy from 1 J to 4 J, for a gain of 3 J;
  • at 10 m/s, starting with a kinetic energy of 100 J, the rocket ends with 121 J, for a net gain of 21 J.

This greater change in kinetic energy can then carry the rocket higher in the gravity well than if the propellant were burned at a lower speed.

 

To illustrate, here's a simple Mun encounter.
At SoI entry, with a hyperbolic periapsis of 20k, it would take me 305 m/s Δv to capture into a 20 km / SoI edge orbit.  This burn requires a radial-out component, otherwise the periapsis will be several km below the Munar surface.

IV1aKQT.png

Capturing into the same (or at least a very similar) orbit at periapsis requires only 74 m/s

aC5mRbk.png

 

4 hours ago, Irihi said:

On a Laythe landing mission, I dropped down to Kerbin through a 100km tylo swing by with just 500m/s or so, so I just assumed I could get a solid kick from the mun at half the altitude, but I guess it's a pretty small body.

Yeah, Tylo is great for gravity assists (you can capture into Joolian orbit for just the cost of the course correction to set it up and no additional Δv), because of it's high mass.  The Mun... not so much :)

Link to comment
Share on other sites

  • 0

from the Mun you can get 50 m/s of extra speed with a gravity assist, 100 at most. as others have said, the more massive the body, the stronger the gravity assist. in fact, the cheapest trajectory to jool entails using a mun gravity assist to leave Kerbin's SoI, to get an orbit with an exact duration of one year so that you intersect kerbin's orbit the next year and get a gravity assist from it, sending you on the opposite direction from jool: to eve. all because eve, with its great mass, is wonderful at gravity assisting.

that trajectory requires a few more flyby of kerbin and eve, and a transfer done like that takes over 30 years. but you can reach jool from LKO with less than 900 m/s.

as for mun, i try to use it whenever i leave kerbin, because saving 100 m/s is better than not saving them. but it's not enough to make a significant impact on the mission

Link to comment
Share on other sites

  • 0

Thanks! I WAS controlling from some weird point on the ship. That fixed my problem.

I split my 22m burn into 3, and it did take less deltaV. I also learned how to use the maneuver tools mo bettah.

I did Munar flybys at 15-30km (I have 3 ships headed to the Joolian system), and, yeah, I only got a Kerbol Apoapsis about halfway to Duna. Since I'm in the window for a Hohmann transfer to Jool, and I had already spent 900m/s to boost up to Munar orbit (in the wrong direction for Jool), I used the Mun flyby to redirect my Kerbin SOI escape 90 degrees. 

 

I don't have the patience to wait for the alignments to do a Kerbin slingshot, so I'm just going to brute-force a Hohmann transfer all the way up to Jool. It's about 2500m/s of DeltaV. 

 

I have 3 components on the way to Jool, each pushed by a 14km/s (no cargo) nuclear transfer vehicle:

1. Laythe base, which is quite massive; includes 2 labs, and an integrated harvester and refinery. 

2. A Laythe SSTO lander which lands on wheels and can trundle over to the base to refuel. Also for hops around Laythe. Also, IT FLOATS upright!

3. A Bop mining ship and refinery which I'm going to use for in-space refueling of the SSTO and nuclear transfer ships.

 

My problem is my ill-advised Mun slingshot and Kerbol transfer leaves me with just 500m/s of delta-V to capture 1. (Laythe Base) into the Jool system. I have an inflatable heat shield up front, so I can try aerobraking at Jool itself. I managed to make that happen last time with a smaller vehicle (after sploding lots of times before I picked a good periapsis). Otherwise I can try gravity braking or powered gravity braking around Jool, Tylo, or Laythe. I think 500m/s should be plenty, considering some people on this forum seem to be able to ping-pong around the entire kerbolar system with a kitten fart worth of propellant.

 

All I need to do is capture Laythe base into the Jool system, then I can play musical chairs with my transfer vehicles and Bop refueling ship. The SSTO and refueling ship are very light and will have plenty of margin when they reach the Joolian system.

 

Here's another question that could save me several hours; If I wanted to put more big stuff on a ship (like more NERV's or something) that won't fit in cargo containers, can I just blast a big ball of tanks and engines into orbit, rendezvous, and then have an engineer assemble the whole thing? I did this with MechJebs after I got that mod, but those are small. What's the biggest thing an engineer can move around/attach/detach in free-fall?

Link to comment
Share on other sites

  • 0
1 hour ago, Irihi said:

Here's another question that could save me several hours; If I wanted to put more big stuff on a ship (like more NERV's or something) that won't fit in cargo containers, can I just blast a big ball of tanks and engines into orbit, rendezvous, and then have an engineer assemble the whole thing? I did this with MechJebs after I got that mod, but those are small. What's the biggest thing an engineer can move around/attach/detach in free-fall?

Unfortunately the largest/heaviest thing kerbals can move/assemble in freefall is the aerospike.

Link to comment
Share on other sites

  • 0
2 hours ago, Irihi said:

Here's another question that could save me several hours; If I wanted to put more big stuff on a ship (like more NERV's or something) that won't fit in cargo containers, can I just blast a big ball of tanks and engines into orbit, rendezvous, and then have an engineer assemble the whole thing? I did this with MechJebs after I got that mod, but those are small. What's the biggest thing an engineer can move around/attach/detach in free-fall?

yeah, one ton seem to be about the upper limit.

you can actually see it in the VAB: right-click on a part, and a bunch of extra information appears. among them, whether it can be manipulated during eva construction. if it does not explicitly say yes, then it cannot be manipulated.

also, you can't manipulate root parts. caused me some grief when i decoupled a lander and it authomatically assigned as root the docking port.

Link to comment
Share on other sites

  • 0
10 hours ago, Irihi said:

Otherwise I can try gravity braking or powered gravity braking around Jool, Tylo, or Laythe. I think 500m/s should be plenty, considering some people on this forum seem to be able to ping-pong around the entire kerbolar system with a kitten fart worth of propellant.

Gravity braking around Tylo/Laythe is probably your best bet; it's a lot less trial and error than aerobraking, and costs about the same amount. If done correctly, you can get a free Laythe intercept right after the gravity capture too.

 

10 hours ago, Irihi said:

Here's another question that could save me several hours; If I wanted to put more big stuff on a ship (like more NERV's or something) that won't fit in cargo containers, can I just blast a big ball of tanks and engines into orbit, rendezvous, and then have an engineer assemble the whole thing? I did this with MechJebs after I got that mod, but those are small. What's the biggest thing an engineer can move around/attach/detach in free-fall?

Since they're too large for engineers to build, you can try assembling it space-station style; put your NERV cluster on a docking port (or multiple ports, if the angle's important) and dock the whole assembly at once. 

Link to comment
Share on other sites

  • 0
12 hours ago, Irihi said:

My problem is my ill-advised Mun slingshot and Kerbol transfer leaves me with just 500m/s of delta-V to capture 

You can still easily refuel the craft before leaving Kerbin. Get a fuel tanker large enough so that after expending about 900m/s dv, it still has enough to refuel your mothership. Get both of these in an elliptical orbit (almost ready to leave Kerbin's SOI), then rendezvous, refuel, and continue with only the full mothership.

* The tanker can be re-usable, just aerocapture it in Kerbin's atmosphere later
* If the whole docked craft is stable enough, you can dock while in low orbit and just burn off fuel from the fuel tanker first
* If you have enough open docking ports, you can add a bunch of drop tanks for the same effect (good luck cleaning that up)

Link to comment
Share on other sites

  • 0
13 hours ago, Blaarkies said:

You can still easily refuel the craft before leaving Kerbin. Get a fuel tanker large enough so that after expending about 900m/s dv, it still has enough to refuel your mothership. Get both of these in an elliptical orbit (almost ready to leave Kerbin's SOI), then rendezvous, refuel, and continue with only the full mothership.

* The tanker can be re-usable, just aerocapture it in Kerbin's atmosphere later
* If the whole docked craft is stable enough, you can dock while in low orbit and just burn off fuel from the fuel tanker first
* If you have enough open docking ports, you can add a bunch of drop tanks for the same effect (good luck cleaning that up)

Good idea, but my mission is already gone bye-bye. The ship in question is now orbiting kerbol, and scheduled for a burn to a Jool transfer orbit in a few days. I'd have to rendezvous in deep space... 

At this point I think it's less effort to just deal with a (relatively) low-deltav jool capture.

I can actually ram my refueling tanker into the Laythe Base transfer vehicle and use it to transfer a few thousand delta-v of propellant from its transfer vehicle. It is only a few hours behind Laythe Base on essentially the same trajectory. Since it's just a big empty gas can  its transfer taxi has nuke fuel to spare.

 

 

Link to comment
Share on other sites

  • 0
2 hours ago, Irihi said:

Good idea, but my mission is already gone bye-bye. The ship in question is now orbiting kerbol, and scheduled for a burn to a Jool transfer orbit in a few days. I'd have to rendezvous in deep space... 

In the future, don't forget that you can quicksave and revert to an earlier save! (In this case, it would let you start over and use the Oberth effect too :D)

Link to comment
Share on other sites

  • 0
16 hours ago, AlpacaMall said:

In the future, don't forget that you can quicksave and revert to an earlier save! (In this case, it would let you start over and use the Oberth effect too :D)

I usually remember to quicksave about halfway through my irrevocable burns. :p

Link to comment
Share on other sites

  • 0

So I had a TON of extra Delta-V in the end. I don't know how KSP was calculating it, but I ended up with my Landers and their transfer vehicles in low Laythe and Bop orbits with 1/4 to 1/3 of their fuel remaining. No aerobraking, just powered gravity braking around Jool and moons. 

I did a stupid and built my Laythe Landers upside down. The 10M heat shield is on top of my base. So I have to reenter zenith first, then flip the ship nadir down (at which point my heat shield will be backwards at the lander/bases zenith like an umbrella), jettison the heat Shields and land via chutes and thuds. I should have just built it with the heat shield in nadir position and launched/pushed it to Laythe backwards. 

I also have no idea what the aerodynamic characteristics of my lander are. I think I built it cg forward of cp when flying zenith first (the orientation in which the heat shield will work), with the idea it will flip upright when I deploy the chutes. I have a lot of RXN wheels and RCS to help point it where I want, but no guarantee It won't get stuck zenith down and Kerbalsplat.

Here's a pic. The base/lander is on the right. Heat shield Zenith, Thuds and landing legs Nadir where it's docked to it's transfer vehicle. Any tips on how to land?

I also am using Mechjeb to land. Since it likes landers that weren't built by insane sorceresses, I'll have to quickly flip it on and off at various times during retro burns, reentry and landing while I do manual adjustments and swap control directions. 

 

LaytheBaseandITV.png

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
Answer this question...

×   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.

 Share

×
×
  • Create New...