Periapsis Dropping at Interplanetary Transfer

[Dill]

I've been playing KSP for a few months, so I am not a total noob. Today I have been trying to perform manned missions to Moho and Jool (Tylo specifically), but every time I have tried to perform my transfer out of a Kerbin orbit, I find myself plummeting into Kerbin every time.

By the way, I have been to Jool about a dozen times (and Moho twice), so It's not like I have no idea what I am doing, but obviously something is going wrong this time. For further information, I have tried performing the transfers with Mechjeb with no success and I have tried performing them manually, according to phase angles given by the ksp.olex.biz calculator, with the same results (crash and burn). I'm not sure what I am doing wrong because, like I have mentioned, I have performed this same burn a dozen times with no problems.

There is probably a simple explanation for this, but I have given up at this point.

Edited September 29, 2014 by Dill

[Scarecrow88]

I guess the simplest explanation is tht you are mistakenly burning retrograde instead of prograde, but I don't imagine MJ would make that mistake. The only time I had something similar was when I needed to make a really long burn, starting some time before I reached the manoeuvre node, which meant that when I started my burn I was actually pointing at Kerbin.

[Epthelyn]

What do you mean by "plummeting into Kerbin"? Are you performing the burn and in doing so pushing your trajectory back into the atmosphere?

[Dill]

The only time I had something similar was when I needed to make a really long burn, starting some time before I reached the manoeuvre node, which meant that when I started my burn I was actually pointing at Kerbin.

You know what, I think this may be the case, but I am confused as to why this has never happened before, because I have always followed the "half-and-half" method when performing burns.

Edit: So yes, that was the case. When I reach Kerbin orbit, I am left with half an orange tank (w/Mainsail) followed by two nuclear engines on a smaller tank. The original interplanetary burn is calculated to be one minute long by the maneuver node system, but Mechjeb was burning five minutes before the node. I think Mechjeb factored in the extra burn time required for using the nukes (after the Mainsail stage has been dumped); thus, burning too early. Isn't this somewhat of a flaw on Mechjeb's part?

Edited September 28, 2014 by Dill

[Vanamonde]

Yes, this can happen if you try to start a longer burn from an overly low starting orbit. You're building up speed, but also pushing your PE down, possibly into atmo and disaster.

[Red Iron Crown]

A bit tough to call it a bug in MechJeb, it is executing the node as ordered. A couple of possible workarounds:

- Start from a higher parking orbit, that will give more leeway for a dropping periapsis and the slower orbital speed means the burn will be completed closer to the actual node.

- Break the burn up into two or more nodes. This is efficient but can require a bit of planning to make the final burn happen at the right time to hit the transfer window.

- Instead of burning in the maneuver node's direction, burn prograde instead. The periapsis won't fall at all, but the resultant orbit will be slightly different than the one predicted by the node, it is likely a bigger correction burn will be required.

- Rework your ship to have a higher TWR. This will shorten the burn time for the same dV expenditure, so dropping periapsis will be less of an issue.

[Kryxal]

If you're going to have to stage in the middle of the burn, I'd strongly suggest splitting that burn. Run the tank dry or nearly so, drop your stage, kill the old node and make a new one, finish your transfer burn. Bonus points if you can get the dropped stage into atmosphere to deorbit.

[Dill]

Thanks everyone. I guess I'll just have to work on my ship or use a different burn technique.

[RexKramer]

I think the above posters have probably covered what is going on. However, there's another possibility, although it may not apply to you:

If you have a probe core (or manned core) mounted upside down by accident, then your ship will be thrusting in the opposite direction. When the core is pointed prograde, the engines will be pushing retrograde. I've had this happen to me in a couple of different situations.

1. I accidentally mounted a probe core upside down. The cores can be tricky to determine which way should be facing up. To avoid this, I try to limit how much I rotate cores when placing them.

2. I've docked two ships together prior to a burn, and selected the incorrect side of the ship to control from. (Or just forgot to even check which core was controlling the ship. In this case, again the core was facing the opposite direction of the engines in use. So again, while the core in use was pointed prograde, the engines in use were pointing retrograde.

I mention this because it's very surprising that a prograde burn would cause you dip low enough into the atmosphere to crash, if you start from a stable orbit.

[Starwaster]

I mention this because it's very surprising that a prograde burn would cause you dip low enough into the atmosphere to crash, if you start from a stable orbit.

That's because it's not a prograde burn throughout the entire burn. Strictly speaking it's only prograde at the moment you reach periapsis. And you don't start at periapsis, you start a bit early. The lower the TWR of the rocket, the earlier you have to start. And that can and will lower your periapsis if you start too early.

So there's a point where when your TWR is too low to complete a high delta-V burn in a timely fashion that you have to break it up. Otherwise you're either screwing your orbit up or not burning enough in the right direction and putting yourself off course to where you don't reach your target.

That's why India's Mars probe had to do multiple burns to achieve escape velocity

[Hannu]

I think that Vanamonde is right. Interplanetary burn takes minutes if you have typical TWR and it is started minutes before periapsis. Then burn pushes you to lower altitude and if you are too low, you fall back to atmosphere. It is good idea to start interplanetary burns from higher orbits. It takes little (or not so little but boosters are cheap) larger boosters, but give more accurate results. Even if you do not fall to atmosphere you turn larger angle around Kerbin during burn which gives more error compared to predicted instantaneous impulse.

Other possibility is to program helper software which calculates burns as finite long prograde burns. I have thought it but it takes time and it is easy to achieve enough accuracy from higher orbit. 300 km is good if you have TWR about 0.25-0.35.

[Ace Eman]

I guess the simplest explanation is tht you are mistakenly burning retrograde instead of prograde, but I don't imagine MJ would make that mistake. The only time I had something similar was when I needed to make a really long burn, starting some time before I reached the manoeuvre node, which meant that when I started my burn I was actually pointing at Kerbin.

You are mistaken. Recently when I programmed mechjeb to lower my periapsis in preparation for my re-entry to Kerbin (I would have done it manually, but it interferes with my drinking tea and smoking) mechjeb went crazy and burned prograde instead of retrograde and put me on an escape trajectory. Luckily i had enough fuel to return manually and didn't suffer any casualties. The newest mechjeb is buggy.

[TrooperCooper]

Just raise your initial orbit before you do the kickout-burn and your problem will be fixed.