Ascent Auto doesn't circularize and begins early

[Lillz]

What Happens? - Ascent works as normal, reaches desired height, makes node to circularize. However, It will start the burn early AND not use a corrective angle to prevent losing altitude after you pass the apoapsis. Never had a problem with this before in mechjeb, but now I do.

[Anatid]

Just replied to this on the forums, but this is a better place:

This is how it is supposed to work in MJ 2. MJ 1 would angle up to maintain altitude after reaching apoapsis. MJ 2 uses a different strategy where it always burns toward the maneuver node it creates, but tries to time the start of the burn so as to end up in a good circular orbit. The hope is that this reduces some problems the old strategy had with vessels with low TWR or poor attitude control authority. The new strategy does result in a slightly less circular orbit, but it should still be quite circular, and if you need precision you can use the maneuver planner to circularize precisely once in orbit.

Are you ending up with a bad final orbit? If so, if you give us the .craft file we'll try it out and see what's going wrong.

[Lillz]

Pretty much every time with the new ships I've been making. I'll say I want a 100km orbit and, on my most recent test, I got a 138km x 68km orbit.. One thing that it does that doesn't seem to work is it doesn't even try to face the node after passing the edge of the atmosphere. I had to take it off auto and point it towards the node myself.

Here's my ship (I'm not sure how to upload to github): https://mega.co.nz/#!OYFVhBqK!TP39XjoqWrbYW7nZSFPlhh_ydhoJoAVVj_8R0W1ZrWs

[Anatid]

Thanks for the craft file!

I tried it out and it got a pretty nice orbit for me (99,995 m x 100,016 m!).

I thought at first that MJ wasn't orienting toward the node, like you said, but then I realized that it was: it's just that the vessel has only a tiny amount of available torque. When the engines are off, only the probe core provides torque, and not very much of it. So it takes a while to get oriented toward the node.

It's possible we've fixed some bug in our development version that is making it work better for me. If you want, you can try the most recent dev version by replacing the MechJeb2.dll in your plugins folder with the one at http://jenkins.mumech.com/job/MechJeb2/

[Lillz]

I just tried the latest dev build and had the same problem. It didn't get pointing at the node till about +12s. Could it be possible that it's my flight path? I have it to start turn at 20km and end at 80km. Final flight angle is at 0 and the shape is set at 24.4

Anatid notifications@github.com wrote:

Thanks for the craft file!

I tried it out and it got a pretty nice orbit for me (99,995 m x 100,016 m!).

I thought at first that MJ wasn't orienting toward the node, like you said, but then I realized that it was: it's just that the vessel has only a tiny amount of available torque. When the engines are off, only the probe core provides torque, and not very much of it. So it takes a while to get oriented toward the node.

It's possible we've fixed some bug in our development version that is making it work better for me. If you want, you can try the most recent dev version by replacing the MechJeb2.dll in your plugins folder with the one at http://jenkins.mumech.com/job/MechJeb2/

— Reply to this email directly or view it on GitHub.

[Lillz]

Ok changed my turn to 20 to 100km and turn angle to 45... Got an orbit of 106km x 93km, so it's better now but still pretty eccentric

[Tallinu]

Better ascent settings can indeed improve your results. For such a high acceleration rocket and low target orbit, try an ascent path like 0 km, 40 km, 5 degrees, 75%, with corrective steering off and limit to terminal velocity on. Looks like the max acceleration peaks around 60m/s/s, but the rocket seems to hold together fine despite wobbling a bit (you may still want to try a lower limit such as 40m/s/s and see if you get to orbit with more fuel or less that way). With less fuel in the second stage tanks and more time before apoapsis, and a much shorter circularization burn required, the craft was easily able to get oriented for a short time-warp to the start of the burn, resulting in a perfect 100km orbit.

You might also want to add a launch clamp - that thing REALLY wanted to fall over on the pad or break its engine off. ;)

[gmcnew]

This is a duplicate of #63, so I'm closing it.