[1.0.5] GravityTurn version 1.3.1 - Automated Efficient Launches (1.1 pre-release available)

[linuxgurugamer]

I just tried this.

I think it has promise, but:

the entry fields are too small, when I did the best guess, it put in number which are totally cut off.  I would double the size, and right-justify the entry fields to make it look more even.

With the best guess option, the ascent was way too shallow, I ended up going horizontal at 2300m/sec at 45,000m altitude. Took 1/2 an orbit to get to 80k altitude, and when it got to 70km, the window went away.  Not sure if that is intentional or not, but the orbit still dipped to 40KM until I circularized manually

[StahnAileron]

38 minutes ago, linuxgurugamer said:

With the best guess option, the ascent was way too shallow, I ended up going horizontal at 2300m/sec at 45,000m altitude. Took 1/2 an orbit to get to 80k altitude, and when it got to 70km, the window went away.  Not sure if that is intentional or not, but the orbit still dipped to 40KM until I circularized manually

The dev specifically stated this mod plugin doesn't do the circularization. It only does exactly what it says it is: an automated gravity turn mod for efficient launches. He did add support for MechJeb to do the final circularization if you have MJ installed and have it on your craft.

2300m/s @ 45km isn't that bad for a 80km apoapsis, I think. Really, that's a low orbital height, so of course you need to hit a high horizontal speed in a fairly shallow climb. Otherwise you run the risk of too high (miss your target height), too slow (more dV needed for the circularization burn because you don't have enough horizontal velocity), or you just won't reach space. (DISCLOSURE: I fly mainly SSTOs, so I'm used to seeing and flying shallow-ish launch profiles. So yes, my statement above is probably biased due to this.) Remember, getting to space (vertical velocity) is the easy part; staying in space (horizontal velocity) is the hard part.

As far as I can tell, the thing to look at isn't, "What is up with this profile?" or, "Why is this taking so long?" You should be looking at "Did using this save me fuel?" So I ask: How much fuel/dV did you spend on the launch itself? On top of that, how much dV was needed for the circ burn at the end? Did you use up more, less, or about the same amount of fuel/dV as your normal launch methods? IMHO, this is how you should be evaluating this mod. "Efficient" doesn't equate to "convenient" after all. (It's often the opposite.)

Edited January 14, 2016 by StahnAileron

[Overengineer1]

@linuxgurugamer: I'll work on the formatting.  But that ascent is not too shallow.  Yes, it should take 1/2 an orbit to get to 80k altitude.  That is the most efficient path, because overall even with these extremely shallow ascents you lose a LOT more speed to gravity than you do to aerodynamic drag.  Even though it looks like there's a lot of air up there while the flames shoot off your ship, the actual numbers tell us there is almost zero drag at that altitude.

Here's a sample of the latest build version that shows actual itemized losses, and you can see even for these shallow ascents the atmospheric loss is much lower than the gravity loss.  And this is on a ship that you can see clearly didn't give a lot of thought to aerodynamics.  Forgive some of the formatting, but I have verified the numbers.

[linuxgurugamer]

Interesting.  I didn't actually compare the dv used, but will wait for the next version because I like things looking pretty.

Re. the too shallow comment, I withdraw that :-)   I'm not used to such shallow ascents, will play with it after the next release.

which is when?

[Overengineer1]

I'm not sure.  As you can see the formatting needs some work, and I hate to think of some of the comments if I release it like that.  Maybe I'll have time tonight to get it cleaned up and tested.

The latest version will also include persistent saved settings on a per-ship-per-planet basis, for people the who are asking for that.

[vardicd]

I've been using this mod and playing around with it, and my rockets keep flipping out on those shallow ascents. Is there any guidelines to rocket building to make this mod work? every rocket I've built that flips out with this mod, I can get to orbit with a steeper ascent profile, so not sure what's going on. Any suggestions you can offer would be great.

[Overengineer1]

The easy answer is to put fins on the bottom of your rocket.

If your rocket is aerodynamically stable, it will be very easy to hold to prograde, because the aerodynamics will do it for you.  You can see how it will behave by going into the VAB and enabling the Center Of Mass and Center Of Lift icons.  You want the blue Center Of Lift icon to be below/behind the yellow Center Of Mass icon.  This way as you fly into the wind, the wind will catch on those fins and pull the back of your ship towards the back, instead of catching on the front of your ship and pulling the front of your ship to the back.

[NathanKell]

14 hours ago, mcirish3 said:

But I do have one thing.  In digging around I found out that there is almost 0 info about the pst 1.0 ksp atmo.  This is something I can may be do something about.  It would be nice if I had some kind of guidance from squad but well the creators of mecheb did not have so I guess I can figure it out.

 

[vardicd]

55 minutes ago, Overengineer1 said:

The easy answer is to put fins on the bottom of your rocket.

If your rocket is aerodynamically stable, it will be very easy to hold to prograde, because the aerodynamics will do it for you.  You can see how it will behave by going into the VAB and enabling the Center Of Mass and Center Of Lift icons.  You want the blue Center Of Lift icon to be below/behind the yellow Center Of Mass icon.  This way as you fly into the wind, the wind will catch on those fins and pull the back of your ship towards the back, instead of catching on the front of your ship and pulling the front of your ship to the back.

well, I do put fins on the bottom, though does it need to be steering fins? I'm still in the early part of a career and haven't unlocked the steering fins yet, so, does that make the difference?

[Overengineer1]

@vardicd Post a picture of your ship in the VAB.  F1 button will take a screenshot for you.

[vardicd]

This is the one in question that's given me the most trouble. if I let the mod use its defaults, the rocket tends to start flipping out somewhere around, 3500m to 5000m, usually on the lower end of that spectrum, before the side boosters detach. below are the figures I have to use to get that ship to orbit, in order:

 

48.5 m/s start turn
15.0 angle of turn
60 hold ap start
40 hold ap finish
75 apoapsis
0 roll
 

[Overengineer1]

That ship is very long, and has lots of stuff up top for the wind to catch.  In particular I really don't think you need those radiator panels.

It also looks like it will wobble quite a bit.  That wobbling will amplify the drag caused by those radiator panels.

If you just want to go to orbit, you have overbuilt that by quite a lot.  Take off the boosters, lose the radiators, and cut the lower stage in half.  If you want to go to Mun, just add fuel to the upper stage, not to the lower stage that you reduced by half.

[tempsgk]

@vardicd

 Could you give me that craft file? want to check.  It could be fixed by just  placing a docking port below the centre tank and control from there could solve the issue. 

Edited January 14, 2016 by tempsgk

[Superfluous J]

2 hours ago, Overengineer1 said:

You can see how it will behave by going into the VAB and enabling the Center Of Mass and Center Of Lift icons.  You want the blue Center Of Lift icon to be below/behind the yellow Center Of Mass icon.

This is extremely important and more importantly, your ship must remain that way throughout the flight.

Next time your ship flips out, cut the engines and right click the fuel tanks to see which ones are empty and which ones are full. Then to back into the VAB and empty those tanks which are empty, and turn on the COM icon. Note that the COM icon will have shifted downward in your rocket, causing the instability.

The easy fix here is to not make your rocket so tall. A different fix is to disable the top couple tanks until all the bottom tanks drain, keeping some mass up top where you want it (then, enable them again as you run out of fuel in the bottom tanks). SEE BELOW FOR AN EDIT: The best fix (imo) is to have an upper stage with a less powerful engine (The terrier is great for this) and using the top 2-3 tanks instead of your lower Reliant/Swivel. This will not only stabilize your rocket through launch but frequently give it MORE dV, without sacrificing much launch TWR.

I also agree that your rocket has far too much TWR at launch. 3 engines is way overkill.

EDIT: On a 2nd look I see you DO have a Terrier. I'd remove that one small fuel tank, and move one or two of the larger tanks from the center bottom stack into place instead. That should both fix your launch problems and give you more dV. After you fix the launch TWR problem

Edited January 14, 2016 by 5thHorseman

[vardicd]

55 minutes ago, Overengineer1 said:

That ship is very long, and has lots of stuff up top for the wind to catch.  In particular I really don't think you need those radiator panels.

It also looks like it will wobble quite a bit.  That wobbling will amplify the drag caused by those radiator panels.

If you just want to go to orbit, you have overbuilt that by quite a lot.  Take off the boosters, lose the radiators, and cut the lower stage in half.  If you want to go to Mun, just add fuel to the upper stage, not to the lower stage that you reduced by half.

It actually does not wobble much at all, and I have tried this design without the radiator panels. I added them because a previous design, which did work at getting into orbit, on a shallow ascent trajectory generated so much heat the command pod would explode before making it to orbit. If I remove the side boosters and cut down on the fuel tanks in the main stage it does not have enough DV to make orbit.

54 minutes ago, tempsgk said:

@vardicd

 Could you give me that craft file? want to check.  It could be fixed by just  placing a docking port below the centre tank and control from there could solve the issue. 

Will try to do this when I have time, but this is also early game career, no docking port researched yet.

13 minutes ago, 5thHorseman said:

This is extremely important and more importantly, your ship must remain that way throughout the flight.

Next time your ship flips out, cut the engines and right click the fuel tanks to see which ones are empty and which ones are full. Then to back into the VAB and empty those tanks which are empty, and turn on the COM icon. Note that the COM icon will have shifted downward in your rocket, causing the instability.

The easy fix here is to not make your rocket so tall. A different fix is to disable the top couple tanks until all the bottom tanks drain, keeping some mass up top where you want it (then, enable them again as you run out of fuel in the bottom tanks). SEE BELOW FOR AN EDIT: The best fix (imo) is to have an upper stage with a less powerful engine (The terrier is great for this) and using the top 2-3 tanks instead of your lower Reliant/Swivel. This will not only stabilize your rocket through launch but frequently give it MORE dV, without sacrificing much launch TWR.

I also agree that your rocket has far too much TWR at launch. 3 engines is way overkill.

EDIT: On a 2nd look I see you DO have a Terrier. I'd remove that one small fuel tank, and move one or two of the larger tanks from the center bottom stack into place instead. That should both fix your launch problems and give you more dV. After you fix the launch TWR problem

I agree with your assessment, however bear in mind this rocket flips out at 3500m to 5000m tops, within 10-30 seconds after launch. it doesn't have time to drain that much fuel. Not even a single tank is empty when it loses control. the center of mass can't have moved far at all.

A big problem is I'm still learning how to build with the new aero system. I'm still used to the old system. I still haven't gotten my head around how to build rockets that work well in the new system. and well, unlearning what I've learned is proving more difficult than expected.

Edited January 14, 2016 by vardicd

[mcirish3]

4 hours ago, NathanKell said:

 

As per usual you have just the info I need.  Thank You.

Edit: though the info is very useful it is not exactly what I was looking for but I think I can get it myself.

Edited January 15, 2016 by mcirish3

[Superfluous J]

1 hour ago, vardicd said:

I agree with your assessment, however bear in mind this rocket flips out at 3500m to 5000m tops, within 10-30 seconds after launch. it doesn't have time to drain that much fuel. Not even a single tank is empty when it loses control. the center of mass can't have moved far at all.

A big problem is I'm still learning how to build with the new aero system. I'm still used to the old system. I still haven't gotten my head around how to build rockets that work well in the new system. and well, unlearning what I've learned is proving more difficult than expected.

It doesn't HAVE to move that far at all, if it starts out low. An empty bathtub can hold lots of water. One that's already full can spill over if you just add a drop.

Still I'm surprised a bit that it flips before you even drop the side boosters. If I have time I'll try to build it and see what it does. One thing I noticed,you said you tilt 15 degrees at start I usually tilt 5 degrees at a time, and only slowly go over as I rise. I usually don't rock that much engine at the back though, and it may just be that the drag at the front is too much for your poor rocket. My rockets almost always do this:

• 1.5ish TWR (you can tell this at launch by the G meter on the right of the navball. Whatever that says when you launch, that's your TWR)
• 50m/s, tilt to 85 degrees (off Vertical)
• Switch to map mode and watch your Apoapsis. Fly with the navball and map mode. Or install KER so you can see Ap in normal mode.
• Tilt over as your rocket rises. Hold 85 until Ap is 4000, then tilt 5 degrees every 2000 meters. Again, I can't stress these are Ap meters, not rocket height meters.
• If you follow this, you should be at about 45 degrees when your Ap is at about 20km, and your rocket is at about 10km. Stop following this rule.
• Now you want to be at 40 degrees, and go down 5 degrees for every 5000 meters. So 35 degrees when Ap is at 25km, 30 degrees at 30km, 25 at 35, 20 at 40, 10 at 50, and 0 at 60.
• You can actually cut it down to 0 degrees when your Ap gets to 50 or so.

If you do that and your rocket flips out, it's either back-heavy (yours is), not aerodynamic enough (yours looks fine from that perspective), or has too much TWR which is causing too much drag at the front (I think yours is).

[mcirish3]

On 1/13/2016 at 9:38 PM, NathanKell said:

density is a function of pressure (in KSP that is constant for a given height and latitude and time of day) and temperature (in KSP that is constant for height but varies by latitude and time of day). So no, you can't trust that at x km the density is y kg/m^3.

I know you linked info from the Claw on the 1.0 change but I did not see (sorry if I missed it) any info on how pressure and temp change for a given height, latitude, and time of day.  in other words how fine in the step function that governs this?  Or I should say, I know  how both of these change with Altitude above sea level (thanks to graphotron 2000),but how do day and night effect these; is it also a nearly smooth gradient, like the altitude one, as Kerbin rotates or is it very  dramatic?  Also is the change smooth as I head north or south; so say for example at 10 degrees north will I see a sudden change in pressure and temp at a given altitude where up to that point it had been constant.  Or will I rather see a gradual change as I head north or south.

In other words are the functions governing this close approximations to smooth continuous functions thus mimicking the real world or are they corse step functions?

[Overengineer1]

@mcirish3 I whipped this up from the API.  X-axis is altitude, y-axis is pressure.

[Superfluous J]

@vardicd I built your ship as best I could. It had a decent number of non-stock parts so I just made a stock equivalent. As I suspected, the COM starts out quite low. Also, it has a TWR over 2 at launch which - as stated before - is too high. It launches *okay* but was a bit squirrely until the two side tanks go. I suspect your mod parts - and your tweaking that engine into the bottom tank - may have more to do with the flip than this mod.

One thing: The parachutes on the drop stages are essentially useless unless you've modded that too. By the time they trigger you're going so fast they get ripped off.

[mcirish3]

1 hour ago, Overengineer1 said:

@mcirish3 I whipped this up from the API.  X-axis is altitude, y-axis is pressure.

Cool share the csv?

 Also Atmo is tricky since as NathanKell says these graphs shift depending upon where  and when you take the measurements. So yes I can say that at 20min past noon near the equator on Kerbin the best approximate formula I can come up with describing change in pressure vs altitude is 

Pressure(?? units)= 104.39e(-2E-04)Altitude(m)
R² = 0.9987

 

BUT what it it at 40min past noon at 10 degrees north latitude?

Edited January 15, 2016 by mcirish3

[vardicd]

9 hours ago, 5thHorseman said:

It doesn't HAVE to move that far at all, if it starts out low. An empty bathtub can hold lots of water. One that's already full can spill over if you just add a drop.

Still I'm surprised a bit that it flips before you even drop the side boosters. If I have time I'll try to build it and see what it does. One thing I noticed,you said you tilt 15 degrees at start I usually tilt 5 degrees at a time, and only slowly go over as I rise. I usually don't rock that much engine at the back though, and it may just be that the drag at the front is too much for your poor rocket. My rockets almost always do this:

• 1.5ish TWR (you can tell this at launch by the G meter on the right of the navball. Whatever that says when you launch, that's your TWR)
• 50m/s, tilt to 85 degrees (off Vertical)
• Switch to map mode and watch your Apoapsis. Fly with the navball and map mode. Or install KER so you can see Ap in normal mode.
• Tilt over as your rocket rises. Hold 85 until Ap is 4000, then tilt 5 degrees every 2000 meters. Again, I can't stress these are Ap meters, not rocket height meters.
• If you follow this, you should be at about 45 degrees when your Ap is at about 20km, and your rocket is at about 10km. Stop following this rule.
• Now you want to be at 40 degrees, and go down 5 degrees for every 5000 meters. So 35 degrees when Ap is at 25km, 30 degrees at 30km, 25 at 35, 20 at 40, 10 at 50, and 0 at 60.
• You can actually cut it down to 0 degrees when your Ap gets to 50 or so.

If you do that and your rocket flips out, it's either back-heavy (yours is), not aerodynamic enough (yours looks fine from that perspective), or has too much TWR which is causing too much drag at the front (I think yours is).

No, no I meant, when I set this mod to tilt to 15 degrees, with the other numbers I worked out it can make it work, but when I launch it myself, I go with much less tilt. I'm home from work at the moment, just finished recording 2 launches of this rocket, one how I launch it, and one letting the mod do it, and that's saving now, soon as I get home in the morning i'll upload it to youtube, and share it here. I really appreciate all the help you guys are trying to give me, and I'm sorry I don't seem to be getting it, but maybe if you see how I launch it, vs how the mod does it, you can give me specific pointers instead of trying to build my ship yourself, and guess at it?

[linuxgurugamer]

6 hours ago, 5thHorseman said:

@vardicd I built your ship as best I could. It had a decent number of non-stock parts so I just made a stock equivalent. As I suspected, the COM starts out quite low. Also, it has a TWR over 2 at launch which - as stated before - is too high. It launches *okay* but was a bit squirrely until the two side tanks go. I suspect your mod parts - and your tweaking that engine into the bottom tank - may have more to do with the flip than this mod.

One thing: The parachutes on the drop stages are essentially useless unless you've modded that too. By the time they trigger you're going so fast they get ripped off.

If he is using StageRecovery, the parachutes don't deploy at all, but SR will use them to calculate how fast the stages would be dropping IF they were deployed safely when the stages finally hit the ground or ocean

14 hours ago, vardicd said:

This is the one in question that's given me the most trouble. if I let the mod use its defaults, the rocket tends to start flipping out somewhere around, 3500m to 5000m, usually on the lower end of that spectrum, before the side boosters detach. below are the figures I have to use to get that ship to orbit, in order:

 

48.5 m/s start turn
15.0 angle of turn
60 hold ap start
40 hold ap finish
75 apoapsis
0 roll
 

Please provide a list of the mods you have installed.

[linuxgurugamer]

One more comment on the mod:  I notice that the window is not draggable, can you please enable that so that it can be positioned manually?

 

Also, if I hide the window, it seems to disable the mod???

Another bug:  If I hide the window during the ascent, it disables the mod (already mentioned).  If I then reshow the window, and manually attain orbit, the window stays, and the button is gone, so I can't hide it anymore

Edited January 15, 2016 by linuxgurugamer

[Nansuchao]

16 hours ago, vardicd said:

This is the one in question that's given me the most trouble. if I let the mod use its defaults, the rocket tends to start flipping out somewhere around, 3500m to 5000m, usually on the lower end of that spectrum, before the side boosters detach. below are the figures I have to use to get that ship to orbit, in order:

 

48.5 m/s start turn
15.0 angle of turn
60 hold ap start
40 hold ap finish
75 apoapsis
0 roll
 

Your issue is caused by low RW power and low atmospheric control, the winglets aren't moving during ascent, the engines has not gimbal and the only reaction wheel from the pod is too feeble to steer the rocket. Also, with this high TWR you'll encounter high drag on the first part of ascent. For this kind of rockets, it's better to use the old manual control, with a steep trajectory on the first part of the ascent and then, after 20/25 km you can steer it and gain orizontal speed.

Edited January 15, 2016 by Nansuchao