How do you launch an unbalanced rocket?

[Loren Pechtel]

The imbalance was tiny--a scientific instrument on one side but not the other (there was an antenna in the opposite position, though.)  I tried many ways of launching that but no matter what I did it would spin out of control at Mach 1, whether I flew it manually or let MechJeb run it.  Putting a fairing over the instruments didn't help, either.  Medium reaction wheel, engines with a bit of gymbal and I even slapped a set of vernier engines on it to see if they could hold it stable--nope.

I finally gave up and wasted the money on putting a duplicate on and it flew to Minmus without a hiccup.

[regex]

Missing gimballing on the engine?  Fins?  Lack of control authority probably killed it.  Also, any time you ask a question like this post a picture of the craft.

[Snark]

I assume that you're talking about the Mystery Goo, yes?  'Coz that's the only one that would do that.

The other radially attached science instruments are physicsless, meaning that they add their drag and mass to the parent part, so you don't have to worry about placing them symmetrically.

And if you placed a goo unit on one side... placing a physicsless part on the other side (such as a Communotron-16 antenna) to try to balance it won't work.  So you pretty much have to use goo units in pairs.

Edited March 12, 2016 by Snark

[Loren Pechtel]

1 hour ago, regex said:

Missing gimballing on the engine?  Fins?  Lack of control authority probably killed it.  Also, any time you ask a question like this post a picture of the craft.

I do agree it had to have been a lack of control authority but why did such a tiny imbalance do it?

And while I misspelled gimballing the engines had it.  I tried a bunch of fins, no change.  My original design used two orange tanks and an engine that gave me about 1.5 TWR, I tried several designs, including as many as three strap-ons, each with two fins on it.  No matter what I did it flew acceptably up to Mach 1, then spun out.  When I tried flying it manually it wanted to pull towards the side with the scanner (really, now--a scanner I could pick up is enough to wig out a rocket weighing many tons?!) but then spun out in the opposite direction at Mach 1.

I didn't take a screenshot because pairing the scanners was more a lets-try-this-because-I-haven't than an expectation of success.  I did post the control equipment on the bird.

[Jhawk1099]

I doubt it is the science instrument itself but rather a un aerodynamic rocket

 

[Loren Pechtel]

1 hour ago, Snark said:

I assume that you're talking about the Mystery Goo, yes?  'Coz that's the only one that would do that.

The other radially attached science instruments are physicsless, meaning that they add their drag and mass to the parent part, so you don't have to worry about placing them symmetrically.

And if you placed a goo unit on one side... placing a physicsless part on the other side (such as a Communotron-16 antenna) to try to balance it won't work.  So you pretty much have to use goo units in pairs.

No, it was the good altitude scanner from the ScanSat mod.  The real purpose was to send up a survey scanner in preparation for ore mining but since I hadn't done an altitude survey yet why not?

Goo is cheap, I normally simply pair it but that scanner is a 25k part.  I didn't realize the other things didn't need pairing, the only time I've used them unpaired on a rocket was with a lander where I paired them with another instrument on the other side.

Just now, Jhawk1099 said:

I doubt it is the science instrument itself but rather a un aerodynamic rocket

 

When I paired the radially-attached stuff it flew fine.

[Snark]

1 minute ago, Loren Pechtel said:

No, it was the good altitude scanner from the ScanSat mod.  The real purpose was to send up a survey scanner in preparation for ore mining but since I hadn't done an altitude survey yet why not?

Goo is cheap, I normally simply pair it but that scanner is a 25k part.  I didn't realize the other things didn't need pairing, the only time I've used them unpaired on a rocket was with a lander where I paired them with another instrument on the other side.

Ah, okay.  It really helps if you include a screenshot or say specifically what you were dealing with-- it's pretty important.  If you're running something from a mod, especially so.

Unbalanced rockets = problematic, if the size of the imbalance is significant relative to the rocket, yes.

[Loren Pechtel]

4 minutes ago, Snark said:

Ah, okay.  It really helps if you include a screenshot or say specifically what you were dealing with-- it's pretty important.  If you're running something from a mod, especially so.

Unbalanced rockets = problematic, if the size of the imbalance is significant relative to the rocket, yes.

The thing is the size of the imbalance seems trivial here.  I would have expected trouble had I tried to hang something heavy on sideways.

[regex]

3 minutes ago, Loren Pechtel said:

The thing is the size of the imbalance seems trivial here.  I would have expected trouble had I tried to hang something heavy on sideways.

Might have something to do with atmospheric drag rather than mass.

[Renegrade]

I have to go with the aerodynamic side - the "near mach 1" thing is almost always an aero red flag.  Perhaps it's a very draggy antenna as regex implied.

[Superfluous J]

It's likely that your rocket's center of mass is far backward (this happens as you burn fuel yet have heavy engines at the back). and maybe the rocket is JUST aerodynamic enough to make it through the worst part of the atmosphere without flipping, but adding this little bit of extra drag was enough to tip the scales.

The 2 easiest ways to make the rocket more stable are to disable the top fuel tank of the first stage  until the bottom ones run out, and to add fins to the very bottom so it wants to keep going forward. I'd try them in that order, not shying away from doing both. Another option is to redesign the rocket so there's more weight up front all the way up, perhaps splitting that launch stage into 2 stages or adding more fuel to the upper lifting stage and taking it away from the bottom one.

As others have said, though, without pics we can only guess.

[Aetharan]

3 hours ago, regex said:

Might have something to do with atmospheric drag rather than mass.

This.  A thousand times this.  If you've strapped something onto the side of your rocket asymmetrically that isn't a physicsless part, then that means you have more drag on that side of your rocket.  Done very, very carefully it could result in a natural gravity turn, but most of the time it's just going to result in your ship spinning out of control unless you've massively overpowered its control authority.

You have two basic options to make your craft fly more easily under these circumstances.  The first is to remove the aerodynamic imbalance by concealing the unpaired part within a fairing / service bay.  The second is to pair all drag-inducing parts.  Never hurts to have a spare anyway-- you never know when a collision is going to wreck one of 'em.

[Renegrade]

If it is indeed an aero problem (and I really do think it is), an alternate to fixing it is to simply use an old souposphere-style "gravity turn" after climbing very slowly.  That is, go up to like 25km-30km at subsonic speed, and then turn over hard east and speed up.

It's terrible, inefficient, horrible, awful, and all sorts of bad things, but it can get most broken rocket designs out there into orbit.

Mind you I'd only do that as a last resort (or with a particularly inappropriate payload that can't be modified for some reason)...

[Rocket In My Pocket]

I run into this problem sometimes with early career rockets before you have service bays, rotating the rocket can sometimes equalize the stability issue until you get far enough out of the atmo where it doesn't matter.

So as it starts to roll over to one side, you rotate that side 180 degrees so it starts heading back the other way, rinse, repeat.

Your ascent should resemble a big elongated series of "S's" however be warned that this isn't terribly efficient or safe.

[Snark]

6 hours ago, 5thHorseman said:

It's likely that your rocket's center of mass is far backward (this happens as you burn fuel yet have heavy engines at the back). and maybe the rocket is JUST aerodynamic enough to make it through the worst part of the atmosphere without flipping, but adding this little bit of extra drag was enough to tip the scales.

+1 to this likely explanation.  Especially after seeing this comment:

9 hours ago, Loren Pechtel said:

My original design used two orange tanks

...by which I'm guessing you had one of them stacked on top of the other?

Stacking fuel tanks vertically often leads to woe.  The problem is that the engine drains the stack from the top tank down, which is the exact opposite of what you want, because it drastically lowers your CoM and causes aerodynamic instability.

What makes that particular problem especially insidious is that people who get bitten by it often don't realize that it has nothing to do with how fast you're going.  Yes, the problem that it causes gets worse if you're going really fast, but the fundamental problem is not that "my rocket went too fast and then flipped," but rather "my rocket burned a certain amount of fuel and then flipped."  It's hard to distinguish these two cases because your rocket is in fact going faster-and-faster, and it is in fact burning more-and-more fuel.  Debugging a problem when the actual reason is something different from what you think it is, is an exercise in frustration.

And, to add insult to injury:  rockets for which that is the problem often aren't helped by adding fins.  Even big fins. Even steerable fins.  Even lots of fins.  Because a fin only helps you if it's far from the CoM, and if your CoM is sitting down at the bottom of the rocket, all the fins in the world aren't going to help you.  Nor will making the front end of the rocket more streamlined.

Anyway:  If you had a rocket whose aero stability was marginal to begin with, it may be that just a fairly small imbalance would be enough to push it over the edge.

Edited March 12, 2016 by Snark

[Loren Pechtel]

34 minutes ago, Snark said:

+1 to this likely explanation.  Especially after seeing this comment:

...by which I'm guessing you had one of them stacked on top of the other?

Stacking fuel tanks vertically often leads to woe.  The problem is that the engine drains the stack from the top tank down, which is the exact opposite of what you want, because it drastically lowers your CoM and causes aerodynamic instability.

What makes that particular problem especially insidious is that people who get bitten by it often don't realize that it has nothing to do with how fast you're going.  Yes, the problem that it causes gets worse if you're going really fast, but the fundamental problem is not that "my rocket went too fast and then flipped," but rather "my rocket burned a certain amount of fuel and then flipped."  It's hard to distinguish these two cases because your rocket is in fact going faster-and-faster, and it is in fact burning more-and-more fuel.  Debugging a problem when the actual reason is something different from what you think it is, is an exercise in frustration.

And, to add insult to injury:  rockets for which that is the problem often aren't helped by adding fins.  Even big fins. Even steerable fins.  Even lots of fins.  Because a fin only helps you if it's far from the CoM, and if your CoM is sitting down at the bottom of the rocket, all the fins in the world aren't going to help you.  Nor will making the front end of the rocket more streamlined.

Anyway:  If you had a rocket whose aero stability was marginal to begin with, it may be that just a fairly small imbalance would be enough to push it over the edge.

I tried several configurations of the rocket, all flipped at Mach 1.

[Snark]

1 hour ago, Loren Pechtel said:

I tried several configurations of the rocket, all flipped at Mach 1.

Fair 'nuff, just wanted to make sure that it wasn't the vertically-stacked tanks problem.

Really hard to say, without seeing a screenshot.

[Red Shirt]

Take the offending part off the craft and see if it flies. That will help narrow down the problem.

[Corw]

Don't do Mach 1 before 35 km and don't do proper gravity, go straight up to thin out atmosphere? Not optimal but has worked for me before.

[oglommi]

17 hours ago, Loren Pechtel said:

The imbalance was tiny--a scientific instrument on one side but not the other (there was an antenna in the opposite position, though.)  I tried many ways of launching that but no matter what I did it would spin out of control at Mach 1, whether I flew it manually or let MechJeb run it.  Putting a fairing over the instruments didn't help, either.  Medium reaction wheel, engines with a bit of gymbal and I even slapped a set of vernier engines on it to see if they could hold it stable--nope.

I finally gave up and wasted the money on putting a duplicate on and it flew to Minmus without a hiccup.

Post a picture or the a file with the rocket or something

[Dispatcher]

When I fly imbalanced craft, I place other parts (usually science parts) to balance the overall mass.  It works well enough.

[Wemb]

On 12/03/2016 at 7:24 AM, 5thHorseman said:

It's likely that your rocket's center of mass is far backward (this happens as you burn fuel yet have heavy engines at the back). and maybe the rocket is JUST aerodynamic enough to make it through the worst part of the atmosphere without flipping, but adding this little bit of extra drag was enough to tip the scales.

Might be possible to counter it by changing your roll during ascent - if the drag is going to produce torque that coupled with gravity flips you, it may be worth rolling to ensure the instrument is on the other side of the rocket, producing torque that counters gravity.  But having said that, I've launchd these scanners before now without problems - also take a look at your TWR and CoM. A picture of your rocket would really help.

Wemb

[Warzouz]

On 12/3/2016 at 8:07 AM, Renegrade said:

I have to go with the aerodynamic side - the "near mach 1" thing is almost always an aero red flag.  Perhaps it's a very draggy antenna as regex implied.

Yes it it was mass imbalance it would flip right at take-off.

[Madscientist16180]

On 11/3/2016 at 8:57 AM, Snark said:

+1 to this likely explanation.  Especially after seeing this comment:

...by which I'm guessing you had one of them stacked on top of the other?

Stacking fuel tanks vertically often leads to woe.  The problem is that the engine drains the stack from the top tank down, which is the exact opposite of what you want, because it drastically lowers your CoM and causes aerodynamic instability.

I never ran into this problem to such an extant that careful piloting could not get me out of it, si never tried anything to rule it out.

So this might be quite a stupid suggestion. Would puting the tank "upside down" cause them to drain from the bottom up?

[Snark]

15 minutes ago, Madscientist16180 said:

So this might be quite a stupid suggestion. Would puting the tank "upside down" cause them to drain from the bottom up?

Not a stupid suggestion at all!  But a wrong one. 

The flow is irrelevant to the tank's orientation.  The fuel flow system is designed to "draw fuel from the farthest connected thing", and in the case of vertical stacks of tanks, that's from the top down.

A workaround for this problem, if you really have to stack tanks, is to disable the top tank in the VAB so that it can't drain at all; that forces the bottom tank to drain.  It means some more micromanagement during flight, since you'll have to remember to re-activate the top tank when the bottom tank is nearly empty, but hey, any port in a storm, right?