Rockets withot crew unstable, why?

[MotherRussia]

Hi There!

This thing was giving me some rages a lot. Whenever I make a satellite, my rocket can't get into space because it either keeps spinning, or it already flips in 5-8km... But the rocket base I used for it was the same as the one that I landed on mun, and got back with (I carried a scientist kerbal there). Why is it that when I have a kerbal in the rocket it flies good, but when I give it a probe it becomes uncontrolable?

Edited November 11, 2017 by MotherRussia

[Snark]

Hello, and welcome to the forum!

1 hour ago, MotherRussia said:

This thing was giving me some rages a lot. Whenever I make a satellite, my rocket can't get into space because it either keeps spinning, or it already flips in 5-8km... But the rocket base I used for it was the same as the one that I landed on mun, and got back with (I carried a scientist kerbal there). Why is it that when I have a kerbal in the rocket it flies good, but when I give it a probe it becomes uncontrolable?

The general reason why rockets flip is that they're aerodynamically unstable.  Usually it's because the CoM (center of mass) is too low.  The way to fix it is to move the CoM forward; make sure the front end is nice and streamlined (i.e. pointy); and put fins on the back, which need to be as far behind the CoM as possible.

That doesn't answer the question of why a rocket would work with a pilot but wouldn't work without one.  That will come down to the physical design of the rocket.  For example:

• Suppose that your crewed ship had a crew pod on the front end, like a Mk1 command pod.
• And to make it uncrewed, suppose you ditched the Mk1 pod, and replaced it with a probe core.
• Result:  The probe core has a much lower mass than the heavy pod.  Which moves your CoM significantly lower on the rocket.  Which makes the ship unstable.

That's just one example of what could happen.  Another possibility would be if you were using a probe core that doesn't have SAS available, so it's not as good at holding attitude as a crewed ship with a pilot.

Would help a lot if you could post a screenshot or two-- then we could do a better job of diagnosing your actual problem instead of just a bunch of what-ifs.  Could you post a screenshot of your crewed ship, and then another one of the probe ship?  (Ideally in the VAB, with the CoM display turned on.)

[Draalo]

When you launch, what probecore do you use? The Stayputnik has no SAS stability, you are forced to steer yourself.

Try OCTO probecore, it can hold, HECS can steer on prograde on its on.

 

And as allways: pictures please

(upload them to a hoster like imgur.com , post the link here)

[Gordon Fecyk]

My thoughts are that the mass of a Mk1 command pod is considerably more than that of a typical probe core, and also it's a little more aerodynamic as long as there's something pointy on top like a Mk16 parachute.

Here's a thought: Take a rocket with a Mk1 pod that you know is already stable, stick a Stayputnik or OKTO probe core on the top of it, remove the kerbal from it, and see if it's any less or more stable than the version without the probe core. This is what Snark is trying to get at.

If you can match the mass of the Mk1 pod at the top, you should get a similar end result. The thing is 800 kg by itself without monopropellant. I can cram fourteen parts into a similar package with the same aerodynamics, and have three to five unique science experiments.

 

[MotherRussia]

Here are the screenes of the ship and the satellite. Trying to get the polar orbit so I can gather info about the Kerbin's surface, and to try out the new maps.

 

I just noticed I left the science thing there. Please ignore it The satellite rocket is supposed to be without it

 

Edited November 11, 2017 by MotherRussia

[Gordon Fecyk]

Thanks for posting that. Those solar panel arms are not very aerodynamic; they would rip off, or at best cause your craft to tumble on ascent.

Try using a 1.25 m service bay to tuck your probe core into, and if you're far enough along in the tech tree, use deployable solar panels instead of that truss structure. You should be able to hide the panels in the service bay until it's in space, then open the service bay and then deploy the panels. You can also tuck batteries in there.

[Snark]

5 hours ago, MotherRussia said:

Here are the screenes of the ship and the satellite. Trying to get the polar orbit so I can gather info about the Kerbin's surface, and to try out the new maps.

Ah, okay, thanks!  Makes things much clearer.

So, with these screenshots at hand, it's clear that you've got a lot of problems there.  Fortunately they're all pretty straightforward to address. 

• First and foremost... holy cow that thing on the top of the ship is a monstrosity.  Seriously, the drag on that's gonna be hideous.  How to fix:  get rid of those "arms", they're nightmarishly draggy and completely unnecessary.  Just use the deployable solar panels instead, or even just put a trio of OX-STAT panels around the sides of the probe core.  You don't need anywhere near that many solar panels, and those arms sticking out are gonna just kill you.  Also, put a fairing around that top science stage.
• Looks like you're using Reliants as your main boost engines.  Nice workhorse of an engine, sure... but it's got no gimbal and therefore no control authority.  Suggest making sure that at least the central stack is a Swivel, if it's not already-- engine gimbal really helps a lot for controllability during ascent.
• Put fins on it.  This is a really easy thing to add.  Stick 'em on your radial boosters, as low down as you can possibly put them.  You may want to consider steerable fins, such as the AV-R8, since you're not using gimbaled engines.
• I see that you're using vertically stacked fuel tanks.  Make sure that you set the fuel flow priority so that the bottom tanks will drain first.  If you don't do that, the top and bottom tanks will drain together.  Why is this so important?  Because by making the bottom tanks drain first, the rocket's CoM will rapidly rise higher as the fuel drains, which will greatly aid your aerodynamic stability.

My guess as to why you have better luck with the crewed version is that it has that heavy pod up top (raising your CoM), and is also much more aerodynamic.  Less drag in the front = flies better.

[Jim DiGriz]

It looks like you don't have any SAS modules and don't have fins.

A Mk1 command pod has 5.0 kN·m of torque.  An OKTO core only has 0.3 kN·m of torque.  The flat OKTO2 has exactly 0 kN·m of torque.

You probably just need one of these:

https://wiki.kerbalspaceprogram.com/wiki/Advanced_Inline_Stabilizer

Stick that on the last stage before your probe.

Also make sure your core sustainer is an LV-T45 with a gimbal and that your outer boosters are LV-T30's without them (to start with, as a general rule, always follow this pattern of having a core engine that gimbals and outer stages that do not gimbal or which you've manually locked the gimbals).

(And, as far as your aerodynamically poor top of your rocket -- meh, have fun and ignore the haters....  I've launched worse...)

Edited November 12, 2017 by Jim DiGriz

[Snark]

35 minutes ago, Jim DiGriz said:

A Mk1 command pod has 5.0 kN·m of torque.  An OKTO core only has 0.3 kN·m of torque.  The flat OKTO2 has exactly 0 kN·m of torque.

You probably just need one of these:

https://wiki.kerbalspaceprogram.com/wiki/Advanced_Inline_Stabilizer

Stick that on the last stage before your probe.

I doubt that would help much with the OP's problem. Aerodynamically unstable rockets tend to generate aero torque *far* higher than reaction wheels can cope with.  Fins, CoM placement, and aerodynamic front ends, on the other hand, tend to be much more helpful, since their effects scale with increasing aero forces as the rocket accelerates.

(Would a reaction wheel be helpful later in the flight, when the ship is in vacuum or near-vacuum? Absolutely. But it probably won't solve the actual problem under discussion here, which is aerodynamic instability during the high-Q part of the ascent.)

35 minutes ago, Jim DiGriz said:

(And as far as your aerodynamically poor top of your rocket -- meh, have fun and ignore the haters....  I've launched worse...)

It's a factor worth considering. Can one launch stuff with draggy contraptions on the front end? Sure, especially when it's done by an experienced player who understands how to design the rest of the rocket for stability. But such issues do make life more difficult, and if a relatively inexperienced player is having trouble building a stable craft, it can make the difference between "works fine" and "doesn't work".

In this particular case, the OP has built a thing on the front that is both lightweight *and* extraordinarily draggy, which is definitely an issue. As shown, trying to fly that design is basically going to be like trying to throw a badminton birdie with the feathered end in front: i.e., more difficult than it needs to be.

I might also suggest that "helpful, friendly people offering reasonable advice" might be a more apt term than "haters", though admittedly that doesn't roll off the tongue quite as briskly.

[Jim DiGriz]

There's still an order of magnitude less control going from the Mk1 to the HECS (which i see is what the OP is using which is 5.0 nM to 0.5 nM.  If the engines are LV-T30's then the OP would have basically gone to zero torque authority.  I'd rule that out first before redoing the whole rocket and payload.  Then add some fins.  Then, yes, consider the aerodynamic monstrosity you've built -- but this is KSP.

[Snark]

12 minutes ago, Jim DiGriz said:

There's still an order of magnitude less control going from the Mk1 to the HECS (which i see is what the OP is using which is 5.0 nM to 0.5 nM.

Sure. But aero forces are at least an order of magnitude bigger than that. For a sizable rocket like that, I expect that the paltry 5 Nm of torque from the Mk1 pod is practically irrelevant, other than perhaps being useful to give the rocket an initial eastward nudge right off the pad to start the gravity turn.

It's kind of like my climbing a mountain while lugging a 1-ounce weight or a 10-ounce weight. One is an order of magnitude bigger than the other, sure... but both are sufficiently dwarfed by my own weight that if I'm having trouble climbing the mountain, "the load was 9 ounces too big" is fairly unlikely to be the reason.

[Spricigo]

5 hours ago, Jim DiGriz said:

There's still an order of magnitude less control going from the Mk1 to the HECS (which i see is what the OP is using which is 5.0 nM to 0.5 nM.   If the engines are LV-T30's then the OP would have basically gone to zero torque authority.  I'd rule that out first before redoing the whole rocket and payload.  Then add some fins.  Then, yes, consider the aerodynamic monstrosity you've built -- but this is KSP.

Since he also have a small Reaction Wheel the probe actually has slight more torque.  And 5kNm was enough to launched the crewed vessel without SAS as OP reported (landed in the Mun with a scientist). I'd say is pretty obvious is not lack of control authority.