Question about slow rotating craft (gyro? reaction wheel? what is gimbal?)

[fastbikkel]

I have searched the forum before, but i still find it confusing.

I was initially looking for something that would help my craft rotate around its axis in space, without using RCS or thrust.

The bigger they get, the slower they move.

So i thougt, let me add a reaction wheel, but is that really the trick? I did not seem to notice a difference.

Then i find words like "gimbal" here. What is that in relation to a gyro?

In short, what do i have to do to help my craft rotate quicker. I have trouble rotating in time for my maneuever node.

I really do not get this info from within KSP, rightclicking items in the assembly hall does not give all these details.

[Dodgey]

Reaction Wheels add more torque to the vessel which in turn means that you can turn faster at the cost of electric charge. With regards to turning in time for the maneuver nodes try turning so that your vessel lines up with the node before you time warp. And when you see the word Gimbal it is most likely referring to the Nav Ball that you see at the bottom centre of the screen.

[bsalis]

Reaction wheels should help to rotate a craft quicker. They consume electricity, so make sure you have a supply.

When approaching a maneuver node, you can orient the craft to the node marker well before arrival. You can also access Physics warp in space by holding the Alt key, then hit the warp keys, in order to make it happen faster.

[Taki117]

explains a little bit about how reaction wheels work. But yes, adding more will make your spacecraft rotate faster. Also,

[blizzy78]

And when you see the word Gimbal it is most likely referring to the Nav Ball that you see at the bottom centre of the screen.

No. Gimbals refer to engines that have thrust vectoring, meaning, they can divert their thrust from the center line so that the whole rocket would spin in either pitch or yaw direction. Using multiple engines would also let you roll.

The nav ball at the bottom is just that - the nav ball.

[KN_Namida]

Alright, seems like there's more than one people here confused about the terminology. So, you want your crafts to rotate quicker.

Using RCS is recommended, stability is improved and it provides sharper controls of your vessel, both in rotation and movement. If you don't want to use that, the only other way to rotate your ships is using reaction wheels. Bear in mind, all of the command pods and probe cores have reaction wheels (except the external seat), but with bigger rockets you may need to add proper in-line reaction wheels, or SAS units, to provide more torque (rotating power). Reaction wheels consume electricity, so adding more of them could be risky if you don't have any batteries or solar panels. Also, the more the reaction wheels are closer to the center of mass of the vehicle, the faster the vehicle will turn.

Gimbal is a completely different thing: many of the engines have thrust vectoring, the ability of changing the direction the nozzle is facing, thus diverting the exhaust gasses. This helps in maneuvering the rocket, just that, so you can't use it for rotating the spacecraft.

[fastbikkel]

Thanks!

But do the reaction wheels need to be attached to the lead module? Or can they be part of a stage that i willl later drop?

I remember fixing a big reaction wheel to a stage, but i really did not notice a difference.

[fastbikkel]

I have also looked up "gimbal", this is a part of the gyro.

"A gimbal is a pivoted support that allows the rotation of an object about a single axis."

[Silverchain]

Thanks!

But do the reaction wheels need to be attached to the lead module? Or can they be part of a stage that i willl later drop?

I remember fixing a big reaction wheel to a stage, but i really did not notice a difference.

Reaction wheels affect the whole ship they are attached to, so if you have extra reaction wheels added as part of your first stage, they will help you turn up until you drop the first stage.

I often add a reaction wheel to each stack in a stage, the extra mass is negligible by comparison with the extra torque.

[Red Iron Crown]

I have also looked up "gimbal", this is a part of the gyro.

"A gimbal is a pivoted support that allows the rotation of an object about a single axis."

That's the dictionary definition, and gimbals are used in the real world for many things like artificial horizons, boat compasses, etc.

When used in reference to KSP, it's generally understood to be the thrust vectoring feature of some engines. The exhaust nozzle in a vectoring engine has a gimbal for pitch and another for yaw.

[blizzy78]

I have also looked up "gimbal", this is a part of the gyro.

"A gimbal is a pivoted support that allows the rotation of an object about a single axis."

Right. So technically, yes, a hardware nav ball is using gimbals to rotate the surface. But that is a technical aspect that is irrelevant to its use. Whereas for the engine, moving the nozzle to divert thrust is also known as "gimbaling." So that should explain why the nav ball is generally not known as a "gimbal."

[fastbikkel]

That's the dictionary definition, and gimbals are used in the real world for many things like artificial horizons, boat compasses, etc.

When used in reference to KSP, it's generally understood to be the thrust vectoring feature of some engines. The exhaust nozzle in a vectoring engine has a gimbal for pitch and another for yaw.

Thanks, i understand now.

Although i would use "thrust vectoring" myself for clarity.

[Streetwind]

No. Gimbals refer to engines that have thrust vectoring, meaning, they can divert their thrust from the center line so that the whole rocket would spin in either pitch or yaw direction. Using multiple engines would also let you roll.

I hope you guys don't mind if I hijack this real quick... this is something I've been quietly scratching my head about for quite a while and just never got around to asking.

Basically, I have a rocket that is mostly steered via thrust vectoring during the gravity turn, because the ascent stage is so heavy that adding reaction wheels shows little difference and RCS would just become dead weight as soon as the ascent stage is discarded (the upper stage doesn't need it). And I have found that this rocket has a severe tendency to roll around its axis whenever I steer with the engines this way. Makes controlling the turn rather tricky as you can imagine.

The thrust vectoring is provided by four 1.25m engines on a 2.5m quad adapter; other engines are present but don't have gimbals. Now that you say "multiple engines would also let you roll", could that be the reason the ship rolls itself inadvertedly? Mind you, I am not touching ya or roll controls at all, I am simply pitching along a symmetry axis between those four auto-synchronized engines. All four of them should divert their identical thrust by an identical amount in an identical direction. But still it starts rolling, even against the reaction wheels (which are very good at controlling roll despite struggling with yaw and pitch).

Could you try help me understand where this rolling force is coming from?

[Claw]

NAV devices have gimbals, but would rater not use them. Ideally the gyro of a nav device would be free spinning as gimbals actually interfere with its operation. So I would not say they "use" gimbals, but they do have them. As far as I have seen, KSP does not simulate them (thankfully).

I hope you guys don't mind if I hijack this real quick... this is something I've been quietly scratching my head about for quite a while and just never got around to asking.

Basically, I have a rocket that is mostly steered via thrust vectoring during the gravity turn, because the ascent stage is so heavy that adding reaction wheels shows little difference and RCS would just become dead weight as soon as the ascent stage is discarded (the upper stage doesn't need it). And I have found that this rocket has a severe tendency to roll around its axis whenever I steer with the engines this way. Makes controlling the turn rather tricky as you can imagine.

The thrust vectoring is provided by four 1.25m engines on a 2.5m quad adapter; other engines are present but don't have gimbals. Now that you say "multiple engines would also let you roll", could that be the reason the ship rolls itself inadvertedly? Mind you, I am not touching ya or roll controls at all, I am simply pitching along a symmetry axis between those four auto-synchronized engines. All four of them should divert their identical thrust by an identical amount in an identical direction. But still it starts rolling, even against the reaction wheels (which are very good at controlling roll despite struggling with yaw and pitch).

Could you try help me understand where this rolling force is coming from?

A picture would help. If yor motors are misaligned on the decouplers or the connections aren't strong enough, your boosters might be bending which can cause yor rocket to roll.

Also, sometimes the SAS does some goofy things which might cause some swaying and twisting which can lead to unwanted rolls. Check if the roll trim indicator is bouncing around or if it stays stuck to one side.

[fastbikkel]

@streetwind

Just curious, you really mean rolling or rotating around it's axis?

I noticed the rocket rotating around its axis during takeoff a few times yes, but i could not understand where it came from. Only on the first stage, as soon as i separated it, it was gone.

I always add winglets on the bottom of the rocket to compensate for rotating around axis/lack of control.

But the winglets never really work.

I never discovered automatisch rollling so to speak, if i do not do anything it will just go up straight in a light corkscrew motion.

Edited February 4, 2014 by fastbikkel

[SSSPutnik]

Could you try help me understand where this rolling force is coming from

Multiple gimballing engines can introduce rotation. Often I just have a central engine with this enabled and the others with vectoring disabled.

With a quad this is more tricky. You could try disabling on diagonal opposites.

SRBs and uneven strutting can do this too or uneven mass distribution.

AV8 winglets can help, they will fight rotation.

[DrMonte]

SAS - I'm not sure about putting them in the center of mass. Usually everyone is trying to put SAS as far to the rocket ends as possible (front or back). Test it! - that's the fun of KSP. One or two is enough to have in space on average rocket. You might get quite funny effects by putting more SAS on big rockets in various places. SAS-Autopilot gets useless in that cases. So if you can't rotate the ship with somewhat 3 SAS in timely manner I would suggest using RCS.

Talking about liftoff - don't overdo with SAS.

Winglets! - don't forgets about this magical part: it is very light, adds a good amount of stability. Note that there are few types of winglets: some rotates, some don't (Try "AV-R8 Winglet" or "Delta-Deluxe Winglet"). Winglets on lower stages can hold up any size rocket without adding mass to it (somewhere up to 50km).

RCS is not for liftoff. You might use it in upper stages for super heavy, super unstable rockets.

[SSSPutnik]

Read somewhere here that in theory, SAS should be near center of mass, but in game, you can whack em anywhere, no difference.

[DrMonte]

Read somewhere here that in theory, SAS should be near center of mass, but in game, you can whack em anywhere, no difference.

Oh it has. SAS can even smack your ship to pieces if you put it wrong.

[SSSPutnik]

Well yes, if you place poorly or on weak parts, but I believe it won't affect turning speed.

[Streetwind]

A picture would help. If yor motors are misaligned on the decouplers or the connections aren't strong enough, your boosters might be bending which can cause yor rocket to roll.

Also, sometimes the SAS does some goofy things which might cause some swaying and twisting which can lead to unwanted rolls. Check if the roll trim indicator is bouncing around or if it stays stuck to one side.

I'm using a quad adapter, so the alignment is perfectly even. I've also added Kerbal Joint Reinforcement and the rocket behaves no different than before.

I'll keep an eye on the roll indicator, but gut feeling says it rolls with SAS off too (I may have turned it off midflight once or twice in order to have an easier time wrestling the rocket around while coasting).

Just curious, you really mean rolling or rotating around it's axis?

I noticed the rocket rotating around its axis during takeoff a few times yes, but i could not understand where it came from. Only on the first stage, as soon as i separated it, it was gone.

I always add winglets on the bottom of the rocket to compensate for rotating around axis/lack of control.

But the winglets never really work.

I never discovered automatisch rollling so to speak, if i do not do anything it will just go up straight in a light corkscrew motion.

Yes, I mean rolling, as in the same motion induced by (or countered by) the Q and E buttons.

My rocket flies perfectly straight with SAS on; it's only when I try to pitch into the gravity turn that it induces roll.

Multiple gimballing engines can introduce rotation. Often I just have a central engine with this enabled and the others with vectoring disabled.

With a quad this is more tricky. You could try disabling on diagonal opposites.

SRBs and uneven strutting can do this too or uneven mass distribution.

AV8 winglets can help, they will fight rotation.

I'm using the quad cluster of engines in this case because they give more thrust than a single skipper equivalent engine and have much better Isp than a mainsail. There's a significant time during which the rocket climbs using the already slightly underpowered central stack only, after shedding its liftoff boosters. Hence I can't afford to lose about 25% thrust by dropping down to a skipper. Meanwhile the fuel usage is such that the rocket finishes circularizing a 70km orbit exactly when the fuel on the ascent stage runs out, which leaves the decoupled transfer stage to go wherever it wants with a full tank. A mainsail would guzzle too much. Too bad there's no 2.5m engine with ~1000 kN thrust in stock or the mods I have... Maybe I can tweak one of KW Rocketry's some to fill that niche.

The roll happens both while boosters are still present and after they have been shed (but they're liquid boosters, not SRBs).

That said, the thing is strutted all over the place, and there's some OX-STAT panels that are in 3-part symmetry and not 4/8-part like the rest of the rocket, and two spotlights in a 2-part symmetry on the lander. But, those things clock in at 5 kg each, on a rocket that weighs 150 tons.

Winglets I've considered; I usually include them, but in this design they don't fit well because the bottom of the rocket is surrounded by no less than 8 boosters (I blame Jeb). Because of the way winglets attach to circular surfaces, it's almost impossible to attach them without having them clip into a booster, and they would definitely be awkwardly offset from the normal axes of symmetry I am using for the rest of the build.

As you can see, I've already put a lot of thought into that rocket, and trying to change it pretty much boils down to starting over fresh anyway. That's okay, I don't need help fixing it, it does work reliably after all. It's just a little willful during ascent. My question was more directed at gaining more insight into the nature of things that cause a rocket to roll while trying to steer it.

To that end I've gotten some good info here.

Edited February 4, 2014 by Streetwind

[fastbikkel]

I am also reading things i have never seen before.

Like being able to disable vectoring, where the heck do i do that? :-)

No hurry, i might be able to find it myself.

[blizzy78]

Like being able to disable vectoring, where the heck do i do that? :-)

Right click engine, select "Lock gimbal" or some such. It also works in the editor and is saved into the craft file.

[Claw]

If you have control surfaces on your rocket in addition to SAS, that might be part of it depending on the design. (That's why I was curious about a picture.)

If the rocket starts rolling when you pitch, try disabling the yaw and pitch axis on your control surfaces (via tweakables) and see if that helps. Sometimes the SAS doesn't do an effective job of stopping motion when too many things are fighting.

As for weak parts, they can throw off your mass balance even if they aren't the pieces attached to the engines. Winglets can cause bending/twisting too, even if struts are installed. So without seeing a picture I hesitate saying "just add more SAS and winglets." That could fix the problem, but could also make it worse.

Edited February 4, 2014 by Claw

[Streetwind]

Now that I am home, I went ahead and made a test. As expected, the mainsail sucked my fuel tank dry before I could circularize even with careful throttling to stay at terminal velocity... but here's the thing: the rocket flew straight as an arrow the entire time. Not even the tiniest hint of rolling while steering. In fact it was one of the most well behaved rockets I've ever flown (with the caveat that the main engine had to be running in order to steer, but that's what it was designed for anyway).

It seems then that it's really the quad-cluster of engines that induces rolling, even when perfectly in symmetry. Kind of a bummer, I like the ability to build a thrust profile not available in a 2.5m engine, or benefit from better Isp.