[1.1.x] Space Shuttle Engines (2016-07-03)

[Silpion]

Love the new big engine. I'm having an issue with shuttles though: because these engines are well off the side from the center of mass, there is a convolution between yaw and roll: When the engines yaw, the craft both yaws and rolls. This often leads my shuttles to roll out of control, especially late in the ascent, out of the atmosphere.

So question: is the axis that the engines do their yaw gimbal relative to along the main thrust axis, or the direction that the craft is pointing? If it is along the thrust axis then this shouldn't be happening (because presumably that axis is through the COM).

Suggestions: If it is relative to "forward", is it possible to make it relative to the engine's thrust axis? Separately, can the yaw gimbal range be kept less free than the pitch? Like keep the 10 degree pitch range, but decrease yaw to a more conventional 1 degree? That could slow some of this down.

[Silpion]

Followup: I'm pretty certain now that the yaw is done relative to the spacecraft axis and not the thrust axis, because I made a shuttle which has the engines nearly perfectly aligned with the CoM (I can take my hands off with SAS off and it flies nearly straight), and the yaw command sends it rolling like mad.

[dtobi]

Followup: I'm pretty certain now that the yaw is done relative to the spacecraft axis and not the thrust axis, because I made a shuttle which has the engines nearly perfectly aligned with the CoM (I can take my hands off with SAS off and it flies nearly straight), and the yaw command sends it rolling like mad.

I noticed the same. I typically try to minimize all yaw and roll operations during the ascent. I align the launch position so that I only need to pitch to get into orbit. I'll see if a limited yaw angle is possible.

Thanks for testing.

[Silpion]

I noticed the same. I typically try to minimize all yaw and roll operations during the ascent. I align the launch position so that I only need to pitch to get into orbit. I'll see if a limited yaw angle is possible.

Yeah, but the problem with that is that it requires perfect performance of the spacecraft. Boosters have a habit of causing roll particularly as they run short on fuel, and this triggers a feedback effect with SAS.

I think if these shuttle engines are every going to perform like the user expects, the yaw angle must be relative to the engine-COM axis.

[dtobi]

Yeah, but the problem with that is that it requires perfect performance of the spacecraft. Boosters have a habit of causing roll particularly as they run short on fuel, and this triggers a feedback effect with SAS.

I think if these shuttle engines are every going to perform like the user expects, the yaw angle must be relative to the engine-COM axis.

For me it works just fine with the stock boosters. What kind of boosters do you use?

[Silpion]

For me it works just fine with the stock boosters. What kind of boosters do you use?

Big liquid boosters, or the big boosters from KW. I think it's imperfect symmetry in struts causing the boosters to skew as their TWR's get very high toward the end of the burn.

[bs1110101]

I like this mod, but it has a few problems, one the 2 meter engine looks too short, i think the nozzle should be longer and/or thinner, and the oms engines shouldn't be quite that bad in an atmosphere, though i say this mostly because i like to make probe shuttles. The 1 meter ones work great, but i think they should get more thrust.

Edit! I did some more testing, give them more thrust, and make them heavier, also 2 words, animated gimbaling.

Edited November 2, 2013 by bs1110101

[Dr.Goodbytes]

I really like this mod! I've spent a good chunk of the day building and optimizing my first shuttle. One thing that I would like to see is a small shuttle type drag chute to help with landings.

[dtobi]

Hi everyone! A new version is out. Almost everything mentioned in this thread is improved and fixed now!!! Check the cangelog!

[dtobi]

I like this mod, but it has a few problems, one the 2 meter engine looks too short, i think the nozzle should be longer and/or thinner, and the oms engines shouldn't be quite that bad in an atmosphere, though i say this mostly because i like to make probe shuttles. The 1 meter ones work great, but i think they should get more thrust.

Edit! I did some more testing, give them more thrust, and make them heavier, also 2 words, animated gimbaling.

I fixed all of your issues. Would you care to try it out again?

[dtobi]

Big liquid boosters, or the big boosters from KW. I think it's imperfect symmetry in struts causing the boosters to skew as their TWR's get very high toward the end of the burn.

I limited the yaw so that it might be easier for you now. Would you care to test it again?

[Firov]

I have to admit, I'm surprised by how well these engines work. While it doesn't make it "easy", per se, this does at least make it feasible to build true space shuttle designs.

I do have a couple of observations though. Mostly just related to balance, which is obviously subjective anyway.

1. The gimbaling doesn't actually seem to activate unless the engine is "staged". If, instead, the engines are activated manually or through an action group, they don't gimbal at all. This is only a problem because it's pretty normal to activate the OMS/deactivate the SSME's with an action group.

2. Honestly, the OMS engines seem massively overpowered. They deliver a huge amount of thrust for their size, while being some of the more efficient engines available. A more modest thrust of 30-40 for the single engine and 60 for the double engine would be a bit more balanced I think.

3. The large SSME is pretty well balanced overall, but it could be argued that it's ability to pivot to 0 degrees renders the OMS engines obsolete, since it can serve as it's own massively powerful OMS engine once in space.

Anyway, like I said. Those are just my observations. Overall though, I love the mod. I look forward to seeing where you take this in the future.

[FellipeC]

Wow, pretty tricky to make this work, but at least work well better than stock engines! Now I'm trying to retrofit my small shuttle to use only one external tank, not this abomination: https://copy.com/hljPzSXt1Wkfkz0P

[nothke]

KerbCom avionics can do separate gimbaling for each engine, also thrust balancing for each engine making asymmetric thrust craft possible (like the Shuttle). SRBs are sometimes buggy though, but running liquid fuel only engines is pretty smooth.

Another thing it can't do is change the actual angle of attack when just having an external tank. The real Shuttle's direction of travel is skewed so that the center of thrust is balanced to the total CoG, that is why the engines are angled "up". (at this altitude wings make no lift so it's unimportant). As the fuel from the tank depletes, the CoG moves toward the orbiter and the engines gimbal even more.

Maybe next step for KerbCom avionics is to implement shifting direction of travel. Then we will be able to make true Space Shuttle =)

Edited November 4, 2013 by nothke

[FellipeC]

Now I mastered working with those engines, they are great! But I found a cosmetic bug. For some reason, sometimes when I cut the engine with action groups, the exhaust animation keeps playing, even with no throttle applied:

[Bartybum]

I'll post some pics up in a tad if you'd like to see

[Sugoi]

This is a pretty rad set of engines.

However, after I ditch those three stages, I get a lot of weird gimbaling; whether or not I need to just strut the thing together more is beyond me, though. It could be because I'm letting MechJeb handle the ascent profile, too. The first two images are that odd gimbaling, the third is after I lock the gimbal and the reaction wheels handle the rest. I don't think that the images capture the full gimbal range; the oscillation is disturbing to see in game, but it the orbiter more or less stays pointed in the direction it needs to go.

Javascript is disabled. View full album

Edited November 5, 2013 by Sugoi

[SmiteZero]

I'd like some fairing sections to match the shape of the radial mounts (perhaps a fuselage section and a LFO section) so it can be extended for visual reasons, and a connector part to fit that flares out to standard circular sizes.

Also, this bug: https://www.youtube.com/watch?v=qe_fjOZEPOI

Edited November 5, 2013 by SmiteZero

[Synthesis]

Well, I made a pretty little shuttle, but I'm still stumped when it comes to overcoming the center of mass versus the thrust distribution (and let's throw lift in there while I'm at it). I'm not sure if the solution is distribution of weight/lift or actually figuring out how I need to aim the shuttle immediately after takeoff.

Still, great mod.

[TG_bigboss]

I made a awesome shuttle using your engines! Thanks they really help ease the process of creating a shuttle. I made sure my center of mass was in the middle of the whole rocket itself and the thrust same. Work great and is balanced! No issues of too much power on one side. I think when you have time really make a nice detail engine but still stock looking. The model is good now no offense but i feel like it could use something or be more close to the real thing. Look at my second image below. Just an idea. Its your mod! Good job! Love it and will keep following for updates.

[FellipeC]

I made a awesome shuttle using your engines! Thanks they really help ease the process of creating a shuttle. I made sure my center of mass was in the middle of the whole rocket itself and the thrust same. Work great and is balanced! No issues of too much power on one side. I think when you have time really make a nice detail engine but still stock looking. The model is good now no offense but i feel like it could use something or be more close to the real thing. Look at my second image below. Just an idea. Its your mod! Good job! Love it and will keep following for updates.

Maybe you like NovaPunch and TyberDyne Shuttle. They have engines that look more like this, but I really prefer those.

[TG_bigboss]

Maybe you like NovaPunch and TyberDyne Shuttle. They have engines that look more like this, but I really prefer those.

I like the model it has now but a little more detail I feel could make these engine models even better!

[pina_coladas]

However, after I ditch those three stages, I get a lot of weird gimbaling; whether or not I need to just strut the thing together more is beyond me, though. It could be because I'm letting MechJeb handle the ascent profile, too. The first two images are that odd gimbaling, the third is after I lock the gimbal and the reaction wheels handle the rest. I don't think that the images capture the full gimbal range; the oscillation is disturbing to see in game, but it the orbiter more or less stays pointed in the direction it needs to go.

Nobody else has mentioned it so I will: that's just mechjeb's PID controller. It's a little wonky sometimes. There's some option in mechjeb to adjust the PID values that can make things much smoother if its misbehaving with a particular ship. Play with those a bit (sorry can't help you with specifics) and see what happens.

The new stock SAS code seems to be pretty smooth now, even with large gimbal ranges. It's too bad mechjeb uses its own currently inferior controller.

[Netris]

make a nice detail engine but still stock looking.

I disagree with you. Most of us will use this mod with B9 (just as you did) do it seems logical it's more in the B9 style. And I find the stock parts pretty ulgly (but that's just my personnal tastes though)

[SFJackBauer]

It's too bad mechjeb uses its own currently inferior controller.

There is an option in MJ to use SAS for control, in the Attitude Adjustement page. Its only good for ascent though, for fine maneuvering once in orbit is better to use MJ own controller.