Why align the SSME with the center of mass of the vehicle?

[EzinX]

In another thread, it was mentioned that the space shuttle main engines are aligned with the center of mass of the orbiter. This center of course shifts quite a bit during flight, as the SRBs rapidly shed mass and are jettisoned and the orange fuel tank also rapidly drops in mass.

Ok, so I know why : if the engines are each aligned with the center of mass, then each will create no net torque on the vehicle.

But, do you have to do it this way? While there isn't any net torque, you've got 3 competing velocity vectors and some of the thrust will be canceled out by the other vectors. If, instead, you align the engines such that their thrust is in line with the direction of travel, and just balance the thrust generated with the other 2 engines (so the torques cancel), won't you get more dV for your fuel?

This thread probably needs some drawings. Will add them later if anyone is having trouble visualizing what I'm talking about.

[Bill Phil]

As far as I know that's what they did...

[DaveofDefeat]

I believe there was a thread discussing this a long time ago, and the general consensus was that the delta v losses weren't all that high.

[Brotoro]

If, instead, you align the engines such that their thrust is in line with the direction of travel, and just balance the thrust generated with the other 2 engines (so the torques cancel), won't you get more dV for your fuel?

The engines are on the Shuttle (so they can be recovered and re-used). The center of mass is below the shuttle's belly. If you aim the thrust vectors of the three engines along the long axis of the rocket, you CAN'T use the engines to cancel out the torques (they will all produce pitch-down torques).

Edited March 7, 2015 by Brotoro

[EzinX]

The engines are on the Shuttle (so they can be recovered and re-used). The center of mass is below the shuttle's belly. If you aim the thrust vectors of the three engines along the long axis of the rocket, you CAN'T use the engines to cancel out the torques (they will all produce pitch-down torques).

Ah. Ok, that makes much more sense. So you can only "balance" the thrusts if the center of mass is located at the midpoint between the three engines. Since it isn't, you can't. All our rockets in ksp, we stick the center of mass directly above generally, between all the engines.

[cpast]

Also, having every engine aligned allows you to handle an engine failing. If the engines didn't track the CoM, an engine failing would result in massive torque. If you relied on differential thrust and *couldn't* gimbal an engine to track the CoM, you'd be unable to control this torque, leading to loss of vehicle and crew.

[K^2]

I don't know about that... Loss of critical engine is a pretty typical scenario in multiengine aircraft, and aerodynamic forces are sufficient to maintain controlled flight. I don't see why this wouldn't apply to a hypothetical space plane with balanced thrust. Deorbiting burn would be a pain to perform with unbalanced engines, but there's no reason you wouldn't be able to perform an emergency landing if an engine failed during takeoff or reentry.

[softweir]

Also, having every engine aligned allows you to handle an engine failing. If the engines didn't track the CoM, an engine failing would result in massive torque. If you relied on differential thrust and *couldn't* gimbal an engine to track the CoM, you'd be unable to control this torque, leading to loss of vehicle and crew.

The Saturn 5 first stage was designed to use the same strategy for the same reason - especially important at the time as the gimballing systems were slower and the control technology less sophisticated.

[cantab]

I don't know about that... Loss of critical engine is a pretty typical scenario in multiengine aircraft, and aerodynamic forces are sufficient to maintain controlled flight. I don't see why this wouldn't apply to a hypothetical space plane with balanced thrust. Deorbiting burn would be a pain to perform with unbalanced engines, but there's no reason you wouldn't be able to perform an emergency landing if an engine failed during takeoff or reentry.

Unlike most aeroplanes rockets are lifted by their engines, so the impact of an engine failure is much greater. Also consider that the Space Shuttle had rubbish aerodynamics when empty; with the ET and Boosters attached I expect the control surfaces on the orbiter would be basically useless. Then there's the Shuttle spending much of its ascent so high that aerodynamic control would be completely irrelevant - I believe most of the manoeuvres required for the RTLS abort scenario would have taken place at such heights.

[Armchair Rocket Scientist]

I don't know about that... Loss of critical engine is a pretty typical scenario in multiengine aircraft, and aerodynamic forces are sufficient to maintain controlled flight. I don't see why this wouldn't apply to a hypothetical space plane with balanced thrust. Deorbiting burn would be a pain to perform with unbalanced engines, but there's no reason you wouldn't be able to perform an emergency landing if an engine failed during takeoff or reentry.

Basically, a jet airliner moving at mach 0.8 and 30,000 feet, with wings optimized for flight in that regime and a TWR of 0.3, is VERY different from the space shuttle, which has a TWR over 1, is designed for hypersonic reentry, and may be flying at, say, mach 2 and 60,000 feet, or mach 10 and 200,000 feet, when an engine failure occurs.

An airplane can pretty much always rely on being able to generate a pretty large yaw force with its tail to compensate for an engine failure, as well as throttling down. The Shuttle can't do that without going to a high angle of attack and risking breaking up, and in some cases throttling down would make the problem WORSE because the remaining engines would be unable to overcome the upward pitching moment produced by the SRBs. Its ability to make drastic trajectory changes is extremely limited with the external tank attached.

Plus, in the event of an engine failure close to takeoff, the shuttle would have to either separate from the still-burning SRBs (very dangerous) or ride them all the way to burnout (in which case the vehicle must remain controllable). Longer after takeoff, it would be on a suborbital trajectory over the ocean, with no air to help with control. At best, if a failure resulted in inability to keep the thrust balanced with gimballing, it could ditch the ET and get out of the spin with RCS, but if it can't keep burning the main engines the OMS would probably be too weak to allow a Transatlantic abort. The shuttle's high stall speed means ditching in the ocean would almost certainly be unsurvivable, so crew would have to bail out. Again, this is very dangerous.

http://en.wikipedia.org/wiki/Space_Shuttle_abort_modes

Here's a chart of the shuttle's abort modes:

http://en.wikipedia.org/wiki/File:ShuttleAbortPost51L.png

Basically, anything after SRB sep is a flight regime where there isn't enough air for a vehicle to maintain control via aerodynamics alone, so without gimballing you have to shut the engines off entirely and immediately (equivalent to a 3-SSME failure) and use RCS to stop any rotation that was imparted. For the space shuttle, unless this happened very near the end of the burn you've got a vehicle loss and frequently a crew loss. This would be true for most spaceplanes of a "shuttle-like" design.

AFAIK the only serious spaceplane design that doesn't have the ability to point its thrust vectors at the COM is Skylon. However, Skylon is again different from the space shuttle: an engine failure low in the atmosphere would be similar to one on a normal airplane, while at higher speed and altitude it could potentially shut its engines off, reenter, and wait until it reached a low enough speed and altitude for safe powered flight before returning to the runway on its single engine (most likely dumping all oxidizer and some of the fuel to reduce mass enough that reentry is safe).

[K^2]

Also consider that the Space Shuttle had rubbish aerodynamics when empty; with the ET and Boosters attached I expect the control surfaces on the orbiter would be basically useless.

We aren't talking about Space Shuttle. It can't have balanced thrust by definition with tank attached, and without that tank, it can't use its main engines. It's a moot question.

What we are talking about here is critical engine failure on something like Skylon. Something that's inherently balanced, still has TWR for vertical ascent, has aerodynamic capability for runway landing, and would become out of balance with an engine failure.

And all you are going to do is the same thing you would on an airplane. Cut thrust, maintain airspeed, and continue flight until you can do an emergency landing. The fact that your engines are capable of 1+ TWR doesn't mean you have to use it.

[Armchair Rocket Scientist]

We aren't talking about Space Shuttle. It can't have balanced thrust by definition with tank attached, and without that tank, it can't use its main engines. It's a moot question.

What we are talking about here is critical engine failure on something like Skylon. Something that's inherently balanced, still has TWR for vertical ascent, has aerodynamic capability for runway landing, and would become out of balance with an engine failure.

And all you are going to do is the same thing you would on an airplane. Cut thrust, maintain airspeed, and continue flight until you can do an emergency landing. The fact that your engines are capable of 1+ TWR doesn't mean you have to use it.

Skylon could be interesting, since its engines are pretty far from the center of mass. At low speeds it ought to be fine doing what you said, but at mach 3+ it's possible that an engine failure would push the vehicle into a flat spin before the other engine could be shut down (as has been experienced by pretty much anyone who's flown a multi-engine SSTO in KSP before the update with the RAPIER), which in real life would result in the vehicle breaking up. Here's an example of this happening with an SR-71: http://www.916-starfighter.de/SR-71_Waever.htm

Once Skylon has performed its pitch-up maneuver following the transition to rocket mode, it will have almost no aerodynamic control. If an engine fails, it would have to immediately shut down the other engine completely (RCS would be much too weak to compensate for imbalanced thrust). It would still likely go into a spin, with associated problems from high centripetal accelerations, but wouldn't immediately break up and could slowly stop its rotation using RCS (heck, maybe use the LOX as a cold gas thruster, since it won't be needing it). However, it's possible that for a large part of the burn to orbit, an engine failure would result in a reentry too steep for the vehicle to survive.

However, according to this: http://forum.nasaspaceflight.com/index.php?topic=33648.450;wap2

Skylon has the equivalent of four engines in rocket mode (I guess each SABRE has two turbopumps), so it's possible it could handle a failure in that regime with just a reduction of thrust.

[K^2]

Flat spin is definitely a possibility. The only way to really deal with that is automatic cutoff and RCS response. If critical engine failure is detected automatically and opposing engine is cut off immediately, which is possible to do with a rocket engine, any remaining angular momentum can be absorbed by RCS even in total vacuum.

Re-entry on failure danger is always there, regardless of engine configuration. There are a whole bunch of situations that can cause loss of propulsion during any stage of the ascent. And it is possible to design ascent profile to give a safe re-entry on failure. It would increase delta-V requirement, reducing payload, but if we are going to see regular commercial operation, which is the only scenario where Skylon-like craft make sense, we will need to see such measures taken. 2% fatal failure rate may be acceptable when you only expect a bit over a hundred launches in three decades, but if you plan to have hundreds of launches per year, it will be a PR nightmare. Total engine failures will happen to space-planes, and they will need to be constructed to handle such an emergency.