Inconsistent asymmetric thrust loss with SSTOs

[Shyrka]

As I've learned through many hard lessons, as you climb the thrust available from turbo jets reduces as the intake air decreases. With planes with multiple engines, I've come to accept that this will always be an asymmetric loss: one turbo jet will be starved of oxygen before the other, resulting in a yaw effect and a flat spin if left alone. At this point, you have to throttle back to reduce the air requirements of both engines until the engines' thrust is symmetrical again.

Very occasionally, one of my SSTO designs won't suffer from asymmetric thrust as it climbs for no obvious reason that I can discern. The plane can fly as high as it pleases at full throttle with perfectly symmetrical - albeit gradually reducing - thrust from both engines. Only when one finally flames out does anything bad happen. I've yet to discover whether this something to do with the plane's design or simply a bug. If it's the latter, I'm loathe to exploit it as it makes the designs much more effective but given that I've not seen it called out specifically in any of the SSTO design tutorials, I'm assuming its a bug. The designs I've had this with are not-particularly-air-hogging, i.e. an intake-to-engine ratio of 4:1 or lower. The latest was a twin turbojet with two ram intakes and a radial each.

Has anyone else experienced this effect? Any ideas what might be causing it?

Edited August 13, 2014 by Shyrka

[Wanderfound]

See http://forum.kerbalspaceprogram.com/threads/64362-Fuel-Flow-Rules-%280-23-5%29 for the detailed explanation. You can build an asymmetry-proof multi-engine plane, but it's a hassle.

Much easier is to just use an odd number of engines. When you get up to flameout altitude, shut down everything except for the central turbojet. This gives you two benefits: firstly, you don't have to worry about asymmetric flameouts, and secondly it concentrates all of your intakes into one engine.

It's better to have one engine working properly than to have many engines spluttering. You don't need a lot of thrust to accelerate at 30,000m; multiple engines are counterproductive at that altitude.

[Shyrka]

See http://forum.kerbalspaceprogram.com/threads/64362-Fuel-Flow-Rules-%280-23-5%29 for the detailed explanation. You can build an asymmetry-proof multi-engine plane, but it's a hassle.

Interesting. Reading that, it would appear to be something I've stumbled upon due to the build order. Time to experiment, I guess!

As far as building with an odd number of engines, I agree. I tend to aim for this with the larger craft but most of my recent SSTO designs feature one central nuke and two turbojets.

[Laie]

Very occasionally, one of my SSTO designs won't suffer from asymmetric thrust as it climbs for no obvious reason that I can discern.

See the link above from wanderfound post for the full details, but here's a short primer: it matters in which ORDER you place intakes and engines while building the craft. Placing four intakes, then four engines, will give you asymmetric flameout. Placing one intake, one engine, one intake, one engine, ... leads to the much better performance you described.

You can still use symmetry if you first build an assembly of (intakes + engine + glue parts to hold them together), then place the whole assembly on your plane using symmetry.

If you're the kind of person who wants to do this, I suggest you use a small plane of two or four engines and experiment a little. Once you know what you're looking for, figuring out the details isn't difficult.

[Red Iron Crown]

Surprisingly, the order in which you place the engines and intakes makes all the difference, which means two ships with the exact same parts in the same locations can perform very differently. The thread Wanderfound mentioned describes it in detail, and there's a less technical explanation in this thread. The most important thing to remember is: Do not place engines alone using symmetry.

[cantab]

There's always the option of stacking your engines vertically, a la the Lightning fighter. Though that would probably look silly for any more than three or four.

[Red Iron Crown]

There's always the option of stacking your engines vertically, a la the Lightning fighter. Though that would probably look silly for any more than three or four.

Wouldn't that just make it spin in the pitch axis during asymmetric flameout instead of the yaw axis?

[cantab]

Some quick tests with a modified Aeris 3A suggest it's less of a problem. With vertically-stacked engines immediately cutting one turbojet caused a sudden pitch change but the plane remained controllable and strong pitch trim returned it to stable level flight. With side-by-side engines things were worse though it didn't completely spin out. I admit these are fairly low-altitude tests and things may be different higher up, but I think most planes have stronger pitch control authority which will help them handle vertically off-centre thrust.

[Red Iron Crown]

Some quick tests with a modified Aeris 3A suggest it's less of a problem. With vertically-stacked engines immediately cutting one turbojet caused a sudden pitch change but the plane remained controllable and strong pitch trim returned it to stable level flight. With side-by-side engines things were worse though it didn't completely spin out. I admit these are fairly low-altitude tests and things may be different higher up, but I think most planes have stronger pitch control authority which will help them handle vertically off-centre thrust.

That's an interesting result, I guess the greater horizontal wing area reduces its susceptibility to asymmetric thrust in the pitch axis. Useful tip, thanks for sharing your results.

[Commander Jebidiah]

if you have three engines place two of them on symmetry then place the central one alone.

The reason for this is the way ksp simulates flameouts is by flaming out the last engine placed, first

[Red Iron Crown]

if you have three engines place two of them on symmetry then place the central one alone.

The reason for this is the way ksp simulates flameouts is by flaming out the last engine placed, first

No it isn't, though that's a common result. Read the threads linked earlier, this one for a simple explanation, this one for a more technical one. Order of placement of both intakes and engines determine which engine flames out first, and it needn't necessarily be the last engine placed.

[Commander Jebidiah]

No it isn't, though that's a common result. Read the threads linked earlier, this one for a simple explanation, this one for a more technical one. Order of placement of both intakes and engines determine which engine flames out first, and it needn't necessarily be the last engine placed.

This is what cruzanak said in his quick guides series

[Red Iron Crown]

This is what cruzanak said in his quick guides series

It's a common result, as it happens when you place all the intakes and then all the engines, or vice versa. But it is possible to construct a vessel in which the last engine placed is not the first to flame out. The threads I linked detailed how it actually works, the order of placement is everything.

[Shyrka]

That's an interesting result, I guess the greater horizontal wing area reduces its susceptibility to asymmetric thrust in the pitch axis. Useful tip, thanks for sharing your results.

Indeed, I've experimented with stacked engines myself and found them generally more stable but they tend to dictate the design of your plane a little more than I'd like.