Surefoot

I dont think that would be this mod's responsibility but rather something like Deadly Re Entry. FAR only models aerodynamics, not tearing or other mechanical constraints. While i agree it should be able to go higher (but that's more from a limitation of what you can do with stock parts), the real one is also very unstable and needs the SAS to be activated (yes there is one in the SR-71, exactly named like that). Very few pilots dared to turn it off... Going just a tad too far on AoA means unrecoverable stall. Not sure what you mean by that, FAR is as realistic as it comes before going for something like X-Plane.. The stalling behaviour is quite realistic actually (so are the "expanding spiral" and other aerodynamic features like "dutch roll"). I dont think many sims reach that level of realism.

That is truly great ! Really i was waiting for that. Thanks a lot !

Is the result like variable sweep wings ? As in reduced drag at high speeds ?

They do make sense from a structural point of view, but in-game this is quite limited indeed.

Real planes that can do this have got a very stable aerodynamic configuration, and the first thing you'll notice is the huge tail fin (especially airliners). I managed to do it with my hypersonic jet so i guess it's easier to do with a subsonic one.

You mean the gyroscopic effect is accounted for in the game physics engine ? Or is it a by-product of other physics rules ?

Well yeah that's the point, if you have to run them at reduced thrust you're better off with smaller engines that you can run at full thrust.

Here, a ship that exhibits the behaviour. Mods needed are MFT, stretchytanks, mechjeb2, KJR. You'll have to use hyperedit to put it in orbit somewhere, then turn on SmartASS to point in a fixed direction (prograde / retrograde), max the engine thrust then activate the stage, see it spinning Lots of roll, some yaw too (if you cut off you'll notice MJ compensating for the yaw). Get the craft file there: http://ge.tt/7LsyqgC1/v/0?c it's called BalancedTest.

I have the same issue, using MFT too.

Yeah i edited the cfg file to match the stock NERVA configuration and indeed its max temp is much lower than the stock one for some reason (and heat production is higher). That would explain the quick blow-up with DRE.

I was suspecting a kind of induced shift due to propagating stresses, for example, but the design is totally symmetric, i fail to see any reason why the stress would be unbalanced to start with. Ohh yes i didnt try that yet (yes they have fuel lines). Otherwise nothing in common apart from being a symmetrical radial tank design. I even changed the radial attachment, trying different solutions.. Good idea, i'll have a shot at that one (and report on MFT thread, sorry for the hijack). Good point, i'll see without MFT (i have to remove stretchytanks and toroidal tanks too then do i)

That's not so much, i've done a 1700t ship too These are cargoes, the payload is extra heavy. So the final deltaV with the payload attached is not so huge (in the high 4000's) so i can go to Jool and come back without refuelling (since they are fuel cargoes..). One of these engine blocks is used for my Grand Tour too, in this case i have just over 6500m/s in dV to be able to go everywhere with a safety margin since i only refuel with Kethane (so it's not always a net profit).

Wow, that must be what happened to DresCroffgrin just above.

That could have been explained in 0.12.4 due to the twitch glitch, but since it's gone from 0.12.5 this is very weird. I just tested hypersonic flight (similar to re-entry) with 0.12.5 and had no weird g-force, no "bump" ot twitch, everything was smooth. I suppose you keep a small AoA and low profile for re-entry ? Maybe one part is creating too much drag ?

@Ferram: got that issue with recent patches (0.23 and so on): http://forum.kerbalspaceprogram.com/threads/64117-0-23-Modular-Fuel-Tanks-v4-1?p=888145&viewfull=1#post888145 and http://forum.kerbalspaceprogram.com/threads/64117-0-23-Modular-Fuel-Tanks-v4-1?p=889021&viewfull=1#post889021 Maybe it's MFT, maybe it's KJR, it's not clear to me. Seemed from reading the stretchytanks threads there was an issue with MFT there, but now i'm less sure. All my symmetric engine configurations have off-balance thrust, with random variations upon game reload / ship change. I am using the latest version of everything.

I am using these engines with Mechjeb and have no issue there...

Happens to the best of us XD And yeah a straight engine (even a cluster) will not trigger the issue. Yeah i just did a lot of tests and it's pretty noticeable with Mechjeb's node execution. Do a significant burn with a symmetric tank+engine configuration with MJ, you'll notice that the burn is slightly off (in the best of cases), and when you're about 2m/s left or so, the node target starts escaping and the ship just flips like a dog chasing its tail. Or just turn on SAS and point at prograde (or with SmartASS) you'll notice with a full burn the ship slightly veers on one side. Then it gets worse upon reload, but it seems pretty random... I'm now back with full stretchytanks, i managed to mitigate the issue with an over-engineered setup, but it's not satisfying... I am trying to do a full Eve lander setup and there any off-axis thrust becomes quickly lethal. Or maybe it's KJR acting up, we'd have to PM Ferram to see if he can see the issue.

You are either missing struts, or KJR (which i suggest you try first).

I see. Each engine is attached individually to a 2.5m stretchy tank (which is quite big in my test here). 8 of them, in radial symmetry, around a 5m diameter tank, total ship is around 600t. I'll see if replacing the stretchy tanks with stock-like (Novapunch or KW) changes anything. (edit) i mounted them on KW 2.5m tanks, no change, still overheating very quickly at full thrust, even on a much lighter ship.

I have the issue with stock, KW tanks, and B9 tanks too. Again, these ships fly perfectly at first. One of them comes from v0.22 and has been doing a Grand Tour... i can tell it flies straight, but not anymore. Uses 4x symmetry with B9 tanks. Swapped the engines, same issue. That's why i'm reducing the potential soures to MFT, as it's one of the few changes (with tweakables) to v0.23 that might have affected my ships. (edit) doing more tests with stock parts.. Glitch is not always affecting the ship immediately. Seems i have to dock 2 ships together to exhibit the problem fully... Also some side note, i dont have the 45%-55% LF+Ox button anymore on any tank.

You mean version 0.12.5 Testing the roll twitch now. (edit) yay no more twitch Supersonic and hypersonic are now smooth as silk. Went up to mach 7.3 with the turboramjets, everything was fine.

My tanks are all full. The test ship has 8x stretchy 2.5m tanks radially attached to a central 5m stretchy tank. All are using MFS with stock fuels, everything is full (and just in case i use TAC fuel balancer, but that is not in use for my little test here). There is zero wobble as i use both superstrong struts (from Novapunch) and KJR. To them are attached 8 nuke engines, i tried with NERVA's and FTmN nukes. Also my lander which has 4x 2.5m pods is now spiralling to death, whereas i have been to many places with it previously.. I see 2 rotations induced by firing up the engines: one in roll (immediate), and one in yaw (more progressive but still very large in the end). This happens on ships that used to fly straight, and on new designs just the same. If you want i can create a ship file with stock parts (or any mod you have) that exhibits clearly this issue ? To make a quick test you can create a similar design, hyperedit it in orbit, do a first test (usually flies straight), then go back to space port and back to the ship, then it doesnt fly straight anymore... Also, unrelated, i noticed the LF/Ox ratio is not the same as stock tanks ? When i create a stretchy tank of the same size as a stock one i end up with less LF and more Ox: is that related to MFS or stretchies ?

The big engine (the 280) overheats much more quickly than the NERVA from Novapunch, while in space. These engines have very close specs (the NERVA is a tad more power hungry), on my test ship i never had a NERVA overheat on me in vacuum, the 280 overheated pretty quickly. What is the reasoning ?

Tested extensively: seems the kraken is gone, but the ship is still unbalanced after a reload (goes off one side for no reason, or even rotates around Z axis ..).

Depends on the engine, with the turboramjet / afterburner engines (found on this mod forum too) you can go up to mach 7.3 with the turboramjet at high altitude, and over mach 1 near ground level with the afterburner.