Accordion B: A experimental proof of concept Mun rocket [0.16 + MechJeb]

[r_rolo1]

After the ship I posted yesterday ( see here ), I decided to ditch the jet engines and try to weave a ship that would combine some concepts I've been working on:

- Ships being pulled by the engines instead of being pushed ( with all the issues with the decoupler strength this is actually a good idea, but the main inspiration was a certain RL Mars mission )

- Not ditching engines out ( really, what a waste of tech heavy parts )

- Laterally mounted engines ( in RL this would be problematic combined with the first point, since the exhaust would most likely create heat issues in the lower stages ... but so far we don't have that issue in KSP, and when we do, this can be partly remedied by tilting the engines out a little to the outside )

- Fuel line work that allows you to burn the lower fuel tanks first ( as we don't have stock non-crossfuel capable decouplers ( RLY, Harvester ???? ), this needs some ingenious work )

In shorter words, I wanted a space crane Munar landing module with a lot of tanks attached below and linked by fuel lines in a way that the lower tanks would always dry first.

And after a dozen of prototypes, I got to this awkward looking rocket :

More awkward than the name is the behaviour it has both when the physics kick in and when you fire the rockets. It is not called Accordion by accident

Anyway, some pics on the mission:

Right after ditching the first lower tank. As you can see by the navball, the launch is somewhat problematic: the ship is pretty close of the maximum mass 3 aerospikes can get off the ground of KSP , and to add, the ship is not exactly structurally solid ( again , experimental and proof of concept ). A good human hand would most likely be able to control this, but , as a not stellar pilot ( well I can land on winglets in the mun on stock parts, but this is quite harder ), I prefer to leave that to mechjeb.

Starting the gravity turn. I used the regular MechJeb flight plan, but I'm pretty sure this can be optimized.

Circularizing. I had just ditched the last fuel tank below the lander proper. It had got to 100 km with a sliver of fuel. This was a major improvement on Accordion A ( pretty much the same rocket , but with 3 heavy landing legs and with lower mounted engines ( will explain that change later ) ) that was forced to start burning the final stage still in the atmosphere.

Munar Injection Burn. Pretty good fuel stock. Those landing legs were really dragging out

Orbit after entering Munar SoI. Picked this orbit for two reasons: first, it will get to the ground in a quite vertical direction, allowing a good suicide burn ( Gs be damned ) and it will drop us in a good place for a direct burn to Kerbin without circularization around the Mun.

Suicide burn. I chickened out a little here, since I could had waited 1 to 2 more seconds before burning. I don't leave this exclusively to MechJeb because it has some kind of respect for the Kerbals and tries to avoid high Gs. This unfortunately will make the fuel bug raise it's ugly head in this ship ...

Landed ship ( coordinates in mechjeb landing window ) and postcard. Only when taking this pic I noticed that I had left a now completely useless extendible stair in the lander *facepalm*

Return orbit after final burn. Tried to do a as much as possible full throttle burn, but this needs some fine tuning since I have made a direct burn from the surface, so I made a part of it in in slow burn. The fuel budget ( as you will see in the next pic ) was more than enough to cope in a bug-less environment

Activating parachutes for aerobraking in the highest possible altitude. As you can see, the remaining fuel was more than enough to cope with the fuel bug issues in the return burn

As I mentioned before, the previous version of this ship had lower mounted engines in the Munar landing module. This made that the ship was too much top heavy for MechJeb SAS could avoid it to turn upside down in reentry, a solvable but annoying situation. I could had solved this by putting a extra parachute, but that would add unwanted and unneeded mass, so, putting the parachute anchoring higher was the chosen solution ( I also not want the parachutes anchoring in the capsule in here, since it makes that the capsule detaches itself from the rest of the lander sometimes when the parachutes open. Not a big issue for this, but it would be problematic I needed to fire up again ( as in a Kerbin return from another planet ) and it is better to not learn things wrong ). It did not solved the issue completely, but it mitigated the issue enough for MechJeb SAS to be able to cope with it.

Firing for soft landing. I'm pretty sure the ship could most likely handle a parachute only touchdown ( atleast the capsule would for sure ), but as we don't have landing legs and have fuel enough for it, better to do it right.

Landed ship and end flight stats. I'm not sure if the 7g are from the suicide burn in the Mun or due to the 6 km Pe in the return burn final orbit ( not good... too many g forces and would roast the ship if the reentry heat was on , even if the ship had a heat shield, because it meant a 15º reentry angle, far higher than the RL 6º . Both were consequences of trying to avoid the dreaded fuel bug ...

Again, let me point out that this is a proof of concept ship. It was not made to be easy to pilot ( I'm not sure if I could do it without mechjeb ), and the fuel budget is quite tight, in spite of appearing otherwise. It was made to share some ideas I've been developing since .16 came out and hopefully someone will gain from them ( as I did when borrowed some many other designs from here )

Anyway, craft file here . Have fun and try to not crash this too much

[djnekkid]

That is certainly an interesting design! And i've actually been thinking of doing something similar, but not quite gotten around to it, but my idea were more that i had a pod+tank in the middle, and then some hanging tanks in a 3 or 4 way around, with dropping tanks as they got more and more empty. This one is just as funny tho

[simplemunrockets]

This is the most kerbal rocket EVER.

[r_rolo1]

@djnekkid

The idea you describe is the next logical step. You would probably need to strut stuff ( a thing that I studiously avoided here ) and would need some work to not collapse under it's own weight in KSP ( this one almost does that ), but I'm pretty sure that is doable. If you do that, be sure to post it

@ simplemunrockets

Thanks for the compliment , but I'm pretty sure that there is stuff around that is a lot more kerbal than this

[Maxed-Rockets]

Can't you just go into the part.cfg file and disable fuel crossfeed?

Genius design though!

[r_rolo1]

What I was meaning is that IMHO the decouplers should be by default non-crossfeed able , because the last thing you want to do is to put your precious fuel lines and pumps ( yes you need pumps to get the fuel between the deposits to get the fuel from one side to another ... you can't rely on gravity in a rocket to do that job ) inside a device that is meant to explode I can even see the decoupler we have now existing, but as a pricier and less powerful ( since you can't put so much explosives inside because of the pumps and the fuel lines inside ) version at the side of a non-crossfeed version .But that is just my opinion

Anyway, my philo is always to work with the stock parts until they can't simply cope out with what we want to do. In here, playing with the fuel lines does the job, so why bother with cfg edits ?

Edited August 16, 2012 by r_rolo1

[Kosmo-not]

I radially mounted 2 FL-T400's to the third tank from the bottom and attached LV-T45's to those. This gives a lot of maneuverability and a much better TWR during the first part of the ascent.

I ended up in a 100km orbit with 188.5L remaining in the small radial tanks and a full center tank.

I could put some small landing legs and it and go to the Mun and back.

[r_rolo1]

Thanks for experimenting this one Kosmo-not. I surely respect your minimal ships and your fine pilot skills being applied to my humble ship

I have full conscience that this ship as posted is horribly inefficient at launch and your change makes sense in terms of improving the efficiency of the design. But this was pretty much a concept ship just to show up that particular solution I had weaved out ( and that I've already deployed in other designs ) so I'm ok with inefficiency. And, let's be honest, the crazy simplicity ( or the simple craziness ) of the design is half of the charm

[segaprophet]

Some interesting concepts here. I've been noodling around with attempts to minimize ditching heavy parts unnecessarily, but this has mostly gone into more radial drop tanks.

This is rather like Goddard's pendulum rocket.

[r_rolo1]

True, and it is exactly as unstable Well, this particular ship is not that unstable as that, but my experience with this kind of design so far is that in general ships pulled up by engines are very susceptible to lateral tilting and that it can be hard to balance things. I tend to resort to some kind of hybrid solution ( a engine in the bottom of rocket axis and 3/6/12 smaller engines in the top pulling ) that normally ends working well enough.